INTRODUCTION Research has shown that fear crime

Research has shown that fear of crime ranks high among Americans’ concerns about

major national issues. More specifically, in a recent Gallup poll, respondents ranked feeling safe

from crime ahead of job satisfaction, financial security, and health (Saad, 2011). Surprisingly,

though, with violent and property crime rates at an all-time low in recent history (UCR, 2010),

Americans fear crime much more than other harms likely to affect them. For example, U.S.

citizens are thirty times more likely to die of a heart attack than they are from criminal homicide

(Kochanek et al., 2011), yet coronary infarction is not among the things of which most

Americans are afraid. Such discrepancy between perceived and actual risks suggests an intrinsic

lack of public knowledge regarding the criminal phenomenon. Moreover, it is possible that the

definition of crime used in those opinion surveys helps further obfuscate what the public really

knows about criminal activity.

The vast majority of polls choose to focus on “street” crime (i.e., property and violent

offenses such as burglary, larceny, assault, rape or homicide), which is predominantly committed

by members of the lower class, and not “white-collar” or “suite” crime (i.e., illegal or unethical

acts such as toxic dumping, labor exploitation, large-scale fraud, illegal warfare, etc.), which is

perpetrated by corporations, politicians, and other elite groups. This almost exclusive focus on

street crime is unfortunate, for research has established that white-collar crime greatly exceeds

the impact of street crime on society, both in terms of financial cost and harmfulness (Knowlton

et al., 2011; Landrigan et al., 2002; Leigh, 2011; Lynch & Michalowski, 2006; Herbert &

Landrigan, 2000; Rebovich & Jiandani, 2000; Reiman & Leighton, 2010). More specifically,

while street offenders cost the public about $18 billion each year (UCR, 2010), annual losses due

to financial crime (e.g., fraud) and health costs caused by work-related injuries and illnesses as

well as environmental pollution add up to over a trillion dollars (Knowlton et al., 2011;

Landrigan et al., 2002; Leigh, 2011; Lynch & Michalowski, 2006).

Similarly, whereas murder and negligent manslaughter claim the lives of about 14,000

people annually (UCR, 2010), the number of victims of work-place related deaths (e.g., workers

who sustain accidental injury due to the company’s negligence, illnesses caused by prolonged

exposure to toxic chemicals, etc.), toxic waste dumping and deadly pollutants, faulty consumer

products, and nefarious and addictive substances (e.g., tobacco) exceeds 100,000 a year (Herbert

& Landrigan, 2000; Leigh, 2011; Lynch & Michalowski, 2006). Moreover, another 225,000 have

been estimated to be victims of medical malpractice annually (Starfield, 2000).

However, despite these staggering differences, white-collar crime is still less prosecuted

than street crime. In fact, corporate regulations are weak, the culpability of corporations harder to

prove for prosecutors, and their sentences for corporate offenders have up until recently typically

been more lenient compared to those imposed upon street criminals (Calavita, Tillman, &

Pontell, 1997; Maddan et al., 2011; Payne, Dabney, & Ekhomu, 2011; Tillman & Pontell, 1992).

Nevertheless, because of the way the aforementioned opinion surveys introduce crime to their

respondents, whether the public is informed about such discrepancies remains unclear.

Most of the empirical literature on public response to crime has focused on street

offenders. Moreover, the majority of this body of research taps attitudes about it (i.e., perceived

seriousness) and not actual knowledge. The few studies (e.g., Bohm, 1987; Doob & Roberts,

1983; Kappeler, Blumberg, & Potter, 1996; Macleans, 1995; Maguire & Pastore, 1995; Roberts

& Stalans, 1997; Wilbanks, 1987) that explored public awareness of crime have identified several

“myths” such as increasing crime rates (Macleans, 1995; Maguire & Pastore, 1995), violent

crime rates increasing faster than property crime (Knowles, 1984), overly high recidivism rates

(Doob & Roberts, 1983), and the deterrent effect of the death penalty (Bohm, 1987, 2003). While

these “myths” have been debunked by existing research, they remain solidly anchored in the

public psyche and belie a distorted view of the reality of crime marked by an overestimated

perceived risk of victimization (Roberts & Stalans, 1997).

Because the majority of public knowledge about crime and justice is derived from the

media (Dowler, 2003; Roberts & Doob, 1990; Surette, 1998), it could be that news outlets do not

perform their duty of disseminating scientific knowledge to the public. Moreover, street crime is

given disproportionate coverage, both in newspaper headlines and in local television news,

compared with white-collar crime (Barak, 1994; Barlow & Barlow, 2010; Ericson et al., 1991;

Lynch & Michalowski, 2006; Lynch, Nalla & Miller, 1989; Lynch, Stretesky & Hammond,

2000). Consequently, given the public’s limited exposure to relevant information about

whitecollar crime in the media, Americans might indeed hold “myths” about crimes of the

powerful as they do regarding street crime. To date, no study has examined the extent of public

knowledge regarding white-collar crime. Therefore, this dissertation proposes to assess the

degree to which the public is informed or misinformed about white-collar crime.

The present study is articulated around six chapters. Chapter one was meant to briefly

introduce the topic and provide a rationale for measuring public level of knowledge about

whitecollar crime. Chapter two provides a review of the literature on white-collar crime from its

inception as a sociological construct to the scientific debate regarding its exact definition, and

acquaints the reader with the concept of elite deviance, its typology, the various issues scholars

are confronted with when conducting research on it, as well as empirical evidence on its impact

on society (i.e., financial cost, physical harm), and its perpetrators’ relative legal immunity

compared with their street counterparts. Also reviewed are popular “myths” about traditional

crime to buttress the argument that Americans may share similar misconceptions and erroneous

beliefs about elite deviance, as well as scholarly findings on public attitudes about white-collar

crime. Lastly, a rationale for the present study is reiterated and research questions presented.

The methodology used in the present study is described in detail in chapter three,

including a discussion about the sample, data collection process, measures (popular knowledge

about white-collar crime, correlates of such knowledge, and public sentiments about elite

deviance), and data analytic plan. Chapters four and five present the results of statistical analyses

and provide answers to each of the research questions previously introduced. Finally, chapter six

provides a discussion of these findings, including a summary and interpretations of the analytical

findings, a review of the limitations of the present study, theoretical and practical implications as

well as avenues for future research.

CHAPTER TWO:

LITERATURE REVIEW

The following review of the literature is organized into nine sections. As previously

mentioned, the first section covers the original definition of the white-collar crime concept and

the confusion that ensued within the academe regarding its actual significance and theoretical

scope. Elite deviance is subsequently introduced in the second section as an alternative concept

to remedy such confusion. The third section offers a typology of elite deviance, including

economic domination, government control, and denial of basic human rights. Various

methodological issues inherent in white-collar crime research are exposed in the fourth section.

Sections five and six cover the empirically established greater financial impact and physical harm

of elite deviance compared with traditional crime. Despite these findings, the relative legal

immunity enjoyed by many white-collar offenders due to their status is also highlighted in

section seven. The eighth section introduces popular “myths” about street crime and hints that the

American public might hold similar misconceptions concerning crimes of the powerful.

Section nine summarizes the empirical findings on public attitudes regarding white-collar crime.

Lastly, the tenth section provides a rationale for expanding such research to the study of

knowledge about elite deviance and introduces the research questions that this dissertation will

strive to answer.

Defining White-Collar Crime

First coined in 1939 by the sociologist Edwin Sutherland during his address to the

American Sociological Society annual conference, the term “white-collar crime” was

subsequently defined as any “crime committed by a person of respectability and high social

status in the course of his occupation” (Sutherland, 1949). This provocative conceptualization of

criminals wearing the proverbial white-collar - a symbol of professional success as opposed to

working class employees’ blue collar - was a landmark in the history of criminology. Until then,

the field had maintained its focus on such offenses as burglary, armed robbery, sexual assault and

murder, that is, illicit activities that were generally associated with impoverished neighborhoods

(Akers & Sellers, 2008). Conversely, Sutherland’s argument that large-scale fraud was not only

more socially harmful but also, because of the culprits’ high social standing, had better chances

of going unreported and/or unpunished, provided fertile grounds for new criminological theories.

Over time, though, considerable confusion in the literature has grown regarding the true

nature and extent of white-collar crime. Some scholars have interpreted Sutherland’s original

definition as ambiguous, reasoning that it does not distinguish crimes committed by individuals

from those perpetrated by corporations (Clinard & Quinney, 1973). Other critics have pointed out

that it could encompass both occupational crimes (i.e., acts committed in the course of one’s

occupation for personal gain) and avocational crimes (i.e., acts that are usually unconnected with

one’s profession such as income tax evasion or credit card fraud, Albanese, 1995). Strader (2002)

has called the term “white-collar” a misnomer, arguing that these crimes can be perpetrated

among the working class (e.g., scams, retail crime, tax evasion, etc.) as much as within the upper

class (e.g., antitrust violation). According to her, the term is a useful moniker to differentiate

nonviolent crime for financial gain committed by means of deception (e.g., securities fraud) from

more common (i.e., street) crime in the public mind. Similarly, Brightman (2009) has argued that

the white-collar crime construct should include any non-violent act committed for financial gain,

regardless of one’s social status.

A similar perspective is echoed in the Federal Bureau of Investigation’s Uniform Crime

Reports (UCR), which collects information on all street crimes reported to the police each year.

Interestingly, the eight index offenses included in Part I, which indexes violent crimes (i.e.,

aggravated assault, forcible rape, murder, and robbery) and property crimes (i.e., arson, burglary,

larceny-theft, and motor vehicle theft), do not include any type of white-collar crime. In fact,

embezzlement, forgery, counterfeiting, and fraud are the only potential white-collar crimes

included in Part II, which indexes less serious crimes such as status offenses (e.g., curfew

offenses, loitering, disorderly conduct, runaways, vandalism, etc.) and consensual crimes (e.g.,

drug offenses, prostitution, vagrancy, etc.). This assignment suggests that even law enforcement

agencies nationwide perceive street crimes to cause the most physical and property damage to

American society (Robinson 1994; Rosoff, Pontell, & Tillman, 2010).

The problem is that those four types of crime do not fall under the umbrella of the

whitecollar crime concept as envisaged by Sutherland. Sutherland was clear in his definition of

whitecollar crime that the term described an offense committed by a person of high social status

in the course of their occupation, which clearly leaves out working class crimes and crimes that

are not part of work contexts. However, such misreading of Sutherland’s original work is still

present today in the official data, which many criminologists rely upon in their investigation of

whitecollar crime.

Perhaps the economic changes that took place during the twentieth century were

responsible for causing scholars to misconstrue Sutherland’s message. Because the United States

has shifted from an industrial to a service economy, the nature of the job market is now very

different from what it was back in the 1940s. Today a postal clerk may well don a white collar

but can hardly be considered a person of “high social status”. In fact, an employee embezzling a

few hundred dollars from his/her company may share more similarities with a street offender

than with the powers-that-be denounced by Sutherland. If one defines power in terms of

economic control (i.e., ownership of the means and modes of production) and political influence,

then, respectable but relatively powerless figures (e.g., bank tellers, accountants, etc.) that would

be considered white-collar offenders according to the UCR do not quite fit Sutherland’s original

definition. Since white-collar crime has lost its original meaning, the term “elite deviance” may

be more apt to encapsulate the social class dimension of the construct.

Elite Deviance

Acknowledging the obsolescence of the “white-collar crime” appellation, some scholars

have favored the terms “crimes of the powerful” (Quinney, 1978) and “elite deviance” (Simon &

Eitzen, 1993), and used conflict as a theoretical framework to guide their research. Conflict

theory is a sociological paradigm that views society as being in a constant state of internal

conflict between a small ruling elite and the masses (Bonger, 1916; Chambliss, 1964; Lynch &

Michalowski, 2006; Mills, 1956; Paternoster & Bachman, 2001; Quinney, 1974; Sellin, 1938;

Turk, 1969; Vold, 1958). Following Sutherland’s introduction of the white-collar crime concept,

C. Wright Mills (1956) coined the term “power elite” to describe the unequal balance of power

between the leaders of military, political and corporate entities, and relatively powerless citizens.

Such imbalance is seen as running counter to the principles of fairness and equality before the

law, which form the basis of the American criminal justice system. More precisely, under this

perspective, those in power may manipulate the law to maintain their status and prosper while

keeping others subservient (Siegel & Welsh, 2011).

Criminal and deviant behaviors are by definition relative concepts that may differ from

one society to another and vary over time (Barkan, 2012). As Becker (1963:9) noted, “social

groups create deviance by making the rules whose infraction constitutes deviance, and by

applying those rules to particular people and labeling them as outsiders”. Since illegality is

determined by a distinct elite group of officials (i.e., the legislature), the same elite that labels

and prohibits street crime may overlook white-collar crimes such as war profiteering,

government corruption, large-scale fraud, or toxic dumping to protect powerful interests.

Consequently, corporations can engage in unethical (i.e., intentional, reckless, and negligent)

behaviors that are not legally defined as criminal (e.g., regulatory offenses) - even though they

may kill and injure more people every year than do street crimes - with little fear of legal

consequences (Lynch & Michalowski, 2006).

With elite deviance as a theoretical avenue to guide the present study, the measurement

of public knowledge about white-collar crime will extend well beyond the FBI’s official but

narrow definition and back to Sutherland’s original concept. More precisely, based on Friedrichs

and colleagues’ distinction between “organizational” offenses (e.g. committed by corporations)

and “individualistic” ones (i.e., perpetrated by individuals), this dissertation will mostly focus on

corporate crime because of its greater damaging impact on society. Moreover, to avoid

redundancies, the terms “white-collar crime” and “elite deviance” will be used interchangeably

and should be understood as synonyms of a form of criminal behavior performed by a small

dominant group of individuals in control of a disproportionate amount of power.

Typology

The three power elite groups referred to here - corporate, political, and military

institutions - can harm society through economic domination, government control, and denial of

basic human rights, respectively (Simon & Eitzen, 1993).

Economic Domination

Economic domination includes both financially and physically harmful offenses. Elite

deviance can be financially detrimental to the public when corporations defraud the government

(Theobald, 2009) or engage in tax evasion (DeBacker, Heim, & Tran, 2012). Companies can also

cheat consumers in cases of price-fixing (i.e., establishing the price of a product or service, rather

than allowing it to be determined by supply and demand, Connor, 2008), price-gouging (i.e.,

artificially inflating prices when no alternative retailer is available, Zwolinski, 2008), and false

advertising and misrepresentation of products (Tushnet, 2010). Moreover, competitors can be the

victims of anti-trust law violations (i.e., monopolistic practices, Dogan, & Lemley, 2008), and

insider trading (i.e., buying or selling of a security by someone who has access to nonpublic

information, Meulbroek, 2012). Finally, owners and creditors can be wronged via managerial

fraud (Fairchild, Crawford, & Saqlain, 2009), self-dealing (i.e., a fiduciary acting in his/her own

best interest in a transaction rather than his clients’, Djankov et al., 2008), and strategic

bankruptcy (Moerman & Van De Laan, 2009).

Corporate offenses can also be physically harmful to the public through unsafe

environmental practices such as toxic emissions above the legal limit (Katz, 2012; Lynch, &

Stretesky, 2010), toxic dumping and hazardous waste disposal (Waldo, 2009), and the release of

deadly pollutants (Rao, 2012). Moreover, corporations can harm consumers with the

manufacturing and distribution of unsafe products (Pyke & Tang, 2010). Lastly, employees can

be victimized when subjected to unsafe working conditions (Landsbergis, 2009), which can lead

to preventable occupational diseases, accidents and deaths (MacDonald, Cohen, Baron, &

Burchfiel, 2009).

Similarly, crimes by professionals in the medical, legal, academic, and religious sectors

can also be defined as elite deviance, insofar as the perpetrators operate under the guise of

respectability and the protection of powerful guilds. More specifically, physicians can harm their

patients through medical negligence and malpractice, unnecessary operations, tests, and other

procedures, as well as false and fraudulent billing (Gogos et al., 2011; Miller, 2012). The same is

true about attorneys when they defraud their clients or collude with their crimes (Abel, 2012). In

addition, college faculty can abuse their position through gross negligence in the fulfillment of

their teaching responsibilities, or when using fraudulent data in research (Michalek, Hutson,

Wicher & Trump, 2010). Lastly, televangelists (i.e., a portmanteau term referring to preachers

using television to communicate the Christian faith) sometimes engage in scams to defraud

believers and use offerings or donations for corrupt purposes (Morefield & Ramaswamy, 2011).

Government Control

Government control (i.e., state crime) can occur domestically when offenses are

committed by the legislative, judiciary and executive branches of power as is the case with

corruption (Issacharoff, 2010), corporate tax loopholes (McIntyre, Gardner, Wilkins, & Phillips,

2011), crimes of electioneering and usurpation of power (Christensen & Colvin, 2009), violations

of individual civil rights such as illegal surveillance by law enforcement agencies

(Michelman, 2009), denials of due process of law (Heupel, 2009), and political party infiltration

(Werbner, 2010). Governments can also impact and destabilize foreign nations through coups

d’état (Marshall, 2009), international law violation, unlawful warfare and war profiteering

(Sandholtz, 2009), the threat of nuclear war (Escalona, 1982), state repression and corruption

(Ross, 2010), and even collaboration with organized crime (Armao, 2012). Further, an indirect

negative consequence of government control and state crime is the erosion of public trust in

elites whose behavior fosters demoralization, cynicism, and alienation and deviance (i.e., street

crime).

Denial of Human Rights

State crimes may also include denial of human rights. The denial of basic human rights

refers to threats to the dignity and quality of life of humankind and particularly oppressed

minorities. It is logically expected when the unequal distribution of power places non-elites at the

mercy of profit-seeking groups that happen to define and control the law. Human rights violation

can be experienced among workers through economic exploitation and human trafficking, both in

underdeveloped and/or instable nations (Shelley, 2010) and in the United States (Bales, 2004;

Gillmore, 2004), unfair labor practices (e.g., surveillance of employees, Gorman, 2006), sexual

harassment (Murrell, Olson, & Frieze, 2010), and racial and gender discrimination in the

workplace (Beal, 2008; Pinzon-Rondon et al., 2010).

Denial of human rights reaches a culmination point with rape, torture, genocide and

ethnic cleansing (Totten & Parsons, 2012). Importantly, now that the Universal Declaration of

Human Rights - to which the United States is signatory - has been ratified by a sufficient number

of nations (Tsutsui, 2012), such acts should fall under international law. Nevertheless, more than

sixty years after its issue, Amnesty International and other sources report that individuals are still

tortured or abused in at least 81 countries, face unfair trials in at least 54 countries, and are

restricted in their freedom of expression in at least 77 countries (Hopgood, 2010).

Even in the United States, some types of human rights violations that are legally defined

as crimes by American law (e.g., affirmative action prevents discrimination against employees or

applicants for employment, on the basis of color, religion, sex, or national origin) are difficult to

prosecute due to the status of the offenders (Push, 2011). Prosecution is even impossible for

related offenses that have not been criminalized by the legislature (i.e., one of the three branches

of the power elite). For example, the forced labor of inmates by corporations, though arguably

unethical, is authorized by the 13th Amendment to the United States Constitution.

Similarly, while domestic human trafficking and workers’ exploitation does fall under

American law, the use of sweatshops and child labor in underdeveloped countries by U.S.-based

firms - facilitated by corporate globalization - allow these companies to lower costs and increase

profits with relative legal impunity due to their complicity with the local polity (Rosen, 2002).

These discrepancies further exemplify the collusions between state and corporations. In fact, the

interconnections between the political sphere and elite deviance could explain the paucity of

official data regarding white-collar crime.

Issues in Research

While administrative and regulatory agencies such as the Occupational Safety and Health

Administration (OSHA), the Environmental Protection Agency (EPA), and the Food and Drug

Administration (FDA) do collect information on regulatory offenses, they are given considerable

discretionary leeway in defining and responding to these offenses (Friedrichs, 1995). In fact, as

of 2011, OSHA had only secured 12 criminal convictions on the 51 cases the Department of

Justice agreed to prosecute since the administration’s inception in 1971 (Tribe, 2011).

Criminologists often rely on two types of data to examine officially recognized

whitecollar crime: The UCR and victimization surveys. There are several problems related to the

use of the UCR. First, as previously mentioned, the offense reported may not accurately reflect

the definition of white-collar crime proposed by Sutherland. Moreover, white-collar crime

victims are often unlikely to report the crime to authorities (Jesilow, Klempner, & Chiao, 1992;

Titus, Heinzelmann, & Boyle, 1995), either because they feel the police are neither willing nor

able to help or they are unaware of the existence of special fraud units (Friedrichs, 1995).

Moreover, victimized corporations may be reluctant to file reports because they fear negative

publicity or loss of trust from shareholders in the organization (Levi, 1992).

Similarly, using victimization surveys to draw an accurate picture of the prevalence and

impact of elite deviance has several limitations. Among these are the delayed harm white-collar

crime creates as well as its lack of immediacy compared with traditional crime, which may result

in a failure to recognize or report these offenses. More specifically, because there is typically

little direct contact between the offender and the victim in white-collar crime (Friedrichs, 1995;

Weisburd & Schlegel, 1992), many victims may not know they are victimized until later

(Albanese, 1995). Further, victimization surveys do not often directly ask about white-collar

crime victimization experiences. For example, while the National Crime Victimization Survey

(NCVS) conducts yearly household surveys, it mostly collects information about street crimes

such as rape, robbery, assault, burglary, larceny, and motor vehicle theft, and tends to bypass

white-collar offenses.

Moreover, researchers in the social sciences have rarely attempted to employ their usual

data collection methods such as surveys, interviews, direct observations and examination of case

records to explore the problem of white-collar crime. First, it is understandably more difficult to

collect data within the corporate world than from offenders such as juvenile delinquents. Before

one can obtain the cooperation of an organization or corporation, the research proposal has to be

introduced in a non-threatening way and present a potential benefit for the organization (Yeager

& Kram, 1990). Further, peer review panels for government research funding tend to favor

studies relevant to conventional crime and its control more than projects that intend to investigate

the wrongdoings of major corporations and government organizations (Galliher, 1979).

There are, however, alternative ways to conduct research on elite deviance. For example,

numerous databases maintained by a variety of state agencies address white-collar or elite crimes

(e.g., injury, disease, employment, and inflation data from the U.S. Bureau of Labor Statistics

and the Centers for Disease Control and Prevention, costs data from the National Council on

Compensation Insurance and the Healthcare Cost and Utilization Project, the National Academy

of Social Insurance, the U.S. Environmental Protection Agency, Centers for Disease Control and

Prevention, the National Center for Health Statistics, the Bureau of Labor Statistics, the Health

Care Financing Agency, the Practice Management Information Corporation, etc.). However,

because these data are not collected uniformly nor centralized, it remains difficult to estimate the

exact extent of the white-collar crime problem. Still, despite difficulties in aggregating these

data, it has been established that the financial and physical impact of elite deviance on society

overshadows that of traditional crime - even if, perhaps, the true extent of elite crime has been

underestimated in prior research.

Financial Impact

When he devised his conceptualization of white-collar offenses, Sutherland (1977:5)

already suspected that “the financial cost of white-collar crime is probably several times as great

as the financial cost of all the crimes which are customarily regarded as the crime problem”. He

also noted, however, that the financial harm resulting from white-collar crime was still less

important than the damage it caused to “social relations” through violation of trust as well as its

physical harmfulness. As previously mentioned, the exact impact of elite deviance is difficult to

ascertain. What is known, however, is that whatever figure white-collar crime represents in terms

of financial cost to society, that figure greatly exceeds the physical and financial harms caused by

traditional crimes. More precisely, it has been estimated that street offenders cost the public an

average $18 billion each year (UCR, 2010). This amount includes the total economic loss to

victims, that is, $1.19 billion for violent crime (e.g., loss of revenue due to homicide, health costs

as a result of assault, rape, etc.) and $16.21 billion for property crime (BJS, 2007). More

specifically, in 2010 the average dollar loss due to robberies reported to the police was $456

million, $6.1 billion were lost to larceny/thefts, $1.4 billion to arsons, $4.6 billion to burglary

(UCR, 2010), and another $499.9 million in victim compensation programs (NACVCB, 2011).

Figures related to the cost of street crime, although considerable, are still relatively

modest compared to the combined costs of corporate crime such as fraud, tax evasion,

pricefixing, price-gouging, false advertising (see, e.g., Simon, 1999; Simon & Eitzen, 1993),

anti-trust violations, and embezzlement (see, e.g., Weisburd et al., 1991), which add up to over

$500 billion dollars annually (Lynch & Michalowski, 2006; Rebovich & Jiandani, 2000; Reiman

& Leighton, 2010). Of course, this figure does not include occasional scandals such as the Enron

debacle, which cost its shareholders an outstanding $74 billion (Pava, 2010). Even an individual

white-collar crime such as Bernard Madoff’s Ponzi scheme added up to $65 million in economic

losses for investors (Arvedlund, 2009), and alone far exceeds the cost of all street crimes that

year.

Health costs due to corporate misconduct must also be included. The number of fatal and

nonfatal injuries in the workplace in 2007 was estimated to cost $6 billion and $186 billion,

respectively. Similarly, the cost of fatal and nonfatal illnesses was estimated at more than $46

billion and $12 billion, respectively. For injuries and diseases combined, cost estimates were $67

billion, and indirect costs were almost $183 billion. Overall, the total estimated medical costs

were approximately $250 billion (Leigh, 2011).

Further, total annual health costs associated with environmental pollutants (e.g., toxic

chemicals of human origin in air, food, water, and communities) were estimated to be $43.4

billion for lead poisoning, $2.0 billion for asthma, $0.3 billion for childhood cancer, and $9.2

billion for neurobehavioral disorders. Alarmingly, this sum constitutes almost 3 percent of total

U.S. health care costs (Landrigan et al., 2002). Moreover, besides medical and insurance

administration expenses, indirect categories such as lost earnings, lost home production, and lost

fringe benefits further add up to the economic burden of elite deviance (Leigh et al., 1997).

Finally, the impact of corporate activity on the environment also comes at a price. More

specifically, costs associated with predicted climate change–related events such as ozone

pollution, heat waves, hurricanes, infectious disease outbreaks, river flooding, and wildfires have

been estimated to be as high as $14 billion. Ninety-five percent of this figure can be attributed to

the value of lives lost prematurely. Moreover, subsequent health care costs due to these climate

change-related events add up to $740 million (Knowlton et al., 2011). If we add up all these

figures, it appears white-collar crime may cost society over a trillion dollars a year compared

with the $18 billion lost to street crime. Perhaps more disconcerting than the gargantuan

economic impact of elite deviance is the evidence that it causes more violent and physical harm

than does street crime.

Physical Harm

Official statistics reveal that about 14,000 people are victims of murder and negligent

manslaughter every year (UCR, 2010). In contrast, reasonable estimates place the number of

deaths due to toxic waste dumping and deadly pollutants, preventable occupational diseases,

accidents and deaths, faulty consumer products (e.g., automobiles, toys, food, drugs, etc.), and

nefarious and addictive substances (e.g., tobacco) at 100,000 a year, that is, eight times as much

as the number of street crime victims (Lynch & Michalowski, 2006; Reiman & Leighton, 2010).

As of 2000, approximately 65,000 workers died each year of work-related injuries (e.g., motor

vehicle accidents, machinery-related events, manslaughter, falls, and electrocution) and

occupational diseases such as cancers, asbestosis, and silicosis (Herbert & Landrigan, 2000). In

2007, the numbers of fatal and nonfatal injuries were estimated to be more than 5,600 and almost

8,559,000, respectively. Further, the numbers of fatal and nonfatal illnesses were estimated at

more than 53,000 and nearly 427,000, respectively (Leigh, 2011).

Moreover, iatrogenic effects (i.e., injuries caused by a physician, including both error and

nonerror adverse events) have made medical care the third leading cause of death in the United

States, after heart disease and cancer (Starfield, 2000). An estimated 12,000 people die each year

from unnecessary - but lucrative - surgeries such as silicon breast implants, circumcision,

coronary artery bypass, hysterectomies, and cesarean (Lynch & Michalowski, 2006), 7,000 of

medication negligence in hospitals, 20,000 of other errors in hospitals, 80,000 of infections in

hospitals, and 106,000 of negative effects of drugs, for an estimated total of 225,000 victims of

medical crime (Starfield, 2000).

The harm caused by the premature - and profit-driven - release of potentially deleterious

drugs should also be included. One such example was the Fen/Phen drug case (Mundy, 2001) in

which the injurious effects of a supposedly harmless medicine were silenced until they had

caused the death of more than 100 people. Another similar scandal was the Thalidomide

prescription disaster where a sleeping-pill-tranquilizer caused 8,000 babies to have been born

deformed (Pontell & Geis, 2007). Due to the ties between the pharmaceutical industry and the

polity, such examples qualify as elite deviance. Indeed, the Federal Drug Administration (FDA)

sometimes overlooks international regulations regarding potentially harmful drugs. For instance,

it approved a livestock drug - beta agonist ractopamine (i.e., a repartitioning agent that increases

protein synthesis) - which had been banned in 160 nations due to its correlation with

hyperactivity, muscle breakdown and ten percent mortality in pigs (Rosenberg, 2010).

Finally, the harmfulness of environmental pollution cannot even be clearly calculated.

Attaching an economic or physical cost to that kind of behavior becomes difficult when the end

result is the destruction of the environment, extensive human and animal misery. Nevertheless, if

the intent is simply to compare elite deviance with traditional crime in terms of physical harm,

currently available figures suggest that white-collar crime may harm or injure over 8,986,000

people every year and lead to the untimely death of another 283,600 people, that is, more than 20

times as many as street crime. Contrasting these figures to the 24,330 homicides reported by the

FBI when homicide rates nationwide peaked in 1990 (Lynch & Michalowski, 2006) further puts

the potential public “myth” that street crime is more harmful than elite deviance back in

perspective.

Relative Legal Immunity

Nevertheless, because of the abovementioned difficulty to investigate these offenses, this

figure may well underestimate the magnitude of the problem. Again, many acts that endanger

human lives without being clearly defined as illegal or criminal (e.g., state and government

crime) inevitably remain unreported and/or unpunished. Worse yet, even white-collar offenses

that fall under American law are statistically less likely to be prosecuted than street crime. In fact,

despite efforts to bring white-collar offenders to justice (e.g., citizen suits by private parties who

have been injured or threatened, or class action suits by a group of directly injured parties,

Friedrichs, 1995), corporate regulations are weak and corporate offenders have the financial

means to provide effective defenses, delay judgments, avoid criminal sanctions and receive more

lenient sentences compared to those imposed upon street criminals (Calavita, Tillman, & Pontell,

1997; Maddan et al., 2011).

