Business Law paper

Odd Jobs, Bad Habits, and Ethical Implications: Smoking-Related Outcomes of Children’s Early Employment Intensity

Amy L. Bergenwall • E. Kevin Kelloway •

Julian Barling

Received: 4 December 2012 / Accepted: 8 May 2013 / Published online: 1 June 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Considerable interest has long existed in two

separate phenomena of considerable social interest, namely

children’s early exposure to employment outside of any

organizational, legislative, or collective bargaining pro-

tection, and teenage smoking. We used data from a large

national survey to address possible direct and indirect links

between children’s early employment intensity and smok-

ing (both intent and history) because of significant long-

term implications of the link between work and well-being

in a vulnerable population. Fifth to ninth grade children’s

(N = 19,018) informal employment intensity (number of

odd jobs) was related to both smoking history and smoking

intent, and these effects were partially mediated by the

amount of weekly spending money and self-esteem. The

number of one’s parents and friends who smoke separately

moderated the relationships between self-esteem and

weekly spending money, and smoking behavior. These

findings illustrate the complex consequences of young

children’s exposure to early employment, and emphasize

the need for research on the ethical implications of pre- and

early teenagers who are employed outside of the formal

employment context.

Keywords Self-esteem � Substance use � Young workers � Youth employment

Young workers continue to participate in the paid labor

force at an unprecedented rate (Bureau of Labor Statistics

2010; Kelloway and Barling 1999; Rothstein 2001). This

should perhaps not be surprising as adults and parents have

long valued and encouraged youth employment (see

Furnham and Thomas 1984; Greenberger and Steinberg

1986; Phillips and Sandstrom 1990), viewing early

employment intensity (the frequency and duration of work)

and experiences as a critical socialization process (Kello-

way and Harvey 1999) that leads to the development of

desirable personal characteristics such as responsibility,

maturity, and appropriate money management. Neverthe-

less, available data suggest that the outcomes related to this

vulnerable population are complex, with potentially

important implications both for the youth involved and

society at large (Kelloway and Barling 1999) and that raise

critical ethical issues. While young people may learn skills

through early employment that will be crucial for later

success (Mortimer et al. 1991), they are exposed to the

hazards of adult employment including issues of work

stress (Loughlin and Lang 2005), workplace aggression

(Dupre et al. 2006; Tucker and Loughlin 2006), and

occupational health and safety (Castillo 1999; Frone 1999);

employment intensity may increase the likelihood of

graduating high school, but it also decreases the probability

of pursuing post-secondary education (Lee and Orazem

2010), and can negatively affect academic performance

(see Frone 1999).

While research slowly accumulates on the nature and

effects of youth employment, the studies are dominated by

one feature that has limited knowledge of the phenomenon

A. L. Bergenwall (&) � J. Barling Queen’s School of Business, Queen’s University, Kingston,

ON K7L 3N6, Canada

e-mail: [email protected]

J. Barling

e-mail: [email protected]

E. K. Kelloway

Department of Psychology, Saint Mary’s University, Halifax,

NS B3H 3C3, Canada

e-mail: [email protected]

123

J Bus Ethics (2014) 122:269–282

DOI 10.1007/s10551-013-1743-x

of youth employment. Specifically, researchers continue to

focus almost exclusively on children between the ages of

15 and 24 (Barling and Kelloway 1999), and a paradox

exists: While most of the research on youth employment

focuses on children between the ages of 15 and 24, much of

employment is conducted by children younger than

15 years old. This omission might be a function of several

factors, including the false assumption that individuals

under the age of 15 do not participate in the labor market,

the fact that child labor is illegal in most jurisdictions

(Tucker and Loughlin 2006), ethical dilemmas for

researchers in accessing young children involved in legally

questionable employment, or researchers being unwilling

to address vulnerable populations operating at the edge of

legally questionable behavior both in and for organizations.

Whatever the reason, the omission is unacceptable and can

no longer continue: Youth employment does exist under

the age of 15 (Covell and Howe 1999; Ferman 1990;

Rothstein 2001). Young children are often engaged in

paper routes, agriculture, construction, or restaurant work,

where they are responsible to a formal employer who

represents an organization. More often, however, children

are engaged in yard work and baby-sitting, for example,

where their employment relationship is informal and work

is unstructured (Weller et al. 2003). This work is not being

performed by 18-year-olds protected by labor regulations,

but by 8-year-olds where it passes unnoticed, unregulated,

and barely understood.

Thus, continuing to exclude children below the age of

15 from research on early employment might have con-

siderable organizational, developmental, and ethical con-

sequences. First, despite the lack of visibility accorded to

this group in public policy initiatives and empirical

research, the number of children involved is considerable

and extensive. For example, pre-teenagers are active par-

ticipants in paid employment in the USA (Covell and

Howe 1999; Mortimer et al. 1990), and the United King-

dom (White and O’Donnell 2001). Moreover, this phe-

nomenon is widespread: Data from a large-scale

representative sample in the USA show that 56 % of a

representative sample of 3,008 sixth through eighth grade

school children were employed on a part-time basis

(Weller et al. 2003), working an average of 7.7 h per week,

with 4.9 % employed for more than 20 h per week. Ferman

(1990) reported that children as young as 8 years of age

were participants in economic activities. Even though these

‘‘hidden’’ workers remain unregulated, it is clear that

society needs and uses these workers, resulting in a highly

complex and ethically challenging context.

Second, young children involved in employment engage

in activities that potentially have significant short- (e.g.,

Frone 1999) and long-term socializing and developmental

effects (both positive and negative), thus leaving a

meaningful gap in our understanding of a potentially vul-

nerable group of workers. Formal employment is almost

invariably preceded by employment in the informal econ-

omy and short-term, temporary jobs in the formal econ-

omy. This is of significant concern, because children’s

early employment experiences (whether in the formal or

informal economy) shape later expectations about, and

reactions to, their work as adults in formal businesses

(Barling and Kelloway 1999).

Last, young children’s employment is not without

physical consequences. Weller et al. (2003) analyzed the

effects of employment intensity for 6th through 8th grade

school children. Of the 3,008 school children who were

employed, 773 (25 %) experienced a workplace injury.

Moreover, these injuries were not trivial: 30 % of the

injured school children were hurt badly enough to require

medical attention, with 40 % of those experiencing an

injury missing between 1 and 7 days of school, and 7 %

losing more than 8 days of school due to the injury.

While relevant data are scarce, what Weller et al.’s

findings highlight is how little we know about the nature

and consequences of young children’s early employment.

The majority of young pre-teen children, who are among

the most vulnerable members of our society, are being

exposed to employment that might have positive benefits

but could leave them physically and psychologically

compromised, without the formal legislative, organiza-

tional or collective bargaining protection routinely offered

to employees aged 15 and above. The care of vulnerable

groups presents a significant ethical challenge, and the goal

of this study is to provide additional data to guide

evidence-based policy decisions regarding children’s

employment.

The Current Study

One frequently investigated consequence of youth

employment that is of considerable social importance is

substance use (see Frone 1999, 2003; Mihalic and Elliot

1997; Valois et al. 1999; Wu et al. 2003). In this study, we

focus on the link between early employment intensity and

one particular substance, namely tobacco use. Several

factors motivated this specific focus. First, most research

on teenagers’ employment has focused on the quantity of

employment, invariably in terms of the number of hours

worked per week. Our measure of employment intensity is

the number of odd jobs per week in which the young

children were involved (e.g., yard work, baby-sitting, paper

route). Second, while it is illegal to sell tobacco to our

entire study population (young persons under the age of 18

in Canada; Tobacco Act of Canada 1997), it is arguably

one of the most likely substances to be available to young

270 A. L. Bergenwall et al.

123

children because of its price and accessibility relative to

other illicit drugs. Third, access to illicit drugs would

presumably be extremely difficult for young children.

