discussion board questions/ experimental design activity

missruf

VSUResearchCh_3Autosaved.ppt

Home >Law homework help >Criminal homework help >discussion board questions/ experimental design activity

General Issues in Research Design

Chapter 3

Introduction

Science is an enterprise dedicated to finding out

No matter what we want to find out there are many ways to go about it

All aspects of research design are interrelated

Foundations of Social Science

The two pillars of science logic/reasoning and observation
This means that –

A scientific understanding of the world must make sense and

Must agree with what we observe

These pillars relate to the three key aspects of scientific research:

Theory

Data collection

Data analysis

Foundations of social science

Scientific theory deals with the logical aspect of science

Data collection deals with observational aspect

Data analysis look for patterns in what is observed eg. whether there is a link between 2 variables or whether certain group are more likely to show certain attributes

Theory, Not philosophy or belief

Social Scientific theory has to do with what is, not what should be

A theory is a systematic explanation for the observed facts and laws that relate to a particular aspect of life eg juvenile delinquency (strain theory)

Theory

Theory: a set of concepts/ideas and the proposed relationship among these concepts;

a structure that is intended to represent or models something about the world

Used to guide research and develop a hypothesis

A researcher can use “routine activity theory” to generate a hypothesis about patterns of burglary

Hypothesis is an expectation about the nature of things derived from a theory

Hypothesis testing: involves finding out if observations are consistent with the hypothesis

Regularities and Exceptions

Goal of theory is to find patterns of regularity in social life. We assume that life is regular.

Norms and rules in society create regularity eg

teenagers commit more crimes than adults; students go to school, judges receive higher salaries than social workers

Exceptions do not negate regularities – patterns still exist

Social regularities represent probabilistic patterns: patterns need not be reflected in 100% of observations

Probabilistic prediction: women are less likely to murder anybody, but when they do, their victims are usually males

Aggregates, Not Individuals

Social scientists study social patterns, or aggregates/entire sum, not individual behavior

Although researchers study motivations that affect individuals, aggregate are the subject of SS research

EX. A researcher studying whether White inmates tend to be assigned to more desirable jobs than non-white inmates wld be interested in patterns of job assignment

Social scientific theories care about why aggregated patterns of behavior are regular even when participating individuals change over time

The focus is on patterns and variables

Variables and Attributes

Social scientists study variables & the attributes that compose them

A Variable is an element or number that increases or decreases over time, or changes in different situations.
Attributes - characteristics that describe some object/person eg. old fashioned, employed, married

Variables - logical groupings of attributes

Male and female are the attributes of the variable gender

Gender is the variable composed of logical grouping of those two attributes. see

Variables and Relationships

Causation – an attribute of one variable is expected to cause, predispose, or encourage an attribute of another variable eg private attorney caused a person to be sentenced to probation

Independent variable: “cause”, “influencer” – can cause corresponding changes in other variables

Dependent variable: “effect”, “depends” – can take different values in response to an independent variables; eg. sentence depends (DV) on the type of attorney (IV)

Eg. Type of defense attorney (IV) effects prison or probation (DV)

Idiographic and Nomothetic Explanations – the way we explain things

Idiographic explanations: seek a full and detailed understanding of a single case or situation
Eg. different explanations of why you have done poorly in an exam. Case Study

Nomothetic explanations: are partial explanations that explain a class of situations or events rather than a single one.

Eg. you do better in an exam if you study alone than if u study in a group

You get more speeding tickets on weekends than on week days

Example of Idiographic Explanations

Clifford R. Shaw developed idiographic explanations to understand the lives of juvenile delinquents.

These case studies, offered researchers a comprehensive view of the effect that variables such as family factors, peer influences, and community circumstances have on the engagement in juvenile delinquency.

Data collection involved interviews with the youths and their parents, analyzing diaries kept by the youths, and reviewing juveniles medical, school, and arrest records.

Inductive and Deductive Reasoning

Deductive reasoning: moves from the general to the specific

From a logically or theoretically-expected pattern to observations that test the presence of the pattern.

From general statements towards observations. See pg. 59

Inductive reasoning: moves from the specific to the general

From a set of observations to general statement or the discovery of a pattern among them. See pg. 59

Qualitative and Quantitative Data

All observations are qualitative at the outset

Qualitative = non-numerical; greater detail, greater richness of meaning

e.g., What constitutes “maturity” or “being rich”?

Quantitative = numerical; carries a focusing of attention and specification of meaning

Makes it easier to summarize data and allows for statistical analysis

e.g., a person’s age

Both are useful and legitimate – choose based on topic or combine aspects of both

Causation in Social Sciences

One of the goals of social scientific research is to explain why things are the way they are

We do that by specifying the causes for the way things are

Some things are caused by other things

Cause is probabilistic eg. certain factors make delinquency more or less likely

Victims of Childhood abuse are more likely to report alcohol abuse as adults;, or upbringing, peer influence

Criteria for Causation

What criteria must be satisfied before we can infer that something causes something else

2 Criteria for assessing an idiographic explanation (Maxwell 2013):

How credible and believable it is ie. logic, make sense

Whether alternative explanations (rival hypothesis”) were seriously considered

Criteria for causation

3 Criteria for assessing an Nomothetic explanation (Shadish, Cook & Campbell, 2002):

1. In a causal relationship between two variables the cause must precede the effect in time

Which comes first: drug abuse or crime

2. The two variables must be empirically correlated with each other – they must occur together

3. The observed empirical correlation between the two variables cannot be explained by the influence of some other variable

Validity and Causal Inference

When we are concerned with whether we are correct in inferring that a cause produced an effect, we are concerned with validity – whether it is true and valid

When we say something is valid, we make judgment about the extent to which relevant evidence supports that interference as being true or correct

Validity threats: reasons we might be incorrect in stating that some cause produced some effect

Validity threats - a way that you might be wrong

Does drug use cause crime?