Further, given the difficulty in establishing criminal intent when dealing with a complex,

hierarchized organization, prosecutors are often reluctant to take on corporate crime cases

(Braithwaite, 1982; Braithwaite and Geis, 1982; Sinden, 1980) and usually prefer to focus on

individual white-collar offenders. Business organizations undeniably profit from the “rotten

apple” perspective on occupational crime, which favors an individualistic rather than systemic

explanation of elite deviance (Ashforth et al., 2008; Gottschalk, 2012). Based on such a

perspective, Bernard Madoff ’s surprisingly severe prison sentence could be understood as

symbolic degradation (Levi, 2009), that is, an attempt to soothe public outrage at other, more

harmful white-collar offenses such as the British Petroleum gulf oil spill, which resulted in a $7.8

billion class action settlement but no criminal prosecution and no prison time for the company’s

executives (Thomas, 2012). Perhaps more alarming is the unwillingness of economically

depressed communities to cooperate and press charges against local corporations if such legal

action could result in massive unemployment in the area (Friedrichs, 1995).

Given the wide gap between the deleterious impact of elite deviance on society and that

of street crime, the relative legal immunity enjoyed by white-collar offenders should logically

cause indignation among the populace. It is not certain, however, to what extent the public is

aware of the body of knowledge that has been amassed by the empirical literature on white-collar

crime. In fact, research shows that Americans are not well informed about crime in general.

Popular “Myths” about Traditional Crime

National and state surveys conducted in the United States (e.g., Knowles, 1984, 1987;

Maguire & Pastore, 1995), Canada (e.g., Doob and Roberts, 1982) and Australia (e.g., O’Connor,

1978; Indermaur, 1987) have revealed that the public holds several misconceptions about crime

rates (Robert & Stalans, 1997). One such misconception is crime being rampant (Barkan, 2012).

More specifically, Knowles (1984) reported that when the annual rate of violent crime in Ohio

was fewer than four incidents per 100 residents, only 5% of respondents perceived the rate to be

that low. Similarly, in a 1993 Harris poll that asked Americans whether they thought crime rates

had increased, decreased or remained the same, more than 50% of respondents believed that

crime had increased when in fact significant declines had been observed for a variety of crimes

(Maguire & Pastore, 1995). Official statistics reveal that crime has been falling sharply and

unexpectedly since the mid-1990s. For example, between 1993 and 2000, murder and robbery

rates both declined over 40% (Blumstein, 2006), and the trend is ongoing today (Barkan, 2012;

UCR, 2010). Nevertheless, a 2010 Gallup poll revealed that 67% of respondents believed that

crime rates were increasing (Albanese, 2012).

Another popular misconception seems to be the particularly violent nature of crime

(Indermaur, 1987; Macleans, 1995; Warr, 1980, 1982) when in reality official data and

victimization surveys suggest that violent crime (i.e., homicide, rape, robbery, and assault) only

represents 12% of all serious crimes reported to the police and 20% of all crimes counted by

victimization surveys, whether reported or not (Albanese, 2012; Barkan, 2012; Robert & Stalans,

1997). The belief in high recidivism rates constitutes yet another public misconception. Doob and

Roberts (1983) found that 60% of their respondents over-estimated the recidivism rate for

property offenders, while 79% over-estimated the figure for violent offenders. Other “myths”

include the belief that most victims are Whites and targeted by African American criminals

(when in fact most violent crime is intraracial, and Blacks are more likely to be victimized than

Whites, Lundman, 2003), and that most violent offenders are youths (when in reality, only about

14% of violent crime is committed by teenagers, Dorfman & Schirardi, 2001).

Perhaps the most salient crime “myth” is the deterrent effect of “get tough” policies (e.g.,

“three strikes and you’re out” laws, longer sentences for less serious crimes, use of the death

penalty, etc.), which are all corollaries of a more punitive approach to crime control. Although

several studies have found some evidence for the purported deterrent effect of incarceration on

crime rates, they also vary widely in their conclusions about its strength (Blumstein, 2006;

Blumstein & Wallman, 2000, 2006; Levitt, 2004). For example, Doob and Webster (2003) found

some inconclusive or, at best, weak evidence of marginal deterrence. Similarly, Pratt and

colleagues (2006) maintained that the effects of severity estimates and deterrence/sanctions

composites, even when statistically significant, are too marginal to suggest practical policy

implications. More recently, Paternoster (2010) critiqued deterrence research and effectively

suggested that the evidence is so weak that criminologists should stop examining this point

altogether. Lastly, despite popular belief, the vast majority of studies on capital punishment have

concluded to its lack of deterrent effect (for a review, see Bohm, 1987, 2003).

How to explain such discrepancy between actual crime statistics and common perceptions

among the public? Why are people so misinformed? Where do they get their information?

Research suggests that Americans derive most of their knowledge about crime from the news

media (Dowler, 2003; Roberts & Doob, 1990; Surette, 1998). Unlike street crime, which is

overemphasized on local news stations (Barlow, Barlow, & Chiricos 1995; Chiricos 1995; Marsh

1991; Ruel 1994; Welch et al. 1998), white-collar crime is not widely reported (Barak, 1994;

Barlow & Barlow, 2010; Ericson et al., 1991; Lynch & Michalowski, 2006, Lynch, Nalla &

Miller, 1989; Lynch, Stretesky & Hammond, 2000). Further, when it is, the perpetrators are often

portrayed in a less negative light than are street offenders (Geis & Meier 1977; Welch et al. 1998;

Williams 1992). Although a case could be made that the treatment Bernard Madoff received in

the wake of his Ponzi scheme scandal was nothing short of media lynching, news networks are

less prone to vilifying well-known and respected companies such as Microsoft, which

nonetheless engage in equally reprehensible white-collar offenses (Lande & Hawker, 2011).

Some scholars view this disparity as the result of powerful entities exerting control over

the media (Barkan, 2012; Lynch & Michalowski, 2006). In keeping with conflict theory, which

denounces the imbalance of power between a ruling elite and the rest of the populace (Mills,

1956), corporations should logically benefit from the public’s ignorance of white-collar crime.

Absent ubiquitous awareness of the deleterious effects of elite deviance, these corporations can

persist with their wrongdoings without fear of legal consequences. Consequently, the media’s

almost exclusive focus on traditional crime may serve to distract citizens away from other and

arguably more serious types of criminality, leaving the American public with distorted views

concerning the reality of street crime and, expectedly, very limited knowledge about elite

deviance. In fact, people may very well harbor “myths” about white-collar crime as they do

regarding traditional crime. Nevertheless, such hypothesis has never been directly tested since

most of the empirical research on elite deviance has focused on attitudes and not knowledge.

Public Attitudes about White-Collar Crime

The literature on perceived seriousness of street crime generally reveals high levels of

consensus among the American public (Grabosky, Braithwaite, & Wilson, 1987; Hauber,

Toonvliet, & Willemse, 1988; Newman, 1976; Scott & Al-Thakeb, 1977; Warr, 1989; Wolfgang

et al., 1985). More specifically, a majority of respondents tend to view conventional crime as a

serious social issue that requires harsh sanctions. Conversely, early research on public

perceptions of elite deviance suggested that Americans did not consider white-collar crime as an

important social problem, especially in comparison to crimes committed against a person or the

public (Geis, 1973; Sutherland, 1949; Wheeler et al., 1988). Two probable factors for such

apathy could be the aforementioned delayed harm and lack of immediacy of white-collar crime.

For example, work-related illnesses and deaths may occur years after exposure to toxic

substances in the workplace and may not shock public opinion as much as homicide (Albanese,

1995).

Nevertheless, Braithwaite (1982:732-733) noted that “contrary to a wide spread

misconception, there is considerable evidence to support the view that ordinary people

subjectively perceive many types of white-collar crime as more serious than most traditional

crime.” Similarly, Conklin (1977:27) posited that there is a “greater degree of public

condemnation of business violations than is thought to exist by those who claim that the public is

apathetic or tolerant of business crime”. In fact, several studies (Cullen, Link, & Polanzi, 1982;

Cullen, Clark, Mathers, & Cullen, 1983; Cullen et al., 1985; Grabosky, Braithwaite, & Wilson,

1987; Geis, 1972; Hauber, Toonvliet, & Willemse, 1988; Holtfreter et al., 2008; Meier & Short,

1982; Rebovich & Jiandani, 2000; Rebovich & Kane, 2002; Rossi et al., 1974; Schrager & Short,

1980; Sinden, 1980; Wolfgang et al., 1985) have suggested that the traditional wisdom about

public apathy regarding white-collar crime might be erroneous.

It could be that elite deviance draws more attention when it mimics street crime’s

violence and physical harmfulness (Albanese, 1995). For example, Holtfreter and colleagues

(2008) found that while fraud might be viewed more favorably than robbery (perhaps because the

latter implies a violent act), white-collar offenses such as work-related deaths that could have

been prevented are usually perceived more negatively. Importantly, their respondents were not

opposed to severe sanctions for white-collar offenders if evidence of harm to society could be

provided. Nevertheless, harmfulness may not be the decisive factor in explaining perceived

severity of elite deviance. In fact, Schoepfer, Carmichael and Piquero’s 2007 study of public

perceptions of sanction certainty and severity indicates that both robbery and fraud are perceived

to be equally reprehensible.

National research efforts reveal similarly surprising public perceptions of white-collar

crime’s seriousness relative to street crime. Every five years since 1999, the National

WhiteCollar Crime Center (a congressionally funded non-profit corporation) has been surveying

public attitudes about the seriousness and impact of elite deviance. More precisely, the National

Public

Survey on White-Collar Crime has been conducted and published on three consecutive occasions

(Rebovich et al., 2000; Kane & Wall, 2006; Huff, Desilets, & Kane, 2010). The first iteration,

conducted by Rebovich and colleagues (2000), compared respondents’ perceptions of the

likelihood of apprehension of street and white-collar offenders with their opinions on how they

should be sanctioned. Participants were presented with a scenario that compared the chances of

apprehension of someone stealing $1,000 in a robbery with someone obtaining $1,000 through a

fraudulent action. Only 22% of the sample believed the fraudster had a greater likelihood of

being apprehended. Even fewer respondents (16%) believed that the convicted fraudster would

be punished more severely by the criminal justice system. Although public knowledge about elite

deviance was not directly measured, these results nonetheless suggest that people may be partly

aware of the relative legal immunity enjoyed by white-collar offenders compared to their street

counterparts.

A comparison of the respondents’ perception of who would be arrested with their beliefs

about who should be punished more severely revealed an interesting difference. More

specifically, less than one third thought that the robber should be punished more severely, while

higher percentages believed the fraudster deserved greater punishment and that both should be

punished with equal severity. Again, these results run counter to the old perception that people

are apathetic about or more lenient with elite deviance.

The second effort by the National White-Collar Crime Center (Kane & Wall, 2006)

expanded upon the original by proposing a more comprehensive definition of the white-collar

crime construct that included illegal or unethical acts violating fiduciary responsibility or public

trust for personal or organizational gain. This new definition incorporated high-tech crimes and

crimes committed both inside and outside of the occupational setting. Importantly, the survey

focused on three areas of public experience with white-collar crime: victimization, reporting

behaviors, and perceptions of crime seriousness. Participants were presented with twelve

scenarios depicting both white-collar offenses and more traditional types of crime. These

scenarios were dichotomized into the following categories: white collar/traditional crime,

physical harmful/financially costly crime, organizational/individual offenders, and

highstatus/non-status offenders. Respondents were found to rate white-collar crime as equally

serious as street crime. Moreover, physically harmful crimes were perceived to be significantly

more serious than those that involved monetary loss. Interestingly, organizational offenders were

rated more negatively than individual offenders. Finally, participants reacted more harshly

against crimes committed by high-status offenders in a position of trust.

These results were echoed in the third and latest National Public Survey on White-Collar

Crime (Huff, Desilets, & Kane, 2010). Once again, measures of public opinion about and

perceived seriousness of various types of elite deviance were collected. As was the case before,

researchers developed a series of scenarios that were first incorporated into different categories

expressing specific variables and then grouped on the basis of common attributes to compare the

perceived seriousness (1) of a white-collar crime to that of street crime, (2) of a crime involving

harm to that of a crime involving financial loss, (3) of a crime involving an organizational

offender to that of a crime involving an individual offender, and (4) of a crime committed by a

high-status offender (e.g., an individual in a position of trust) to that of a crime committed by a

low-status offender.

This time, survey results indicated that the public tended to view white-collar crime as

slightly more serious than street crime. Not surprisingly, crimes that involved direct physical

harm to individuals were once again found to be more serious than the crimes that resulted in

monetary loss. As was the case in 2006, participants reacted more harshly to cases involving

organizational instead of individual offenders. Similarly, crimes involving high-status offenders

were statistically deemed more serious than those involving low-status offenders. Although the

National Public Survey on White-Collar Crime examined only a limited number of white-collar

and street crime scenarios, these findings suggest that Americans may see elite deviance as a

more serious social issue than was previously hypothesized and might even sometimes take it

more seriously than traditional crime.

Perhaps, then, punitiveness is a function of knowledge. After all, maybe the 2010 British

Petroleum oil spill in the Gulf of Mexico met with such public indignation and outrage because

of the visible, tangible damage it provoked. The greatest limitation of the National White-Collar

Crime Center survey may have been its failure to control for public knowledge about elite

deviance. While respondents’ fear and negative attitudes about corporate offenders suggest a

reasonable amount of knowledge about this form of deviance, the true extent of information

among the public - particularly in regards to the physical costs of white-collar crime - remains

uncertain. Hypothetically, there could be increased demand for tougher sanctions against

highstatus offenders if people were better informed about the tremendous impact of elite

deviance on society. Then again, this suggestion is mere conjecture since no study, to date, has

provided a measure of what people really know about white-collar crime. Measuring public

knowledge about elite deviance and comparing it to sentiments about white-collar crime would

therefore represent an important contribution to the field.

The Present Study

The present study expands research on public opinions about white-collar crime by

providing a much-needed measure of popular knowledge regarding elite deviance. The following

research questions will be addressed:

1) Is the public informed about elite deviance? If it is, to which extent are Americans

informed about it?

The field of epistemology (i.e., the branch of philosophy concerned with the study of

knowledge) differentiates truth (evidence-based information) from belief (non-evidence-based

opinion). One proposition is that knowledge may be better understood as justified true belief

(Turri, 2012). That is, a belief becomes knowledge if it is both believed to be true and currently

supported by extant research. Knowledge about elite deviance will therefore be conceptualized

and operationalized as statements that according to extant research are commonly regarded as

valid (e.g., superior harmful costs to society compared with street crime). As previously

mentioned, since most Americans derive their information from the news media (Dowler, 2003;

Roberts & Doob, 1990; Surette, 1998), and because the news media do not report on cases of

white-collar crime at the same rate they do on street crime (Barak, 1994; Barlow & Barlow,

2010; Ericson et al., 1991; Lynch & Michalowski, 2006; Lynch, Nalla & Miller, 1989; Lynch,

Stretesky & Hammond, 2000), it is plausible to expect that the public is not well educated about

the problem of elite deviance.

2) Is there a gap between the public’s subjective and objective knowledge about whitecollar

crime?

By “subjective” knowledge, one must understand what people think they know about

elite deviance. Conversely, “objective” knowledge refers to what they actually know about

white-collar crime. Several studies in the social science have observed a gap between the public’s

subjective and objective knowledge about various issues related to crime and the criminal justice

system. Phrased differently, there can be a discrepancy between what individuals believe they

know and how informed they truly are. As previously mentioned, extant research on public

knowledge about street crime has revealed gross differences between the people’s perceived and

actual risks of victimization (e.g., Doob & Roberts, 1983; Kappeler, Blumberg, & Potter, 1996;

Macleans, 1995; Maguire & Pastore, 1995; Roberts & Stalans, 1997; Wilbanks, 1987). Such

discrepancy was also observed with measures of public information about capital punishment

(Bohm, 1987, 2003; Cochran & Chamlin, 2005). More specifically, those studies revealed that

Americans were not as informed about issues related to the death penalty as they thought they

were. Therefore, there might be a similar gap between what individuals think they know about

white-collar crime and their actual level of information regarding elite deviance.

3) Does the public hold common “myths” about elite deviance like they do regarding street

crime?

With knowledge about elite deviance conceptualized and operationalized as statements

that according to extant research are commonly regarded as valid (e.g., superior harmful costs to

society compared with street crime), unsubstantiated beliefs and arguments that have been

debunked by science but which the American public might still commonly hold therefore

constitute white-collar crime “myths”. As previously mentioned, research has identified public

“myths” about street crime and the criminal justice system (e.g., Bohm, 1987; Doob & Roberts,

1983; Kappeler, Blumberg, & Potter, 1996; Macleans, 1995; Maguire & Pastore, 1995; Roberts

& Stalans, 1997; Wilbanks, 1987). Given the limited coverage allotted to various forms of

whitecollar crime in the news media (Barak, 1994; Barlow & Barlow, 2010; Ericson et al., 1991;

Lynch & Michalowski, 2006; Lynch, Nalla & Miller, 1989; Lynch, Stretesky & Hammond,

2000), it is very likely that Americans harbor specific “myths” regarding elite deviance. Recall

that some studies have shown that people tend to rate white-collar crime as less serious than

street crime until they are presented with evidence of the harmfulness of elite deviance

(Albanese, 1995; Huff, Desilets, & Kane, 2010; Holtfreter et al., 2008). Consequently, the

perceived harmlessness of white-collar crime compared to its street counterpart could be one of

many prevalent “myths” about elite deviance. This study will therefore attempt to identify such

public “myths”.

4) What are the correlates of knowledge about white-collar crime?

Because white-collar crime is such a complex social phenomenon, information about elite

deviance might correlate positively with one’s general level of education. That is, those with

higher degrees may have had previous exposure to relevant information about white-collar crime

and possess better knowledge and understanding of it. Still, it is not clear what other

characteristics may be associated with knowledge about elite deviance, either positively or

negatively. Does knowledge about white-collar crime vary by age, race, gender, profession,

socioeconomic status, or other sociodemographic characteristics? Research suggests significant

variations in perceived seriousness of white-collar crime among the public. More specifically,

some studies found that older people and people of lower socioeconomic status tend to view elite

deviance as somewhat more serious than conventional violent crime and narcotic offenses

(Grabosky, Braithwaite, & Wilson, 1987; Hauber, Toonvliet, & Willemse, 1988). Further, Blacks

tend to rate white-collar crimes directed at consumers (e.g., fraud, health threats and deception

from the production and sale of goods and services, etc.) as somewhat more serious than do

Whites, who seem more sensitive to white-collar crimes directed at businesses (e.g., forgery,

embezzlement, etc., Miethe, 1984). However, it is not known whether such attitudes were

dictated by prior knowledge of the problem.

5) Is knowledge about elite deviance correlated with public opinion regarding whitecollar

crime?

Several studies have found that because business executives, managers, criminal justice

bureaucrats, and lawyers are aware of the complexity of white-collar crime compared to street

crime, they tend not to consider harsh penal sanctions to be the most effective way to deal with it

(Cole, 1983; Frank et al. 1989; Hartung, 1953; McCleary et al., 1981). Of course, it could be that

such individuals are unlikely to support severe sentences for white-collar offenses because they

are themselves part of the elite groups that engage in those acts. It is uncertain, however, whether

and how knowledge about elite deviance may affect public opinion regarding its seriousness as

well as the measures that should be taken against its perpetrators (e.g., type of legal actions,

choice of punishment, sentence severity, etc.).

CHAPTER THREE:

METHODS

Sample

The subjects in this study were recruited on Amazon’s Mechanical Turk, a web service

that coordinates the supply and demand of human intelligence tasks (HIT). More precisely,

requesters post HITs that can be done on a computer (e.g., a survey) and workers volunteer to

complete them based on the size of the reward (i.e., a small payment) and maximum time allotted

for the completion. Mechanical Turk has recently become popular in the social sciences,

particularly in psychology (e.g., Berinsky, Huber, & Lenz, 2011; Buhrmester, Kwang, & Gosling,

2011; Horton, Rand, & Zeckhauser, 2011; Paolacci, Chandler, & Ipeirotis, 2010), but also in

criminology (Nadler & McDonnell, 2012; Robinson, Goodwin, & Reisig, 2010) as a source of

data collection due to its low cost and convenient recruitment.

Mechanical Turk presents several advantages compared to other data collection methods.

Recruitment is made easy by the increasing popularity of crowdsourcing platforms (i.e., websites

that outsource jobs to an undefined group of individuals in the form of an open call).

Crowdsourcing offers a large, stable pool of people willing to participate in experiments for very

low pay (Mason & Suri, 2011). Since there are no travel costs, and because workers choose when

they want to complete tasks, the effort to participate is much lower than in lab-based

experiments. As a result, rewards typically range from $0.01 to $1, with most HITs being paid

$0.10, which makes it a relatively inexpensive way to collect data. While compensation rate and

task length do impact participation, payment levels do not appear to affect data quality. For

example, mean alphas computer by Buhrmester and colleagues (2011) for data collected at three

levels of compensation (2, 10, and 50 cents) were within one hundredth of a point across the

three compensation levels. Further, research suggests that data collected on Mechanical Turk are

as reliable as those obtained via traditional methods. More specifically, Paolacci and colleagues

(2010) replicated standard judgment and decision-making experiments among subjects recruited

on Mechanical Turk, online discussion boards, and at a large university and found the results to

be qualitatively identical. In addition, Mechanical Turk workers are at least as representative of

the U.S. population as traditional subject pools, with gender, race, and age all matching the

population more closely than college undergraduate samples and Internet samples in general

(Berinsky, Huber, & Lenz, 2011; Buhrmester, Kwang, & Gosling, 2011; Paolacci, Chandler, &

Ieirotis, 2010).

Moreover, Amazon’s terms of service follow standard guidelines in conducting research

with human subjects. For instance, both requesters and workers must be at least 18 years of age.

In addition, the system does not allow requesters to ask for identifying information, thus

protecting workers’ anonymity. More specifically, workers’ IDs are anonymized strings that do

not contain personally identifiable information. Subjects are further protected in that they can

read brief descriptions and see previews of the tasks before accepting to work on them.

Informed consent is provided via a statement on the preview page of the HIT that explains

the purpose, risks and benefits of the task, and where to contact the researcher (and/or IRB). The

working conditions and hours are entirely determined by the worker and there is no direct or

indirect obligation on the workers to do any unwanted work. Once they choose to complete a

task and effectively do so, the researcher who supplied that task pays them. Payment is denied if

the task is not entirely completed, which helps ensure data quality and minimize attrition.

Furthermore, confidentiality is facilitated by the use of external HITs. More specifically, by

providing a survey link to another website instead of uploading the survey directly on

Mechanical Turk, the data go straight from the worker to the external website (e.g., Qualtrics, a

web-based survey service that provides advanced security and confidentiality for results with

password protection) and are never available to Amazon.

Data Collection Process

Once approved by the IRB, a questionnaire was uploaded on Qualtrics and a survey link

was made available on Mechanical Turk. The preview page included the title of the study and a

brief description of the purpose, risks and benefits of the project. The name and contact

information of the lead investigator (myself) and institutional review board were included. The

description of the study stated that subjects’ participation was voluntary, that in accordance with

Amazon’s terms of service their identity would remain anonymous, and that the information they

would provide would not be tied to them in any way.

Also included was the payment amount and expected completion time. These two

elements are important in determining sample size. Prior research suggests that recruiting 500

subjects on Mechanical Turk for a social science experiment is a realistic goal (see, e.g.,

Berinsky, Huber, & Lenz, 2011; Buhrmester; Kwang, & Gosling, 2011). Paolacci and colleagues

(2010) posted a task that required workers to answer a 5-minute survey for $0.10 and were still

able to attract 131 workers. Because the instrument to be used in the present study was twice as

long (pilot-testing showed that completion time was about 10 minutes) and admittedly more

complex, a monetary incentive of $2.00 per respondent was proposed to maximize completion

rate.

Prospective subjects were also informed that they could take the survey only once.

Restricting a task to one attempt was made easy by Amazon’s policy, which prevents workers

from having multiple accounts. Moreover, while Mechanical Turk is available in several nations,

it was possible to restrict the desired sample to one country only. This is a non-negligible

advantage, given that the present study proposes to measure public knowledge and attitudes

about elite deviance in the United States. Lastly, because it was uncertain how fast quality data

would be obtained, it was decided that the survey would be initially available for one month and

could be extended for another until the sample size goal had been reached.

Data collection took place on April 1st, 2013. The sample size goal of 500 participants

was reached within only three hours. As previously mentioned, piloting of the instrument

indicated that at least 10 minutes were necessary to read all questions carefully and provide

honest answers. As a result, every survey completed in less than 10 minutes was considered

unusable and subsequently deleted. Eliminating incomplete and/or dubious surveys yielded a

final sample of 408 respondents. Overall participants rated the experience favorably. More

specifically, several respondents took the time to email the lead investigator to comment on the

interest they took in the survey’s topic and expressed their desire to seek out further information

about white-collar crime.

The demographic characteristics of the sample were somewhat representative of the

overall American population. The 2010 United States Census indicates that the national median

age is 36.8, that 50.8% of Americans are females, that 78% identify as Whites, 13.1% as Blacks,

1.1% as Middle Eastern, 1.2% as American Indians or Alaskan Natives, 5% as Asians, 0.2% as

Native Hawaiians or Pacific Islanders, and 16.7% as Hispanics. Comparatively, the median age

in this study was 31, 49.8% of the respondents were females, 83.6% of them identified as

Whites, 8.8% as Blacks, 0.5% as Middle Eastern, 0.5% as American Indians or Alaskan Natives,

5% as Asians, 0.2% as Native Hawaiians or Pacific Islanders, and 6.9% as Hispanics. Further, the

median annual household income was between $40,000 and $49,000 (i.e., slightly lower than the

national estimate of $52,762). College graduates (Bachelor’s degree or higher) accounted for

49.5% of the total sample (against only 28.2% at the national level), and 41.7% of the

respondents were employed full-time (against 44.1% of the overall population).

It therefore appears that the largest discrepancies with national rates involve age,

race/ethnicity, and education. These gaps can be explained by the choice of data collection

method. Despite its increasing popularity, Amazon’s Mechanical Turk is still a relatively new

market place for work that is restricted to a specific category of individuals with knowledge of

and interest in computer technologies and online crowdsourcing. Not surprisingly, these

individuals are more likely to be younger and better educated. It is not clear, however, why racial

and ethnic minorities were not more represented. A likely explanation is that while the proportion

of Black and Hispanic Internet users has nearly doubled between 2000 and 2010,

African-Americans and immigrant Spanish speakers are still less likely than Whites to go online

(Smith, 2010). Despite its imperfections, the sample in this study is still an acceptable proxy for

the American public, following national trends closely in terms of gender, household income and

employment status distribution. Table A (see Appendix A) summarizes the descriptive statistics

of the variables and measures that follow.

Measures

The questionnaire used in this study included five sections: (1) Respondent

sociodemographic characteristics (items 1 to 17), (2) measures of public knowledge about elite

deviance (items 18 to 27), (3) perceived seriousness of white-collar crimes compared with a

baseline street crime (items 28 to 38), (4) punitiveness, including perceived seriousness of

whitecollar crimes involving physical risks compared to harmful street crimes with choice of

prosecution process, sentence determination, and sentence severity (items 39 to 43), and (5)

choice of attribution style (i.e., perceptions of white-collar offenders’ motives; items 44 to 51).

Sociodemographic Control Variables

A sociodemographic questionnaire was used to control for certain variables that might

account for variation in public information and opinions concerning elite deviance. As previously

mentioned, these potential correlates of knowledge about white-collar crime include gender (0 =

female, 1 = male), age (coded in years), and race (1 = White, 2 = Black or African American, 3 =

Asian, 4 = Middle Eastern, 5 = Native Hawaiian or other Pacific Islander, 6 = American Indian or

Alaskan Native, and 7 = Other). The following dummy variables were then created: Whites,

Blacks, and Other race. Further, ethnicity was measured via the following options: Hispanic,

Latino, or Spanish origin (1 = yes, 2 = no) and then dichotomized (0 =NonHispanic, 1 =

Hispanic).

In an effort to control for cultural differences in knowledge and opinions about

whitecollar crime, the region where the respondents grew up was also included (1 = North, 2 =

East, 3

= South, 4 = West, 5 = Midwest, 6 = Other) and then dummy coded (1 = Northeast, 0 = Other).

Also included was participants’ current residence (1 = A large central city (over 250,000), 2 = A

medium size central city (50,000 to 250,000), 3 = Suburb of a large central city, 4 = Suburb of a

medium size central city, 5 = An unincorporated area of a large central city (e.g., township,

division, 6 = An unincorporated area of a medium central city, 7 = A small city (10,000 to

49,999), 8 = A town or village (2,500 to 9,999), 9 = An incorporated area less than 2,500 or an

unincorporated area (1,000 to 2,499), and 10 = Open country within larger civil divisions (e.g.,

township, division). This variable was then dummy coded (0 = Rural, 1 = Urban).

Again, household income (1 = Under $10,000, 2 = $10,000-$19,999, 3 = $20,000-

$29,999, 4 = $30,000-$39,999, 5 = $40,000-$49,999, 6 = $50,000-$69,999, 7 = $70,000$89,999,

8 = $90,000-$119,999, 9 = $120,000-$149,000, and 10 = More than $150,000), and completed

education (1 = Grade school or less, 2 = Some high school, 3 = High school graduate, 4 = 1 or

more years of technical, vocational, or trade school, 3 = Some college, 6 = College graduate, 7 =

1 or more years of graduate, law, or medical school, and 8 = Advanced degree (e.g., Master’s,

Ph.D., J.D., M.D., etc.) were also included.

Further, a measure of profession seemed warranted. Recall that in previous studies,

criminal justice elites did not consider harsh penal sanctions to be the most effective way to deal

with such a complex phenomenon as white-collar crime (Cole, 1983; Frank et al. 1989; Hartung,

1953; McCleary et al., 1981). Measures of employment status and occupation were also included

to determine whether individuals in different lines of work shared such opinion. Employment

status was measured by asking subjects whether they were disabled, employed full-time,

employed part-time, self-employed, or unemployed (including students, homemakers and

retirees), and then dummy coded (0 = Not employed full-time, 1 = Employed full-time),

Respondents were then asked which option best described their occupation. Importantly, to avoid

long response options the question was deliberately left open-ended. The coding system of the

2010 US census was subsequently used to provide a comprehensive measure of occupation.

Moreover, political ideology was also included as they could potentially affect the

subjects’ perception of seriousness of elite deviance, particularly at the corporate level. More

specifically, conservative subjects might support elements of neoliberal economics such as

market deregulation, which has been shown to facilitate the commission of certain white-collar

crimes (see, e.g., Lynch & Michalowski, 2006; Rosoff, Pontell & Tillman, 2010). Such support,

in turn, might lead them to overlook corporate crime. Conversely, those leaning toward the left

end of the political spectrum might have had greater exposure to information about elite deviance

through their favorite media source and be more critical of it. Political ideology was measured by

asking the respondents to describe their personal social and political views (1 = Very liberal, 2 =

Liberal, 3 = Somewhat liberal, 4 = Somewhat conservative, 5 = Conservative, and 6 = Very

conservative). Political affiliation was measured via the following: “Republican Party”,

“Democrat Party”, “Independent Party”, “Reform Party”, “Other”, “I am not registered”, and “I

do not identify with any political party”. The following dummy variables were then created:

“Republican”, “Democrat”, “Other party”, and “No party”.