Fourth, the long-term health effects of tobacco smoking are

well documented (Bartechhi et al. 1994; Mathers et al.

2006). Fifth, tobacco has been conceptualized as a ‘‘gate-

way drug,’’ as young people who begin smoking are more

likely to use other illicit drugs (Bailey 1992; Henningfield

et al. 1990), and are more likely to abuse alcohol (Ellickson

et al. 2001). Sixth, youth’s early tobacco use has been

related to several other behavioral and health outcomes

such as mental health problems (McGee et al. 2000),

depression and anxiety (Brook et al. 2002), frequent sleep

problems (Patten et al. 2000), and crime (Ellickson et al.

2001).

Research shows that cigarette smoking, and both alcohol

and illicit drug use, have been linked to youth employment

(Frone 1999). Aside from the immediate adverse social-

and health-related consequences of early substance use,

there is also evidence that adolescent employment may

result in substance use that persists into adulthood (Mihalic

and Elliot 1997). Most studies show a significant link

between youth employment intensity and cigarette and

substance use (Frone 1999; Ramchand et al. 2007; Wu

et al. 2003), and there is some evidence that this relation-

ship is stronger among younger teenagers (Breslin and

Adlaf 2002). Our study extends findings from prior

research by focusing on the possible link between early

employment intensity and smoking behavior in younger

children in grades five through nine—a group ignored in

this prior research.

Throughout our research, we focus on two aspects of

smoking behavior, namely smoking history (whether the

child had ever smoked previously), and future smoking

intention (among those children who had not previously

smoked). Previous research shows consistently that smok-

ing intentions predict smoking behaviors (Conrad et al.

1992). We hypothesize that:

H1a The number of odd jobs a child participates in will

be positively associated with prior smoking history.

H1b The number of odd jobs a child participates in will

be positively associated with intent to smoke in the future.

While consistent with prior research, the first hypothesis

merely proposes the existence of a link between young

children’s early employment intensity and tobacco smok-

ing; empirically based explanations of how employment

intensity affects behavioral outcomes remain sparse. In this

study, we investigate the processes underlying the link. To

do so, we first introduce the possibility of two different

mediators between young children’s early employment and

their smoking behavior, namely weekly spending money

and self-esteem. In doing so, we highlight the complexity

of this relationship, as these two mediators exert opposite

effects on smoking history and intentions. The possible role

of both of these mediators can be better understood by

placing them in the context of Jahoda’s (1982) notion of

the manifest and latent consequences of employment;

financial outcomes are the manifest functions of employ-

ment, and psychological outcomes (e.g., self-esteem, time

structure) constitute the latent functions of employment.

Like adults, young children would seek employment to

fulfill both their needs for financial and psychological

outcomes; in turn, how employment intensity relates to the

manifest and latent consequences of children’s employ-

ment will affect their smoking behaviors.

With respect to the mediating role of spending money,

we argue that there will be a positive relationship between

employment intensity and spending money. Compensation

for odd jobs completed by young children would most

likely be based either on the number of odd jobs, or on the

amount of time taken to complete the work. In either event,

there would be a meaningful link between employment

intensity and the amount of spending money available. This

is important, as cigarettes are costly, perhaps especially for

young children with their relatively limited access to

money. Thus, over-and-above any regular allowance, any

income earned could be available for the purchase of cig-

arettes; among adults there is a positive relationship

between income and substance use (Frone 2008). Perhaps

the most prosaic explanation for the hypothesized link

between young children’s early employment and smoking

behavior is that employment intensity, reflected in the

number of odd jobs completed, results in increased dis-

cretionary spending money. The availability of discre-

tionary spending money acts as a mechanism through

which early employment intensity influences cigarette use.

H2 Weekly spending money will partially mediate the

relationship between young children’s early employment

intensity and smoking behavior.

Our second explanation for the link between young

children’s early employment intensity and cigarette

smoking is consistent with Jahoda’s (1982) latent (or

psychological) consequences of employment. Jahoda

(1982) identified self-esteem as one of the major latent

consequences of employment. Self-esteem is associated

with adults’ employment and unemployment experiences

(Croghan et al. 2006). In addition, young adults and teen-

agers who are engaged in employment report higher self-

esteem than those who are not (e.g., Barling et al. 1995;

Prause and Dooley 1997), and adolescent employment

intensity is consistently associated with self-esteem (e.g.,

Conrad et al. 1992; Ploeger 1997). What gives this finding

added importance is that any benefits to self-esteem may be

Ethical Implications of Children’s Employment 271

123

long-lasting. Based on cross-sectional and retrospective

data, Kelloway and Thun (2006) reported a significant

correlation between the number of odd jobs held as a child

and self-efficacy as a young adult.

At the same time, adolescents’ poor self-esteem predicts

their involvement in illicit substance use. Voelkl and Frone

(2000) showed that even after taking demographic, per-

sonality, academic, and expectancy variables into account,

adolescents’ self-esteem was negatively associated with the

frequency of marijuana (but not alcohol or smoking) use. In

addition, analysis of a large longitudinal database showed

that adolescents’ low self-esteem predicted tobacco

dependence some 10 years later (Trzesniewkski et al.

2006). Thus, we hypothesize that:

H3 Self-esteem will partially mediate the relationship

between young children’s early employment intensity and

smoking behavior.

Thus far we have focused on how two mediating vari-

ables will aid in understanding the relationship between

young children’s employment intensity and their smoking

behavior. To understand how young children’s early

employment results in smoking behavior, we also accord a

moderating role to significant others that is consistent with

young children’s psychosocial development. Specifically,

both parents and peers are crucial influences on children’s

behavior, and are also among the most significant early

influences in the decision to initiate and maintain smoking

behavior (Simons-Morton et al. 2004). First, and consistent

with the tenets of social learning theory (Bandura 1973),

individuals learn from watching the behaviors enacted by

salient role models, and when they do not see those same

behaviors punished, they are more likely to enact the

behaviors themselves (Bandura 1965). It should not be

surprising then, that parents’ smoking is associated with

their children’s initiation and progression of smoking

(Conrad et al. 1992). From this, we suggest that parental

smoking status will moderate the relationships between

weekly spending money and self-esteem, and smoking

behavior. Specifically, in the presence of parents who

smoke cigarettes, the relationships between both discre-

tionary spending money and poor self-esteem, and smoking

behavior, will be greater than in the presence of non-

smoking parental models.

Second, as children develop, the influence of peer

groups increases (Harris 1995), and peers are a significant

influence on smoking behavior (Simons-Morton et al.

2004). Considerable research exists investigating what

aspects of peer groups influence individuals so powerfully.

Research has pointed to descriptive norms rather than

normative appeals as powerful influences over individual

behavior (Cialdini 2006; Goldstein et al. 2008). Within the

smoking context, this might help explain why exhortations

by authority figures about the negative personal effects of

smoking are not more effective. In contrast, descriptive

norms provide specific information about what behaviors

salient others engage in, and importantly, individuals do

not need to be aware of this influence for it to be effective

(Nolan et al. 2008), which might be especially important in

influencing children’s behavior. Thus, we suggest that

descriptive social norms will moderate the relationships

between both spending money and poor self-esteem, and

smoking behavior. Specifically, the relationships between

poor self-esteem and spending money, and smoking

behavior will be greater the higher the proportion of one’s

peer group that engages in smoking. With a larger pro-

portion of peers involved in smoking, excess spending

money is more likely to be used to purchase cigarettes as

the individual yields to social norms. A higher proportion

of peers involved in smoking might also impact the effects

of self-esteem on smoking; while self-esteem is generally

negatively related to substance use (e.g., Voelkl and Frone

2000), the effects may be lessened in the presence of peers

who smoke; when self-esteem is already low and an indi-

vidual is exposed to peer smoking, they are much more

likely to succumb to social influence (McGuire 1968).