Lastly, because it is possible that one’s religious affiliation might influence personal

opinions regarding the ethicality of corporate crime, the subjects were asked which religion, if

any, they identified with (1 = Catholicism, 2 = Protestantism, 3 = Judaism, 4 = Buddhism, 5 =

Islam, 6 = Hinduism, 7 = Other, and 8 = None). Based on the respondents’ answers, the

following dummy variables were then created: “Catholic”, “Protestant”, “Other religion”, and

“No religion”. Further, those subjects who identified themselves as Protestant were asked to

choose a particular denomination (1 = Baptist, 2 = Assembly of God, 3 = Church of Christ, 4 =

Lutheran, 5 = Methodist, 6 = Presbyterian, 7 = Episcopalian, and 8 = Other). Based on previous

classifications (e.g., Cochran & Beeghley, 1991; Smith, 1990), participants’ answers were then

included into one of the following dummy-coded categories: “Conservative Protestant” (i.e.,

Baptists, Assembly of God, Church of Christ, and other), “Moderate Protestant” (i.e., Lutherans

and Methodists), and “Liberal Protestant” (Presbyterians and Episcopalians).

Attribution Style

Because societal response might be determined by perceptions of the offender’s motives

as much as by the nature of the crime itself, controlling for attribution styles seemed warranted.

Attribution theory (Heider, 1958) addresses how people explain their own behavior and the

behavior of others. Behaviors are generally attributed to two different causes: internal

(dispositional) or external (situational). Research suggests that those who employ a dispositional

attribution style consider that criminals choose to commit crime, which makes them morally

culpable and, in turn, leads to more severe sentence options. Conversely, those who employ a

situational attribution style tend to blame the system and view criminals as the victims of

external social forces, which makes them less morally culpable and, in turn, leads to more lenient

sentence options (Blatier, 2000; Carroll, 1978; Cochran, Boots, & Heide, 2003; Cullen, Clark,

Cullen, & Mathers, 1985; Debuyst, 1985; Grasmick & McGill, 2004).

Applied to the present study, attribution theory might explain respondents’ choice of

societal response to the three white-collar crime scenarios involving harmfulness. The

participants were presented with a series of items measuring both dispositional and situational

attribution styles. Those items that measure the dispositional attribution style focus on the

personal motive and characteristics (Wheeler et al., 1979) of the offender. They include “Most

white-collar offenders are greedy individuals”, “Most white-collar offenders have bad characters

and no personal ethics because they place profit above public safety”, “Most white-collar

offenders choose to violate the law when the perceived benefits of their actions outweigh the

perceived costs” and “Most white-collar offenders have the inability to control themselves”.

Conversely, those items that measure the situational attribution style focus on the

business/environment type motivation (Wheeler et al., 1979), that is, systemic pressures or forces

acting upon the offender. They include “Most white-collar offenders’ business environment

promotes competition and encourages the commission of white-collar crimes”, “Most whitecollar

offenders are pressured/coerced by their superiors to reach business goals”, “Most whitecollar

offenders have a fiduciary responsibility (i.e., a legal or ethical relationship of trust) to their

company’s shareholders”, and “Most white-collar offenders are otherwise law-abiding citizens

who do not think that their business practices are really wrong”. The participants were asked the

extent to which they agreed with each of these items. Response options for all these Likert-type

items were as follows: 1 = “Strongly agree”, 2 = “Somewhat agree”, 3 = “Somewhat disagree”,

and 4 = “Strongly disagree”.

The first four items tapping dispositional attribution were entered into a principal

components factor analysis from which a single factor solution best fit these data (eigenvalue =

1.73). This factor explained 43.2 percent of the variation among those items and produced factor

loadings from .22 and .78. After discarding the low self-control item, factor loadings from .65

and .8 emerged. Nevertheless, the Cronbach’s alpha reliability for this additive scale is .61, which

is under the commonly accepted threshold of .7 indicating a moderately reliable scale

(Nunally, 1978).

The second four items tapping situational attribution did not fare any better. Once again a

single factor solution best fit these data (eigenvalue = 1.51). This factor explained 37.7 percent of

the variation among those items and produced factor loadings from .54 and .75. After dropping

the fiduciary item, factor loadings from .62 and .74 were produced. However, the Cronbach’s

alpha reliability for this additive scale is .44, substantially lower than what would be considered

acceptable.

Unnever and colleagues (2010) have suggested that people may not fall into either one

category and might in fact evince both dispositional and situational attribution styles depending

on the type of crime presented to them. Thus, using a bipolar scale (i.e., with dispositional items

on one end and situational items on the other) or two separate scales simultaneously may not be

the best way to measure complex opinions about crime. Consequently, attribution was measured

via eight different variables, each reflecting a potential cause people regard white-collar crime as

resulting from. In the tables presented in chapter 4 and 5, these variables are designated as

“greed”, “moral”, “control”, “choice”, “influence”, “fiduciary”, “pressure”, and “no wrong”.

Knowledge about White-Collar Crime

Knowledge about elite deviance was measured via a two-pronged approach. More

specifically, the instrument included measures of both subjective and objective knowledge.

Again, since most Americans have been shown to rely on the news media as their main source of

information (Dowler, 2003; Roberts & Doob, 1990; Surette, 1998), and because white-collar

crime is not as widely reported as street crime (Barak, 1994; Barlow & Barlow, 2010; Ericson et

al., 1991; Lynch & Michalowski, 2006), there might be a considerable gap between the

participants’ perceived and actual knowledge about elite deviance.

Subjective knowledge

Subjective knowledge was measured several ways. First, respondents were asked to

selfassess the degree to which they felt they were informed about white-collar crime (1 = Not

informed, 2 = Somewhat informed, 3 = Informed, 4 = Very informed). Importantly, the

participants were also asked about their primary sources of information (1 = Television news

stations, 2 = Radio news stations 3 = Newspapers, 4 = Magazines, 5 = Books, 6 = Internet, and 7

= Other). Based on the subjects’ answers, the following dummy variable was then created (0 =

traditional media, 1 = Internet). Second, subjects were asked whether they previously had been

exposed to relevant information about white-collar crime, and if so what medium they used to

educate themselves (1 = College course, 2 = Movie/TV series, 3 = Documentary, 4 = Television

news report, 5 = Newspaper article, 6 = Book, 7 = Other, and 8 = I have not been exposed to such

information). Lastly, participants were asked how confident they felt about their answers to an

objective knowledge questionnaire (1 = Not at all confident, 2 = Somewhat confident, 3 =

Confident, and 4 = Very confident). This questionnaire is described in greater detail in the

following section.

Objective knowledge

Objective knowledge was measured via a 10-item questionnaire that includes

multiplechoice and true or false questions largely derived from the bank of test items developed

for

Rosoff, Pontell and Tillman’s text Profit Without Honor: White-Collar Crime and the Looting of

America (2010). One caveat when developing valid and reliable measures of public knowledge is

to craft questions that are reasonably understandable to a large number of people with response

options that are not too specific. Rosoff and colleagues’ test bank was originally supposed to be

administered to students enrolled in a college course on white-collar crime. As such, it may not

work well with a more heterogeneous population with probably little previous exposure to

relevant information about elite deviance. To address this problem, only those items that tapped

broad dimensions were selected while those that focus on specific examples and use precise

figures in their response options were deliberately excluded. Moreover, several questions and

answer options had to be rephrased to make them more accessible to a non-educated audience.

This questionnaire taps ten different dimensions of elite deviance.

1) Meaning of the term “White-Collar” Crime

The first dimension is the meaning of white-collar crime. It is not at all certain that the

public even knows and understands Sutherland’s implicit reference to upper class professionals.

This dimension was measured by asking respondents about the origin of the term (response

options include “The types of victims”, “The occupations of the perpetrators”, and “The

offenders’ association with religion”).

2) Financial Cost

The second dimension is the financial cost of elite deviance, which was measured by

asking respondents how much they thought street crime cost the American public compared to

white-collar crime. Response options included “Significantly less”, “Somewhat less”, “The costs

are about the same”, “Somewhat more”, and “Significantly more”. As previously mentioned,

research indicates that street crime costs approximately $18 billion to the public while the

financial impact of white-collar crime on society due to fraud and various medical costs resulting

from workplace injuries and illnesses (due to corporate negligence) and environmental pollution

exceeds a trillion dollars every year (Knowlton et al., 2011; Landrigan et al., 2002; Leigh et al.,

2011; Lynch & Michalowski, 2006; Rebovich & Jiandani, 2000; Reiman & Leighton, 2010).

3) Harmfulness

The third dimension is elite deviance’s harmfulness and was measured by asking subjects

how likely street crimes like assaults, murders, and muggings were to injure or kill people

compared with white-collar crime. Response options were: “Significantly less likely”,

“Somewhat less likely”, “As likely”, “Somewhat more likely”, and “Significantly more likely”.

Again, research indicates that the physical danger caused by elite deviance greatly exceeds that

of street crime (Herbert & Landrigan, 2010; Kramer, 1984; Langrin, 1988; Leigh, 2010; Lynch &

Michalowski, 2006; Reiman, 1998; Reiman & Leighton, 2010; Starfield, 2000). As previously

mentioned, white-collar crime may harm or injure over 8,986,000 people every year and lead to

the untimely death of another 283,600 people, that is, more than 20 times as many as street

crime.

4) Legal Immunity

The fourth dimension is the relative legal immunity enjoyed by white-collar offenders

compared with street criminals, and was measured by asking subjects how likely they believed

someone who committed a street crime like burglary and stole $1000 was to be convicted and to

receive a similar sentence as someone who committed a white-collar crime like fraud and stole

$1000. Response options were: “Significantly less likely”, “Somewhat less likely”, “As likely”,

“Somewhat more likely”, and “Significantly more likely”. Again, research suggests that

whitecollar offenders are more likely to avoid criminal convictions and to receive more lenient

sentences compared to those imposed upon street criminals (Calavita, Tillman, & Pontell, 1997;

Maddan et al., 2011; Tillman & Pontell, 1992). Although recent research suggests a toughening

of white-collar crime prosecution (Payne, Dabney, & Ekhomu, 2011), individual offenders are

still easier targets than are business organizations, possibly because establishing criminal intent

for a corporation is a difficult task (Ashforth et al., 2008; Gottschalk, 2012).

5) Reckless Disregard

The fifth dimension is reckless disregard. While corporations typically escape convictions

of purposeful intent to cause harm, they can be found guilty of engaging in acts they know to be

dangerous while ignoring potential harmful consequences (Treiman, 1981). Reckless disregard

was measured by asking respondents whether the following narrative was true or false:

“Although Ford knew their Pinto model’s gas tank represented a safety defect, they chose not to

invest in an inexpensive and safer design, reasoning that it would be cheaper to pay out expected

wrongful death lawsuits. As a result, several people died in fiery crashes.” The answer is “true”.

6) Medical Crime

The sixth dimension is medical crime and was measured by asking participants how many

people in the U.S. they believed died from medical malpractice each year compared with

criminal homicides. Response options included the following: “More”, “An equal number”, and

“Fewer”. Recall that approximately 14,000 people die from homicide annually (UCR, 2010).

Conversely, the empirical literature estimates that 225,000 victims die each year from medical

crime (Starfield, 2000).

7) Human Trafficking

The seventh dimension is human trafficking in the United States. Subjects were asked

whether they believed the statement “Human trafficking is more common in underdeveloped

countries than in developed nations” was true or false. While the public may associate human

trafficking with squalid living and working conditions in U.S. company-owned Southeast Asian

sweatshops, the existence of domestic slavery among foreign workers on American soil has been

documented.

8) State-Corporate Crime

The eighth dimension is state-corporate crime and was measured by presenting subjects

with the following statement and asking them whether they believed it was true: “Private

American military companies have been accused of engaging in a number of human rights

violations including the abuse and torture of detainees, shootings and killings of innocent

civilians, destruction of property, and sexual harassment and rape”. The same statement can be

found on the Amnesty International website to illustrate allegations of involvement in human

rights violations such as torture at the American prison camp of Abu Ghraib, Iraq and the 2007

shootings of Iraqi civilians in Nisoor Square by private U.S. security contractor Academi

(formerly Blackwater), which resulted in 17 deaths and 24 people injured.

9) Toxic Dumping

The ninth item taps the environmental crime dimension and asked participants to

determine whether they believed landfills and toxic waste disposal sites were most likely to be

located near African American communities (response options included “true” and “false”). As

previously mentioned, research indicates that those sites do tend to be found close to

disadvantaged, predominantly African American neighborhoods with very limited legal recourse

(Lynch, 2004; Pueschel, Linakis, & Anderson, 1996; Roderick, 1992; Wargo, 1998; Colborn,

Dumanoski, & Myers, 1997; Needleman et al., 1996; Pihl & Ervin, 1990; Dietrich et al., 2001;

Lynch & Stretesky, 2001; 2004; Denno, 1990; Barnette, 1999).

10) Toxic Emissions

The tenth and final item also taps environmental crime by asking respondents whether

they believed it was true that current levels of toxic emissions had not been reduced as much as

they easily could have been (response options include “true” and “false”). A relatively new

paradigm concerned with the impact of environmental crime, green criminology (Benton, 1998;

Frank & Lynch, 1992; Groombridge, 1998; Lane, 1998; Lynch, 1990; Lynch & Stretesky, 2003;

South, 1998; South & Beirne, 2006) interprets America’s refusal to sign the Kyoto Protocol (an

international treaty pledging to reduce greenhouse gas emissions), or to invest in modern,

sustainable forms of energy (e.g., solar roofing in Florida) as being driven by corporate interests,

simply because these progressive ventures are deemed financially detrimental to American

corporations.

Each correct answer to those ten items was entered into an overall knowledge scale so

that a percentage of correct answers could be calculated. The intent was to attribute a percent

score to subjects, as would be the case if they had taken a test based on a 100-point grading scale.

As with subjective knowledge, the following grading policy was used: 90-100 = “Very

informed”, 80-90 = “Informed”, 70-80 = “Somewhat informed”, and below 70 = “Not informed”.

Further, a measure of “truth” acceptance and “myth” adherence was provided. Whitecollar

“myth” variables were created whenever subjects felt either “confident” or “very confident”

about their answer to a knowledge item even though they chose the wrong response option.

Similarly, white-collar “truth” variables were created each time participants picked the correct

answer to a knowledge item while feeling “confident” or “very confident” about it. Subjects who

felt “not confident” or only “somewhat confident” about their response yet answered correctly

are hereupon classified as “lucky guessers”. Conversely, those who answered incorrectly while

feeling either “not confident” or ”somewhat confident” are arbitrarily referred to as “honestly

uninformed”. Table 1 presents the classification of subjects based on the 2X2 cross-tabulation of

answer correctness and confidence.

Table 1. Classification of Subjects Based on the 2X2 Cross-Tabulation of Answer

Correctness and Confidence (N= 408)

______________________________________________________________________________

Answer Correctness

Yes No

Answer Confidence

Yes

“Truth” Accepters “Myth” Adherers

Lucky Guessers Honestly Uninformed

______________________________________________________________________________

Sentiments about White-Collar Crime

The amount and quality of knowledge about elite deviance may help fashion public

opinion regarding crimes of the powerful. As previously mentioned, while early research

suggested that white-collar offenses such as fraud were not viewed as serious by the public as

violent crimes like robbery (Geis, 1973; Rossi et al., 1974; Sutherland, 1949; Wheeler et al.,

1988), studies that provided respondents with examples of corporate wrongdoings involving

injury or death have found support for more severe sanctions against such crimes (Huff, Desilets,

& Kane, 2010; Holtfreter et al., 2008). These findings indicate that an informed audience might

view white-collar crime more negatively than do uninformed citizens.

Public opinion about elite deviance was measured in several ways. First, the subjects

were presented with a series of items derived from the 2010 National Public Survey on

WhiteCollar Crime (Huff, Desilets, & Kane, 2010). These items consisted of eleven short

scenarios involving two street crimes (i.e., burglary and assault) and nine different types of

white-collar crime. The first of these scenarios described a burglar stealing $10,000 worth of

jewelry from a private residence while the owner was away on vacation. The second one tapped

embezzlement by describing a bank teller befriending a customer and stealing $10,000 out of his

personal account over the course of two years. The third scenario described identity theft by a

computer hacker who stole personal patient information from a healthcare clinic’s database and

then sold this information to a third party for $10,000.

Scenario number four dealt with false charges added by a large manufacturing company

to an invoice, costing a small business owner $10,000. In scenario number 5, robbery and assault

were described with someone attempting to rob several joggers in the park, and - despite the

robbery being foiled - the joggers sustaining non-fatal injuries and receiving treatment at the

hospital. Scenario number six described hacking, with an individual sending out viruses on the

Internet and infecting many personal computers with software that allowed the hacker to

distribute millions of spam. A pharmaceutical company falsely advertising as safe an

antidepressant drug it knew to be unsafe was described in scenario number seven; importantly, it

was indicated that the drug was later found to be related to a string of random violent acts,

costing the lives of several people.

Espionage was introduced in scenario number eight, with a former employee of a U.S.

defense contractor selling nuclear secrets and other classified information he acquired during his

employment to foreign governments. The ninth scenario tapped market rigging by describing a

Wall Street financial firm that conspired to manipulate the precious metals market, profiting at

the expense of other traders and owners of precious metals who were unaware of the price-fixing

scheme. In the tenth scenario, counterfeiting was introduced by describing a person selling a

counterfeit antique bracelet on an online auction site, misrepresenting its true value and making

an extra $1,000. Lastly, insurance overcharge was the main topic in the eleventh scenario; more

specifically, an insurance agent was described as selling an insurance policy at an inflated price

to an unsuspecting customer and pocketing an extra $20,000.

Participants were then asked to compare the seriousness of each offense described in

these scenarios with someone stealing a parked car worth $10,000 (1 = Much less serious, 2 =

Somewhat less serious, 3 = About as serious, 4 = Somewhat more serious, and 5 = Much more

serious). Interestingly, with the exception of counterfeit sales, previous respondents rated these

scenarios as being more serious than motor vehicle theft (Huff, Desilets, & Kane, 2010). These

results tend to indicate that, when presented with several concrete examples of elite deviance, the

people’s level of indignation about white-collar crimes might increase relative to traditional

crime. As such, they merit replication.

However, it could also be that the baseline crime scenario was not shocking enough.

Perhaps a violent - instead of property - street crime would have invited a lower degree of

perceived seriousness for the various white-collar crime scenarios. Recall that the National

Public Survey on White-Collar Crime proposed to measure respondents’ perceived offense

seriousness by contrasting the abovementioned scenarios with a single baseline vignette

describing a property crime (i.e. motor vehicle theft). With the exception of robbery (which only

resulted in injuries) and false drug label, all the other scenarios described financially costly but

not physically harmful offenses. This could explain why subjects only rated three of them as

much more serious than the baseline scenario. To remedy this limitation, participants were also

asked to read another five vignettes partly derived from Kennedy’s ethics scenarios (2010), some

describing violent street crimes and others harmful white-collar crimes.

Street crime scenarios included homicide (“Someone attempts to rob a couple while they

are walking back to their car at night. The husband tries to disarm the attacker, but is shot by him.

He later dies of his injuries.”), and forcible rape (“Someone breaks into a dorm at night and

forcibly rapes a female student.”). Harmful white-collar crime scenarios included consumer

safety violations endangering children (“Because of cost reductions, the materials used by a

company to build a popular toy will present a potential hazard to the product’s users. The

company decides to manufacture and distribute the toy regardless of the risks.”), illegal toxic

waste disposal (“In order to increase profits and meet production goals, a manufacturing

company uses production processes that allow for the release of pollutants into the water and air

and exceed legal limits. Several people become seriously ill as a result.”), and denial of risk and

peril by failing to enforce safety measures on the workplace and to take responsibility for

employees’ toxic contamination (“A mining company fails to ensure safety measures such as

proper ventilation and the use of masks, goggles and gloves among its workers, and covers up

evidence regarding the link between asbestos exposure and lung cancer deaths.”).

For each scenario, the respondents were asked to rate the seriousness of the offense (1 =

Not very serious, 2 = Somewhat serious, 3 = Serious, and 4 = Very serious), decide how the case

should be handled (1 = By some non-legal means, 2 = In a non-criminal court, 3 = In a criminal

court), and determine the proper societal response (i.e., punishment) to them. Response options

included fine, financial compensation for the victims, or imprisonment. Finally, subjects were

also asked to choose specific dollar amounts for monetary sanctions via a 5-point ordinal scale (0

= No fine/compensation, 1 = Under $100,000, 2 = $100,000-$499,000, 3 = $500,000-$999,999, 4

= Above $1,000,000) as well as the number of years for incarceration via a 7-point ordinal scale

(0 = No prison, 1 = 1-5 years, 2 = 6-10 years, 3 = 11-20 years, 4 = 21-30 years, 5 = 31-40 years,

6 = 41 years-life).

Data Analytic Plan

The first research question asks two different things: (1) Whether the public is informed

about elite deviance, and (2) the extent to which it is. To answer these questions, the percentage

of correct answers to the objective knowledge questionnaire will first be compared to the

aforementioned total knowledge index. As previously mentioned, a percent score of 70 or below

will indicate a lack of information. Descriptive statistics will also be used to describe the

distribution of correct answers for each of the ten knowledge questionnaire items.

The second research question asks to compare objective and subjective knowledge about

white-collar crime. Any discrepancy between perceived and actual levels of information will be

revealed by comparing the number of correct answers to the objective knowledge questions with

two measures of subjective knowledge: (1) Subjects’ answers to the question asking how

informed they felt they were about elite deviance, and (2) their answers as to how confident they

were that they answered the knowledge items correctly. Further, respondents’ primary sources of

information will be correlated with objective knowledge to determine which types of media have

the highest educational value relative to white-collar crime.

Examining subjects’ answers to the ten dimensions of elite deviance tapped by the

knowledge questionnaire will answer the third research question asking whether the public holds

common “myths” about elite deviance like it does regarding street crime. More precisely,

statistically significant variation in the answers will indicate that knowledge about these issues is

normally distributed among the American people. However, if the distribution is skewed toward

incorrect answers, it could be that a majority of respondents share certain misconceptions about

white-collar crime. Once again, the term “myth” refers to more than a simple assumption and

involves a certain amount of confidence that a particular belief is true. As previously mentioned,

white-collar crime “myths” will be operationalized as every knowledge item that subjects

answered incorrectly while feeling confident that they were right. Descriptive statistics will be

used to determine which dimensions of the knowledge questionnaire seem to be popular “myths”

about elite deviance.

The fourth research question as to whether public information about elite deviance has

specific sociodemographic correlates will be answered via simple bivariate zero-order

correlations between the objective knowledge variable and the abovementioned

sociodemographic characteristics including gender, age, race/ethnicity, education level,

employment status, household income, area of residence, region where the subjects grew up,

political ideology, religious affiliation, source of information, and attribution style (i.e.,

dispositional [attributing guilt to the offender] or situational [attributing guilt to external

factors]).

Finally, the fifth research question as to whether knowledge about white-collar crime is

correlated with attitudes towards elite deviance will be answered by correlating objective

knowledge with several measures of opinion: 1) Perceived seriousness of financially costly and

harmful white-collar crimes, as well as violent and property street crimes, and 2) respondents’

punitiveness. Punitive scales include choice of prosecutorial process (i.e., by some non-legal

means, in a non-criminal court, or in a criminal court) for the perpetrators, choice of punishment

(i.e., fine, monetary compensation, or prison), and punishment severity (i.e., amounts in dollars

for the monetary sanctions and number of years in prison).

CHAPTER FOUR:

PRELIMINARY RESULTS

This chapter presents the results of statistical analyses and provides answers to each of the

five research questions. As previously mentioned, those questions include (1) the extent of public

knowledge about elite deviance, (2) whether a gap exists between subjective (perceived) and

objective (actual) knowledge, (3) the existence of popular “myths” about elite deviance, (4) the

correlates of knowledge about white-collar crime, and (5) whether such knowledge is correlated

with attitudes towards elite deviance. Such attitudes comprise (a) participants’ perceived

seriousness of financially costly and harmful white-collar crimes as well as property and violent

street crimes, and (b) respondents’ punitiveness, including choice of prosecutorial process and

punishment (i.e., monetary compensation, fine, and/or prison sentence), and punishment severity.

Public Knowledge about Elite Deviance

Subjective vs. Objective Knowledge

The first two research questions pertain to the extent of lay knowledge about white-collar

crime. In other words, how much does the American public really know about elite deviance?

Further, is there a gap between participants’ perceived and actual knowledge? To answer these

questions, participants were first asked to self-assess their level of familiarity with the topic.

Subsequently, a measure of actual knowledge about white-collar crime was provided via a

10item multiple-choice and true/false questionnaire, which then served as a 100-point knowledge

scale. Again, the following grading policy was used: 90-100 = “Very informed”, 80-90 =

“Informed”, 70-80 = “Somewhat informed”, and below 70 = “Not informed”. As is evident in

Table 2 (which compares percent subjective and objective knowledge about whitecollar crime),

respondents tended to overestimate their actual level of information about elite deviance. While

75.5% were found to be “not very informed” about the subject, only 12.5% clearly admitted

lacking knowledge in this area. Further, whereas 73.5% estimated being “somewhat informed”, a

mere 14.7% objectively deserves to be referred to as such. It should be noted, however, that a

mere 14% estimated being either “informed” or “very informed”. These findings suggest that

although participants overestimated their true level of knowledge about white-collar crime, they

did not feel confident enough to rate their knowledge of elite deviance highly.

Table 3 provides a more in-depth report of participants’ objective knowledge about

white-collar crime by presenting the percentage of subjects with scores on the overall

knowledge scale. Recall that based on the classification that was adopted in this study, a score

of at least 70% (i.e., 7 correct answers) was necessary to be deemed “somewhat informed”.

Only about one fourth of the sample scored above that cut-off point. Further, only 7.4%

answered enough questions correctly to be considered “informed” and a mere 2.4% were

found to be “very informed” about elite deviance. Since respondents were found to be, at

best, superficially informed about white-collar crime, understanding their primary source of

information seemed warranted. Eighty-one point one percent cited the Internet as their

medium of choice for keeping informed of important issues, far above television news

stations (15%) and other traditional forms of media. However, when asked whether they had

been previously exposed to relevant information about white-collar crime, only 3.2%

mentioned the Web as their prior source of knowledge. Instead, 38.5% reported having

received some form of information about the subject by watching television news reports,

12% by reading newspaper articles, 10.3% by watching documentaries, 9.3% by watching

movie/TV series, and only 6.6% by taking a college course. Further, 18.1% admitted having

never received any form of information about white-collar crime.

As previously mentioned, white-collar crime is generally underreported by the news

media compared with street crime (Barak, 1994; Barlow & Barlow, 2010; Ericson et al.,

1991; Lynch & Michalowski, 2006; Lynch, Nalla & Miller, 1989; Lynch, Stretesky &

Hammond, 2000). Admittedly, so little time allotted to the coverage of elite deviance may not

suffice to thoroughly educate television audiences about this multi-faceted social issue. Since

a majority of participants rely on TV news reports as their main source of information about

white-collar crime, their apparent lack of knowledge about the topic is therefore not at all

surprising. What remains to be seen is whether subjects were more informed about certain

aspects of elite deviance than others, and whether their level of confidence in their answers to

questions tapping those specific dimensions matched their degree of knowledge.

Recall that this study’s conceptualization of knowledge relies on a fourfold

classification based on the intersections of answer correctness and confidence: (1) “Truth”

accepters (i.e., answered questions both correctly and confidently), (2) lucky guessers (i.e.,

answered questions correctly but not confidently), (3) “myth” adherers (i.e., answered

questions incorrectly but confidently), and (4) honestly uninformed (i.e., answered questions

incorrectly and without confidence). Table 4 presents the percentage of subjects falling in each

of these four categories along with the mean score on the overall knowledge scale, and percent

correct, percent incorrect and percent confident on the ten items that comprise it. The

following is a description of these data.

Percent Correct

The second column in Table 4 presents the mean score on the overall knowledge scale

and the percent correct on the ten items that comprise it. With an average overall score of

54.9 out of 100, the sample in this study was far from reaching the cut-off point of 70 meant

to represent “somewhat informed” subjects. Nevertheless, it appears that participants’ level of

knowledge varied greatly depending on which aspects of the topic they were addressing. For

example, a large proportion (89.2%) of respondents correctly indicated that the term “white-

collar crime” is based on the occupation of the perpetrators. However, only 23.8% answered

that street crime costs significantly less to the American public than does white-collar crime.

Interestingly, very few (3.2%) estimated that statistically, street crimes like assaults, murders,

and muggings are significantly less likely to injure or kill people than is white-collar crime.

Further, only a third (32.8%) indicated that someone who commits a street crime like

burglary and steals $1,000 is significantly more likely to be convicted than someone who

commits a white-collar crime like fraud and steals the exact same amount of money.

Moreover, while a large percentage (75.5%) deemed the description of the Ford Pinto

case accurate, only 38.2% seemed to agree that more people in the U.S. die each year from

medical malpractice than from criminal homicide. In addition, although less than a third

(30.9%) correctly indicated that the statement pertaining to human trafficking was false,

66.7% got the toxic dumping question right. Lastly, an overwhelming majority of respondents

correctly answered those questions that tapped the dimensions of state-corporate crime

(91.2%) and toxic emissions (97.1%). Though necessary, participants’ correct responses are

nonetheless not sufficient to provide evidence of knowledge about white-collar crime. Only

by comparing subjects’ answers to how confident they felt about them can we (1) provide a

valid indicator of knowledge (Hunt, 2003) and (2) determine which of the four

abovementioned categories to which subjects belong (i.e., “truth” accepters, lucky guessers,

“myth” adherers, and honestly uninformed).