Thus, we predict that:

H4a Parental smoking will moderate the relationship

between young children’s spending money and smoking

behavior, strengthening the relationship.

H4b Parental smoking will moderate the relationship

between young children’s self-esteem and smoking

behavior, weakening the relationship.

H5a The proportion of one’s peer group who smoke will

moderate the relationship between children’s spending

money and smoking behavior, strengthening the

relationship.

H5b The proportion of one’s peer group who smoke will

moderate the relationship between children’s self-esteem

and smoking behavior, weakening the relationship.

A visual representation of our model can be found in

Fig. 1.

Method

Respondents and Procedure

We used data from the Youth Smoking Survey (Statistics

Canada 2002), which provided interviews with over 19,000

students between Grade 5 and Grade 9, of a large and

nationally (Canadian) representative sample. Complete

details of the sampling procedure and data collection

272 A. L. Bergenwall et al.

123

methods are provided in Statistics Canada (2002). The

sample was selected through a three-level stratification

process in which the first stratification was to select school

boards within provinces, the second was to select schools

within school boards, and the third was to select classes

within each school. In total 1,070 classes in 982 schools

situated in 327 school boards participated in the study.

Approximately equal numbers were sampled from each

grade (i.e., Grade 5 = 3,544; Grade 6 = 3,717; Grade

7 = 3,725; Grade 8 = 3,960; Grade 9 = 4,072). The

sample was evenly split between male (50.1 %, n = 9,530)

and female (49.9 %, n = 9,488) respondents. Respondents

were drawn from all provinces of Canada, and the sample

is representative of the most populous areas of the country;

the source publication (Statistics Canada 2002) notes that

respondents in remote communities and those in small

(N \ 10) classes are underrepresented in the dataset.

Measures

All items for the current study were taken from the Youth

Smoking Survey (Statistics Canada 2002). (a) Employment

intensity was assessed with one item asking, on a weekly

basis ‘‘how often have you done odd jobs (a paper route,

babysitting, etc.).’’ Responses were on a 4-point scale

(1 = never; 2 = less than 1/week; 3 = 1–3 times/week;

4 = four or more times a week). This is comparable to a

measure used by Mortimer et al. (1994) categorizing hours

worked per week. (b) Weekly spending money was also

assessed with one item, which asked, ‘‘about how much

money do you usually get each week to spend on yourself or

to save? (include money from allowances and from jobs)’’;

(1 = $0–$9; 2 = $10–$24; 3 = $25–$49; 4 = $50–$74;

5 = $75 or more). (c) Children completed a five item

(a = .80) measure of self-esteem (e.g., ‘‘I like the way I am’’, ‘‘When I do something, I do it well’’) adapted for this survey

from the National Longitudinal Study of Children and

Youth). Each item was rated along a five-point response

scale (1 = false, 2 = mostly false, 3 = sometimes false/

sometimes true, 4 = mostly true, 5 = true). (d) Students

were asked about the smoking behaviors of their parents

(‘‘Does your father smoke?’’, ‘‘Does your mother smoke?’’

1 = Yes, 0 = No for each), from which the number of

parents who smoke was derived. (e) Students were also asked

to indicate the number of close friends they have, and the

number of close friends who smoke; we used these data to

determine the proportion of close friends who smoke.

(f) Last, two questions asked respondents about smoking.

The first asked, ‘‘Have you ever tried cigarette smoking, even

just a few puffs?’’ (Yes = 4,488; No = 14,530). Thereafter,

those who responded ‘‘no’’ to having tried smoking were

asked, ‘‘Have you ever seriously thought about trying

smoking?’’ (Yes = 1,459, No = 13,004). These questions

represented smoking history and smoking intent in the cur-

rent analyses. While the use of these single-item measures is

less than ideal, similar measures have been validated and

successfully used in similar youth smoking research (e.g.,

Gidwani et al. 2002; Patrick et al. 1994).

Finally, a number of control variables were used

throughout our analysis: gender, age, and province. Age

and gender are common control variables when predicting

youth smoking (e.g., Gidwani et al. 2002; Mermelstein and

the Tobacco Control Network Writing Group 1999). Geo-

graphic location has been shown to be a significant con-

textual factor in youth smoking initiation, thus province

was also controlled for (Frohlich et al. 2002).

Results

Table 1 presents descriptive statistics on the distribution of

child employment in the current sample. Approximately

62 % of the sample participated in odd jobs. Female students

held more odd jobs than did male students (M = 2.31,

SD = .96 vs. M = 2.14, SD = 1.03, t(18,884) = 11.79,

p \ 0.01), but reported less weekly spending money (M = 1.89, SD = 1.01 vs. M = 2.01, SD = 1.18,

t(14,529) = 6.59, p \ 0.01). As would be expected, employment intensity increased across grade level: Children

in elementary school (Grades 5 and 6; M = 1.99,

SD = 1.01) reported having fewer jobs than did children in

junior high (Grades 7–9, M = 2.37, SD = 0.96, t(18,

885) = 25.92, p \ 0.01). Table 2 presents the descriptive statistics and correlations for all study variables.

To assess the proposed hypotheses, we followed

Preacher et al.’s (2007) method for testing moderated

mediation using Hayes’s (2012) PROCESS computational

procedure in SPSS to simultaneously test both proposed

mediators and moderated mediators. The PROCESS pro-

cedure improves upon many of the recently highlighted

drawbacks of Baron and Kenny’s (1986) tests for

Proposed conceptual model

% Smoking Friends

Odd Jobs

Weekly Spending Money

Self-Esteem

Smoking Behavior

# Smoking Parents

Fig. 1 Proposed conceptual model

Ethical Implications of Children’s Employment 273

123

mediation (Hayes 2009; Zhao et al. 2010), produces con-

ditional indirect effects in moderated mediation models,

and bootstrapped confidence intervals, and detects and

implements logistic regression for dichotomous outcomes

(i.e., smoking history and intent). In each analysis, relevant

demographic variables were controlled. Smoking history

and smoking intent were assessed in separate models due to

the nature of the data. Mediation was tested in each using

bootstrapping with 1,000 bootstrapped samples.

Tables 3, 4, and 5 present the results of the analyses. All

variables were centered to control for multicollinearity

among interaction terms and improve the interpretability of

the results. Employment intensity was significantly and

positively associated with both smoking history and smoking

intent after controlling for sex, grade, and province, sup-

porting hypothesis 1 (Table 3). Early employment intensity

was positively related to both self-esteem and spending

money (Table 4), self-esteem was negatively related and

spending money was positively related to both smoking

history and intent in the final model (Table 5), and all con-

fidence intervals produced by bootstrapping for spending

money and self-esteem did not include zero for both smoking

history and smoking intent, supporting mediation (H2 and

H3).

We also assessed whether the mediating relationships

were moderated by the extent to which parents smoked,

and the proportion of the children’s friends who smoked.

Thus four separate interaction variables were included in

the final regression (see Table 5). Partial support for

hypotheses 4 and 5 was found. Parental smoking did not

moderate the mediating path of spending money for either

outcome, but did moderate the mediating path of self-

esteem for smoking intent (H4). Friend smoking did not

moderate the mediating path of self-esteem for either

outcome, but did moderate the mediating path of spending

money for both smoking history and intent.