Percent Confident

The fourth column in Table 4 presents participants’ percent confident in their answers

to the knowledge scale. Interestingly, subjects were not very confident about their choices,

even when they did respond correctly. Recall that the average overall score on the knowledge

scale was 54.9. Comparatively, the average overall level of confidence was only 49.9. A closer

look at each individual item reveals further gaps. While 89.2% of the sample correctly

answered the question pertaining to the meaning of the term “white-collar crime,” fewer

subjects (71.8%) felt confident about their choice. Similarly, subjects evinced little confidence

in their answer to the item that tapped reckless disregard (30.1%) compared to the 75.5% who

chose the right answer. A similar finding is echoed in the question about medical crime. More

specifically, while 38.2% picked the correct answer, only 25.7% felt confident about their

choice.

Moreover, compared to the 91.2% who correctly indicated as true the statement that

private American military companies have been accused of engaging in a number of human

rights violations, only 61.2% were confident in their answer. Likewise, while 66.7% correctly

answered the question that asked whether landfills and toxic waste disposal sites are more

likely to be located near African American communities, only 39.2% were confident about

their choice. Further, while almost three fourth of the sample (74.8%) were confident in their

answer to the question that asked whether toxic emissions could be reduced much more if

industries agreed to employ appropriate technologies, a much larger percentage (97.1%)

answered that question correctly.

Nevertheless, a reverse gap between answer correctness and confidence could be

observed in regard to four items. More precisely, whereas only 23.8% seemed to agree that

white-collar crime is significantly more financially costly to society than is street crime, a

somewhat larger percentage (27.7%) felt confident in their answer. A similar gap emerged

with the item that tapped the legal immunity of white-collar offenders compared to street

criminals. More precisely, while a mere 32.8% found the right answer, 58.5% were positive

about their choice. Likewise, while 43.4% were certain that they answered the item that

tapped human trafficking correctly, only 30.9% actually did. The greatest gap that could be

observed had to do with the item tapping the harmfulness of elite deviance. Whereas very few

(3.2%) subjects correctly indicated that white-collar crime claims more lives annually than

does street crime, 66.2% of the sample were certain that they chose the correct answer. This

outstanding discrepancy suggests that participants in this study had difficulty ascribing the

concept of physical harm to crimes of the powerful.

Two important findings emerge from this analysis. First, as far as knowledge is

concerned, participants seemed more informed about certain dimensions of white-collar crime

than they are about others. More specifically, a majority of respondents were familiar with the

term “white-collar crime” and its actual meaning. Further, subjects were found to be quite

knowledgeable about some of the harmful activities that some corporations undertake

(e.g., reckless disregard for customers’ safety, human rights violations in occupied countries, and

reluctance to implement pollution-reducing policies). Nevertheless, respondents were less likely

to acknowledge the overwhelming physical harmfulness of white-collar crime compared to street

crime, and tended - albeit to a lesser degree - to underestimate the former’s considerable financial

burden on society. Similarly, respondents were reluctant to recognize the fact that medical crime

claims more lives every year than do criminal homicides, or to realize that white-collar offenders

are statistically more likely than street criminals to avoid criminal prosecutions.

Second, except for a few items (tapping the harmfulness of elite deviance, the relative

legal immunity of white-collar offenders, human trafficking in developed nations and - to a

lesser degree - the financial cost of white-collar crime), the percentage of questions answered

confidently was systematically lower than that of correct answers. This finding suggests that

respondents may not be familiar enough with the subject and might have chosen the right

answers by chance alone. The next step is thus to determine the percentage of participants who

qualify as “truth” accepters rather than lucky guessers.

“Truth” Accepters vs. Lucky Guessers

Columns 5 and 6 in Table 4 present the percentage of “truth” accepters and lucky

guessers, respectively. Once again, while this classification refers to subjects who answered

correctly, the main difference between these two categories lies in how confident participants

felt about their answers. Phrased differently, whereas “truth” accepters responded both

correctly and confidently, lucky guessers did not evince such confidence and may have picked

the right answers by chance alone. First of all, the overall percentage of “truth” adherers

(32.8%) is larger than that of lucky guessers (22.1%). That is, the proportion of subjects who

answered correctly while feeling confident about their choice was generally greater than that

of participants who chose the right answers as a result of a guess. Such gap is particularly

noticeable in regard to the items tapping the meaning of the term “white-collar crime” (66.9%

of “truth” adherers vs. 22.3% of lucky guessers), legal immunity (27.2% vs.

5.6%, respectively), state corporate crime (60.3% vs. 30.9%), and toxic emissions (73.8% vs.

23.3%).

However, the gap is reversed with the items tapping reckless disregard (27.5% of

“truth” accepters vs. 48% of lucky guessers), medical crime (13.2% vs. 25%), human

trafficking (11% vs. 19.9%), and toxic dumping (31.4% vs. 35.3%). That is, for these items, it

appears that luck more than actual knowledge can explain correct answers. Discrepancies are

nevertheless far less visible with the items that tapped the financial cost (13.7% vs.

10.1%) and harmfulness (2.7% vs. 0.5%) of elite deviance. Recall that participants scored

particularly poorly on the question pertaining to the greater physical harmfulness of elite

deviance compared to street crime (only 3.2% answered it correctly). However, 66.2% were

confident about their answer, a finding consistent with the “myth” adherence taxon used in this

study. The next step is therefore to distinguish “myth” adherers from those subjects who were

honestly uninformed.

“Myth” Adherers vs. Honestly Uninformed Subjects

The third research question asks whether the American public adheres to “myths”

about white-collar crime as it does with street crime. Again, “myth” adherence in this study is

operationalized as an incorrect answer held with confidence. Columns 7 and 8 of Table 4

present the percentage of “myth” adherers and honestly uninformed subjects, respectively.

The overall proportion of respondents who gave incorrect answers without feeling confident

about their choice (28.6%) was greater than that of respondents who adhered to “myths”

(16.5%). Such gap was larger for those items that tapped the financial cost of white-collar

crime (62.2% vs. 14%), legal immunity (42% vs. 25.2%), reckless disregard (21.8% vs.

2.7%), medical crime (49.3% vs. 12.5%), and toxic dumping (25.5% vs. 7.8%), and smaller in

regard to state corporate crime (7.8% vs. 1%), the meaning of the term “white-collar crime”

(5.9% vs. 4.9%), human trafficking (36.7% vs. 32.4%), and toxic emissions (1.9% vs. 1%).

However, the gap was reversed with the item that tapped the harmfulness of elite deviance;

more specifically, the percentage of subjects who gave a wrong answer while stubbornly

sticking to their positions was almost double that of participants who answered incorrectly yet

with no confidence (63.5% vs. 33.3%, respectively).

It therefore appears that the crux of the concept of “myths” about white-collar crime

lies within the dimensions of harmfulness, human trafficking, and legal immunity. One may

already discern two interesting patterns from these findings. First, an important number of

subjects seem to share a deeply rooted notion that elite deviance represents more of a

financial threat to society than a physical one. Second, some answers suggest trust in the

institutions of the American criminal justice system belied by subjects’ reluctance to admit

that U.S. corporations, though believed to engage in unethical acts abroad, can do the same in

the United States with relative impunity. These conjectures will be more thoroughly discussed

in chapter six.

Conclusion

In summation, it appears that participants in this study may not be sufficiently

knowledgeable about elite deviance. On average they scored well below the cut-off point meant

to represent “somewhat informed” respondents. Further, the gap between correct answers and

actual “truths” about white-collar crime suggests a lack of confidence among participants in

their knowledge about the subject. In fact, while they overestimated their actual level of

information about elite deviance, few of them considered themselves very informed, and almost

one fifth of the sample confessed to having never received any kind of information about it.

Lastly, the existence of popular “myths” about white-collar crime is seemingly supported by

respondents’ reluctance to acknowledge the greater harmfulness of elite deviance over street

crime and - to a certain extent - that specific white-collar offenses believed to be more common

in underdeveloped nations can be committed in America with little to no legal repercussion.

Now that a measure of public knowledge about white-collar crime has been established,

determining its correlates seems warranted as it is uncertain which specific demographic

variables may be associated with either the acceptance of

“truths” or the adherence to “myths” regarding elite deviance.

Correlates of Knowledge about Elite Deviance

The fourth research question asks what are the correlates of knowledge about

whitecollar crime. Table 5 presents the results of one-way between subjects ANOVAs in the

effect of sociodemographic predictors of knowledge about white-collar crime, “truth”

acceptance and “myth” adherence. F-tests and effect sizes (eta squared) are reported. Analysis

of variance showed significant differences in knowledge about elite deviance in regard to the

region where subjects grew up [F (4, 403) = 2.44, p < .05, η2 = .024], their level of education

[F (6, 401) = 4.48, p < .000, η2 = .06], their political ideology [F (5, 402) = 2.87, p < .05, η2 =

.03], their religious affiliation [F (7, 400) = 2.57, p < .05, η2 = .03], as well as attribution of

blame to choice [F (3, 404) = 2.99, p < .05, η2 = .02] and to outward influences [F (3, 404) =

5.84, p < .001, η2 = .04]. However, the eta squared statistics indicated small to moderate

effect sizes.

Post hoc comparisons using Tukey’s test failed to find any statistically significant mean

group differences for the region where subjects grew up and attribution of blame to choice.

Subjects who held an advanced degree (M = 6.27, SD = 1.46) were found to be more

knowledgeable about white-collar crime than those who only completed high school (M =

4.96, SD = 1.37), one or more years of technical, vocational, or trade school (M = 4.67, SD =

1.61), and some college (M = 5.44, SD = 1.52). Further, participants who identified as “very

liberal” (M = 6.02, SD = 1.57) were statistically more knowledgeable about elite deviance than

were “somewhat conservative” subjects (M = 5.12, SD = 1.47). In addition, Protestant

respondents (M = 5.17, SD = 1.56) were less knowledgeable about white-collar crime

compared to subjects who reported belonging to no religion (M = 5.73, SD = 1.54). Lastly,

those who strongly agreed that white-collar crime is mainly the result of business

environmental influences (M = 5.96, SD = 1.52) were more knowledgeable than subjects who

somewhat agreed (M = 5.38, SD = 1.54), somewhat disagreed (M = 5.34, SD = 1.45) and

strongly disagreed (M = 4.87, SD = 1.52).

With respect to “truth” acceptance, significant differences were found as far as

subjects’ level of education [F (6, 401) = 3.47, p < .01, η2 = .05], political views [F (5, 402) =

4.16, p < .001, η2 = .05], political affiliation [F (5, 402) = 2.91, p < .05, η2 = .03], as well as

attribution of blame to low self-control [F (3, 404) = 4.33, p < .01, η2 = .01] and to outward

influences [F (3, 404) = 5.24, p < .001, η2 = .04]. However, it should be noted that eta

squared statistics once again indicated only small to medium effect sizes. Tukey post hoc

comparisons showed that subjects with an advanced degree (e.g., master’s, Ph.D., M.D., J.D.,

etc.) were more likely to accept “truths” about elite deviance (M = 3.81, SD = 1.99) than were

those who only completed high school (M = 2.64, SD = 1.77) and one or more years of

technical, vocational, or trade school (M = 3.58, SD = 1.68).

Moreover, participants who identified as “very liberal” were more likely to be “truth”

accepters (M = 4.33, SD = 1.85) than were all other subjects. Further, Republicans were less

likely to accept “truths” (M = 2.75, SD = 1.67) compared with Independents (M = 3.75, SD =

1.92). In addition, those who somewhat disagreed with the notion that white-collar crime

results from low self-control were less likely to be “truth” accepters (M = 2.86, SD = 1.62)

than were those who strongly disagreed (M = 3.56, SD = 1.95) and somewhat agreed (M =

3.58, SD = 1.97). Lastly, those who strongly agreed that elite deviance is a consequence of

negative business environment influences were more likely to accept “truths” (M = 3.87, SD =

1.92) compared to those who somewhat disagreed (M = 3.03, SD = 1.83) and somewhat

agreed (M = 3.08, SD = 1.75).

In regard to “myth” adherence, significant group differences were found for race [F (6,

401) = 3.24, p < .01, η2 = .05], but post hoc comparisons could not be performed due to one

category (“Native Hawaiian or pacific Islander”) having less than two cases. An independent

sample t test using the dichotomized variable “White” showed a significant difference, t (406)

= -2.65, p < .01, with white subjects (M = 1.56, SD = .1.28) being less likely than their non-

white counterparts (M = 2.00, SD = .165) to adhere to “myths” about elite deviance. However,

Cohen’s effect size (d = .30) suggested modest practical significance. Similar differences

emerged in regard to the region where subjects grew up [F (4, 403) = 2.93, p < .05, η2 = .03].

More specifically, “myth” adherence was higher among subjects who grew up in northern

states (M = 2.06, SD = 1.43) than among those who grew up in the Midwest (M = 1.25, SD =

.98). Although small but significant differences emerged for income [F (9, 398) = 2.41, p <

.05, η2 = .05], post hoc comparisons using Tukey’s test failed to find any mean group

differences.

Significant differences emerged in regard to the type of information source that

subjects used [F (6, 401) = 2.63, p < .05, η2 = .04]. Once again, post hoc comparisons could

not be completed due to several groups having less than two cases. An independent samples t

test using the dichotomized variable “Internet” yielded a significant difference, t (406) = 3.15,

p < .01, with web users (M = 1.55, SD = 1.29) being less likely to adhere to “myths” about

elite deviance than are those who relied on traditional media (M = 2.09, SD = 1.63). Cohen’s

effect size (d = .37) suggested moderate practical significance. Further differences emerged

for attribution of blame to outward influences [F (3, 404) = 4.39, p < .01, η2 = .03]. More

precisely, subjects who strongly disagreed that the commission of elite deviance is encouraged

by white-collar offenders’ business environment (M = 2.49, SD = 1.70) were more likely to be

“myth” adherers than all other subjects.

Finally, small but significant mean group differences could be observed for attribution

of blame to pressure to succeed [F (3, 404) = 2.68, p < .05, η2 = .02]. More specifically, those

who strongly disagreed that white-collar offenders are coerced by their superiors to reach

business goals (M = 2.53, SD = 1.77) were more likely than all other participants to adhere to

“myths”. Taken together, these results suggest small yet statistically significant differences in

knowledge about white-collar crime as well as in the acceptance of “truths” and adherence to

“myths” about elite deviance. As such, they warrant further investigation.

Table 6 presents zero-order correlations between sociodemographic characteristics,

knowledge about elite deviance, “truth” acceptance and “myth” adherence. Correlation

coefficients reported are Pearson’s r when using dichotomous predictors and Spearman’s rho

when using multinomial nominal and/or ordinal predictors. Importantly, it should be noted that

the majority of the correlations failed to attain statistical significance. Further, significant

associations tended to be weak (i.e., less than +/- 0.20). Although no difference was found for

gender in terms of overall knowledge about white-collar crime, statistically significant

differences emerged for “truth” acceptance and “myth” adherence. More specifically, men were

not only more likely to accept “truths” (r = .127) but also to adhere to “myths” (r = .160). Recall

that “truths” and “myths” were operationalized as subjects’ confidence in correct or incorrect

answers, respectively. It therefore appears that males were more confident in their response

choices than were their female counterparts who showed more reservation. These findings are

concordant with previous research suggesting that women tend to express less confidence in their

self-assessments (Clark & Zehr, 1993; Smith,

Morrison, & Wolf, 1994), whereas men evince greater belief in their scholarly abilities (Sax &

Harper, 2007). As for race, Whites were more knowledgeable (r = .154) and less likely to

adhere to “myths” (r = -.131) while Blacks (r = .114) were more likely to adhere to them.

Further, Hispanics were less likely to accept “truths” than non-Hispanics (r = -.110). Moreover,

education was positively correlated with knowledge (r = .219) and “truth” acceptance (r =

.175).

Statistically significant relationships were also found for political ideology. Recall that

measures of political views and affiliation were included as potential correlates of knowledge

about white-collar crime due to the fact that right-wing politics tends to support elements of

neoliberal economics such as market deregulation, which has been shown to facilitate the

commission of certain white-collar crimes (see, e.g., Lynch & Michalowski, 2006; Rosoff,

Pontell & Tillman, 2010). In turn, such attitudes may lead conservatives to discard

information that equates corporations with street criminals. As expected, more politically

conservative subjects and Republicans were less knowledgeable about whitecollar crime (r =

- .158 and 113, respectively) and less likely to accept “truths” (r = -.142 and -.129,

respectively). Results for conservative Protestants mirror those findings. More specifically,

they were also less knowledgeable (r = -.142) and less likely to admit “truths” about elite

deviance (r = .142). In fact, subjects who reported not having any religion were more

knowledgeable (r = .140) and more likely to accept “truths” (r = .118).

Perhaps more interestingly, those who selected the Internet as their main source of

information relative to other traditional media were less likely to adhere to “myths” (r = .155).

Such finding suggests that the Web represents a formidable educational platform, at least for

active Internet users. From social networks to non-profit organizations disclosing classified

information (e.g., WikiLeaks), websites may provide their users with a wider variety of

sensitive topics than do official news media, which are usually owned by corporations and

sometimes serve a specific political agenda.

The results are more ambiguous in regard to attribution of blame. While those subjects

who believed that white-collar crime is the result of low moral standards were both more

knowledgeable (r = .140) and more likely to accept “truths” about elite deviance (r = .118),

similar findings were found among those participants prone to believe white-collar offenders

are otherwise law-abiding citizens who see no wrong in their actions (r = .182 and .151,

respectively). Similarly, respondents who agreed that pressure to succeed causes elite

deviance (i.e., a situational attribution style) were also found to be more knowledgeable about

the topic (r = .119).

In summation, despite very weak correlation coefficients no greater than +/- 0.20 and

most failing to attain statistical significance, it appears that respondents’ gender, race/ethnicity,

income, education, political ideology, religious affiliation, source of information, and

attribution style were associated with their level of general knowledge about white-collar

crime, as well as their “truth” acceptance and “myth” adherence. Having established generally

what participants know about elite deviance, what remains to be seen is how they feel about it.

In other words, what are respondents’ attitudes and opinions about white-collar offenses and

their perpetrators? How serious do they perceive those acts to be relative to street crime?

Moreover, what do they believe is the appropriate societal response (i.e., how should society

punish such offenders)? Further, do the same sociodemographic variables predict both

knowledge and sentiments about white-collar crime? Lastly, are there any associations

between subjects’ knowledge, “truth” acceptance, and “myth” adherence, and their perceived

seriousness of and punitiveness about elite deviance?

Opinions about Elite Deviance

Perceived Seriousness

The fifth and last research question asks whether knowledge about white-collar crime

is correlated with attitudes towards elite deviance. In order to answer this question, one must

first identify such sentiments. An important attitude about white-collar crime is respondents’

perceived seriousness of such activities. The third National Public Survey on White-Collar

Crime (Huff, Desilets, & Kane, 2010) features a valid measure of perceived seriousness of

elite deviance and street crime. Its results are used in the present study for comparative

purposes. Table 7 presents mean crime seriousness scores for white-collar and street crime

compared with motor vehicle theft in the National White-Collar Crime Center Survey and the

present study. Several findings are noteworthy. First of all, perceived seriousness was lower

in this dissertation’s sample than among the participants of the National White-Collar Crime

Center’s survey (M = 3.9 and 4.2, respectively). That is, with the exception of espionage (M

= 4.8), subjects in the present study rated all scenarios as less serious than did those in the

Huff and colleagues’ sample.

The largest discrepancy between both samples was observed with the scenario

describing a hacker who infects computers with spam (mean difference = 1.2). Subjects in the

National Public Survey on White-Collar Crime: (1) only rated counterfeit sales (M = 2.9) as

equally serious as motor vehicle theft, (2) rated burglary (M = 3.7), invoice charges (M =

3.8), and hacking (M = 3.9) as somewhat more serious, and (3) considered embezzlement (M

= 4.1), identity theft (M = 4.3), robbery (M = 4), false drub label (M = 4.8), espionage (M =

4.8), market rigging (M = 4.3) and insurance overcharge (M = 4.4) as much more serious.

Conversely, in the present study, participants: (1) considered both hacking (M = 2.7) and

counterfeit sales (M = 2.3) less serious offenses than stealing a parked car worth $10,000,

(2) considered that burglary (M = 3.2), embezzlement (M = 3.7), identify theft (M = 3.9),

invoice charges (M = 3.5), robbery (M = 3.6), and insurance overcharge (M = 3.7) were

only somewhat more serious, and (3) recognized only false drug labeling (M = 4.7),

espionage (M = 4.8), and market rigging (M = 4.2) as much more serious issues.

Despite their relatively lenient rating of crime seriousness, it is important to note that

subjects in the present study deemed false drug labeling, espionage, and market rigging as

more serious offenses than a violent crime such as robbery that resulted in victims’

hospitalization. Such attitudes might be a corollary to the acceptance of “truths” about elite

deviance. Recall that “truth” accepters were more likely to accept information relative to

state-corporate crime and the kind of unethical activities some large, powerful firms are

alleged to engage in (e.g., environmental pollution, violating human rights in invaded

countries, etc.). In fact, hostility toward organizational and high-status rather than individual

and non-status offenders in this study’s sample mirrors the National White-Collar Crime

Center’s findings. That is, in both samples, scenarios that described the unlawful activities of

corporations or high-status offenders (i.e., false drug label, market rigging, espionage, etc.)

were rated more negatively than those that depicted relatively powerless individuals (i.e.,

hacking, counterfeit sales, etc.). In summation, it can be said that, despite being less critical

of crime in general (a finding to be discussed in chapter six), subjects in this dissertation rated

various instances of white-collar crime more negatively than they did street crime.

How do such attitudes hold when comparing white-collar crimes with more harmful

street crimes? Table 8 presents the results of paired samples t-tests to compare mean

perceived seriousness of scenarios describing both physically injurious white-collar crimes

(i.e., knowingly manufacturing a potentially dangerous toy, releasing deadly pollutants in a

river, and knowingly exposing workers to asbestos) and violent street crimes (i.e., murder and

forcible rape). All scenarios have a mean above 3.00, which is the score meant to represent

“serious” offenses. As could be expected, homicide (M = 3.91, SD = 0.36) and forcible rape

(M = 3.94, SD = 0.24) were perceived to be more serious than the white-collar crimes

described. More precisely, the mean for the murder scenario was statistically higher than

those for the defective toy vignette (t = 12.10, p < .01, d = 0.60), the deadly pollutants

scenario (t = 11.50, p < .01, d = 0.57), and the asbestos exposure scenario (t = 6.68, p < .01, d

= 0.33). Similarly, rape was statistically perceived as more serious than consumer safety

violation (t = 13.12, p < .01, d = 0.65), toxic dumping (t = 12.89, p < .01, d = 0.64), and

reckless endangerment of employees (t = 8.13, p < .01, d = 0.40). In all cases, Cohen’s effect

sizes suggested small to moderate practical significance. Conversely, the means for the two

street crime scenarios were the only ones to not statistically differ from one another.

Of the three harmful white-collar crime scenarios, the one describing the deliberate

manufacturing of a defective toy (M = 3.43, SD = 0.76) was considered less serious than the

toxic dumping scenario (M = 3.51, SD = 0.64, t = -1.98, p < .05, d = -0.10) and the asbestos

exposure scenario (M = 3.72, SD = 0.52, t = -7.76, p < .01, d = -0.38). However, Cohen’s

effect sizes were this time smaller. Similarly, releasing deadly pollutants was considered less

serious than the reckless endangerment of employees (t = -7.07, p < .01, d = -0.35). This

should come as no surprise since the defective toy scenario only alluded to a potential risk

whereas the toxic dumping vignette referred to people falling “seriously ill”, and the words

“cancer” and “deaths” were mentioned in the asbestos exposure scenario. Perhaps a clearer

mention of harm resulting from consumer safety violation might have invited higher

perceived seriousness.

Still, the forcible rape scenario was judged more negatively than all three examples of

white-collar crime, including the asbestos exposure vignette in which employees die from a

lethal disease contracted in the workplace. This is surprising since, despite the violent nature

of sexual assault, no mention of death was made. It is possible that contextual details

influenced respondents’ attitudes. The rape victim was assaulted in her own bedroom, which

might have made the crime appear even more frightening. Because delayed victimization is a

common characteristic of white-collar crime (e.g., work-related diseases may take years

before being diagnosed and attributed to one’s professional activity), elite deviance may not

elicit the same amount of shock and fear, which might in turn explain this study sample’s

lower level of perceived seriousness of such offenses. It is not certain, however, whether

similar differences exist in punitiveness.

Punitiveness

The second attitude about elite deviance measured was subjects’ level of punitiveness

for the abovementioned harmful white-collar offenses compared with the two violent street

crimes. Measures of punitiveness included (a) choice of prosecution process (i.e., by some

non-legal means, in a non-criminal court, or in a criminal court), (b) punishment for their

perpetrators (i.e., fine, monetary compensation, and/or prison) and (c) sentence severity (i.e.,

in dollar amounts and/or number of years in prison). Table 9 presents the results of paired

samples t-tests to compare subjects’ choice of prosecutorial process for white-collar crime and

street crime. First of all, no subject chose the non-legal means alternative for any of the five

scenarios. Conversely, homicide was the only scenario for which every participant

recommended the perpetrator be tried in a criminal court (M = 3.00, SD = 0.00).

In fact, the murder scenario was the only one to be statistically different from all other

instances of crime described, including white-collar offenses such as consumer safety violation

(M = 2.74, SD = 0.47, t = -11.13, p < .01, d = 0.55), toxic dumping (M = 2.70, SD = 0.51, t =

11.90, p < .01, d = 0.59), and asbestos exposure (M = 2.71, SD = 0.50, t = 11.75, p

< .01, d = 0.58), but also - although to a lesser degree - forcible rape (M = 2.99, SD = 0.99, t =

2.01, p < .05, d = 0.10). Further, subjects were statistically more likely to recommend a

harsher prosecution process for the rapist than they were for white-collar offenders in the toy

scenario (t = 10.37, p < .01, d = 0.51), the deadly pollutants scenario (t = 11.39, p < .01, d =

0.56), and the asbestos exposure scenario (t = 11.34, p < .01, d = 0.56). Nevertheless, there

was no statistical difference between subjects’ choice of prosecutorial process for these three

white-collar crimes. It therefore appears that relative consensus emerged in this study’s

sample regarding the best way to try white-collar and street offenders. That is, participants

were more inclined to select a non-criminal court for the perpetrators of offenses they

perceived to be less serious than murder and forcible rape.

Similar differences emerged when asking subjects how much, if any, of a fine should

be imposed to the offenders in each scenario. Table 10 presents the results of paired samples

t-tests to compare subjects’ choice of fine amount for white-collar crime and street crime.

Once again, murder (M = 2.25, SD = 1.75) and rape (M = 0.35, SD = 0.96) did not statistically

differ from one another. While means for both street crimes are well under 1.00 -

i.e., the score meant to represent a fine under $100,000 - white-collar offenses such as selling

customers a hazardous product (M = 1.83, SD = 1.72), dumping toxic waste above the legal

limit (M = 2.25, SD = 1.75), or being negligent in implementing proper safety measures in the

workplace and denying risk and peril (MD = 1.78, SD = 1.82) elicited average fine amounts

ranging between $100,000 and $499,000. Large and statistically significant differences were

found between the murder scenario and those that described the defective toy (t = -16.99, p <

.01, d = -0.84), deadly pollutants (t = -21.51, p < .01, d = -1.06), and asbestos exposure (t = -

15.77, p < .01, d = -0.78).

Similar differences were found between rape and consumer safety violation (t = 17.44,

p < .01, d = -0.86), toxic dumping (t = -21.68, p < .01, d = -1.07), and the reckless

endangerment of employees (t = -16.99, p < .01, d = -0.84). While the defective toy scenario

slightly differed from the deadly pollutants one (t = -5.21, p < .01, d = -0.26), it was not

statistically different from the asbestos vignette. Conversely, the work-related disease scenario

elicited a smaller fine amount compared with the toxic dumping scenario (t = -5.53, p < .01, d

= -0.27). In short, it seems subjects were more inclined to choose a higher fine amount against

white-collar offenders (particularly the company responsible for polluting over the legal limit)

than they were against the murderer and rapist.

Slightly similar findings emerged when asking subjects how much, if any, of a

monetary compensation should be granted to the victims and their families. Table 11 presents

the results of paired samples t-tests to compare subjects’ choice of compensation amount for

white-collar crime and street crime. As was the case with fine, a majority of respondents did

not consider such form of punishment adequate when dealing with offenders described in the

murder and forcible rape scenarios. More specifically, the means for homicide (M = 0.97, SD =

1.52) and rape (M = 0.77, SD = 1.30) did not even reach the cut-off point of 1.00 that indicates

the first range of amount (i.e., anything under $100,000).

While means for both street crimes only slightly differed from one another (t = 3.90, p

< .01, d = 0.19), larger statistical differences emerged between those offenses and all three

instances of elite deviance. Subjects were less inclined to recommend monetary

compensation in the homicide scenario than they were in the defective toy scenario (t = 7.00,

p < .01, d = -0.35), the deadly pollutants scenario (t = -9.72, p < .01, d = -0.48), and the

asbestos exposure vignette (t = -15.73, p < .01, d = -0.78). Similarly, participants were less

prone to support a high monetary compensation amount against the rapist than they were

against those responsible for violating consumer safety (t = -10.18, p < .01, d = -0.50),

polluting over the legal limit (t = -13.04, p < .01, d = -0.64), and lying to their employees

regarding the risk of contracting lethal diseases in the workplace (t = -19.16, p < .01, d = -

0.95).

Consensus did not emerge regarding the appropriate punishment against white-collar

crime. More specifically, while the mean for the asbestos exposure scenario (M = 2.42, SD =

1.54) is above the cut-off point meant to represent a monetary compensation amount ranging

between $100,000 and $499,000, those for the defective toy scenario (M = 1.58, SD = 1.48)

and toxic dumping (M = 1.85, SD = 1.59) are below that threshold. In fact, the mean for the

asbestos vignette statistically differed from those for the consumer safety scenario (t = 10.71,

p < .01, d = 0.53) and the deadly pollutants scenario (t = 7.30, p < .01, d = 0.36). Further, the

defective toy scenario invited a slightly less monetary compensation amount than did the toxic

dumping vignette (t = -3.35, p < .01, d = -0.17).

In summation, whereas toxic dumping elicited higher fine amounts than did all other

crime scenarios, respondents in this study were more inclined to recommend higher monetary

compensation against those responsible in the asbestos exposure vignette. While supporting

greater economic sanctions against white-collar crime than street crime seems logical, what

remains to be seen is whether respondents were prone to punish white-collar offenders and street

offenders with equally long prison sentences.

Table 12 presents the results of paired samples t-tests to compare mean prison sentence

severity for white-collar crime and street crime. In other words, does the nature of the crime

described in each scenario (i.e., elite deviance or traditional offense) influence subjects’

decision regarding how much, if any, prison time the perpetrators should serve? Although not

originally given as a response option, capital punishment is nonetheless included here since a

few subjects were punitive enough to require a death sentence for murder (M = 4.68, SD =

1.63), corporate negligence and denial of risk and peril in the case of asbestos exposure (M =

1.35, SD = 1.79), and rape (M = 3.26, SD = 1.57). Such recommendation belies a lack of

knowledge about the criminal justice system since sexual assault is no longer punishable by

death.

Once again, both street crimes elicited longer prison sentence lengths ranging between

11 and 30 years than did white-collar crimes for which the average prison sentence did not

exceed 5 years. Compared with homicide, a majority of respondents did not perceive

incarceration as the appropriate punishment for the offenses involving the defective toy (M =

0.84, SD = 1.18, t = -40.20, p < .01, d = -1.99), illegal toxic dumping (M = 0.94, SD = 1.32, t

= -38.32, p < .01, d = -1.90), and even lying about the link between unprotected asbestos

exposure and lung cancer (t = -29.45, p < .01, d = -1.46). Further, Cohen’s effect sizes

suggested large practical significance.

Mean prison sentence severity was also statistically higher in the murder scenario than

in the rape vignette (t = 16.75, p < .01, d = 0.83), although the difference is less pronounced

than with white-collar offenses. Rape invited higher prison sentence severity than did

consumer safety violation (t = 26.39, p < .01, d = 1.31), toxic dumping (t = 24.79, p < .01, d =

1.23), and the reckless endangerment of employees (t = 18.13, p < .01, d = 0.90). Further

differences emerged between the three instances of elite deviance, with the defective toy

scenario eliciting less prison severity than the asbestos exposure vignette (t = -5.90, p < .01, d

= -0.29), but not statistically differing from the deadly pollutants scenario. Conversely, toxic

dumping invited a shorter prison sentence than did the reckless endangerment of employees (t

= -5.48, p < .01, d = -0.27).

Conclusion

To conclude, respondents were generally less punitive about elite deviance than they were

regarding street crime. Phrased differently, they were more likely to perceive murder and forcible

rape as offenses of greater seriousness to be prosecuted in a criminal court and punished with

longer prison terms compared with all three examples of white-collar crime. Such findings hold

after controlling for harm intensity. For example, even the scenario that described a corporation

failing to protect its workers from dangerous toxic contamination and denying its responsibility

when they develop and die from fatal diseases contracted in the workplace met with less popular

disapproval than did sexual assault, regardless of the fact that no mention was made of the rape

victim dying.

Further, while subjects tended to consider economic sanctions such as fine and

monetary compensation more appropriate punishments against elite deviance, it is noteworthy

that no scenario describing white-collar crimes generated mean financial penalties even close

to the maximum amount range (i.e., $1,000,000 and above). Despite their relative lack of

knowledge about the issue (and particularly in regard to corporate offenders’ greater legal

immunity compared with street criminals), perhaps some subjects were already aware of the

difficulty in prosecuting and punishing a firm. Companies are shielded by their employees

whom bankruptcy and dissolution would impoverish, causing tremendous unpopularity for

the prosecutor and judge and therefore possibly jeopardizing their reelection. In addition,

corporations have the power and financial means to avoid criminal prosecution by reaching

financial settlements with victims’ families. Moreover, as previously mentioned, corporations

are monitored and more likely to be dealt with by administrative and regulatory agencies such

as the Occupational Safety and Health

Administration (OSHA), the Environmental Protection Agency (EPA), and the Food and Drug

Administration (FDA). Maybe those subjects more inclined to accept “truths” about elite

deviance were also more likely to recognize these obstacles and, unenthusiastically, to not

recommend sanctions that they believe would be ineffective. The next logical steps are thus to

determine (a) if the sociodemographic correlates of the attitudes under investigation (i.e.,

perceived seriousness and punitiveness) match those of knowledge, acceptance of “truths” and

adherence to “myths” about white-collar crime, and (b) and whether such knowledge in turn

influences public sentiments towards elite deviance.

Sociodemographic Correlates of Perceived Seriousness

Table 13 presents the sociodemographic correlates of perceived seriousness of the

National White-Collar Crime Center scenarios compared with motor vehicle theft. As was the

case for associations between these correlates and knowledge, “truth” acceptance and

“myth” adherence, statistically significant correlation coefficients are both scarce and weak.

Overall, men - relative to women - rated embezzlement (r = -.139), identity theft (r = -.102),

false charges (r = -.184), counterfeit sales (r = -.104), and insurance overcharge (r = -.160) as

less serious than motor vehicle theft, and only found market rigging (r = .105) to be a more

serious offense. While market rigging can hypothetically affect a lot more people, it is

surprising to note that robbery - which led to victim’s hospitalization - did not attain statistical

significance.

As one might expect, age was positively associated with robbery (r = .222), hacking

(r = .192), false drug label (r = .144), espionage (r = .128), and counterfeit sales (r = .128).

That is, older subjects were more inclined to consider these offenses to be of greater

seriousness compared with motor vehicle theft. Few differences could be found between races.

More specifically, while Whites deemed hacking (r = -.131) and counterfeit sales (r = .102) to

be less serious than stealing a car, Blacks considered hacking (r = .142) to be a more serious

offense. Such discrepancy may be explained by racial gaps in criminal involvement.

Whereas African Americans are disproportionately represented in the criminal justice system

for street offenses, Whites - who enjoy greater educational and professional opportunities - are

more likely to engage in elite deviance (Barkan, 2012). It is therefore possible that Blacks

harbor more hostile feelings for crimes of greed (which they are statistically excluded from)

than they are for crimes of need.

Though still premature and requiring further investigation, the hypothesis that social

inequalities may predict attitudinal differences in regard to perceived seriousness of elite

deviance also finds support with the variable measuring subjects’ annual household income.

More specifically, respondents with higher income levels only found burglary (r = .118) and

robbery (r = .098) - i.e., the only two street crimes of the list - to be more serious than stealing

a car. Similarly, those subjects currently employed rated burglary (r = .103) but not hacking (r

= -.122) as being more serious than auto theft. Such choices could be due to a perception that,

unlike their unemployed counterparts, spending most of the day away from home and being

expectedly more affluent puts them at a higher risk of being burglarized, a notion that may

elicit more fear than spam and viruses.

Analogously, hacking was considered less serious than motor vehicle theft by members

of other political parties, by subjects with no religious affiliation, and by those who reported

the Internet as their medium of choice (r = -.112, -.115, and -.102, respectively). Further, those

subjects who grew up in the Northeast only rated false charges (r = .101) as more serious than

motor vehicle theft. Interestingly, more conservative respondents evinced lower perceptions of

seriousness for the false drug label scenario (r = -.122). Similarly, conservative Protestants

were less inclined to recognize market rigging (r = -.144) as more serious an offense than auto

theft, unlike counterfeit sales (r = .144).

These findings are puzzling. Recall that the market-rigging scenario described a Wall

Street financial firm that conspires to manipulate the precious metals market, profiting at the

expense of other traders and owners of precious metals who are unaware of the price-fixing

scheme. Conversely, the counterfeiting scenario described a person selling a counterfeit

antique bracelet on an online auction site, misrepresenting its true value and making an extra

$1,000. While no dollar amount was given for the market-rigging scheme, the resulting losses

for traders and owners could be expected to be significantly higher than the $1,000 made by

the counterfeiter. How then can the conservative Protestants preference that deemed the

former to be less serious and the latter more serious than motor vehicle theft be explained?

Perhaps the historical link between Protestantism and the spirit of capitalism (Weber, [1905]

2002) can account for such tolerance toward the competitiveness displayed by the Wall Street

firm compared with the sale of a fake bracelet, an offense that any petty street thief could

engage in. Conversely, those subjects with other religions rated embezzlement (r = .122) as

being more serious than car theft while those with no religious affiliation at all found hacking

(r = -.115) and counterfeiting (r = -.108) to be less serious than the baseline scenario.

A similar finding can be observed among those subjects who used the Internet over

other traditional media as their primary source of information. More precisely, these subjects

found hacking (r = -.102), counterfeit sales (r = -.137) and insurance overcharge (r = -.135)

to be less serious than car theft. Recall that Internet users were statistically less likely to

espouse “myths” about white-collar crime. It could be that, being aware of more serious cases

of elite deviance (e.g., state-corporate crime, human trafficking, etc.), they found these white-

collar offenses to pale in comparison.

Lastly, a number of differences in perceived seriousness of crime were observed in

regard to attribution style. More specifically, those participants inclined to believe that such

offenses are the result of greed were more likely to consider identity theft, false drug label,

espionage, counterfeit sales and insurance overcharge as more serious offenses than car theft (r

= .110, .212, .131, .124 and .124, respectively). These results are mirrored among those

respondents who attributed white-collar crime to bad moral character in regard to false drug

label, espionage, market rigging and counterfeit sales (r = .177, .113, .126, and .124,

respectively). Likewise, those more inclined to attribute elite deviance to a rational choice

selected identity theft, false drug label, and market rigging as more serious offenses than motor

vehicle theft (r = .098, .195 and .121, respectively).

Those who attributed white-collar crime to business environmental influences and to a

fiduciary responsibility to shareholders both rated embezzlement (r = .140 and .104), identity

theft (r = .117 and .123), false charges (r = .105 and .102), false drug label (r = .104 and .137),

market rigging (r = .220 and .129), and insurance overcharge (r = .121 and .099) as more

serious offenses. It should be noted, however, that the latter group of subjects also rated

robbery (r = .103) and espionage (r = .162) as being more serious than car theft. Further, those

more inclined to attribute elite deviance to pressure to succeed selected embezzlement (r =

.240), identity theft (r = .188), false charges (r = .239) and market rigging (r = .177) as more

serious than car theft.

Participants who evinced a more situational attribution style were less likely to perceive

white-collar crime scenarios as more serious than car theft. More precisely, those more

inclined to believe white-collar offenders see no wrong in their actions were less likely to rate

embezzlement (r = -.109) as a serious offense. However, the attribution of blame to low self-

control is more equivocal. More specifically, those inclined to believe white-collar offenders

simply lack the ability to refrain from crime were less likely to rate false drug label (r = -

.102) and espionage (r = -.128) as serious offenses compared with car theft. It is therefore

unclear whether low self-control in this sample was deemed an aggravating or a mitigating

factor in the commission of white-collar crime.

In summation, despite weak coefficients, statistically significant correlations were

observed between certain sociodemographic variables (including gender, age, race, region

where subjects grew up, income, employment, political ideology, religious affiliation,

information source, and attribution style) and perceived seriousness of the scenarios used in

the National White-Collar Crime Center survey. Overall, it appears the greatest gaps might be

due to attitudinal differences toward capitalism. That is, those subjects with higher income

levels, religious and political beliefs favorable to free market economics, and who explained

elite deviance as the result of external pressures in a competitive environment were less

inclined to rate white-collar crime as more serious a problem than street crime.

If differences in politico-religious attitudes create popular dissensus regarding perceived

seriousness of elite deviance, such findings should hold even after controlling for harm

intensity. Recall that this study’s sample was found to be less critical of white-collar crime

(even physically harmful offenses such as deadly pollutants release) when presented with

examples of violent street crime. Table 14 presents sociodemographic correlates of perceived

seriousness of consumer safety violation, illegal toxic dumping, and denial of risk and peril

after workers die from unprotected asbestos exposure in the workplace compared with two

violent crimes (i.e., murder and forcible rape). As was mentioned earlier, no baseline was

used for comparison purposes. Rather, participants were asked to rate the seriousness of the

case on a scale of 1 (not very serious) to 4 (very serious).

Once again, very few correlation coefficients attained statistical significance. Further,

such coefficients are - at best - weak to modest in strength. Compared with their female

counterparts, men were less inclined to rate toxic dumping (r = -.130) and asbestos exposure (r

= -.107) as serious offenses. There were no differences for gender for other scenarios.

Conversely, with the exception of rape, older subjects rated all scenarios as serious crimes.

Interestingly, the correlation coefficient for the asbestos vignette (r = .280) is even stronger

than that for murder (r = .247). Subjects with higher income levels were less likely to rate the

toxic dumping case a serious crime (r = -.104), as were more politically conservative subjects

(r = -.133) and Republicans (r = -.146). Similarly, those subjects belonging to no political party

were less inclined to rate the toy scenario as a serious offense (r = -.108), even though children

were the potential victims.

The only other statistically significant associations were found with information

source and attribution style. More specifically, those subjects who named the Internet as their

medium of choice were less likely to rate the three physically harmful white-collar crimes as

serious offenses (r = -.137, -.162, and -.118, respectively) relative to those who relied on more

traditional news media. While this finding is puzzling, it could be that those subjects were

already acquainted with other forms of state-corporate crimes that resulted in much greater

harm, which in turn may have influenced their perception of seriousness.

With the exception of rape, those subjects who attributed white-collar crime to greed were

more likely to rate every offense described as serious crimes (r = .179, .117, .188, and .147,

respectively). Further, those who attributed blame to low moral character were more likely to

rate all five scenarios as serious offenses (r = .165, .137, .249, .111, and .213, respectively).

Moreover, even those more inclined to believe fiduciary responsibility to shareholders forces

otherwise law-abiding citizens to commit white-collar offenses rated all three examples of elite

deviance as serious crimes (r = .175, .206, and .146, respectively). Further, those who saw elite

deviance as the result of a rational choice were more likely to perceive murder (r = .114) and

asbestos exposure (r = .098) as serious offenses. Lastly, those who explained elite deviance as

resulting from pressure to succeed were more inclined to consider the release of dangerous

pollutants a serious offense (r = .098), whereas those who believed white-collar offenders see

no wrong in their actions were less likely to consider rape a serious crime (r = -.099).

In summation, some showing of a mild form of consensus regarding the perceived

seriousness of white-collar crime emerges after controlling for harm intensity. Phrased

differently, with the notable exception of age and attribution style, very few differences in

perception of offense gravity are discernable. In fact, most statistically significant relationships

with white-collar crimes were this time negative, which could be due to their comparison with

murder and rape, two street crimes that may cause more fear than safety violations.

Nevertheless, it is worth reiterating that more conservative subjects and

Republicans were less likely to perceive illegal toxic dumping resulting in serious illness as a

serious offense, which once again suggests that political identity may influence perceptions of

severity. It is unclear, however, whether subjects’ perceived seriousness influences their levels

of punitiveness and if the latter varies as a function of the same sociodemographic variables.

Sociodemographic Correlates of Punitiveness

Recall that punitiveness was operationalized as participants’ choice of prosecutorial

process, fine, monetary compensation, and/or incarceration, as well as sentence severity (i.e.,

dollar amounts and number of years in prison). Table 15 presents the sociodemographic

correlates of respondents’ choice of prosecutorial process for the five scenarios. Again, these

choices are of increasing severity. That is, subjects could decide whether the case ought to be

dealt with by some non-legal means (e.g., out-of-court financial settlement), in a noncriminal

court (i.e., civil court), or in a criminal court.

First of all, the effect for murder was a constant, and 100% of the sample

recommended a criminal prosecution for the perpetrator. With respect to white-collar offenses,

as was the case before, very few correlation coefficients attained statistical significance.

Further, their strength is once again very weak. Men were less inclined to support a severe

prosecution style (i.e., in a criminal court) for the firm described in the asbestos exposure

scenario (r = -.139). Perhaps men were more likely to believe a criminal conviction would not

only impact the company and its employees but would also be very difficult to obtain. After

all, recall that despite being more likely to adhere to “myths”, males were also more inclined

to accept “truths” such as white-collar criminals’ relative legal immunity compared with street

offenders. Conversely, older subjects were more likely to support a harsher prosecution style

for those responsible in the asbestos exposure scenario (r = .124).

No notable differences could be observed as far as race is concerned. Surprisingly,

though, Hispanics were less inclined to demand a severe prosecution process for the rapist (r =

-.170). It could be that ethnic minorities, whom are disproportionately represented in the

criminal justice system (Hagan, Shedd, & Payne, 2005; Unnever & Cullen, 2007) but

relatively excluded from opportunities to commit white-collar crime, may perceive such gap

as racial discrimination. In turn, perceived injustice might lead them to evince more tolerance

for street offenders and increased severity toward elite deviance. In fact, though not

statistically significant, all associations between Hispanic subjects and white-collar crime

scenarios are positive.

More educated respondents were also less inclined to choose a severe prosecution

method for the rapist (r = -.135), as were conservative Protestants (r = -.108). This last finding

is surprising and runs counter to preconceptions regarding religious differences in

punitiveness relative to street crime (Grasmick et al., 1993). Subjects from the Northeast

were more likely to demand harsher prosecutorial processes against offenders depicted in the

illegal dumping and asbestos exposure scenarios (r = 145 and .101, respectively).

On the other hand, those participants with higher income levels, as well as those who

identified as Republicans, were less likely to support criminal prosecution against the

offenders in the scenarios describing illegal toxic dumping and asbestos exposure (r = -.103

and -.129, respectively. More conservative subjects were also less likely to condone harsh

prosecutorial processes for these two offenses (r = -. 147 and -.111, respectively). These

findings once again buttress the argument that attitudinal differences toward free market

economics may influence sentiments about elite deviance. Similarly, those subjects inclined to

believe white-collar offenders see no wrong in their actions were less likely to require a severe

prosecutorial method for decision-makers in the toxic dumping and asbestos exposure

scenarios (r = -.130 and -.109, respectively).

Conversely, those who believed that bad moral character is an important factor of elite

deviance recommended harsher prosecutorial processes against offenders described in all

three white-collar crime scenarios (r = .121, .133, .165, respectively). Similarly, those who

attributed crimes of the powerful to fiduciary responsibility to shareholders recommended a

harsher prosecutorial process against those responsible in the defective toy scenario (r =

.135), the toxic dumping scenario (r = .142) and the asbestos exposure vignette (r = .155).

With the exception of the toxic dumping scenario, these results are mirrored among those

who believed that greed is the root of elite deviance (r = .098 and .104, respectively). Lastly,

those who surmised that white-collar offenders choose their actions were more likely to select

a harsher prosecutorial process for those responsible in the asbestos exposure scenario (r =

.098).

Overall, while absolute consensus was reached in regard to the best way to prosecute

the murder case, disparities in punitiveness could be observed for the white-collar crime

scenarios. As previously mentioned, perhaps some participants considered a criminal

prosecution to be detrimental to business, a hypothesis seemingly supported by right-leaning

respondents’ leniency. It once again appears that subjects with higher income levels and with

political views favorable to capitalism were statistically less likely to recommend that offenders

in the white-collar crime vignettes be tried in a criminal court. Still, it is worth reiterating that

few statistically significant differences emerge, and those that did were weak.

Though still very weak in strength, more correlation coefficients attained statistical

significance when examining the sociodemographic correlates of participants’ choice of fine

amount (see Table 16). Men were more likely to demand a higher fine against those

responsible for illegal toxic dumping (r = .134). Such finding is consistent with previous

research that found greater levels of retributiveness among men (Bohm, 1992; Gilligan, 1977,

1982).

Likewise, older subjects were more likely to recommend a fine for the asbestos

exposure vignette (r = .099). Ambiguous findings emerged for race, occupational status and

political affiliations. More specifically, while Whites were more likely to choose a greater fine

amount for the toxic dumping scenario (r = .101), Blacks were less inclined to do so against

those responsible for selling a potentially dangerous toy (r = -.099), as did those subjects

currently employed (r = -.097) and Democrats (r = -.101).

Further, subjects from other political parties were more inclined to choose a more severe

monetary sanction against toxic dumping (r = .155), unlike Democrats (r = -.136). Conservative

Protestants were less likely to consider a fine a relevant form of punishment against the

companies described in the toy and asbestos exposure scenarios (r = -.109 and .108,

respectively). Moreover, conservative participants were more likely to support a higher fine

against the perpetrators in both street crime scenarios (r = .108 and .119, respectively).

Very few correlation coefficients attained statistical significance in regard to

attribution style. More specifically, subjects who believed white-collar offenders have little

self-control were more likely to require a higher fine in all five scenarios (r = .175, .116, .101,

.129, and .109, respectively). Those who considered fiduciary responsibility to be a cause of

elite deviance selected the same form of sanction for those responsible in the asbestos

exposure case (r = .134). Other than these, no further statistically significant differences could

be found. It might be that participants felt that a fine would be a disservice to the company

and hurt its business, and favored a more lenient punishment (i.e., monetary compensation for

the victims and their families). Alternatively, they may also have considered that a more

severe sanction (i.e., incarceration) would be more appropriate.

Table 17 presents the sociodemographic correlates of participants’ choice of monetary

compensation amount. Once again, few correlation coefficients attained statistical

significance and their strength is quite weak. All associations between age and white-collar

crime scenarios are positive (r = .118, .182, and 183, respectively). While no difference

could be found between Whites and Blacks, subjects of other races were less likely to support

this form of punishment against the company responsible for distributing the potentially

dangerous toy (r = -.120). Further, participants from the Northeast were less inclined to

support a monetary compensation in the case of the defective toy scenario (r = -.098).

While no major differences could be found for political affiliation - with the exception

of subjects of other parties supporting monetary compensation for the victims of illegal toxic

dumping (r = .111) - a few notable gaps were observed with religious identity. More

specifically, moderate Protestants favored this form of sanction for both murder and rape (r =

.133 and .117, respectively) and also for the toxic dumping scenario (r = .113). Conversely,

subjects who identified with other religions besides Catholicism and

Protestantism were less inclined to support monetary compensation for the victims of deadly

pollutants (r = -.107).

As could be expected, those inclined to blame elite deviance on greed were more

likely to support damages for the victims and the families in both the toxic dumping and

asbestos exposure scenarios (r = .121 and .138, respectively). Conversely, those who believed

white-collar crime derives from low self-control deemed monetary compensation a fair

sanction in all cases, with the exception of the asbestos scenario (r = .139, .116, .119, and

.118, respectively). Further, those inclined to cite fiduciary responsibility to shareholders as a

motive for elite deviance supported the payment of damages in all scenarios, except for the

release of deadly pollutants (r = .163, .117, .114, and .124, respectively). Those more likely to

agree with the idea that white-collar crime stems from bad moral character only supported

monetary compensation in the asbestos case (r = .107). A likely explanation is their

perception of white-collar offenders being inherently immoral led them to seek a more severe

form of punishment. If this is the case, then positive relationships with prison sentence

severity should emerge.

Table 18 presents the sociodemographic correlates of prison sentence severity against

white-collar crime and street crime. As was the case before, few correlation coefficients

attained statistical significance, and those that did were weak. With the exception of the

asbestos exposure scenario, older subjects were once again found to be quite punitive against

elite deviance as they supported long prison time for the culprits in the toy and toxic dumping

scenarios (r = .110 and .105, respectively) as they did for the murderer (r = .166) and rapist (r =

.109).

An interesting gap emerged between races. More precisely, while Whites were more

likely to demand prison time for the homicide perpetrator (r = .150), Blacks and subjects of other

races were less inclined to do so (r = -.099 and -.132, respectively). As previously mentioned, it

could be that an acute perception of racial discrimination within the criminal justice among

minorities - suggested by their greater likelihood of being convicted and incarcerated (Hagan,

Shedd, & Payne, 2005; Unnever & Cullen, 2007) - led them to display more leniency toward

street offenders.

Similarly, more educated subjects recommended longer prison time for those responsible

in the defective toy scenario (r = .104) but not for the rapist (r = -.131). Likewise, perhaps

because incarceration is after all a tax-financed punishment, the rape scenario did not elicit

demand for a harsh prison sentence among employed subjects (r = -.114). Moderate Protestants

were also less likely to demand a longer sentence length for the culprit in the murder case (r = -

.103).

Conversely, more politically conservative subjects and Republicans were inclined to

impose a more severe sentence on the murderer (r = .116 and .130, respectively). With the

exception of Catholics supporting prison time for denial of risk and peril in the asbestos

exposure scenario (r = .106), no notable differences could be found in regard to religion.

Further, subjects affiliated with no political party and those who use the Internet as their

primary source of information were less inclined to support prison time for manufacturing and

selling a potentially dangerous product destined for children (r = -.144 and -.161,

respectively).

In regard to attribution style, the defective toy scenario elicited stronger prison

sentence severity among those who attribute white-collar crime to bad moral character (r =

.110), choice (r = .109) and fiduciary responsibility (r = .101). Further, the toxic dumping

vignette generated greater punitiveness among those inclined to believe greed is the cause of

elite deviance (r = .106) as well as among those who cite low moral standards (r = .175) and

fiduciary responsibility to shareholders (r = .128) as white-collar crime factors. In addition,

those who blamed elite deviance on negative business environmental influences required a

longer prison sentence for those responsible in the asbestos exposure scenario (r = .106).

Conversely, as could be expected, those inclined to believe white-collar offenders see no

wrong in their actions were also less likely to demand tougher sentences for the culprits in the

toxic dumping and asbestos exposure scenarios (r = -.137 and -.114, respectively).

In summation, few but consistent gaps in punitiveness among this study’s subjects could

be observed, suggesting significant popular dissensus in societal response to elite deviance.

More specifically, older subjects were found to be more punitive both with street and white-

collar offenders. This finding may be in line with the curvilinear relationship between age and

punitive attitudes; that is, people may become more punitive as they grow older until they reach

a tipping point, whereupon punitiveness starts to decrease (Schwartz et al., 1993). In fact, Rossi

and Berk (1997) found that those between the age of 35 and 64 were the most punitive, whereas

those under 35 and above 64 years old were the least punitive. Moreover, recall that several

studies found that older people tend to view elite deviance as somewhat more serious than

conventional violent crime and narcotic offenses (Grabosky,

Braithwaite, & Wilson, 1987; Hauber, Toonvliet, & Willemse, 1988). With a mean age of

33.58, this study’s sample counted few very young or very old subjects, thus limiting the ability

to draw any significant conclusions about the curvilinear findings from prior studies.

Perhaps more interestingly, it is also worth reiterating that men, Whites, those with

higher income levels, more politically conservative subjects, Republicans, conservative

Protestants, and those who believed white-collar offenders see no wrong in their actions were

often more lenient in their attitudes towards elite deviance, both in terms of perceived

seriousness and punitiveness, compared with street crime. More specifically, Table 14 reports

lower perceived seriousness for white-collar crimes among conservative participants and

Republicans. Further, Table 15 shows that males, wealthier and right-leaning subjects were

less likely to rate harmful white-collar crime scenarios as serious offenses. Moreover, Table 16

suggests that currently employed and politically conservative participants as well as

conservative Protestants were less inclined to favor a fine for white-collar offenders. In

addition, Table 17 reveals that moderate Protestants were more likely to support compensating

the victims and their families, a somewhat lenient choice compared with the harsher sanction

options proposed to them. Lastly, Table 18 reports that moderate Protestants were less inclined

to require a long prison sentence for those companies’ executives, and that Whites, politically

conservative subjects as well as Republicans were more likely to demand prison time for the

street offender in the murder scenario. Conversely, those inclined to believe that white-collar

offenders are otherwise law-abiding citizens who see no wrong in their actions exhibited

greater tolerance for those responsible in the toxic dumping and asbestos exposure scenarios.

Importantly, recall that, with a few exceptions, those subjects generally scored poorly on the

knowledge questionnaire and were also more likely to adhere to “myths” about white-collar

crime. It is therefore possible that knowledge (or lack thereof) about elite deviance might

influence general attitudes regarding this social issue.

Table 19 presents zero-order correlations between knowledge about elite deviance,

“truth” acceptance and “myth” adherence, and perceived seriousness of the eleven scenarios

used in the National White-Collar Crime survey. Though weak in strength, positive associations

between knowledge and the false drug label, espionage, and market rigging scenarios were

statistically significant (r = .156, .129, and .176, respectively). These three scenarios - as well as

embezzlement, identify theft, and insurance overcharge - were also positively correlated with

“truth” acceptance (r = .128, .155, .158, .120, .107, and .101, respectively). However, no

coefficient attains statistical significance in regard to “myth” adherence.

A few statistically significant findings emerged when controlling for more physically

injurious instances of elite deviance. Table 20 presents zero-order correlations between

knowledge about elite deviance, “truth” acceptance, “myth” adherence and perceived seriousness

of the five harmful white-collar crime and street crime scenarios. Knowledge was positively

correlated with perceived seriousness of the asbestos exposure scenario (r = .126), as was “truth”

acceptance with the toxic dumping and workers’ endangerment vignettes (r = .101 in both cases).

Conversely, “myth” adherence was negatively associated with perceived seriousness of denial of

risk and peril described in the third white-collar crime scenario (r = .101).

Similar findings emerged when comparing knowledge about elite deviance, “truth”

acceptance and “myth” adherence with participants’ choice of prosecutorial process. Table 21

reveals that those subjects who accepted “truths” regarding elite deviance were more likely to

demand harsher prosecution for the company accused of knowingly selling a potentially

dangerous toy (r = .138). In contrast, “myth” adherers were less inclined to support a harsh

prosecution process for the rapist (r = -.116).

Although - as is evident in Table 22 - only one correlation coefficient attained statistical

significance when comparing knowledge about elite deviance, “truth” acceptance and “myth”

adherence with respondents’ choice of fine amount, knowledge is once again positively

associated with the asbestos exposure scenario (r = .121). That is, those subjects who were

more knowledgeable about white-collar crime were also more likely to support fining the

company responsible for causing the deaths of several of its employees.

Interestingly, Table 23 reveals that subjects with knowledge about elite deviance and

“truth” acceptance were more inclined to support monetary compensation for the rapist (r = .131

and .121, respectively). Further, “truth” believers were more likely to condone such punishment

against the perpetrators described in the illegal toxic dumping (r = .144). Lastly, Table 24

reports positive associations between knowledge about elite deviance, “truth” acceptance, and

prison sentence severity for all three white-collar crime scenarios (r = .144,

.110 and .124 for more knowledgeable subjects, and .139, .102 and .137 for “truth” believers).

Conclusion

In summation, it appears that knowledge about white-collar crime does influence

attitudes towards elite deviance. That is, while probably being aware of the numerous

challenges inherent in successfully prosecuting a corporation, knowledgeable subjects were

nevertheless more likely to perceive the white-collar infractions presented to them as more

serious offenses compared with street crimes and to demand harsher sentences against their

perpetrators. Conversely, less educated participants and “myth” adherers were generally less

critical of elite deviance as a whole and more lenient in their choice of sanctions against it

relative to traditional crime. Further, more knowledgeable subjects were found to be those

who identified as Whites, with higher education levels, without any religious affiliation, and

who used the Internet as their main source of information. In comparison, less knowledgeable

participants turned out to be predominantly male, politically more conservative, Republican,

conservative Protestant, who relied on traditional media sources rather than the Internet and

who attributed white-collar crime to situational rather than dispositional factors.