The PROCESS computational tool probes the moder-

ated mediation by producing bootstrapped confidence

intervals and total effect sizes at varying combinations of

moderator levels. Results of this analysis appear in

Table 6. Illustrations of the directional impact of the sig-

nificant moderations appear in Fig. 2. The number of par-

ents who smoked moderated the relationship between self-

esteem and smoking intent, such that the negative rela-

tionship between self-esteem and smoking is lower when

the number of parents who smoke is higher. The proportion

of friends who smoked significantly moderated the rela-

tionships between spending money and both smoking his-

tory and intent, but in the opposite direction than was

expected: In both cases, the relationship between spending

money and smoking behavior was weaker (but still posi-

tive) when more of the friends smoked.

Discussion

Understanding young children’s decisions to engage in

early employment intensity or smoking are of considerable

social and health-related interest, as both early employment

intensity or smoking could exert significant immediate and

long-term developmental effects. Thus, our goal in the

current study was to investigate whether, and how, chil-

dren’s early employment intensity is associated with

smoking intention and history. The results of the current

study offer full support for the first three hypotheses. Early

employment intensity was positively related to smoking

history and smoking intent (H1); the greater the number of

odd jobs that children complete is positively related to the

probability that they have smoked in the past or intend to

smoke in the future. Mediated relationships were found for

both spending money (H2) and self-esteem (H3), as the

number of odd jobs was positively related to both spending

money and self-esteem, and spending money was posi-

tively related to smoking behavior and negatively related to

self-esteem. This begins to shed light on the highly com-

plex effects of early employment intensity, as odd jobs

maintained not only an overall positive relationship with

smoking behaviors, but also a mitigating mediating rela-

tionship with self-esteem. Hypothesis four was partially

supported, as parental smoking moderated the relationship

Table 1 Employment intensity by grade level and gender

Ever done odd jobs (% responding)

Never \1/week 2–3/week 4?/week N

Grade 5

Male 49.1 24.5 15.7 10.8 1,758

Female 45.0 28.1 17.0 9.9 1,758

Overall 47.1 26.3 16.3 10.3 3,493

Grade 6

Male 41.0 28.9 19.1 11.0 1,860

Female 30.8 34.6 23.6 11.0 1,813

Overall 36.0 31.7 21.3 11.0 3,673

Grade 7

Male 32.6 33.0 22.2 12.2 1,835

Female 17.9 36.3 31.7 14.1 1,871

Overall 25.2 34.7 27.0 13.1 3,706

Grade 8

Male 24.7 35.1 25.0 15.1 1,981

Female 12.4 38.1 34.9 14.6 1,967

Overall 18.6 36.6 30.0 14.9 3,948

Grade 9

Male 24.1 33.3 28.0 14.6 2,024

Female 12.2 39.4 35.5 12.8 2,043

Overall 18.1 36.3 31.8 13.7 4,067

274 A. L. Bergenwall et al.

123

Table 2 Intercorrelations for all study variables

1 2 3a 3b 3c 3d 3e 3f 3g

1. Grade

2. Sex 0.00

3. Prov

3a. Alb 0.00 -0.01

3b. Sask 0.02* 0.02* -0.09**

3c. Man 0.01 -0.01 -0.09** -0.09**

3d. Ont -0.02* 0.02* -0.12** -0.12** -0.12**

3e. Que -0.02* 0.00 -0.13** -0.14** -0.13** -0.18**

3f. NB 0.01 -0.01 -0.09** -0.10** -0.09** -0.12** -0.14**

3g. NS 0.00 0.00 -0.09** -0.10** -0.10** -0.13** -0.15** -0.10**

3h. PEI 0.00 -0.01 -0.07** -0.08** -0.07** -0.10** -0.11** -0.08** -0.08**

3i. NFLD -0.01 0.00 -0.09** -0.09** -0.09** -0.12** -0.14** -0.09** -0.10**

4. Smoke history 0.31** -0.01 -0.03** 0.01 -0.03** -0.07* 0.15** 0.00 0.01

5. Smoke intent 0.10** 0.00 -0.01 -0.02* -0.01 0.02* 0.03** 0.00 0.01

6. Weekly money 0.29** -0.06** 0.00 0.02* 0.00 -0.02* -0.05** 0.03** 0.00

7. Self-esteem -0.15** -0.06** 0.00 0.03** 0.01 -0.01 -0.03** -0.01 0.01

8. # Odd jobs 0.19* 0.09** -0.01 0.04** -0.01 0.00 0.02** 0.00 -0.01

9. # Parent smoke 0.02* 0.02* -0.01 0.02* -0.02** -0.04** 0.03** 0.02** 0.02**

10. % Friends smoke 0.27** 0.01 -0.06** -0.02** -0.02** -0.06** 0.13** 0.02** 0.01

Mean 7.07 – – – – – – – –

SD 1.42 – – – – – – – –

3h 3i 4 5 6 7 8 9 10

1. Grade

2. Sex

3. Prov

3a. Alb

3b. Sask

3c. Man

3d. Ont

3e. Que

3f. NB

3g. NS

3h. PEI

3i. NFLD -0.07**

4. Smoke history -0.04** 0.03**

5. Smoke intent -0.02 0.00 – a

6. Weekly money 0.04** 0.04** 0.21** 0.08**

7. Self-esteem 0.03** -0.01 -0.23** -0.16** -0.05**

8. # Odd jobs 0.03** -0.01 0.11** 0.04** 0.25** 0.01

9. # Parent smoke 0.00 0.03** 0.21** 0.05** 0.10** -0.11** 0.03**

10. % Friends smoke -0.03** 0.05** 0.48** 0.18** 0.22* -0.19** 0.10** 0.18**

Mean – – – – 1.95 12.13 2.23 0.52 0.12

SD – – – – 1.10 2.80 1.00 0.72 0.24

** p \ 0.01, * p \ 0.05 a

Correlation cannot be computed, children were only asked about intent if they had no smoking history

Ethical Implications of Children’s Employment 275

123

between self-esteem and smoking behavior (H4b); the

greater number of parents who smoke, the less impact that

self-esteem has on a child’s intention to smoke. Put another

way, parents who smoke significantly reduce the positive

effects of self-esteem, increasing the odds that their child

will smoke. Finally, hypothesis 5 was partially supported as

peer smoking moderated the relationship between spending

money and smoking behavior but opposite the hypothe-

sized direction (H5a)—the more smoking friends that

children have, the weaker the mediating effect of spending

money. These results, discussed further below, further our

understanding of the process underlying the link between

children’s early employment intensity and smoking inten-

tions and behaviors, and achieve considerable importance

given the large-scale nature of the nationally representative

sample, and the vulnerable population on which they are

based.

We began this research with the observation that

although very young children are engaged in economic

activities (e.g., Ferman 1990), little is known about the

contemporaneous or long-term consequences of these

experiences. The current results suggest that this form of

economic participation is an appropriate focus of future

research: Early employment intensity of young people

exerts complex effects, both positive (i.e., the enhancement

of self-esteem) and negative (i.e., the development of

smoking behavior). Moreover, these effects likely carry-

over into later adolescent and adult life in many cases,

suggesting the need to understand the nature and conse-

quences of early employment experiences more fully.