As previously mentioned, these last sociodemographic variables are usually correlated

with greater support for elements of neoliberal economics such as market deregulation, a

natural consequence of laissez-faire capitalism that has been shown to facilitate the

commission of certain white-collar crimes (Lynch & Michalowski, 2006). It is therefore

possible that pro-capitalism attitudes may have led these individuals to discount relevant

information about elite deviance and biased their sentiments towards it. In light of such

hypothesis, these sociodemographic variables will be further analyzed in chapter five.

Tables

Table 2. Percent Subjective and Objective Knowledge about White-Collar Crime (N=408)

________________________________________________________________________

Percent Subjective Percent Objective Difference

Not very informed

12.5

75.5

-63.0

Somewhat informed

73.5

14.7

58.8

Informed

11.3

7.4

3.9

Very informed

2.7

2.4

.30

________________________________________________________________________

*p < .05.

Table 3. Percentage of Subjects with Scores on the Overall Knowledge Scale (N=408)

________________________________________________________________________

Overall Knowledge Score Percentage of Subjects Cumulative Percent

0 0 0

10 0 0

20 2.8 2.8

30 6.6 9.4

40 16.9 26.3

50 24.0 50.3

60 25.2 75.5

70 14.7 90.2

80 7.4 97.6

90 2.2 99.8

100 0.2 100

________________________________________________________________________ Table

4. Percent Correct, Percent Incorrect, and Percent Confident on the Overall

Knowledge Scale and the Ten Items that Comprise It, and “Truth” Accepters (N=134),

Lucky Guessers (N=90), “Myth” Adherers (N=67), and Honestly Uninformed Subjects (N=117)

___________________________________________________________________________

Percent

Correct

Percent

Incorrect

Percent

Confident

“Truth”

Accepters

Lucky

Guessers

“Myth”

Adherers

Honestly

Uninformed

Overall

54.9

45.1

49.9

32.8

22.1

16.5

28.6

Meaning of the Term

“White-Collar Crime”

89.2

10.8

71.8

66.9

22.3

4.9

5.9

Financial Cost of White-

Collar Crime

23.8

76.2

27.7

13.7

10.1

14.0

62.2

Physical Harmfulness of

White-Collar Crime

3.2

96.8

66.2

2.7

0.5

63.5

33.3

Legal Immunity (relative

to street crime)

32.8

67.2

58.5

27.2

5.6

25.2

42.0

Reckless Disregard (Ford

Pinto case)

75.5

24.5

30.1

27.5

48.0

2.7

21.8

Medical Crime (vs.

homicides)

38.2

61.8

25.7

13.2

25.0

12.5

49.3

Human Trafficking (in the

U.S. vs. abroad)

30.9

69.1

43.4

11.0

19.9

32.4

36.7

State Corporate Crime

(private military firms)

91.2

8.8

61.2

60.3

30.9

1.0

7.8

Toxic Dumping (African

American communities)

66.7

33.3

39.2

31.4

35.3

7.8

25.5

Toxic Emissions

97.1

2.90

74.8

73.8

23.3

1.0

1.9

(Reluctance to invest in

clean technologies)

___________________________________________________________________________

Table 5. One-Way Between Subjects ANOVAs in the Effect of Sociodemographic

Predictors of Knowledge about White-Collar Crime, “Truth” Acceptance and “Myth”

Adherence (N = 408)

___________________________________________________________________________

Knowledge

“Truth” Acceptance

“Myth” Adherence

F (df)

η2

F (df)

η2

F (df)

η2

Age

.96 (49, 358)

1.16 (49, 358)

.89 (49, 358)

Race

1.50 (6, 401)

1.07 (6, 401)

3.24* (6, 401)

.05

Region

2.44* (4, 403)

.02

.90 (4, 403)

2.93* (4, 403)

.03

Residence

.85 (9, 398)

.67 (9, 398)

1.37 (9, 398)

Education

4.48* (6, 401)

.06

3.47* (6, 401)

.05

.311 (6, 401)

Employed

1.24 (4, 403)

1.77 (4, 403)

1.89 (4, 403)

Income

1.41 (9, 308)

.89 (9, 398)

2.41* (9, 398)

.05

Pol. Ideology

2.87* (5, 402)

.03

4.16* (5, 402)

.05

1.79 (5, 402)

Pol. Affiliation

1.59 (5, 402)

2.91* (5, 402)

.03

2.06 (5, 402)

Religion

2.57* (7, 400)

.04

1.91 (7, 400)

1.35 (7, 400)

Information

1.13 (6, 401)

.54 (6, 401)

2.63* (6, 401)

.04

Greed

.79 (3, 404)

1.23 (3, 404)

1.08 (3, 404)

Moral

.97 (3, 407)

1.50 (3, 404)

.15 (3, 404)

Control

1.56 (3, 404)

4.33* (3, 404)

.01

1.42 (3, 404)

Choice

2.99* (3, 404)

.02

1.99 (3, 404)

.085 (3, 404)

Influence

5.84* (3, 404)

.04

5.24* (3, 404)

.04

4.39* (3, 404)

.03

Pressure

2.16 (3, 404)

1.69 (3, 404)

2.68* (3, 404)

.02

Fiduciary

1.78 (3, 404)

2.33 (3, 404)

1.87 (3, 404)

No wrong

.26 (3, 404)

.76 (3, 404)

.75 (3, 404)

___________________________________________________________________________

Note. η2 = effect size.

*p < .05.

Table 6. Zero-Order Correlations between Sociodemographic Characteristics, Knowledge about

Elite Deviance, “Truth” Acceptance and “Myth” Adherence (N=408)

___________________________________________________________________________

Knowledge

“Truth” Acceptance

“Myth” Adherence

Male

.910

.127*

.160**

Age

.031

-.025

-.048

White

.154**

.078

-.131**

Black

-.074

.051

.114*

Other Race

-.087

-.072

-.002

Hispanic

-.086

-.110*

.090

Northeast

-.081

-.083

.065

Urban

-.014

.035

.037

Income

-.036

.076

.091

Education

.219**

.175**

.008

Employed

-.026

.074

.090

Pol. Ideology

-.158**

-.142**

.078

Republican

-.113*

-.129**

.120

Democrat

.017

.021

.026

Other Party

.070

.079

-.120*

No Party

-.013

-.061

-.071

Catholic

-.075

-.070

. 029

Cons. Protestant

-.142**

.084

Mod. Protestant

-.044

-.003

.002

Lib. Protestant

-.027

-.016

.008

Other Religion

.099

.068

.004

No Religion

.140**

.118*

-.085

Internet

.075

.005

-.155**

Greed

.099

.068

.004

Moral

.140**

.118*

-.085

Control

.075

.005

-.155**

Choice

.065

.087

.037

Influence

.068

.071

-.022

Fiduciary

-.033

-.017

.082

Pressure

.119*

.082

-.024

No Wrong

.182**

.151**

-.097

___________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, ** p < .01.

Table 7. Mean Crime Seriousness Scores for White-Collar and Street Crime Compared with

Motor Vehicle Theft in the National White-Collar Crime Center Survey and the Present Study

(N=408)

___________________________________________________________________________

NW3C Survey

Present Study

Difference

Burglary

3.7

3.2

.05

Embezzlement

4.1

3.7

.04

Identity Theft

4.3

3.9

.04

False Charges

3.8

3.5

.03

Robbery

4.0

3.6

.04

Hacking

3.9

2.7

1.2

False Drug Label

4.8

4.7

.01

Espionage

4.8

.00

Market Rigging

4.3

4.2

.01

Counterfeit Sales

2.9

2.3

.06

Insurance Overcharge

4.4

3.7

.07

Overall

4.2

3.9

.03

___________________________________________________________________________

Note. A mean of 1 = Much less serious, 2 = Somewhat less serious, 3 = About as serious, 4 =

Somewhat more serious, and 5 = Much more serious.

Table 8. Results of Paired Samples t-Tests to Compare Mean Perceived Seriousness of

White-Collar Crime & Street Crime (N=408)

___________________________________________________________________________

Mean (SD)

Toy

Murder

Pollutants

Rape

Toy

3.43 (0.76)

Murder

3.91 (0.36)

-12.10**

Pollutants

3.51 (0.64)

-1.98*

11.50**

Rape

3.94 (0.24)

-13.12**

-1.67

-12.89**

Asbestos

3.72 (0.52)

-7.76**

6.68**

-7.07**

8.13**

___________________________________________________________________________

Note. A mean of 1 = Not very serious, 2 = Somewhat serious, 3 = Serious, and 4 = Very serious.

*p < .05, **p < .01.

Table 9. Results of Paired Samples t-Tests to Compare Subjects’ Choice of Prosecutorial

Process for White-Collar Crime & Street Crime (N=408)

___________________________________________________________________________

Mean (SD)

Toy

Murder

Pollutants

Rape

Toy

2.74 (0.47)

Murder

3.00 (0.00)

-11.13**

Pollutants

2.70 (0.51)

1.45

11.90**

Rape

2.99 (0.99)

-10.37**

2.01*

-11.39**

Asbestos

2.71 (0.50)

1.12

11.75**

-0.47

11.34**

___________________________________________________________________________

Note. A mean of 1 = Non-legal means, 2 = Civil court, and 3 = Criminal court. *p < .05, **p

< .01.

Table 10. Results of Paired Samples t-Tests to Compare Subjects’ Choice of Fine Amount for

White-Collar Crime & Street Crime (N=408)

___________________________________________________________________________

Mean (SD)

Toy

Murder

Pollutants

Rape

Toy

1.83 (1.72)

Murder

0.35 (0.98)

16.99**

Pollutants

2.25 (1.75)

-5.21**

-21.51**

Rape

0.35 (0.96)

17.44**

0.22

21.68**

Asbestos

1.78 (1.82)

0.65

-15.77**

5.53**

-16.99**

___________________________________________________________________________

Note. A mean of 0 = No Fine, 1 = Under $100,000, 2 = $100,000-499,000, 3 =

$500,0001,000,000, and 4 = Above $1,000,000.

*p < .05, **p < .01.

Table 11. Results of Paired Samples t-Tests to Compare Subjects’ Choice of Compensation

Amount for White-Collar Crime & Street Crime (N=408)

___________________________________________________________________________

Mean (SD)

Toy

Murder

Pollutants

Rape

Toy

1.58 (1.48)

Murder

0.97 (1.52)

7.00**

Pollutants

1.85 (1.59)

-3.35**

-9.71**

Rape

0.77 (1.30)

10.18**

3.90**

13.04**

Asbestos

2.42 (1.54)

-10.71**

-15.73**

-7.30**

-19.16**

___________________________________________________________________________

Note. A mean of 0 = No Compensation, 1 = Under $100,000, 2 = $100,000-499,000, 3 =

$500,000-1,000,000, and 4 = $Above 1,000,000.

*p < .05, **p < .01.

Table 12. Results of Paired Samples t-Tests to Compare Mean Prison Sentence Severity for

White-Collar Crime & Street Crime (N=408)

___________________________________________________________________________

Mean (SD)

Toy

Murder

Pollutants

Rape

Toy

0.84 (1.18)

Murder

4.68 (1.63)

-40.20**

Pollutants

0.94 (1.32)

-1.56

38.32**

Rape

3.26 (1.57)

-26.39**

16.75**

-24.79**

Asbestos

1.35 (1.79)

-5.90**

29.45**

-5.48**

18.13**

___________________________________________________________________________

Note. A mean of 0 = No prison, 1 = 1-5 years, 2 = 6-10 years, 3 = 11-20 years, 4 = 21-30 years,

5 = 31-40 years, 6 = 41-Life, and 7 = Death.

*p < .05, **p < .01.

Table 13. Sociodemographic Correlates of Perceived Seriousness of the National White-Collar Crime Center Survey Scenarios

Compared with Motor Vehicle Theft (N=408)

____________________________________________________________________________________________________________

Burglary

Embezzle

ID Theft

Charges

Robbery

Hacking

False Label

Espionage

Market

Counterfeit

Overcharge

Male

-.005

-.139*

-.102*

-.184*

.020

-.003

-.090

-.055

.105*

-.104*

-.160*

Age

.024

-.057

.033

-.007

.222*

.192**

.144**

.128**

.093

.128**

.027

White

-.020

-.048

-.043

-.036

.034

-.131*

.013

.069

.007

-.102*

-.013

Black

-.005

.063

.049

-.011

-.007

.142**

.032

-.028

.004

.020

-.010

Other Race

.001

-.007

-.002

.033

-.026

.091

.004

.000

.030

.069

.002

Hispanic

.037

.017

.020

.037

-.022

-.027

-.059

-.082

-.043

.077

.029

Northeast

.014

.070

.041

.101*

-.002

-.028

.005

.012

.011

-.049

Urban

.040

.021

-.030

-.009

.057

.004

-.050

-.063

.033

.089

.044

Income

.118*

-.005

.080

-.056

.090

.098*

.059

-.062

-.022

-.050

-.031

Education

-.038

-.010

-.015

-.032

.077

.066

.076

.006

.087

.012

-.024

Employed

.103*

.032

-.058

-.026

-.027

-.122*

-.056

-.034

.003

-.074

-.062

Pol.Ideology

.040

.013

.006

-.041

-.027

-.030

-.122*

.008

-.029

.026

-.042

Republican

.084

-.060

-.009

-.085

.033

.077

.010

.011

-.083

.031

-.040

Democrat

.011

.046

.038

.037

.068

.050

.021

.028

-.013

.031

.035

Other Party

-.077

-.001

-.032

.028

-.096

-.112*

-.029

-.038

.078

-.056

-.005

No Party

-.048

-.016

-.017

-.022

-.069

-.081

-.061

-.039

-.017

-.090

-.056

Catholic

.057

.014

.031

.075

.008

.030

.001

.038

.092

.057

.021

Cons. Prot.

-.053

.019

-.036

.019

-.032

.034

-.038

-.044

-.144*

.144**

-.052

Mod. Prot.

-.049

-.007

.066

-.016

-.004

.043

-.026

-.015

-.048

-.023

.025

Lib. Prot.

.053

-.057

.013

-.014

.056

.058

.032

.044

.019

.046

.021

Other Rel.

-.050

.122*

.032

.085

.055

.024

-.021

.046

.073

-.063

.002

No Religion

.036

-.060

-.069

-.096

-.036

-.115*

.042

-.031

.020

-.108*

-.008

Internet

.051

.031

-.046

-.045

-.033

-.102*

.027

-.006

.047

-.137*

-.145*

Greed

.075

.081

.110*

.045

.026

.041

.212*

.131**

-.064

.124*

Moral

.087

.038

.062

.000

.069

.028

.177**

.113*

.126*

.051

.124*

Control

-.027

-.010

-.031

-.049

-.061

.026

-.102*

-.128*

-.017

.041

.039

Choice

.004

.035

.098*

.096

.034

.063

.195**

.073

.121*

.044

.088

Influence

-.047

.140**

.117*

.105*

-.014

.040

.104*

.012

.220**

.035

.121*

Fiduciary

.032

.104*

.123*

.102*

.103*

.035

.137**

.162**

.129**

.067

.099*

Pressure

.000

.240**

.188**

.239**

.067

.016

.014

-.051

.117*

.052

.067

No Wrong

-.109*

-.047

.032

.004

.042

.076

-.075

-.089

-.022

.042

-.011

_____________________________________________________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous predictors and Spearman’s rho when

using multinomial nominal and/or ordinal predictors. *p < .05, ** p < .01.

108

104

Table 14. Sociodemographic Correlates of Perceived Seriousness of White-Collar Crime &

Street Crime (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Male

-.077

-.040

-.130**

.032

-.107*

Age

.263**

.247**

.172**

.092

.280**

White

.055

.001

-.023

.070

-.011

Black

-.082

.050

.051

.037

.000

Other Race

.016

.009

-.045

-.063

-.039

Hispanic

-.015

-.065

.025

-.092

.054

Northeast

-.005

.024

.053

.021

.044

Urban

-.085

-.051

-.081

.001

-.091

Income

.037

.030

-.104*

.040

.019

Education

.002

.045

-.022

-.044

.015

Employed

-.084

.035

-.028

-.010

-.061

Pol.Ideology

-.049

.003

-.133**

-.049

-.042

Republican

.009

.008

-.146**

.006

-.007

Democrat

.045

.077

.032

.073

.041

Other Party

-.054

-.085

.081

-.081

-.037

No Party

-.108*

-.088

.068

-.058

-.095

Catholic

-.051

.029

.030

.018

.022

Cons. Prot.

.024

-.040

-.086

-.051

-.025

Mod. Prot.

-.051

.018

-.008

-.030

-.039

Lib. Prot.

.020

.059

.049

-.084

.014

Other Rel.

.079

-.052

.015

.001

.058

No Religion

.005

.061

.014

.081

-.008

Internet

-.137**

.038

-.162**

-.019

-.118*

Greed

.179**

.117*

.188**

.062

.147**

Moral

.165**

.137**

.249**

.111*

.213**

Control

-.012

-.060

-.023

-.029

.008

Choice

.081

.114*

.081

.063

.098*

Influence

-.009

-.073

.056

-.018

.023

Fiduciary

.175**

.080

.206**

.031

.146**

Pressure

.064

.018

.098*

-.010

.042

No Wrong

-.096

.008

-.054

-.099*

-.068

_____________________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p <

.05, ** p < .01.

105

Table 15. Sociodemographic Correlates of Choice of Prosecutorial Process against White-Collar

Crime & Street Crime (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Male

-.008

-.083

.000

-.139**

Age

.031

.052

-.034

.124*

White

.012

-.033

.071

-.082

Black

.006

.013

.029

.061

Other Race

-.076

.012

.025

.006

Hispanic

.047

.028

-.170**

.061

Northeast

.013

.145**

.023

.101*

Urban

.066

-.053

-.074

-.026

Income

.073

-.103*

-.050

Education

.069

-.051

-.135**

-.008

Employed

.099

.023

-.002

-.029

Pol.Ideology

-.072

-.147**

-.020

-.111*

Republican

-.015

-.077

-.021

-.129**

Democrat

.012

.052

-.007

.080

Other Party

.000

.006

.024

.018

No Party

-.043

-.022

.038

-.031

Catholic

-.069

.053

-.030

.086

Cons. Prot.

.016

-.050

-.108*

-.047

Mod. Prot.

.009

.006

-.025

-.004

Lib. Prot.

.005

.002

.023

-.072

Other Rel.

-.008

-.052

.030

-.025

No Religion

.034

.020

.088

.019

Internet

-.080

-.078

.016

-.068

Greed

.098*

.070

.006

.104*

Moral

.121*

.133**

.035

.165**

Control

.000

.010

-.054

-.025

Choice

.071

.056

.059

.098*

Influence

.039

.067

-.009

.048

Fiduciary

.135**

.142**

.009

.155**

Pressure

-.008

.064

.063

.018

No Wrong

-.033

-.130**

.083

-.109*

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Table 16. Sociodemographic Correlates of Choice of Fine Amount against White-Collar Crime

& Street Crime (N=408)

______________________________________________________________________________

106

Toy

Murder

Pollutants

Rape

Asbestos

Male

.033

.028

.134*

.026

.044

Age

.079

.073

-.041

.087

.099*

White

.059

-.006

.101*

-.004

.058

Black

-.099*

.016

-.068

-.039

-.085

Other Race

-.017

-.029

-.054

.025

-.037

Hispanic

.027

.021

-.039

.023

.033

Northeast

-.036

-.062

-.049

-.057

-.061

Urban

-.029

-.013

-.022

-.019

-.015

Income

.084

.008

.065

-.039

.049

Education

-.021

.010

-.057

-.027

.003

Employed

-.097*

-.028

-.047

-.028

-.044

Pol.Ideology

.036

.108*

.078

.119*

.004

Republican

.060

.024

-.020

.076

-.006

Democrat

-.101*

-.069

-.136**

-.068

-.080

Other Party

.058

.052

.155**

.011

.087

No Party

.063

.038

.090

.018

.046

Catholic

.049

.001

-.055

.019

-.004

Cons. Prot.

-.109*

-.023

-.056

.033

-.108*

Mod. Prot.

.015

.092

.069

.041

.064

Lib. Prot.

.049

.034

.026

-.046

.003

Other Rel.

-.040

.049

-.001

.024

-.033

No Religion

.029

-.095

.016

-.059

.047

Internet

-.015

-.095

.047

-.035

-.041

Greed

.009

.013

.072

-.008

.009

Moral

.062

.047

.052

.003

.081

Control

.175**

.116*

.101*

.129**

.109*

Choice

-.012

.029

.002

-.002

.066

Influence

-.056

-.088

-.065

-.059

.014

Fiduciary

.085

.077

.066

.058

.134**

Pressure

-.031

-.043

-.041

-.024

.016

No Wrong

.024

.056

.082

.069

.023

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Table 17. Sociodemographic Correlates of Choice of Monetary Compensation Amount against

107

White-Collar Crime & Street Crime (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Male

.064

.029

-.037

.025

.062

Age

.118*

.041

.182**

.077

.183**

White

.078

.058

.053

.082

-.042

Black

-.021

-.072

-.077

-.082

.057

Other Race

-.120*

-.029

-.078

-.042

-.056

Hispanic

.012

-.005

.059

Northeast

-.098*

-.090

-.041

-.073

-.067

Urban

-.051

-.041

-.068

-.009

-.080

Income

.066

.025

.041

.018

.058

Education

-.072

-.017

-.061

-.004

-.073

Employed

.087

-.034

.004

.048

.010

Pol.Ideology

-.025

.055

.028

.053

.005

Republican

.023

.004

.002

.049

.022

Democrat

.020

-.056

-.094

-.038

.001

Other Party

-.038

.055

.095

.001

-.018

No Party

.031

-.017

.111*

-.019

.008

Catholic

-.029

-.043

-.022

.013

-.025

Cons. Prot.

.028

-.035

-.058

-.037

.007

Mod. Prot.

.073

.133**

.113*

.117*

.054

Lib. Prot.

-.043

.027

-.064

-.013

-.032

Other Rel.

-.061

-.030

-.107*

.004

-.063

No Religion

-.001

-.038

.059

-.066

.022

Internet

-.062

-.063

-.093

-.027

-.016

Greed

.069

.060

.121*

.048

.138**

Moral

.045

.053

.095

.073

.107*

Control

.139**

.116*

.119*

.118*

.053

Choice

.019

.012

.081

.002

.077

Influence

-.033

-.064

.006

-.056

-.079

Fiduciary

.163**

.117*

.085

.114*

.124*

Pressure

-.047

-.026

-.020

.019

-.030

No Wrong

.028

.013

.021

.018

.067

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Table 18. Sociodemographic Correlates of Prison Sentence Severity against White-Collar Crime

& Street Crime (N=408)

______________________________________________________________________________

108

Toy

Murder

Pollutants

Rape

Asbestos

Male

.023

.040

.015

-.045

-.016

Age

.110*

.105*

.166**

.109*

.094

White

-.039

.150**

-.005

.074

.003

Black

.055

-.099*

-.049

-.048

-.072

Other Race

.017

-.132**

-.012

-.049

.047

Hispanic

-.013

-.012

.071

-.020

.028

Northeast

.010

-.042

.080

-.050

.021

Urban

.042

.006

.017

-.023

.026

Income

.063

.014

-.062

-.067

-.021

Education

.104*

-.058

.039

-.131**

.037

Employed

-.046

-.003

-.005

-.114*

-.035

Pol.Ideology

.003

.116*

-.060

-.006

-.051

Republican

.089

.130**

-.034

.067

-.081

Democrat

.000

-.096

.000

-.029

.021

Other Party

-.069

-.001

.026

-.023

.041

No Party

-.144**

0.43

-.056

.032

-.020

Catholic

.008

.018

.092

.012

.106*

Cons. Prot.

.053

-.002

-.015

-.033

-.056

Mod. Prot.

-.085

.036

-.103*

.019

-.062

Lib. Prot.

.021

-.067

-.042

-.016

-.063

Other Rel.

-.036

-.024

-.014

-.032

.037

No Religion

.031

.005

.047

.025

.015

Internet

-.161*

-.050

-.055

-.012

-.080

Greed

.079

.058

.106*

.061

.052

Moral

.110*

.059

.175**

.037

.077

Control

.051

.001

.024

-.066

.012

Choice

.109*

.026

.057

.089

.042

Influence

.078

.058

.093

-.029

.106*

Fiduciary

.101*

.000

.128**

-.005

.060

Pressure

.032

-.047

.036

-.061

.045

No Wrong

-.031

.035

-.137**

.007

-.114*

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Zero-

109

Table 19. Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Perceived Seriousness of the National White-Collar Crime

Center Survey Scenarios Compared with Motor Vehicle Theft (N=408)

______________________________________________________________________________

1 2 3 4 5 6 7 8 9 10 11

Knowledge -.093 .090 .085 .061 .055 .044 .156** .129** .176** -.056 .032

“Truth” -.066 .120* .107* .047 .083 .020 .128* .155** .158** -.010 .101*

“Myth” .092 .034 .060 .011 .032 .044 -.076 -.006 -.062 .021 .016

_____________________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p <

.05, ** p < .01.

1=Burglary

5=Robbery

9=Market Rigging

2=Embezzlement

6=Hacking

10=Counterfeit Sales

3=Identity Theft

7=False Drug Label

11=Insurance Overcharge

4=False Charges

8=Espionage

Table 20. Zero-Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Perceived Seriousness of White-Collar Crime and Street

Crime (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Knowledge

.009

-.074

.086

-.012

.126*

“Truth”

.069

-.084

.101*

-.051

.101*

“Myth”

.012

-.046

-.069

-.035

-.101*

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, ** p < .01.

Zero-

110

Table 21. Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Choice of Prosecutorial Process (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Knowledge

.069

.038

.080

.050

“Truth”

.138**

.050

.015

.072

“Myth”

.007

-.015

-.116*

-.010

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, ** p < .01.

Table 22. Zero-Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Choice of Fine Amount (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Knowledge

.073

.038

.079

016

.121*

“Truth”

.065

.074

.083

.068

.087

“Myth”

.023

.003

.038

.017

-.023

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Table 23. Zero-Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Choice of Monetary Compensation Amount (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Knowledge

.061

.056

.144*

.131**

.024

“Truth”

.067

.074

.019

.121*

.069

Zero-

111

“Myth”

.021

.003

-.076

.017

.052

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Table 24. Order Correlations Between Knowledge about Elite Deviance, “Truth”

Acceptance, “Myth” Adherence and Prison Sentence Severity (N=408)

______________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Knowledge

.144**

-.049

.110*

-.027

.124*

“Truth”

.139**

.018

.102*

-.029

.137**

“Myth”

.014

.026

-.033

-.010

-.026

______________________________________________________________________________

Note. Correlation coefficients reported in all tables are Pearson’s r when using dichotomous

predictors and Spearman’s rho when using multinomial nominal and/or ordinal predictors. *p

< .05, **p < .01.

Zero-

112

113

CHAPTER FIVE:

SUPPLEMENTARY RESULTS

Preliminary analyses reported in chapter four revealed a series of noteworthy

considerations, including a positive relationship between knowledge about elite deviance and

sentiments (i.e., perceived seriousness and punitiveness) towards it. Conversely, those subjects

who were found to be less knowledgeable, less inclined to accept “truths” and more likely to

adhere to “myths” about white-collar crime were also less prone to perceive it as a serious social

issue compared with street crime, and were generally more lenient in their prescribed sanctions

against it. Further, dissensus in both knowledge and attitudes about elite deviance in this sample

was found to stem from sociodemographic differences in gender, age, race/ethnicity, education,

socio-economic status, and attribution style. Phrased differently, those who exhibited relatively

more tolerance for white-collar offenses tended to be male, younger, white, less educated yet

wealthier, and to support the notion that elite deviance results from situational rather than

dispositional factors.

Perhaps more importantly, considerable politico-religious divergences were suggested by

the lesser likelihood of politically conservative subjects, Republicans and conservative

Protestants to rate the instances of white-collar crime presented to them as more serious than

traditional offenses and to seek more stringent sanctions against them. As previously mentioned,

empirical research has established that these individuals are usually more critical of and punitive

against traditional crime than are their more liberal counterparts (Grasmick et al., 1993; Unnever

et al. 2005). Further, they have also been found to be more supportive of free market economic

policies (Gallup Poll, 2012), whose deregulation has been showed to facilitate the commission of

114

white-collar crime (Lynch & Michalowski, 2006). The goal of this chapter is to investigate the

hypothesis that knowledge and attitudes about elite deviance (i.e., perception of offense gravity

and punitiveness) vary as a function of indicators of support for capitalism. To this effect,

multivariate regression models were run to determine whether differences found at the bivariate

level would persist after a more stringent analysis.

In the following regression models, reference groups include White for race, Republican

for political affiliation, and conservative Protestant for religious identity. That is, these three

suppressed categories are coded 0 on their respective dummy variables, which allows for

comparing their impact on knowledge and sentiments about white-collar crime relative to other

races, political parties and religions, with all other variables controlled. In each analysis,

standardized regression coefficients (betas) and their corresponding significance levels are

presented. Importantly, with no variance inflator factor greater than 4, excessive multicollinearity

could not be detected.

Knowledge about Elite Deviance, “Truth” Acceptance and “Myth” Adherence

Table 25 presents the regression analysis summary for sociodemographic predictors of

knowledge about elite deviance, “truth” acceptance, and “myth” adherence. These models

predicted the outcome with limited success, with adjusted R2 ranging between .110 and .160.

Further, as was the case with correlation coefficients, statistically significant betas are both

scarce and weak. Compared to Whites, non-Black subjects who reported belonging to “other

racial” groups were less likely to be knowledgeable about elite deviance (b = -.153), as were

Hispanics relative to non-Latinos (b = -.129).

115

Conversely, education and belonging to “other” religions or no religion at all (compared

with conservative Protestants) significantly increased the likelihood of having knowledge about

white-collar crime (b = .268, .162, and .159, respectively). Similar results emerged as far as

“truth” acceptance. More specifically, compared to Whites, members of other races were less

likely to accept relevant information about white-collar crime (b = -.118), as were Hispanics (b =

-.138), and more politically conservative subjects (b = -.142). Once again, however, more

educated participants as well as those belonging to other religious groups were more likely to

agree with empirically validated facts about elite deviance (b = .162 and .137, respectively).

Lastly, those more likely to adhere to “myths” about white-collar crime were males (b = .183)

and Blacks (b = .112), whereas those less likely to espouse such unfounded beliefs included

subjects who used the Internet as their main source of information (b = -.204), as well as those

inclined to cite negative business environmental influences as a leading cause of elite deviance (b

= -.120).