To the extent that early employment intensity (i.e.,

engaging in odd jobs) is associated with adverse behaviors,

the current data suggest that this effect is only partially

mediated by the amount of weekly spending money

available to children, pointing to the need to further

investigate the effects of early employment and how such

experiences may be linked to smoking and other

Table 3 Results of the logistic regression of employment intensity on smoking outcomes

Predictor History Intent

b SE Exp(b) b SE Exp(b)

Grade 0.595** 0.015 1.813 0.217** 0.021 1.242

Sex -0.080* 0.037 0.923 0.024 0.056 1.024

Province

Alberta 0.321** 0.091 1.378 0.006 0.130 1.006

Sask 0.609** 0.084 1.838 -0.113 0.130 0.893

Manitoba 0.312** 0.090 1.366 0.001 0.128 1.001

Ontario 0.104 0.081 1.110 0.296** 0.104 1.345

Quebec 1.359** 0.071 3.894 0.433** 0.104 1.542

NB 0.578** 0.085 1.783 0.195 0.122 1.216

NS 0.664** 0.083 1.943 0.213 0.120 1.237

PEI 0.141 0.103 1.152 -0.067 0.145 0.935

NFLD 0.823** 0.084 2.278 0.237 0.126 1.268

Number of

odd jobs

0.138** 0.019 1.148 0.074** 0.028 1.077

Constant -1.409** 0.021 0.244 -2.172** 0.028 0.114

* p \ 0.05; ** p \ 0.01

Table 4 Results of the regression on mediators

Predictor Mediator: self-esteem Mediator: money

History Intent History Intent

b SE b SE b SE b SE

Grade -0.324** 0.018 -0.226** 0.019 0.196** 0.007 0.166** 0.007

Sex -0.353** 0.048 -0.120** 0.052 -0.167** 0.018 -0.139** 0.020

Province

Alberta -0.133 0.110 -0.039 0.115 0.065 0.041 0.082 0.043

Sask 0.126 0.107 0.250* 0.113 0.041 0.040 0.056 0.043

Manitoba -0.046 0.110 0.104 0.115 0.069 0.041 0.047 0.043

Ontario -0.237* 0.095 -0.221* 0.097 0.012 0.036 0.010 0.037

Quebec -0.395** 0.090 -0.193 0.100 -0.089** 0.034 -0.100** 0.037

NB -0.205 0.107 0.149 0.113 0.131** 0.040 0.130** 0.043

NS -0.110 0.103 0.011 0.110 0.045 0.039 0.052 0.041

PEI 0.167 0.123 0.264 0.127 0.149** 0.046 0.116* 0.048

NFLD -0.350** 0.106 -0.028* 0.114 0.212** 0.040 0.186** 0.043

Number of odd jobs 0.141** 0.025 0.181** 0.027 0.238** 0.009 0.222** 0.010

Constant 0.061* 0.024 0.388** 0.026 -0.017 0.009 -0.098** 0.010

* p \ 0.05; ** p \ 0.01

276 A. L. Bergenwall et al.

123

developmentally significant outcomes. The effects of

spending money on both smoking history and intention

were moderated by the proportion of a child’s friends who

smoke, suggesting that children are less likely to spend

their money on cigarettes when their friends smoke. While

in the opposite direction than what was predicted, we offer

the following explanation based on a post-hoc analysis:

The Youth Smoking Survey included information on

children’s sources of cigarettes, and not surprisingly, the

most common source of cigarettes was purchasing them

from friends (30 %), followed closely by being given

cigarettes by friends (27 %). It could be that the larger

proportion of friends who smoke, the greater the avail-

ability of free cigarettes, while those with a smaller pro-

portion of peers who smoke must purchase them from other

sources. Either way, these results emphasize the strong

effects that a children’s peer group has on their behaviors.

The results also emphasize the importance of early parental

socialization of responsible monetary management for

children (Kelloway and Harvey 1999), and suggest that

monitoring the amount of weekly spending money and how

it is used, might be important from a preventive and

intervention perspective.

The findings of this study also provide some data rele-

vant to the widespread assumption (Goodnow 1988; Kel-

loway and Harvey 1999) that economic involvement has

beneficial effects for children. In the current case, early

employment intensity was associated both with positive

and negative outcomes. Clearly, any policy initiatives on

the topic need to be guided more by research findings than

assumptions.

Despite the potential social significance of these find-

ings, these results may underestimate the effects of early

employment on smoking, as intensity reflects only one

dimension of early employment; and there are some sug-

gestions that employment intensity may not be the most

important dimension. Specifically, youth employment

varies not only in terms of its intensity (number of hours or

odd jobs) but also in terms of the quality of employment

experiences (e.g., role stressors, exposure to leadership,

injuries) as is clear from both empirical research (Barling

and Kelloway 1999; Frone 1999; Mortimer et al. 1992) and

Table 5 Results of full logistic regression analysis

Predictor Full model

History Intent

b SE Exp(b) b SE Exp(b)

Grade 0.423** 0.020 1.527 0.133** 0.027 1.142

Sex -0.214** 0.050 0.808 0.012 0.070 1.012

Province

Alberta 0.502** 0.119 1.652 0.177 0.159 1.194

Saskatchewan 0.630** 0.113 1.878 0.037 0.162 1.038

Manitoba 0.292* 0.123 1.339 -0.065 0.168 0.937

Ontario 0.103 0.108 1.108 0.397** 0.131 1.487

Quebec 1.152** 0.097 3.164 0.304* 0.135 1.355

New Brunswick 0.322** 0.118 1.380 0.19 0.155 1.209

Nova Scotia 0.514** 0.111 1.672 0.184 0.151 1.203

PEI 0.038 0.139 1.039 -0.169 0.190 0.844

Newfoundland 0.475** 0.114 1.608 0.335* 0.151 1.397

Number of parents who smoke 0.517** 0.034 1.678 0.154** 0.050 1.166

Proportion of friends who smoke 3.359** 0.113 27.770 2.045** 0.163 7.728

Number of odd jobs 0.079** 0.026 1.082 0.084* 0.036 1.087

Self-esteem -0.138** 0.009 0.871 -0.181** 0.012 0.834

Weekly spending money 0.200** 0.023 1.222 0.145** 0.033 1.156

Parents 9 self-esteem -0.008 0.012 0.992 0.044** 0.017 1.045

Parents 9 money -0.039 0.028 0.962 -0.007 0.043 1.007

Friends 9 self-esteem -0.073 0.038 0.930 0.034 0.054 1.035

Friends 9 money -0.217* 0.085 0.805 -0.312* 0.127 0.732

Constant -1.498** 0.029 0.224 -2.081** 0.037 0.125

* p \ 0.05; ** p \ 0.01

Ethical Implications of Children’s Employment 277

123

anecdotal reports (Schlosser 2002). Thus, employment

intensity—as reflected by work hours or number of odd

jobs is an incomplete and uni-dimensional reflection of

young children. Odd jobs completed by young children are

also likely to be characterized by minimal training with

potentially large variations in skill use and responsibility

(e.g., baby-sitting), isolation from peers, and working

before sunrise or after sunset (e.g., newspaper delivery). In

short, young children most likely engage in low-quality

jobs for which they receive inadequate preparation and

minimal regulatory protection. Thus, any future research

linking early employment intensity with smoking behavior

might beneficially focus both on early employment inten-

sity, quality, and form of employment. Furthermore, we

emphasize again that this work is largely completed in the

informal economy, where it escapes any and all regula-

tion—the negative outcomes associated with early

employment further highlight the need for organizational

and governmental intervention.

Why has research continued to emphasize employment

intensity at the expense of the quality of the employment

experience? One possible reason is that most prior research

has been conducted by developmental, social, or clinical

psychologists, whose focus is primarily on the outcomes of

early employment (e.g., Greenberger and Steinberg 1986).

In contrast, when research has been conducted by

organizational psychologists (e.g., Barling et al. 1995,

2002; Dekker et al. 1998; Frone 1998, 2000, 2003;

Loughlin and Barling 1999), the focus has understandably

turned more to the quality of employment experiences

given the centrality of subjective work experiences within

organizational psychology. Irrespective of the reason,

future research should focus both on early employment

intensity and experiences, especially because the quality of

employment experiences may be a better predictor of

mediators such as self-esteem than is employment

intensity.