In summation, while several associations were no longer statistically significant, a

number of sociodemographic divergences persisted after running the regressions. Once again,

Whites and better-educated individuals were found to be more knowledgeable about elite

deviance, a disparity that may find its root in racial and ethnic gaps in educational attainment

(U.S. Census, 2012). Conversely, politically conservative subjects and conservative Protestants

proved to be less knowledgeable than their liberal counterparts. It therefore appears that political

and religious affiliations supportive of capitalism are also significant predictors of lack of

knowledge about elite deviance. What remains to be seen is whether such discrepancies are

reflected in subjects’ attitudes about white-collar crime.

116

Perceived Seriousness of White-Collar Crime vs. Auto Theft

Table 26 presents the regression analysis summary for sociodemographic predictors of

perceived seriousness of the National White-Collar Crime Center survey scenarios compared

with motor vehicle theft, controlling for knowledge about elite deviance. There was considerable

difference in the percentage of variance explained in each model, with adjusted R2 ranging from

.040 to .088. Males were less likely to rate embezzlement as more serious than motor vehicle

theft (b = -.162), as were those prone to believe that white-collar offenders see no wrong in their

actions (b = -.108), unlike more politically conservative subjects (b = .158) and those inclined to

blame elite deviance on pressure to succeed (b = .229). Identity theft was rated as more serious

than car theft by subjects with higher income levels (b = .134), and by those who believed that

white-collar crime results from negative business environment influences (b = .077). Males were

also less likely to deem the false charges scenario more serious (b = -.177), unlike those inclined

to agree with the idea that pressure to succeed causes elite deviance (b = .200). Those currently

employed evinced less perceived seriousness toward hacking (b = -.124), unlike older subjects,

Blacks and members of other races (b = .208, .154, and .114, respectively).

Only those who attributed elite deviance to greed rated the false drug label scenario as

more serious than auto theft (b = .171). While wealthier subjects rated market rigging as less

serious than car theft (b = -.105), the perception of offense gravity was reversed among older

subjects (b = .106), Catholics (again, relative to conservative Protestants, b = .191), and those

inclined to blame white-collar crime on negative business environmental influences (b = .142).

Lastly, several significant differences could be observed for the vignette describing counterfeit

sales. More specifically, older subjects and urban residents deemed it more serious than the

baseline crime (b = .130 and .105, respectively). Conversely, moderate Protestants rated it as less

117

serious than did their more conservative counterparts (b =-.152), a finding echoed among

subjects from other religious groups (b = -.158) and those with no religion (b = -.243). In

summation, compared with their respective counterparts, younger, wealthier as well as politically

and religiously conservative participants exhibited less perceived seriousness about these

examples of elite deviance than they did toward instances of street crime and petty white-collar

crimes.

Slightly similar findings emerged after controlling for the acceptance of “truths” and the

adherence to “myths” about elite deviance. Table 27 presents the regression analysis summary

for sociodemographic predictors of perceived seriousness of the National White-Collar Crime

Center survey scenarios compared with motor vehicle theft, controlling for “truth” acceptance

and “myth” adherence. As was previously the case, these models do not account for a large

percentage of the variance, with adjusted R2 ranging from .045 to .093. Perhaps because

embezzlement is not an organizational offense and profits individual employees rather than the

company, it was once again rated as more serious by politically conservative subjects (b = .160)

as well as those who attributed such crime to pressure to succeed (b = .234), but not by those

who believed white-collar offenders see no wrong in their practices (b = -.122).

While males were less likely to rate identity theft as more serious than car theft (b =

.108), those with higher income levels and who blame elite deviance on pressure to succeed

deemed it more serious (b = .117 and .171, respectively). Men were also found to be less critical

of false charges (b = -.190), unlike those prone to attribute white-collar crime to outward

influences (b = .068). Several associations reached statistical significance in regard to hacking.

More specifically, whereas currently employed respondents found it to be less serious than car

theft (b = -.127), males, older subjects, Blacks and Hispanics expressed the opposite opinion (b =

118

.059, .211, .149, and .107, respectively). As was the case before, attribution of blame to greed

was the sole significant predictor of perceived offense gravity in the false drug label scenario (b

= .174). Further, subjects who reported higher income levels once again rated market rigging as

less serious than car theft (b = -.115), unlike older participants (b = .109), Catholics (relative to

conservative Protestants, b = .189), and those prone to attribute white-collar crime to negative

business environmental influences (b = .142).

Results with respect to counterfeit sales were somewhat different after controlling for

“truth” acceptance and “myth” adherence. More specifically, while older subjects and urban

residents once again rated such offense as more serious than car theft (b = .129 and .106,

respectively), those belonging to other races (compared with Whites) and politically conservative

participants shared the same views (b = .101 and .015, respectively). On the other hand,

moderate Protestants were again found to perceive less seriousness than their more conservative

counterparts (b = -.159), an attitude mirrored by members of other religions (b = -.168), those

with no religion (b = -.254), and males (b = -.060).

To conclude, despite weak relationships, these results tend to confirm the few statistically

significant differences in perceived offense gravity between younger, wealthier as well as

politically and religiously conservative participants and their respective counterparts found in the

correlational analysis. We now turn our attention to the five scenarios describing injurious

whitecollar crimes and violent street crimes.

Perceived Seriousness of Injurious White-Collar Crimes vs. Violent Street Crimes

Table 28 presents the regression analysis summary for sociodemographic correlates of

perceived seriousness of white-collar crime and street crime. In an effort to reduce

multicollinearity, each scenario comprises two models. Knowledge about elite deviance is

119

included in model 1 whereas model 2 controls for “truth” acceptance and “myth” adherence.

These models perform relatively better than before, with adjusted R2 ranging from .037 to .148,

but can still be considered weak.

The only differences observed for the vignette describing consumer safety violations (i.e.,

through the manufacturing and selling of a defective and potentially hazardous toy) included age

and situational attribution. More specifically, older subjects were more likely to deem it a serious

offense (b = .213 and .214), unlike those inclined to believe white-collar offenders see no wrong

in their actions (b = -.109 and -.116). Age was also a significant predictor of perceived

seriousness of murder (b = .151 and .141), as was believing that elite deviance is caused by bad

moral character in model 2 (b = .116), contrary to knowledge about white-collar crime (b = -

.137) and “truth” acceptance (b = -.148).

More ambiguous findings emerged in regard to toxic dumping. While politically

conservative subjects were expectedly less inclined to rate it as a serious offense in model 1 (b =

-.177), a similar finding emerged among those who relied on the Internet as their main source of

information (b = -.138), an unexpected finding. Conversely, those unaffiliated with any political

party were more likely than conservative Protestants to perceive the release of deadly pollutants a

serious crime in model 1 (b = .134), as were those prone to link elite deviance to bad moral

character (b = .187, also in model 1). Lastly, while males and “myth” adherers were less inclined

to rate the asbestos exposure scenario as a serious offense (b = -0.76 in model 1 and -.115,

respectively), older subjects, Hispanics and “truth” accepters evinced higher perceived

seriousness in the second model (b = .212, .111, and .111, respectively).

In summation, it appears the largest differences in perceived seriousness of these five

crime scenarios were found in regard to age, race/ethnicity, political views and attribution style.

120

Once again, older subjects were more critical of white-collar crime whereas minority members

perceived street crime to be of lesser gravity. These attitudes are reversed with respect to

rightleaning subjects as well as those prone to believe white-collar offenders see no wrong in

their actions. Let us now move on to predictors of subjects’ punitiveness.

Choice of Prosecutorial Process

Table 29 presents the regression analysis summary for sociodemographic predictors of

choice of prosecutorial process against white-collar crime and street crime. Once again, these

models yield low adjusted R2 ranging from .051 to .059. Further, the models for the murder

scenario could not be run, as it was a constant (i.e., every single participants agreed that the

murderer should be tried in a criminal court). Whereas subjects who grew up in Northeastern

states were more likely to recommend a harsher prosecution style for those responsible in the

toxic dumping vignette (b = .149 and .150), more politically conservative subjects were less

punitive (b = -.211 and -.209), as were those who believed that white-collar offenders see no

wrong in their actions (b = -.139 and -.143). Several noteworthy divergences appeared in regard

to the appropriate prosecution style for rape. More precisely, older subjects in the second model

were less likely to recommend the rapist be tried in a criminal court (b = -.131), as were

Hispanics (b = -154 and -154), but also more educated participants and “myth” adherers in model

2 (b = -.177 and -.177, respectively).

The opposite view was held among those with no religious affiliation (b = .200 and .200),

as well as those inclined to believe white-collar offenders are otherwise law-abiding citizens who

do not think that their business practices are really wrong (b = .103 and .118). The latter group

and males were also less likely to require that those responsible in the asbestos exposure case be

tried in a criminal court (b = -.106 in model 2, and -.120 and -.127, respectively). Once again,

121

despite few and weak statistically significant relationships, consistent patterns continue to

emerge, with right-leaning participants and those more likely to find excuses for elite deviance

being less punitive toward white-collar crime relative to street crime.

Choice of Fine Amount

Table 30 presents the regression analysis summary for sociodemographic predictors of

choice of fine amount against white-collar crime and street crime. As was previously the case,

these models do not perform very well, with adjusted R2 ranging from .039 to .058. While those

currently employed opposed a fine against the company responsible for selling a potentially

dangerous toy (b = -.118 and -.129), the reverse position was observed in regard to religious

views, relative to conservative Protestants. More precisely, among those who supported a harsher

fine were Catholics (b = .154 and .154) and liberal Protestants (b = .113 and .113). Similar results

were found among those inclined to blame elite deviance on low self-control (b = .199 and .202).

In fact, the variable measuring attribution to impulsiveness yielded several significant

relationships in these models. For example, those individuals prone to accept that white-collar

offenders have difficulty controlling themselves were also more inclined to demand a higher fine

amount against the murderer (b = .109 and .119), along with males (b = .050 and .054), and those

disposed to believe white-collar offenders see no wrong in their actions (b = .112 and .110).

Similarly, those subjects attributing elite deviance to self-control were more likely to favor a

harsher monetary sanction against the company responsible for release deadly pollutants in a

river (b = .107 and .108), as were males (b = .116 and .109).

Taken together, these results parallel previous divergences in punitiveness. Males, older

subjects and those blaming elite deviance on low self-control were found to be the most

retributive (at least in their choice of fine amount). Although few differences could be detected

122

with respect to sanction options for the toxic dumping and asbestos exposure scenarios,

conservative Protestants were less likely than other religious groups to support a monetary fine

against the company guilty of distributing a potentially harmful toy. It could be that capitalism

supporters (weary of the firm’s financial well-being) may favor a punishment less harmful to

business. This includes either paying damages to the potential victims or their families (as in the

Ford Pinto case) or sentencing those responsible to prison. We now turn our attention to these

two options.

Choice of Monetary Compensation Amount

Table 31 presents the sociodemographic predictors of choice of monetary compensation

amount against white-collar crime and street crime. Once again, these models produced weak

adjusted R2 ranging from .039 to .058. Males were more likely to recommend the payment of

damages in the defective toy scenario (b = .121 and .126), as were older subjects (b = .117 and

.118), currently employed participants (b = .128 and .124), as well as those blaming elite

deviance on low self-control (b = .102 and .107) and fiduciary responsibility to shareholders (b =

.137 and .137). On the other hand, better-educated subjects deemed monetary compensation an

inadequate sanction (b = -.142 and -.130). Age and blame attribution to low self-control were

significant predictors in regard to the toxic dumping scenario (b = .171 and .172, and .115 and

.116, respectively), as was knowledge about elite deviance (b = .175), and with respect to rape (b

= .166 and .173, and .122 and .127, respectively). Once again, subjects more knowledgeable

about white-collar crime were more likely to support such sanction against the rapist (b = .140),

unlike “myth” adherers (b = -.008).

Lastly, the asbestos case created much dissensus in regard to whether paying damages to

the victims and/or their families is an appropriate form of punishment. More specifically, those

123

inclined to support this relatively more lenient sanction included males (b = .121 and .117), older

subjects (b = .169 and .173), Hispanics (b = .103 and .106), as well as wealthier subjects in

model 1(b = .106). Conversely, those less prone to favor such option were urban residents (b =

.103 and -.103), better-educated subjects (b = -.167 and -.163), and those attributing elite

deviance to negative business environmental influences (b = -.109 and -.108).

In summation, older subjects and those blaming low self-control as a leading factor of

white-collar crime were once again the most punitive. However, participants with higher

education levels were less likely to support monetary compensation against white-collar

offenders. Further, no significant differences emerged in terms of political and religious

affiliations. It is not certain, however, whether these individuals considered paying damages a

more lenient punishment or a financial burden to the company. Being admittedly the harshest

form of sanction, prison sentence might yield less equivocal results.

Prison Sentence Severity

Table 32 presents the regression analysis summary for sociodemographic predictors of

prison sentence severity against white-collar crime and street crime. Once again, these models

yield low adjusted R2 values that explain between only 5.6% and 6.6% of the variance. Fewer

statistically significant differences emerged this time. Compared to Republicans, those subjects

affiliated to no political party were less likely to favor a harsh prison sentence against those

responsible for distributing a potentially dangerous toy (b = -.206 and -.209), as were those who

used the Internet as their main source of information (b = -.154 and -.151), a somewhat odd

finding given that these individuals were also found to be less likely to espouse “myths” about

elite deviance. More expectedly, participants more knowledgeable about white-collar crime and

“truth” believers were more inclined to require a longer prison sentence (b = .135 and .127,

124

respectively). Lastly, further forms of dissensus emerged with respect to toxic dumping. More

specifically, while older subjects and “truth” believers were more inclined to support the

incarceration of those responsible for releasing deadly pollutants in a river (b = .189 and .193,

and .108, respectively), wealthier subjects as well as those who believe white-collar offenders see

no wrong in their actions were more lenient in their decision (b = -.108 in model 2, and -.118 and

-.123, respectively).

In summation, the very few statistically significant coefficients bring only partial support

to the hypothesis under examination. Although public dissensus in prison sentence severity

emerged, it was not due to politico-religious differences. Rather, more knowledgeable subjects

and “truth” believers were more inclined to require a harsher prison sentence for the white-collar

offenders described in the abovementioned scenarios. Conversely, participants who reported

higher income levels and those with situational attribution styles were less likely to recommend a

severe prison sentence for such offenses (all expected findings). It is not certain, however,

whether such tolerance displayed towards the perpetrators of elite deviance was motivated by

lesser perceived fear or, alternatively, disbelief in the effectiveness of the sanctions proposed to

them. These results and their implications are further discussed in the sixth and last chapter.

125

Tables

Table 25. Regression Analysis Summary for Sociodemographic Predictors of Knowledge about

Elite Deviance, “Truth” Acceptance, and “Myth” Adherence (Betas; N =408)

______________________________________________________________________________

Knowledge

“Truth” Acceptance

“Myth” Adherence

Male

.046

.095

.183***

Age

.048

-.009

-.088

Black

-.070

.051

.112*

Other Race

-.153**

-.118*

.041

Hispanic

-.129**

-.138**

.098

Northeast

-.065

-.059

.058

Urban

-.015

-.009

.013

Income

-.096

.037

.065

Education

.268***

.162**

-.020

Employed

-.028

.085

.092

Pol. Ideology

-.141

-.142*

.021

Democrat

-.035

.000

-.063

Other Party

.053

.125

-.139

No Party

-.025

-.038

-.094

Catholic

.056

.060

-.064

Mod. Protestant

.086

.127

-.084

Lib. Protestant

.003

.010

-.039

Other Religion

.162**

.137*

-.022

No Religion

.159*

.158

-.116

Internet

.074

-.013

-.204***

Greed

.018

.011

.057

Moral

.012

.060

-.052

Control

-.032

-.073

.065

Choice

.098

.038

.005

Influence

.070

.038

-.120*

Fiduciary

.035

.081

.056

Pressure

.095

.083

-.087

No Wrong

Intercept

.039

1.551

.094

-1.135

.091

2.099

Adj. R2

.160***

.130***

.110***

______________________________________________________________________________

Note. Reference categories include White for race, Republican for political affiliation, and

conservative Protestant for religious identity.

*p < .05, **p < .01., ***p < .001.

Table 26.

Regression Analysis Summary for Sociodemographic Predictors of Perceived Seriousness of the National White-Collar

Crime Center Survey Scenarios Compared with Motor Vehicle Theft, Controlling for Knowledge about Elite Deviance (Betas; N=408)

____________________________________________________________________________________________________________

Burglary

Embezzle

ID Theft

Charges

Robbery

Hacking

False Label

Espionage

Market

Counterfeit

Overcharge

Male

-.015

-.162*

-.089

-.177*

.062

.060

-.088

-.021

.098

-.055

-.158

Age

.000

-.080

-.025

-.047

.157**

.208**

.062

.074

.106*

.130*

.001

Black

.012

.061

.041

-.019

-.008

.154**

.037

-.021

.028

.005

-.028

Other Race

-.013

-.017

.010

.021

-.050

.114*

.011

.012

.017

.092

.011

Hispanic

.026

.011

.032

.025

-.043

.002

-.038

-.072

-.073

.079

.035

Northeast

.006

.079

.058

.097

.016

-.010

.002

.007

-.005

.012

-.043

Urban

.011

.031

-.037

.007

.052

-.010

-.045

-.037

.048

.105*

.056

Income

.103

.029

.134*

-.013

.087

.072

.007

-.036

-.105*

-.072

-.019

Education

-.078

-.002

-.054

.023

.043

.014

.052

-.029

.060

.008

-.025

Employed

.077

.013

-.063

-.035

-.042

-.124*

-.060

-.034

-.023

-.080

-.065

Pol. Ideology

.018

.158*

.083

.028

-.067

-.117

-.095

.023

.066

.005

.001

Democrat

-.121

.132

.087

.118

-.011

-.133

-.132

.033

.060

.033

.019

Other Party

-.149

.062

.002

.110

-.073

-.122

-.100

-.042

.079

.050

No Party

-.101

.022

-.004

.030

-.060

-.122

-.114

-.035

.023

-.044

-.038

Catholic

.104

-.027

.043

-.003

.026

.010

.044

.097

.191**

-.119

.068

Mod. Prot.

.046

.003

.112

-.027

.024

.001

.021

.010

.080

-.152*

.066

Lib. Prot.

.097

-.057

.034

-.039

.039

.006

.042

.074

.078

-.062

.057

Other Rel.

.017

.088

.039

.040

.059

-.009

-.007

.054

.120

-.158*

.041

No Religion

.152

-.034

.008

-.112

-.002*

-.110

.060

.054

.149

-.243*

.052

Internet

.073

.017

-.023

-.055

.021

-.050

.085

.036

.049

-.102

-.116

Greed

.079

.062

.066

.052

.018

.032

.171**

.080

.067

-.096

.070

Moral

.064

-.023

.004

-.080

.055

-.037

.036

.046

.052

.027

.028

Control

-.036

-.021

-.064

-.055

-.042

.026

-.086

-.089

-.020

.084

.058

Choice

-.036

-.005

.035

.085

-.015

.038

.098

-.044

-.018

.031

.010

Influence

-.022

.088

.077**

.058

-.056

.010

.054

.017

.142**

.053

.099

Fiduciary

-.036

.044

.042

.046

.077

-.036

.065

.132*

.074

.021

.020

Pressure

.034

.229**

.162

.200**

.086

.033

-.005

-.037

.061

.036

-.006

No Wrong

-.079

-.108*

-.019

-.043

.034

.055

-.087

-.045

-.030

.049

-.024

Knowledge

-.062

.083

.091

.041

.010

.051

.095

.114*

.084

-.026

.026

Intercept

3.022

2.196

1.758

2.424

1.846

1.827

3.144

4.135

1.567

2.267

3.156

Adj. R2

.011

.084**

.040*

.080**

.020

.075**

.088**

.022

.080**

.065*

.019

___________________________________________________________________________________________________________

Note. Reference categories include White for race, Republican for political affiliation, and conservative Protestant for religious

identity.

*p < .05, **p < .01.

Table 27.

130

Regression Analysis Summary for Sociodemographic Predictors of Perceived Seriousness of the National White-Collar

Crime Center Survey Scenarios Compared with Motor Vehicle Theft, Controlling for “Truth” Acceptance and “Myth” Adherence

(Betas; N=408)

____________________________________________________________________________________________________________

Burglary

Embezzle

ID Theft

Charges

Robbery

Hacking

False Label

Espionage

Market

Counterfeit

Overcharge

Male

-.031

-.182

-.108*

-.190*

.054

.059**

-.086

-.034

.099

-.060*

-.173*

Age

.005

-.069

-.012

-.038

.161**

.211*

.064

.082

.109*

.129*

.007

Black

.008

.042

.021

-.031

-.013

.149**

.030

-.039

.021

.005

-.039

Other Race

-.017

-.020

.003

.016

-.049

.107*

.008

.012

.015

.101*

.019

Hispanic

.013

.008

.024

.018

-.042

-.004

-.034

-.067

-.070

.088

.044

Northeast

.000

.076

.053

.093

.016

-.014

.003

.008

-.004

.016

-.040

Urban

.010

.030

-.038

.006

.051

-.011

-.045

-.038

.048

.106*

.056

Income

.105

.012

.117*

-.022

.083

.066

-.003

-.054

-.115*

-.071

-.028

Education

-.081

.004

-.043

.029

.040

.027

.062

-.023

.067

-.006

-.035

Employed

.074

-.005

-.080

-.045

-.048

-.127*

-.067

-.052

-.031

-.083

-.078

Pol. Ideology

.014

.160*

.081

.025

-.064

-.123

-.094

.029

.068

.015**

.012

Democrat

-.112

.133

.089

.120

-.009

-.134

-.138

.030

.055

.035

.020

Other Party

-.128

.063

.007

.117

-.073

-.119

-.111

-.054

.069

.044

.043

No Party

-.092

.031

.004

.036

-.056

-.122

-.116

-.030

.022

-.042

-.031

Catholic

.112

-.024

.048

.001

.026

.013

.042

.095

.189**

-.122

.065

Mod. Prot.

.059

.003

.116

-.022

.022

.006

.015

.001

.074

-.159*

.058

Lib. Prot.

.102

-.055

.037

-.037

.040

.007

.040

.073

.076

-.062

.057

Other Rel.

.019

.088

.044

.056

-.002

-.005

.051

.121

-.168*

.031

No Religion

.166

-.029

.018

-.103

-.003

-.102

.057

.049

.145

-.254*

.043

Internet

.090

.040

.002

-.038

.028

-.044

.085

.049

.051

-.103

-.106

Greed

.073

.058

.062

.048

.016

.032

.174**

.079

.069

-.097

.067

Moral

.074

-.025

.004

-.078

.054

-.037

.030

.039

.046

.024

Control

-.046

-.021

-.066

-.058

-.041

.024

-.081

-.082

-.015

.088

.063

Choice

-.040

-.001

.040

.087

-.015

.043

.104

-.039

-.013

.027

.009

Influence

-.011

.098

.090

.068**

-.054

.014

.053

.021

.142**

.050

.100

Fiduciary

-.038

.034

.041

.073

-.036

.063

.123*

.072

.017

.010

Pressure

.044

.234**

.171**

.207

.087

.039

-.007

-.038

.059

.031

-.010

No Wrong

-.085

-.122*

-.031

-.050

.028

.055

-.088

-.057

-.033

.044

-.037

“Truth”

-.073

.109

.088

.033

.036

.008

.091

.158**

.089

.042

.109*

“Myth”

Intercept

.109*

2.845

.074

2.227

.081

1.781

.067

2.392

.027

1.835

.011

1.875

-.036

3.275

.016

4.234

-.026

1.725

.002

2.264

.035

3.191

Adj. R2

.018

.093**

.045*

.082

.017

.070**

.086**

.031

.079**

.064*

.029

Table 28.

___________________________________________________________________________________________________________

Note. Reference categories include White for race, Republican for political affiliation, and conservative Protestant for religious

identity.

*p < .05, **p < .01.

131

Regression Analysis Summary for Sociodemographic Correlates of Perceived Seriousness of White-Collar Crime & Street

Crime (Betas; N=408)

____________________________________________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Male

-.015

-.027

.022

.035

-.070

-.065

.056

.066

-.076*

-.061

Age

.213**

.214**

.151**

.141*

.085

.084

.075

.070

.216

.212**

Black

-.065

-.070

.039

.059

.064

.063

.032

.042

.033

Other Race

.027

.037

-.022

-.017

-.016

-.013

-.072

-.076

-.010

-.008

Hispanic

.019

.029

-.084

.046

.054

-.106*

-.112*

.097

.111*

Northeast

.007

.011

.008

.010

.040

.044

.025

.023

.044

.051

Urban

-.082

-.081

-.063

-.062

-.074

-.073

-.009

-.094

-.093

Table 29.

Income

.066

.064

.006

.027

-.069

-.073

.059

.068

.035

.028

Education

-.038

-.054

.062

.049

.023

.024

-.035

-.032

.001

.008

Employed

-.078

-.086

.036

.055

.004

.001

-.006

.005

-.048

-.049

Pol.Ideology

-.092

-.080

-.089

-.090

-.177*

-.173

-.028*

-.036

-.093

-.089

Democrat

-.016

-.014

.014

.017

.009

.004

-.004

-.003

-.060

-.070

Other Party

.005

-.001

-.025

-.017

.080

.066

-.076

-.068

.003

-.022

No Party

-.087

-.082

-.056

-.060

.134*

.131

-.064*

-.069

-.075

-.084

Catholic

-.092

-.095

.048

.047

.065

.061

.074

.076

-.010

-.019

Mod. Prot.

-.095

-.103

.067

.072

.023

.014

.042

.049

-.029

-.044

Lib. Prot.

-.020

.046

.091

.089

-.051

-.050

-.010

-.015

Other Rel.

.047

.035

.085

.082

.030

.028

.049

.055

.060

.059

No Religion

-.048

-.059

.017

.016

.028

.019

.126

.132

.001

-.014

Internet

-.064

-.060

.034

.016

-.138*

-.144

.032

.024

-.043

-.058

Greed

.079

.077

.028

.093

.097

.013

.015

.020

.027

Moral

.016

.013

.110

.116*

.187*

.180

.090**

.094

.095

.084

Control

.017

.021

-.025

-.030

-.018

-.011

-.037

-.042

.033

.045

Choice

-.006

-.010

.075

.068

-.042

-.039

.033

.032

.086

.093

Influence

-.038

-.039

-.075

-.082

-.041

-.047

-.024

-.026

-.023

-.034

Fiduciary

.078

.071

.021

.030

.093

.091

.000

.008

.028

.029

Pressure

.059

.054

.090

.087

.063

.058

.033

.036

.014

.005

Table 30.

No Wrong

-.109*

-.116*

-.002

.009

-.067

-.086

-.077

-.064

-.060

Knowledge

-.019

-.137*

.053

-.046

.102

“Truth”

.069

-.148*

.075

-.099

.111*

“Myth”

Intercept

2.984

.024

2.975

3.318

-.026

3.249

2.543

-.051

2.657

3.772

-.016

3.772

2.805

-.115*

2.984

Adj. R2

.090**

.092**

.037*

.146**

.148**

.005

.010

.099**

.107**

____________________________________________________________________________________________________________

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p < .05, **p < .01.

132

Regression Analysis Summary for Sociodemographic Predictors of Choice of Prosecutorial Process against White-Collar

Crime & Street Crime (Betas; N=408)

____________________________________________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Male

-.016

-.019

-.048

-.052

.012

.038

-.120*

-.127*

Age

.003

.004

.025

.028

-.124

-.131*

.080

.084

Black

-.004

-.010

.021

.015

.034

.042

.066

.058

Table 31.

Other Race

-.060

-.052

.066

.030

.022

.032

.033

Hispanic

.060

.073

.039

.041

-.154**

.074

.076

Northeast

.057

.063

.149**

.150**

.021

.022

.096

Urban

.059

.060

-.050

-.069

-.036

Income

.064

.058

-.075

-.081

-.002

.001

-.007

Education

.032

.025

-.056

-.053

-.137*

-.117*

.007

.008

Employed

.013

.004

.036

.030

.043

.053

-.026

-.034

Pol.Ideology

-.120

-.109

-.211**

-.209**

.017

.007

-.048

-.045

Democrat

-.082

-.086

-.106

-.107

-.011

-.022

.114

Other Party

-.008

-.024

-.043

-.047

-.053

-.063

.122

.118

No Party

-.049

-.051

-.050

-.021

-.035

.059

.062

Catholic

-.119

-.126

.023

.022

.110

.107

.053

Mod. Prot.

-.017

-.030

.036

.034

.104

.103

.024

.020

Lib. Prot.

-.028

-.031

.019

.018

.112

.108

-.065

Other Rel.

-.030

-.039

-.058

.102

.114

-.018

-.021

No Religion

-.053

-.068

.004

.003

.200*

.028

.026

Internet

-.066

-.067

-.072

-.067

.000

-.018

-.028

-.020

Greed

.012

.013

-.007

.050

.058

-.002

-.003

Moral

.061

.053

.086

.084

.070

.065

.100

.097

Control

.000

.008

.015

.017

-.095

-.091

-.017

-.015

Table 32.

Choice

.015

-.031

-.029

.051

.060

.062

.064

Influence

.017

.012

.081

.082

-.052

-.061

-.001

.001

Fiduciary

.103

.097

.096

.093

.013

.024

.087

.082

Pressure

-.033

-.041

.006

.043

.041

-.004

No Wrong

-.032

-.039

-.139**

-.143**

.103*

.118*

-.099

-.106*

Knowledge

.045

.042

.077

.044

“Truth”

.115*

.055

-.012

.069

“Myth”

Intercept

2.363

-.030

2.440

2.844

.006

2.879

2.899

-.117*

2.924

1.988

.016

2.019

Adj. R2

-.004

.059**

.058**

.051**

.056**

.057**

.058**

____________________________________________________________________________________________________________

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p < .05, **p < .01.

133

Table 33.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Regression Analysis Summary for Sociodemographic Predictors of Choice of Fine Amount against White-Collar Crime &

Street Crime (Betas; N=408)

____________________________________________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Male

.011

.004

.050*

.054*

.116*

.109*

.047

.044

.041

.044

Age

.100

.106

.120

.118

.010

.017

.146*

.146**

.114*

.118*

Black

-.054

-.067

.021

.020

-.070

-.082

-.013

-.019

-.056

-.065

Other Race

-.002

-.006

-.018

-.011

-.048

-.055

.053

.062

-.009

-.015

Hispanic

.035

.034

.041

.054

-.036

-.040

.049

.064

.057

.055

Northeast

-.051

-.053

-.084

-.078

-.050

-.053

-.080

-.073

-.058

-.060

Urban

-.009

-.010

.018

.019

.004

.002

.008

.009

-.008

-.009

Income

.079

.066

-.021

-.025

.089

.076

-.045

-.052

.064

.053

Education

-.042

-.032

.016

.013

-.088

-.075

-.059

-.067

-.057

-.041

Employed

-.118*

-.129*

-.018

-.022

-.037

-.048

-.006

-.015

-.033

-.040

Table 34.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Pol.Ideology

-.059

.128

.136

.043

.039

.080

.092

-.042

-.047

Democrat

-.163

-.165

.108

.102

.057

.055

-.016

-.019

-.055

-.060

Other Party

-.068

-.072

.108

.090

.137

.001

-.017

.033

.027

No Party

.011

.014

.117

.113

.135

.137

.007

.042

.040

Catholic

.154*

.027

.020

-.001

.001

-.004

-.011

.066

Mod. Prot.