In addition to focusing on employment experiences,

several questions remain open to future research. First,

given the broad array of developmental outcomes linked to

adolescents’ employment (see Frone 1999), understanding

whether young children’s employment intensity or quality

influence academic performance or counterproductive

behaviors (e.g., engagement in substance abuse) should be

investigated. Second, the nature of the odd job might make

a difference: Some odd jobs such as baby-sitting utilize

interpersonal skills and require high levels of responsibil-

ity; others (e.g., lawn mowing, snow shoveling) involve

social isolation and monotony. Third, parents serve as more

than external models from whom children learn: the quality

of parent–child communication regarding tobacco use also

influences children’s decisions involvement in tobacco use.

Table 6 Conditional indirect effects at values of moderators

Parents smoke

value

Friends smoke

value

Mediator: self-esteem Mediator: spending money

Effect Boot SE Boot LLCI Boot ULCI Effect Boot SE Boot LLCI Boot ULCI

Outcome: smoking history

-0.715 -0.239 -0.016 0.004 -0.025 -0.010 0.067 0.010 0.048 0.086

-0.715 0.003 -0.019 0.004 -0.027 -0.012 0.054 0.008 0.038 0.068

-0.715 0.245 -0.021 0.005 -0.032 -0.013 0.042 0.010 0.024 0.061

0.009 -0.239 -0.017 0.004 -0.025 -0.011 0.060 0.008 0.043 0.075

0.009 0.003 -0.020 0.004 -0.028 -0.013 0.047 0.005 0.037 0.058

0.009 0.245 -0.022 0.005 -0.032 -0.014 0.035 0.008 0.021 0.050

0.734 -0.239 -0.018 0.004 -0.027 -0.011 0.053 0.010 0.035 0.072

0.734 0.003 -0.020 0.004 -0.029 -0.013 0.041 0.007 0.028 0.056

0.734 0.245 -0.023 0.005 -0.033 -0.015 0.028 0.009 0.013 0.045

Outcome: smoking intent

-0.761 -0.220 -0.040 0.007 -0.057 -0.028 0.046 0.012 0.024 0.070

-0.761 -0.064 -0.039 0.007 -0.054 -0.027 0.036 0.010 0.015 0.056

-0.761 0.091 -0.038 0.007 -0.052 -0.026 0.025 0.011 0.005 0.047

-0.081 -0.220 -0.035 0.006 -0.048 -0.024 0.047 0.010 0.028 0.068

-0.081 -0.064 -0.034 0.006 -0.045 -0.023 0.037 0.008 0.021 0.051

-0.081 0.091 -0.033 0.006 -0.045 -0.022 0.026 0.008 0.010 0.041

0.600 -0.220 -0.029 0.006 -0.043 -0.019 0.049 0.012 0.025 0.072

0.600 -0.064 -0.029 0.006 -0.040 -0.019 0.038 0.010 0.019 0.057

0.600 0.091 -0.028 0.005 -0.039 -0.018 0.027 0.009 0.010 0.046

278 A. L. Bergenwall et al.

123

Notwithstanding the large and nationally representative

sample on which our analyses were based, potential limi-

tations of the current data need to be confronted. First, the

data used were derived from a cross-sectional survey,

limiting any causal inferences that can be made from the

data. Despite the fact that alternative explanations were

tested and excluded, longitudinal data remain necessary in

order to optimally contrast ‘‘selection’’ and ‘‘socialization’’

hypotheses (e.g., Simons-Morton et al. 2004). Second,

while the archival nature of this study allowed us access to

a large and nationally representative sample, we were

necessarily limited to the pre-selected items already

available. Thus, the variable assessing the amount of

money available to the employed children did not provide a

unique assessment of income earned from employment, as

it could have also included money from allowances not

necessarily linked to employment. In addition, given that

the reasons why teenagers choose to work are critical in

predicting subsequent outcomes (Dupre et al. 2006), future

research might assess whether young children choose to

engage in employment specifically to fund this habit.

Third, the effect sizes yielded in the current study are

very small, but we suggest that the applied significance of

the current findings should not be understated by the sta-

tistical effect sizes. Small effect sizes can be meaningful in

some circumstances (Prentice and Miller 1992), and we

suggest that the effect of early employment intensity on

cigarette use is important for several reasons. First, the

serious adverse consequences of tobacco use are well

documented (e.g., Max 2000) and community health efforts

are increasingly aimed at preventing smoking (e.g., Lantz

et al. 2000). Identifying correlates or predictors of young

children’s smoking may result in more focused prevention

and intervention initiatives. Second, early involvement in

tobacco smoking as a function of employment predicts

later involvement (Mihalic and Elliot 1997). Third, the

percentage of explained variance may be a misleading

criterion when the phenomenon under investigation can be

expected to cumulate over time (Abelson 1985); and the

increasing negative health effects of smoking reflects one

such situation.

One last finding that was not a central focus in this study

should not pass without comment. Female students repor-

ted engaging in a greater number of odd jobs—but less

weekly spending money than did male students. While

recognizing that the current data were collected in the fall

of 2002, the early emergence of this gender effect high-

lights the different value assigned to ‘‘male’’ and ‘‘female’’

tasks, even for the youngest of workers (Desmarais and

Curtis 1999). Discrimination based on gender is not only

salient in adult life, but it also exists at a very young age.

To the extent that early employment experience shapes

later expectations (Desmarais and Curtis 1999), this dis-

parity suggest that women may be socialized at a very early

age to expect less remuneration for tasks than is offered to

males, and reinforces the need for further research on the

nature and consequences of early employment experiences.

Conclusion

We investigated the effect of early employment intensity

on tobacco use among an unregulated, vulnerable popula-

tion of young children in Grades 5 through 9. Consistent

with earlier observations, children were found to engage in

employment at a very young age. More than 50 % of the

Grade 5 respondents were involved in some odd jobs; by

Fig. 2 a Smoking history predicted by spending money moderated by proportion of friends who smoke, holding parents who smoke

constant. b Smoking intent predicted by spending money moderated by proportion of friends who smoke, holding parents who smoke

constant. c Smoking intent predicted by self-esteem moderated by number of parents who smoke, holding friends who smoke constant.

Low = 1 SD below the mean, High = 1 SD above the mean

Ethical Implications of Children’s Employment 279

123

Grade 8, over 80 % of the sample was participating in the

irregular economy. Participation in odd jobs was linked

both to using tobacco and intent to use tobacco through

enhanced self-esteem and availability of discretionary

spending money. These findings demonstrate that early

employment intensity has important, ethical, and complex

effects for children, and supports the need for further

research that extends the construct of youth employment to

include early employment experiences, and to include

much younger children than those currently the focus of

research and policy. If the current findings are replicated by

longitudinal research, and extended to include subjective

employment experiences and a broader range of mediator

and moderator variables, valuable conceptual, research and

policy implications will ensue.

References

Abelson, R. P. (1985). A variance explanation paradox: When a little

is a lot. Psychological Bulletin, 97, 129–133. doi:10.1037/

0033-2909.97.1.129.

Bailey, S. L. (1992). Adolescents’ multisubstance use patterns: The

role of heavy alcohol and cigarette use. American Journal of

Public Health, 82, 1220–1224.

Bandura, A. (1965). Influence of models’ reinforcement contingencies

on the acquisition of imitative responses. Journal of Personality

and Social Psychology, 1, 589–595. doi:10.1037/h0022070.

Bandura, A. (1973). Social learning theory. Englewood Cliffs, NJ:

Prentice Hall.

Barling, J., & Kelloway, E. K. (1999). Introduction. In J. Barling & E.