.093

.091

.064

.052

.085

.086

-.014

-.028

.131

.130

Lib. Prot.

.113*

.033

.030

.062

-.077

-.079

.039

.038

Other Rel.

.050

.052

.083

.077

.015

.020

.027

.016

.033

.040

No Religion

.137

.139

-.011

-.026

.006

.012

-.035

-.051

.106

.109

Internet

.034

.045

-.049

-.056

.065

.077

.003

.001

.006

.010

Greed

-.055

-.056

-.025

-.022

-.001

-.097

-.096

-.066

-.064

Moral

.078

.074

.062

.054

.090

.089

-.002

-.011

.096

.092

Control

.199**

.202**

.109*

.119*

.107*

.108*

.110*

.119*

.095

.098

Choice

-.055

-.049

.037

.038

-.041

-.035

.026

.024

.031

Influence

-.073

-.068

-.092

-.100

-.094

-.088

-.037

-.042

-.008

-.007

Fiduciary

.054

.049

.062

.059

.037

.033

.051

.044

.092

Table 35.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Pressure

-.067

-.065

-.096

-.104

-.060

-.055

-.105

-.114*

-.052

-.049

No Wrong

.004

-.003

.112*

.110*

.092

.087

.109*

.102*

.020

Knowledge

.091

.048

.087

.047

.104

“Truth”

.091

.094

.067

.126*

.067

“Myth”

Intercept

1.330

.016

-1.295

-.061

-1.100

-.424

.024

.587

.151

-.032

.319

-.716

-.023

-.388

1.544

Adj. R2

.058**

.056**

.039*

.044*

.040*

.014

.024

.027

.019

____________________________________________________________________________________________________________

< .05, **

134

Regression Analysis Summary for Sociodemographic Predictors of Choice of Monetary Compensation Amount against

White-Collar Crime & Street Crime (Betas; N=408)

____________________________________________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Table 36.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Male

.121*

.126*

.045

.027

-.021

.001

.046

.042

.121*

.117*

Age

.117*

.118*

.109

.120*

.171**

.172**

.166**

.173**

.169**

.173**

Black

-.039

-.043

-.082

-.098

.016

.013

-.079

-.094

.073

.065

Other Race

-.066

-.070

-.009

-.017

-.018

-.039

-.025

-.031

.001

Hispanic

.023

.025

.030

.018

.046

.036

.000

.103*

.106*

Northeast

-.096

-.095

-.087

-.094

-.029

-.034

-.076

-.077

-.075

-.074

Urban

-.057

-.024

-.025

-.041

-.042

.013

.012

-.103*

Income

.087

.079

.058

.045

.065

.053

.048

.030

.106*

.097

Education

-.142*

-.130*

-.021

-.010

-.107

-.065

-.045

-.028

-.167*

-.163*

Employed

.128*

.124*

-.031

-.045

.047

.048

.055

.041

.058

.050

Pol.Ideology

-.127

-.129

.041

.036

-.021

-.040

.031

.029

-.057

-.055

Democrat

-.022

-.028

.101

.105

-.066

-.077

.033

.028

-.026

-.028

Other Party

-.074

-.083

.129

.140

.014

.008

.022

.013

-.034

-.041

No Party

.024

.021

.060

.069

.103

.092

.016

.017

.005

.006

Table 37.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Catholic

-.065

-.066

-.026

-.019

.011

.014

.018

-.043

-.044

Mod. Prot.

-.004

-.008

.110

.118

.105

.110

.103

.098

.025

.020

Lib. Prot.

-.082

-.084

.024

.027

-.046

-.049

-.028

-.029

-.050

Other Rel.

-.087

-.083

-.017

-.011

-.076

-.053

.006

.011

-.062

-.063

No Religion

-.120

-.121

-.034

-.019

.049

.063

-.048

-.047

-.020

-.022

Internet

-.019

-.020

-.002

.022

-.027

-.032

.036

.046

.042

.047

Greed

.003

.006

.015

.010

.053

.061

-.009

-.007

.044

Moral

.011

.006

.061

.065

.056

.052

.048

.042

.066

.062

Control

.102*

.107*

.121*

.116*

.115*

.116*

.122*

.127*

.020

.023

Choice

-.007

-.001

.007

.011

.024

.041

.001

.010

.045

.049

Influence

-.047

-.048

-.071

-.057

-.015

-.014

-.066

-.062

-.109*

-.108*

Fiduciary

.137*

.061

.055

-.017

-.008

.041

.036

.048

.044

Pressure

-.076

-.077

-.066

-.056

-.059

-.052

-.019

-.017

-.050

-.051

No Wrong

.026

.027

.024

.014

.013

.026

.046

.041

.075

.071

Knowledge

.082

.057

.175**

.140*

.060

“Truth”

.062

.037

.024

.124

.077

Table 38.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

“Myth”

Intercept

.938

-.040

1.209

-1.192

.094

-1.288

-.732

-.087

-.215

-1.484

-.008*

-1.185

.544

-.002

.717

Adj. R2

.070**

.066**

.027

.032

.080**

.058*

.037*

.031

.060**

____________________________________________________________________________________________________________

< .05, **

135

Regression Analysis Summary for Sociodemographic Predictors of Prison Sentence Severity against White-Collar Crime &

Street Crime (Betas; N=408)

____________________________________________________________________________________________________________

Toy

Murder

Pollutants

Rape

Asbestos

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Model 1

Model 2

Male

.027

.029

.045

.035

.030

.027

-.013

-.018

-.015

-.021

Age

.079

.083

.099

.098

.189**

.193**

.063

.064

.051

.058

Black

.094

.083

-.110*

-.109*

-.006

-.016

-.050

-.051

-.052

-.068

Table 39.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Other Race

.058

.054

-.137*

-.124*

.017

.018

-.034

-.032

.065

.066

Hispanic

.024

.028

-.026

-.016

.068

.074

-.006

-.005

.028

.036

Northeast

.043

.044

-.046

-.041

.070

.072

-.056

.011

.014

Urban

.046

.045

.044

.045

.040

.010

.032

Income

-.020

-.035

.018

.022

-.098

-.108*

-.038

-.026

-.043

Education

.054

.069

-.047

-.070

.007

.012

-.133*

-.138*

-.031

-.024

Employed

-.053

-.063

.011

.009

-.021

-.031

-.084

-.086

-.057

-.073

Pol.Ideology

-.007

.049

.062

.013

.017

-.034

-.032

.057

.063

Democrat

-.168

-.175

-.126

-.122

-.022

-.026

-.127

-.125

.105

.099

Other Party

-.104

-.118

-.115

-.118

.029

.017

-.130

-.128

.092

.076

No Party

-.206**

-.209**

-.033

-.028

-.034

-.061

-.058

.049

.051

Catholic

-.046

-.049

.047

.043

.036

.033

.091

.113

.109

Mod. Prot.

-.124

-.132

.040

.033

-.105

-.113

.057

-.042

-.054

Lib. Prot.

-.033

-.035

-.052

-.051

-.053

-.055

.038

.039

-.034

-.036

Other Rel.

-.061

-.057

.024

.010

-.039

-.041

.027

.024

.022

.019

No Religion

-.012

-.015

.052

.040

.020

.014

.102

.023

.015

Table 40.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

Internet

-.154**

-.151*

.008

-.035

-.032

.040

.043

-.098

-.090

Greed

.027

.031

.055

.051

.062

.063

.061

.059

.068

.070

Moral

-.015

-.023

.010

.009

.026

.019

-.008

-.007

-.049

-.059

Control

.013

.021

.006

.009

.025

.032

-.067

-.068

-.006

.003

Choice

.023

.031

.017

.009

-.054

-.050

.054

.052

-.053

-.047

Influence

.075

.074

.106

.103

.102

.022

.023

.103

.104

Fiduciary

.056

.053

-.031

-.037

.062

.057

-.021

-.023

.022

.014

Pressure

.027

.025

-.066

-.073

.052

.049

-.090

.015

.011

No Wrong

-.013

-.016

.024

.016

-.118*

-.123*

.033

.030

-.144*

-.153*

Knowledge

.135*

-.074

.083

-.020

.130*

“Truth”

.127*

.023

.108*

.000

.167*

“Myth”

Intercept

-.463

-.042

-.134

3.807

.027

3.642

-.937

-.020

-.699

3.854

.022

3.771

.624

-.017

1.092

Adj. R2

.059**

.056**

.019

.013

.064**

.066**

.000

-.003

.024

.031

____________________________________________________________________________________________________________

< .05, **

Table 41.

Note. Reference categories include white for race, republican for political affiliation, and conservative protestant for religious identity.

*p p < .01.

136

137

CHAPTER SIX:

DISCUSSION

A growing body of research has evidenced formidable discrepancies between the

harmfulness of street crime and that of elite deviance (Knowlton et al., 2011; Landrigan et al.,

2002; Leigh, 2011; Lynch & Michalowski, 2006; Herbert & Landrigan, 2000; Rebovich &

Jiandani, 2000; Reiman & Leighton, 2010). More precisely, while traditional property offenses

such as burglary and theft cost the public about $18 billion each year (UCR, 2010), annual losses

due to white-collar crime (including various forms of fraud and health costs caused by

workrelated injuries and illnesses as well as environmental pollution) exceed a trillion dollars

(Knowlton et al., 2011; Landrigan et al., 2002; Leigh, 2011; Lynch & Michalowski, 2006).

The harms associated with elite deviance include physical harm as well. For example,

compared with the 14,000 people who lose their lives to murder and negligent manslaughter

every year (UCR, 2010), an estimated 300,000 die annually as a result of work-place related

accidental injuries due to the company’s negligence, illnesses caused by prolonged exposure to

toxic chemicals, toxic waste dumping and deadly pollutants, faulty consumer products, nefarious

and addictive substances (Herbert & Landrigan, 2000; Leigh, 2011; Lynch & Michalowski,

2006), as well as medical malpractice (Starfield, 2000).

However, in spite of these astounding differences, street crime continues to overshadow

elite deviance in the news media (Barak, 1994; Barlow & Barlow, 2010; Ericson et al., 1991;

Lynch & Michalowski, 2006; Lynch, Nalla & Miller, 1989; Lynch, Stretesky & Hammon, 2000),

the criminal justice system (Calavita, Tillman, & Pontell, 1997; Maddan et al., 2011; Payne,

138

Dabney, & Ekhomu, 2011; Tillman & Pontell, 1992) and even academia (Lynch, McGurrin &

Fenwick, 2004; McGurrin, Jarrell, Jahn & Cochrane, 2013). Surprisingly, scholarly efforts that

have investigated societal response to crimes of the powerful have limited their field of inquiry to

public opinions about white-collar crime (e.g., Huff, Desilets, & Kane, 2010; Kane & Wall,

2006; Rebovich et al., 2000; Schoepfer, Carmichael & Piquero, 2007, etc.). These studies have

provided valuable empirical evidence of a growing concern among Americans regarding the

danger posed by elite offenses. Their failure to include a valid measure of lay knowledge about

white-collar crime, however, significantly limits our ability to infer the extent to which the public

is familiar with the scope and magnitude of this social issue.

The present study sought to address such limitation by providing the first measure of

public knowledge about elite deviance. This project was designed to explore five research

questions that included (1) the extent of public information about white-collar crime, (2) whether

a gap exists between subjective (perceived) and objective (actual) knowledge, (3) the existence

of popular “myths” about elite deviance akin to public misconceptions regarding street crime

(e.g., crime being rampant, overly violent, etc.), (4) the sociodemographic correlates of

knowledge about white-collar crime, and (5) whether such knowledge is associated with attitudes

towards elite deviance. Such attitudes comprised (a) participants’ perceived seriousness of

financially costly and harmful white-collar crimes as well as property and violent street crimes,

and (b) respondents’ level of punitiveness, including their choice of prosecutorial process (i.e., by

non-legal means, in a non-criminal court or in a criminal court) and of punishment (i.e.,

monetary compensation, fine and/or prison sentence), and punishment severity (in dollar amount

and/or years in prison).

139

Methodology Followed and Key Findings

The subjects in this study were recruited on Amazon’s Mechanical Turk, a web service

that coordinates the supply and demand of human intelligence tasks such as social science

surveys. Four hundred and eight participants completed an online questionnaire that comprised

measures of respondents’ (1) sociodemographic characteristics, (2) subjective and objective

knowledge about elite deviance, (3) perceived seriousness of white-collar crimes compared with

a baseline property crime, (4) perceived seriousness of physically harmful white-collar crimes

compared to violent street crimes, as well as punitiveness with choice of prosecutorial process,

sentence determination, and sentence severity, and (5) choice of attribution style (i.e., perceptions

of white-collar offenders’ motives). Statistical analyses of the data collected via this instrument

provided the following answers to the five research questions:

1) Is the public informed about elite deviance? If it is, to which extent are Americans

informed about it?

Overall, participants’ level of information about elite deviance was low and erratic. While

they seemed knowledgeable about the meaning of the term “white-collar crime”, the reluctance

of some companies to invest in cleaner forms of energy, the calculated endangerment of

consumers for profit, and about corporate human rights violations abroad, they were found to be

rather uninformed about medical crime and the relative legal immunity enjoyed by elite offenders

compared with street criminals. Further, though prone to recognize the greater financial cost of

white-collar crime compared with traditional crime, they had difficulty estimating the true extent

of such disparity.

2) Is there a gap between the public’s subjective and objective knowledge about

whitecollar crime?

140

Subjects tended to overestimate their knowledge about white-collar crime. A comparison

of answer correctness and confidence, however, revealed a relative lack of certitude among

participants regarding their awareness of the problem, suggesting that the concept of elite

deviance and its various dimensions may still be arcane to many Americans. In fact, about 20%

of this study’s sample admitted having never received any kind of information about it.

3) Does the public hold common “myths” about elite deviance like they do regarding

street crime?

Despite their self-doubts regarding their acquaintance with the topic of white-collar

crime, respondents were not inclined to acknowledge hard-earned empirical evidence such as the

greater physical harmfulness of elite deviance over street crime and to recognize that some elite

offenses - which they admit are common in underdeveloped nations (e.g., human trafficking) -

can be committed in the United States with little to no legal repercussion for the perpetrators.

Such reluctance provides support for the hypothesis that the American public may harbor

“myths” about white-collar crime as they do regarding street crime.

4) What are the correlates of knowledge about white-collar crime?

There was significant variation among participants in their level of knowledge about elite

deviance, acceptance of “truths” and adherence to “myths” with respect to gender, race/ethnicity,

income, education, political ideology, religious affiliation, source of information, and blame

attribution style. Despite admittedly small effect sizes, more knowledgeable subjects were found

to be those who identified as Whites, with higher education levels, without any religious

affiliation, and who used the Internet as their main source of information. In comparison, less

knowledgeable participants and “myth” adherers turned out to be predominantly male, politically

more conservative, Republican, conservative Protestant, who relied on traditional media sources

141

rather than the Internet and who attributed white-collar crime to situational rather than

dispositional factors.

5) Is knowledge about elite deviance correlated with public opinion regarding

whitecollar crime?

Knowledge (and lack thereof) was associated with sentiments about elite deviance. More

specifically, less knowledgeable subjects (including men, those with higher income levels, more

politically conservative subjects, Republicans, conservative Protestants, and those who believed

white-collar offenders saw no wrong in their actions) were often more lenient in their attitudes

towards elite deviance, both in terms of perceived seriousness and punitiveness, compared with

street crime. By suggesting profound politico-religious dissensus in the best way to deal with

elite deviance, these findings stand in contrast with several studies on consensus theory which

concluded that widespread agreement exists among all members of society about perceived

seriousness of and response to crime (Blumstein & Cohen, 1980; Carlson & Williams, 1993;

Cullen et al., 1985; Heller & McEwen, 1975; Levi & Jones, 1985; O’Connell & Whelan, 1996;

Roth, 1978).

In fact, previous research has shown that politically conservative subjects, Republicans

and conservative Protestants tend to be more critical of traditional crime than are their more

liberal counterparts and to support tougher policies against street offenders (Grasmick et al.,

1993; Unnever et al. 2005). Further, these individuals have been found to evince more support

for elements of free market economic policies (Gallup Poll, 2012), whose deregulation has been

showed to facilitate the commission of white-collar crime (Lynch & Michalowski, 2006). More

stringent analyses were therefore run to test the hypothesis that perceived seriousness of and

punitiveness against elite deviance vary as a function of socio-demographic characteristics

142

indicative of support for capitalism, with limited success. Overall, age was a stronger predictor of

dissensus regarding the proper way to address the problem of white-collar crime than were

political and religious categories usually associated with pro-capitalism attitudes.

Implications of Findings

To the extent that the sample is representative of the American population, these findings

have significant implications. First of all, they suggest that the American public may not be

sufficiently informed about crime categories that are statistically more likely to harm them than

are traditional offenses. Who is to blame for such lack of information? Surely, the people’s

overreliance on media outlets more intent to report on street crime (Dowler, 2003; Roberts &

Doob, 1990; Surette, 1998) - perhaps due to pressure from corporate governance - does little to

educate them about the ubiquity and peril of elite deviance. If such explanation is relevant, how

then can we account for the fact that crimes of the poor are still given disproportionate academic

attention compared to elite malfeasance (Lynch, McGurrin & Fenwick, 2004; McGurrin, Jarrell,

Jahn & Cochrane, 2013), even when tenure guarantees protection to scholars whose research

may threaten powerful interests?

Could it be that the subversive political dimension of white-collar crime clashes with

conservative beliefs and values of mainstream scholars (Greenberg, 1976; Werkentin, Hofferbert,

& Bauerman, 1974)? Arguably, the reluctance of some criminologists to study the interwoven

relationships of class, structure, and crime - let alone accept them - may belie psychological

contradictions between trust in our capitalist economic system and inconvenient truths about its

negative consequences. Recall that subjects in this study whose sociodemographic characteristics

are usually indicative of support for neoliberal capitalism tended to overlook the seriousness of

elite deviance and to recommend relatively lenient sanctions against it. While the strength of

143

these relationships was admittedly weak, perhaps a direct measure of pro-capitalism attitudes

would have evidenced more robust associations.

In fact, support for capitalism might explain subjects’ rather surprising recalcitrance about

accepting information relative to the physical harmfulness of white-collar crime. Once again, a

majority of respondents perceived the kinds of deleterious activities in which the upper class

engages (e.g., consumer safety violation, toxic dumping, preventable work-related diseases,

medical crime, etc.) to be less injurious than street crimes such as murder and rape.

Hypothetically, supporters of capitalism should have no problem condemning offenses that are

usually committed by disreputable and unsuccessful individuals. Conversely, it may be especially

difficult for them to reconcile their admiration for economic prosperity with the unsettling reality

of American firms sacrificing human lives for profits.

Subjects’ resistance to information highlighting the physical harmfulness of elite deviance

may also have stemmed from a belief that human rights are a primarily Western construct

particular to capitalist societies (Schwab & Pollis, 2000). More precisely, participants’ refusal to

accept evidence of human trafficking or racial discrimination domestically - as in the case of

toxic dumping in African American neighborhoods - suggests the existence of a “myth” that the

United States offers greater protection to its citizens against all forms of crime (including elite

offenses) compared with non-capitalist nations and dictatorships. Recall that subjects in this

sample were critical of U.S. private military corporations such as Academi and willing to accept

information pertaining to the damage they inflict abroad in the name of profit. Yet, they were

also less inclined to acknowledge arguments that undermine American notions of justice and

equality before the law such as white-collar offenders’ relative legal immunity.

144

If sentiments about elite deviance are partly shaped by pro-capitalism attitudes, one may

doubt the effectiveness of scholarly and journalistic efforts meant to increase the dissemination

of knowledge about white-collar crime. Cognitive dissonance theory (Festinger, 1956) suggests

that individuals need to maintain a certain degree of balance between their cognitions (i.e.,

values, beliefs, attitudes, and opinions) and their behaviors. However, such balance may be

threatened by incongruent arguments causing psychological dissonance. More specifically,

messages that are both relevant and contradictory with our personal positions or actions may

result in mental discomfort. As a result, we strive to remedy any inconsistencies by realigning

incongruent messages to make them consistent with our original cognitions (Aronson, 1969;

Brehm & Cohen, 1962; Festinger, 1957; Festinger & Carlsmith, 1959; Kiesler & Pallak, 1976;

Wickland & Brehm, 1976).

Festinger (1957) proposes three types of cognitive dissonance resolution strategies. The

first is to alter the dissonant cognition. This means either accepting the dissonant element and

changing one’s cognitions or denying its validity and rejecting it. For example, pro-capitalism

individuals might dismiss empirical evidence about the greater threat of elite deviance compared

with street crime and the relative legal immunity enjoyed by white-collar offenders by invoking

methodological flaws, unfounded socialist rhetoric, or inflexible trust in this country’s criminal

justice system. The second strategy consists in restoring equilibrium by outnumbering the

dissonant element with more consonant examples so it no longer creates any dissonance. For

instance, the same individuals may retort that in the Ford Pinto case the company’s main priority

was to maximize profit and not to harm its customers, thereby presenting white-collar offenders

as responsible but not culpable. Finally, the third strategy is to accept the dissonant argument but

downplay its importance. This subtler alternative may imply admitting evidence about elite

145

deviance such as ecological damage while at the same time claiming that environmentally

harmful behavior is a necessary evil in an otherwise prosperous economic system. If these

hypotheses were true, disseminating relevant information about white-collar crime might prove

ineffective among those already immune to such arguments.

Limitations and Avenues for Future Research

This study had several limitations. These notable impediments need to be addressed to

inform future research. First of all, the 2010 National White-Collar Crime Center survey included

twelve scenarios, one of which described overbilling (“A company overbills another company it

supplies with heavy equipment, making an extra $10,000 in unwarranted profits.”). This scenario

was accidentally omitted in this dissertation’s analysis. Such omission, coupled with the fact that

Huff and colleagues did not report standard deviations for perceived seriousness of the twelve

crimes they asked their participants to compare with auto theft, precluded a test of statistical

differences between the 2010 and 2013 surveys.

Another limitation could be the arbitrary taxonomy used to categorize subjects in regard

to their level of knowledge about elite deviance (i.e., “truth” accepters, lucky guessers, “myth”

adherers, and honestly uninformed). Such typology does not capture the role unconscious

knowledge may have played among subjects who seemingly guessed correctly on the knowledge

questionnaire. In the field of cognitive psychology, unconscious knowledge is defined as

knowledge we have, and could very well be using, but are not aware of (e.g., Dienes, 2008;

Augusto, 2010). While this hypothesis is beyond the scope of the present study, future research

may want to refine the classification system used here by including measures of metaknowledge

(i.e., knowledge about knowledge, Dienes & Perner, 2002) about elite deviance to determine

146

whether “lucky guessers” might actually be already informed about white-collar crime while

being unable to specify how they acquired such information.

Replications and extensions of the present study should also strive to develop a much

better instrument comprising a greater number of dimensions of elite deviance. The knowledge

questionnaire used in this dissertation only included ten items, which is far from providing an

exhaustive review of a multifaceted construct like white-collar crime. A more comprehensive

scale therefore ought to incorporate multiple examples of economic domination, including more

technical and potentially less known offenses (e.g., price-gouging, price-fixing, insider trading,

strategic bankruptcy, anti-trust violations, etc.). Additionally, further aspects of medical crime

(e.g., medical negligence and malpractice, unnecessary operations, tests, and other procedures,

fraudulent billing, etc.) should be covered since it was found that subjects had difficulty

admitting its greater physical harm compared with homicides.

Moreover, in view of participants’ reluctance to recognize the fact that U.S. citizens are

not necessarily more protected here than is the case in non-democracies and dictatorships,

measures of knowledge about government control should include both crimes committed

internationally (e.g., destabilizing foreign nations through coups d’état, international law

violation, unlawful warfare and war profiteering, etc.), and offenses perpetrated domestically

(e.g., crimes of electioneering and usurpation of power, violations of individual civil rights such

as illegal surveillance by law enforcement agencies, denials of due process of law, political party

infiltration, etc.). Lastly, denial of human rights should also be more thoroughly covered by

incorporating items tapping labor exploitation, sexual harassment, and racial and gender

discrimination in the workplace.

147

Besides its admittedly unrefined measure of information relative to elite deviance, this

dissertation was further limited by the non-random sample used to collect data about the

American public’s knowledge and sentiments about white-collar crime. Far from being truly

representative of the overall U.S. population, the sample comprised a disproportionate number of

relatively well-educated white citizens. Moreover, those subjects were predominantly Democrats

and less likely to identify with any religious affiliation. Lastly, the age group was relatively

young, which may both explain respondents’ predilection for Internet-related activities and have

influenced their definitions and perceptions of elite deviance. Recall although subjects were

found to be less critical of crime in general compared with Huff and colleagues’ sample, older

participants perceived elite offenses to be of greater seriousness and deserving of a more severe

punishment. While such finding is in line with the curvilinear relationship between age and

perceived seriousness of crime (Schwartz et al., 1993), the restricted age group in the present

study impedes our ability to generalize its results.

Further, because of the crude manner in which participants’ profession was asked (i.e., via

an open-ended question), it was impossible to create a measure of occupation. Such limitation is

regrettable since research suggests that occupations that either provide experience about white-

collar crime (i.e., attorney, judge, prosecutor, scholar, journalist, etc.) or facilitate its commission

(e.g., business executive, medical doctor, etc.) influence perceptions of seriousness of elite

deviance (Cole, 1983; Frank et al. 1989; Hartung, 1953; McCleary et al., 1981).

Moreover, controlling for occupational prestige would have been extremely useful in

exploring the purported relationship between support for capitalism and attitudes about

whitecollar crime. Marx ([1867]1967) defined the structural model of capitalism in terms of

one’s relationship to the means and modes of production, with those being owned by an elite -

148

the bourgeoisie - who maximize their rate of surplus value (i.e., profit) by exploiting and

impoverishing the proletariat or working class. Even though wealthier respondents were

generally less critical of white-collar crime, it could not be determined whether class and power

differentials created dissensus regarding the best way to deal with elite deviance since

participants’ social position with respect to the means and modes of production remained unclear.

Consequently, future research should seek to include a measure of occupational prestige

along with other indicators of support for capitalism. Nevertheless, an Internet survey may be

poorly suited to attract the elite given the small monetary incentive offered and potentially

uncomfortable questions. In summation, although Amazon’s Mechanical Turk turned out to be an

acceptable way to recruit subjects, a serious application of Marxist theory to the topic of

knowledge and sentiments about elite deviance would necessitate a much better proxy for the

American public.

In fact, a sample with a wide spectrum of professional activities ranging from lay people,

criminal justice experts, and potential elite offenders may even require different questionnaire

sections with varying levels of complexity. Pilot testing of the instrument with undergraduate

students suggested that several questions had to be rephrased and specific terms omitted or

simplified because respondents did not even understand what was being asked. This implies that

an instrument with questions of increasing difficulty for every correct answer given might be

necessary and that an adaptive computerized assessment system may be a useful alternative to a

simple standardized test.

Perhaps the greatest limitation to the present study is the cross-sectional nature of the

project. Knowledge and sentiments about white-collar crime were only measured at one given

point in time. Although the intent was to examine the effect of knowledge on perceived

149

seriousness of and punitiveness against elite deviance, what remains to be seen is the effect of

exposure to relevant information on changes in such attitudes. Longitudinal studies on the effect

of knowledge on sentiments about capital punishment using a one-group pretest-posttest design

(e.g., Bohm, 1989, 1990; Bohm et al, 1990, 1991; Bohm & Vogel, 1991, 1994, 2002; Bohm et

al., 1993) have evidenced a decrease in support for the death penalty among subjects after taking

a course on the subject.

However, Bohm and colleagues (1994) noted that those participants who held retributive

attitudes were more likely to be immune to change, perhaps due to cognitive dissonance. While

they did not find support for such claim, Cochran and Chamlin (2005) have posited that multiple

previous “doses” of information might in fact enhance the effectiveness of a single exposure to

the topic. In his study on the impact of ethics education on business students’ perceptions of

white-collar crime, Kennedy (2010) suggested that such perceptions could be positively

influenced over time through continued education and training. Consequently, future research on

the relationship between knowledge and attitudes about white-collar crime should seek to

investigate the effect of prolonged education regarding the numerous ramifications of the

whitecollar crime construct, controlling for pro-capitalist cognitions as potential cognitive

dissonance resolution strategies.

To conclude, future replications and extensions of the present study may want to consider

a longitudinal design involving a pre-test of knowledge and sentiments about elite deviance (i.e.,

perceived seriousness and punitiveness) followed by multiple post-tests to study the effect of a

treatment (i.e., exposure to relevant information about white-collar crime) on changes in the

dependent variable, controlling for pro-capitalism attitudes. The treatment could consist of onsite

and/or online material - including both written and video documents - reporting empirical

150

evidence of the dangers posed by elite deviance. Importantly, it may be wise to ask participants

to justify their attitudes about white-collar crime via open-ended questions, particularly if

continuous exposure to pertinent information does not result in changes in perceived seriousness

and punitiveness.

As previously mentioned, the questionnaire should include a measure of occupational

prestige to establish participants’ social position in regard to the means and modes of production

as defined by Marx. This, coupled with a direct measure of attitudes about capitalism compared

with rival economic systems, would help test the hypothesis that knowledge and sentiments

about elite deviance vary as function of pro-capitalism attitudes. Nevertheless, ensuring the

cooperation of elite membership (particularly in view of the time investment which longitudinal

studies require) will necessitate a more ingenious and persuasive recruitment strategy than a

nonprobability Internet sampling method, and provide participants with greater monetary

compensation. It is not certain, however, which research-funding agency will agree to finance

such a project.

Despite evident challenges, those are but a few potential avenues for future research

about knowledge and sentiments regarding white-collar crime. The present study was purely

exploratory and as such should be considered the first step in a series of research projects meant

to unravel the true extent of public information about elite deviance and the effect of awareness

programs on perceived seriousness of and punitiveness against pernicious behaviors shrouded in

Students also viewed