K. Kelloway (Eds.), Young workers: Varieties of experience (pp.

3–16). Washington, DC: APA Books. doi:10.1037/10309-006.

Barling, J., Loughlin, C., & Kelloway, E. K. (2002). Development and

test of a model linking safety-specific transformational leader-

ship and occupational safety. Journal of Applied Psychology, 87,

488–496. doi:10.1037//0021-9010.87.3.488.

Barling, J., Rogers, K. A., & Kelloway, E. K. (1995). Some effects of

teenagers’ part-time employment: The quantity and quality of

work make the difference. Journal of Organizational Behavior,

16, 143–154. doi:10.1002/job.4030160205.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator

variable distinction in social psychological research: Conceptual,

strategic, and statistical considerations. Journal of Personality

and Social Psychology, 51, 1173–1182.

Bartechhi, C. E., MacKenzie, T. D., & Schrier, R. W. (1994). The

human costs of tobacco use (first of two parts). New England

Journal of Medicine, 330, 907–912.

Breslin, F. C., & Adlaf, E. M. (2002). Part-time work and cigarette

use among teenagers. Canadian Journal of Public Health, 93,

356–359.

Brook, D. W., Brook, J. S., Zhang, C., Cohen, P., & Whiteman, M.

(2002). Drug use and the risk of major depressive disorder,

alcohol dependence, and substance use disorders. Archives of

General Psychiatry, 59, 1039–1044.

Bureau of Labor Statistics. (August 27, 2010). News release:

Employment and unemployment among youth—Summer 2010.

Castillo, D. N. (1999). Occupational safety and health in young

people. In J. Barling & E. K. Kelloway (Eds.), Young workers:

Varieties of experience (pp. 159–200). Washington, DC: APA

Books. doi:10.1037/10309-006.

Cialdini, R. B. (2006). Descriptive social norms as underappreciated

sources of social control. Psychometrika, 72, 263–268. doi:

10.1007/s11336-006-1560-6.

Conrad, K. M., Flay, B. R., & Hill, D. (1992). Why children start

smoking: Predictors of onset. Addiction, 87, 1711–1724. doi:

10.1111/j.1360-0443.1992.tb02684.x.

Covell, K., & Howe, R. B. (1999). Working adolescents in

economically depressed areas of Canada. Canadian Journal of

Behavioural Science, 31(4), 229–239. doi:10.1037/h0087092.

Croghan, I. T., Bronars, C., Patten, C. A., Schroeder, D. R., Nirelli, L.

M., Thomas, J. L., et al. (2006). Is smoking related to body image

satisfaction, stress and self-esteem in young adults? American

Journal of Health Behavior, 30, 323–333. doi:10.5993/AJHB.30.

3.10.

Dekker, I., Greenberg, L., & Barling, J. (1998). Predicting union

attitudes in student part-time workers. Canadian Journal of

Behavioural Science, 30, 49–55. doi:10.1037/h0087057.

Desmarais, S., & Curtis, J. (1999). Gender differences in employment

and income experiences among young people. In J. Barling & E.

K. Kelloway (Eds.), Young workers: Varieties of experience (pp.

59–88). Washington, DC: APA Books. doi:10.1037/10309-003.

Dupre, K., Inness, M., Connelly, C., Barling, J., & Hoption, C. (2006).

Workplace aggression in teenage part-time employees. Journal

of Applied Psychology, 91, 987–997. doi:10.1037/0021-9010.

91.5.987.

Ellickson, P. L., Tucker, J. S., & Klein, D. J. (2001). High-risk

behaviors associated with early smoking: Results from a 5-year

follow-up. Journal of Adolescent Health, 28, 465–473.

Ferman, L. A. (1990). Participation in the irregular economy. In K.

Erikson & S. P. Vallas (Eds.), The nature of work: Sociological

perspectives (pp. 119–140). New Haven, CT: Yale University

Press.

Frohlich, K. L., Potvin, L., Gauvin, L., & Chabot, P. (2002). Youth

smoking initiation: Disentangling context from composition.

Health & Place, 8(3), 155–166. doi:10.1016/S1353-8292(02)

00003-5.

Frone, M. R. (1998). Predictors of work injuries among employed

adolescents. Journal of Applied Psychology, 83, 565–576. doi:

10.1037/0021-9010.83.4.565.

Frone, M. R. (1999). Developmental consequences of youth employ-

ment. In J. Barling & E. K. Kelloway (Eds.), Young workers:

Varieties of experience (pp. 89–128). Washington, DC: APA

Books. doi:10.1037/10309-004.

Frone, M. R. (2000). Interpersonal conflict at work and psychological

outcomes: Testing a model among young workers. Journal of

Occupational Health Psychology, 5, 246–255. doi:10.1037/

1076-8998.5.2.246.

Frone, M. R. (2003). Predictors of overall and on-the-job substance

use among young workers. Journal of Occupational Health

Psychology, 8, 39–54. doi:10.1037/1076-8998.8.1.39.

Frone, M. R. (2008). Are work stressors related to employee

substance use? The importance of temporal context in assess-

ments of alcohol and illicit drug use. Journal of Applied

Psychology, 93, 199–206. doi:10.1037/0021-9010.93.1.199.

Furnham, A., & Thomas, P. (1984). Adult’s perception of the economic

socialization of children. Journal of Adolescence, 7, 217–231.

Gidwani, P. P., Sobol, A., DeJong, W., Perrin, J. M., & Gortmaker, S.

L. (2002). Television viewing and initiation of smoking among

youth. Pediatrics, 110(3), 505–508.

Goldstein, N. J., Cialdini, R. B., & Giskevicius, V. (2008). A room

with a viewpoint: Using social norms to motivate environmental

conservation in hotels. Journal of Consumer Research, 35,

472–482. doi:10.1086/586910.

Goodnow, J. J. (1988). Children’s household work: Its nature and

functions. Psychological Bulletin, 103, 5–26. doi:10.1037/0033-

2909.103.1.5.

280 A. L. Bergenwall et al.

123

Greenberger, E., & Steinberg, L. D. (1986). When teenagers work:

The psychological and social costs of adolescent employment.

New York: Basic Books.

Harris, J. R. (1995). Where is the child’s environment? A group

socialization theory of development. Psychological Review, 72,

458–489. doi:10.1037/0033-295X.102.3.458.

Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation

analysis in the new millennium. Communication Monographs,

76, 408–420.

Hayes, A. F. (2012). PROCESS: A versatile computational tool for

observed variable mediation, moderation, and conditional pro-

cess modeling [White paper]. Retrieved May 15, 2012, from

http://www.afhayes.com/public/process2012.pdf.

Henningfield, J. E., Clayton, R., & Pollin, W. (1990). Involvement of

tobacco in alcoholism and illicit drug use. British Journal of

Addiction, 85, 279–292.

Jahoda, M. (1982). Employment and unemployment: A social

psychological analysis. New York: Cambridge University Press.

Kelloway, E. K., & Barling, J. (1999). When children work:

Implications for youth, society and organizations. International

Journal of Management Reviews, 1(2), 159–170. doi:10.1111/

1468-2370.00010.

Kelloway, E. K., & Harvey, S. (1999). Learning to work. In J. Barling &

E. K. Kelloway (Eds.), Young workers: Varieties of experience

(pp. 37–58). Washington, DC: APA Books. doi:10.1037/10309-004.

Kelloway, E. K., & Thun, B. (2006). Subjective norms and lemonade

stands: The effects of early socialization and childhood work

experiences on entrepreneurial intent. Administrative Sciences

and Association of Canada Annual Conference. Banff, Alberta.

Lantz, P. M., Jacobson, P. D., Warner, K. E., Wasserman, J., Pollack,

H. A., Berson, J., et al. (2000). Investing in youth tobacco

control: A review of smoking prevention and control strategies.

Tobacco Control, 9, 47–63. doi:10.1136/tc.9.1.47.

Lee, C., & Orazem, P. F. (2010). High school employment, school

performance, and college entry. Economics of Education Review,

29, 29–39.

Loughlin, C., & Barling, J. (1999). The nature of youth employment.

In J. Barling & E. K. Kelloway (Eds.), Young workers: Varieties

of experiences (pp. 17–36). Washington, DC: APA Books. doi:

10.1037/10309-004.

Loughlin, C., & Lang, K. (2005). Young workers. In J. Barling, E.

K. Kelloway, & M. Frone (Eds.), Handbook of work stress.

Thousand Oaks, CA: Sage.

Mathers, M., Toumbourou, J. W., Catalano, R. F., Williams, J., & Patton,

G. C. (2006). Consequences of youth tobacco use: A review of

prospective behavioral studies. Addiction, 101, 948–958.

Max, W. (2000). The financial impact of smoking on health-related

costs: A review of the literature. American Journal of Health

Promotion, 15, 321–331. doi:10.4278/0890-1171-15.5.321.

McGee, R., Williams, S., Poulton, R., & Moffitt, T. (2000). A

longitudinal study of cannabis use and mental health from

adolescence to early adulthood. Addiction, 95, 491–503.

McGuire, W. J. (1968). Personality and susceptibility to social

influence. In E. F. Borgatta & W. W. Lambert (Eds.), Handbook

of personality theory and research. Skokie, IL: Rand McNally.

Mermelstein, R., & The Tobacco Control Network Writing Group.

(1999). Explanations of ethnic and gender differences in youth

smoking: A multi-site, qualitative investigation. Nicotine &

Tobacco Research, 1, S91–S98.

Mihalic, S. W., & Elliot, D. (1997). Short- and long-term conse- quences of adolescent work. Youth and Society, 28(4), 464–498.

doi:10.1177/0044118X97028004004.

Mortimer, J. T., Finch, M. D., Dennehy, K., Lee, C., & Beebe, T.

(1994). Work experience in adolescence. Journal of Vocational

Education Research, 19, 39–70.

Mortimer, J. T., Finch, M. D., Owens, T. J., & Shanahan, M. (1990).

Gender and work in adolescence. Youth and Society, 22,

201–224. doi:10.1177/0044118X90022002005.

Mortimer, J. T., Finch, M. D., Ryu, S., & Shanahan, M. (1991).

Evidence from a prospective longitudinal study of work expe-

rience, adolescent mental health, and behavior adjustment.

Paper presented at the biennial meeting of the Society for

Research in Child Development, Seattle, Washington.

Mortimer, J. T., Finch, M., Shanahan, M., & Ryu, S. (1992). Work

experience, mental health, and behavioral adjustment in adoles-

cence. Journal of Research on Adolescence, 2, 25–57. doi:

10.1207/s15327795jra0201_2.

Nolan, J. M., Schultz, P. W., Cialdini, R. B., Goldstein, N. J., &

Griskevicius, V. (2008). Normative social influence is underde-

tected. Personality and Social Psychology Bulletin, 34, 913–933.

doi:10.1177/0146167208316691.

Patrick, D. L., Cheadle, A., Thompson, D. C., Koepsell, T., & Kinne, S.

(1994). The validity of self-reported smoking: A review and meta-

analysis. American Journal of Public Health, 84, 1086–1093.

Patten, C. A., Choi, W. S., Gillin, C., & Pierce, J. P. (2000).

Depressive symptoms and cigarette smoking predict develop-

ment and persistence of sleep problems in U.S. adolescents.

Pediatrics, 106, 1–9.

Phillips, S., & Sandstrom, K. L. (1990). Parental attitudes toward

youth work. Youth and Society, 22, 160–183.

Ploeger, M. (1997). Youth employment and delinquency: Reconsid-

ering a problematic relationship. Criminology, 35, 659–676. doi:

10.1111/j.1745-9125.1997.tb01234.x.

Prause, J., & Dooley, D. (1997). Effect of underemployment on

school-leavers’ self-esteem. Journal of Adolescence, 20,

243–260. doi:10.1006/jado.1997.0083.

Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing

moderated mediation hypotheses: Theory, methods, and pre-

scriptions. Multivariate Behavioral Research, 42(1), 185–227.

doi:10.1080/00273170701341316.

Prentice, D. A., & Miller, D. T. (1992). When small effects are

impressive. Psychological Bulletin, 112(1), 160–164. doi:

10.1037/0033-2909.112.1.160.

Ramchand, R., Ialongo, N. C., & Chilcoat, H. D. (2007). The effect of

working for pay on adolescent tobacco use. American Journal of

Public Health, 97, 2056–2062. doi:10.2105/AJPH.2006.094045.

Rothstein, D. S. (2001). Youth employment in the United States.

Monthly Labor Review, 124(8), 8–17.

Schlosser, E. (2002). Fast food nation: The dark side of the all-

American meal. New York: Perennial Books. doi:10.1057/

palgrave.development.1110302.

Simons-Morton, B., Chen, R., Abroms, L., & Haynie, D. L. (2004).

Latent growth curve analyses of peer and parental influences on

smoking progression among early adolescents. Health Psychol-

ogy, 23, 612–621. doi:10.1037/0278-6133.23.6.612.

Statistics Canada. (2002). Microdata users guide: Youth smoking

survey. Ottawa, ON: Statistics Canada.

Tobacco Act of Canada, SC 1997, c. 13, s. 8.

Trzesniewkski, K. H., Donnellan, M. B., Moffitt, T. E., Robins, R. W.,

Poulton, R., & Caspi, A. (2006). Low self-esteem during

adolescence predicts poor health, criminal behavior, and limited

economic prospects during adulthood. Developmental Psychol-

ogy, 42, 381–390. doi:10.1037/0012-1649.42.2.381.

Tucker, S., & Loughlin, C. (2006). Young workers. In E. K. Kelloway,

J. Barling, & J. J. Hurrell Jr. (Eds.), Handbook of workplace

violence (pp. 417–444). Thousand Oaks, CA: Sage.

Valois, R. F., Dunham, A. C. A., Jackson, K. L., & Waller, J. (1999).

Association between employment and substance abuse behaviors

among public high school adolescents. Journal of Adolescent

Health, 25, 256–263. doi:10.1016/S1054-139X(99)00039-7.

Ethical Implications of Children’s Employment 281

123

Voelkl, K. E., & Frone, M. R. (2000). Predictors of substance abuse at

school among high school students. Journal of Educational

Psychology, 92, 583–592. doi:10.1037/0022-0663.92.3.583.

Weller, N. F., Cooper, S. P., Tortolero, S. R., Kelder, S. V., & Hassan,

S. (2003). Work-related injury among South Texas middle-

school students: Prevalence and patterns. Southern Medical

Journal, 96, 1213–1220.

White, L., & O’Donnell, C. (2001). Working children and accidents:

Understanding the risks. Child: Care, Health and Development,

27, 23–34. doi:10.1046/j.1365-2214.2001.00184.x.

Wu, L., Schlenger, W. E., & Galvin, D. M. (2003). The relationship

between employment and substance abuse among students aged

12 to 17. Journal of Adolescent Health, 32, 5–15. doi:10.1016/

S1054-139X(02)00447-0.

Zhao, X., Lynch, J. G., Jr., & Chen, Q. (2010). Reconsidering Baron

and Kenny: Myths and truths about mediation analysis. Journal

of Consumer Research, 37, 197–206.

282 A. L. Bergenwall et al.

123

Copyright of Journal of Business Ethics is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.