Understanding causal statements about drug use and crime requires:

Carefully specifying 2 key concepts –drug use and crime

And considering the different ways these concepts might be related

The 3 criteria for inferring cause –
Time order – which comes first, drug use or crime?

Chaiken (1982) found that 12% of their adult subjects committed crimes after using drugs for at least 2 years

Whereas 15% committed predatory crimes 2 or + years b/4 using drugs

3 criteria for inferring cause cont.

Empirical relationship –

Studies have found that some drug users commit crimes and some criminals use drugs

Bernett (2008) found that drug users are three times more likely to commit crime than non users

This establishes statistical relationship

Eliminating alternative explanations

What about the possible influence of other factors, things that might be related to both drug use and crime

Chaiken concludes that Both drug use and crime participation are weakly related

Presence of other factors like strain suggest that the relationship is not directly causal

Units of Analysis 1

Unit of analysis also known as units of observation: what or who is studied

Individuals - (individual people may be partly defined as police, victims, defendants, inmates, gang members, burglars)

Groups - multiple persons with same characteristics - (gangs, police beats, patrol districts, households, city blocks, cities, counties)

Organizations - formal organizations with established leaders and rules - (prisons, police departments, courtrooms, drug treatment facilities, businesses, agencies)

Unit of Analysis 2

Social artifacts –products of social behavior or social interactions among people –

Egs. stories about crime in newspapers,

video posts on the Internet showing actual crimes,

photographs of crime scenes,

incident reports, police/citizen interactions

E.g. A researcher could analyze whether videos are actual recordings of crime in progress or reports of crime from media resources

Issues/Problems with Unit of Analysis

Ecological fallacy is one of the problems with units of analysis

Ecological fallacy refers to the danger of making assertions about individuals as a unit of analysis based on the examination of groups or aggregations (Poor areas = more crime, therefore poor people commit more crime)

Individualistic fallacy: danger of using hearsay/subjective evidence to make arguments about individuals

Eg. media stories about drug problems in US cities often focus on drug use among African Americans

That does not mean that most African Americans are drug users or that drugs are not a problem among Whites

Time Dimension

Another dimension of research is the treatment of time

The time dimension of research requires careful planning

Observations can either be made more or less at one point, or stretched over a longer period

Cross-Sectional Studies

Cross-sectional studies: observing a single point in time and analyzing that cross section carefully

Simple and least costly way to conduct research

A single US census, is a study aimed at describing the US population at a given time

A police dept study may ask residents about crime problems in their neighborhood in a single time frame

Problems: We cannot see social changes; have to worry if we picked a bad point in time to capture

Longitudinal Studies

Longitudinal studies: permit observations over an extended time, ex. analyses of monthly drug arrests for 5 years

3 types of longitudinal studies –

Trend – studies changes within some general population over time. Eg. comparison of UCR figures over time, it may show an increase from 1960-1995 & a decrease pattern thru 2015

Cohort (age group) – examine more specific populations as they change over time. Could be age group

Cohorts are group of people who enter or leave an institution at the same time

Panel – similar to trend or cohort, but observations are made on the same set of people on two or more occasions (NCVS – subjects are interviewed 7times at 6 months intervals)

Cincinnati Study on Lead Exposure
identify the type of study

Researchers at the University of Cincinnati recruited individuals born from 1979-1985 to take part in a study. During childhood, the research subjects’ blood was tested for lead exposure several times. Interviews and surveys of the subjects and their parents were used to assess the youths’ levels of delinquency. The results showed that youths who were exposed to lead engaged in more acts of delinquency and used marijuana more than those not exposed. At the time of this study, the research subjects were between 15 and 17. The researchers are still studying these subjects, which will allow for more examinations of the effects of lead exposure on adult behaviors.

Dietrich, K.N., Ris, M.D., Succop, P.A., Berger, O.G. & Bornschein, R.L. (2001). Early exposure to lead and juvenile delinquency. Neurotoxicology and Teratology, 23, 511-518.

Answer

longitudinal panel Study

Approximating Longitudinal Studies

It is possible to draw conclusions about processes that take place over time even when only cross-sectional data are available

We can approximate longitudinal studies through logical inferences and retrospective studies

Approximating Longitudinal Studies

Logical inferences: may be possible to draw approximate conclusions about processes that take place over time, even when only cross sectional data is available

Drawing conclusion about processes that take place across time

Eg. a researcher may discover in a cross-sectional study of high school students that males are more likely than females to smoke marijuana

The researcher can conclude that gender affects the tendency to use marijuana

i.e. drawing conclusion from a one-time observation to processes that take place across time

Retrospective Research

Asks people to recall their past for the purpose of approximating observations over time

Eg. in a study of recidivism, we might select a group of prison inmates and analyze their history of delinquency or crime

Problems: People have faulty memories; people lie

Analysis of past records also suffer from problems – records may be unavailable, incomplete, or inaccurate

Time Dimension Summarized

Cross-sectional study = snapshot – an image at one point in time

Trend study = slide show – a series of snapshots in sequence over time, allows us to tell how some indicator varies over time

Panel study is like a motion picture that can capture moving images of the same individuals

– gives information about individual rates of offending over time

END

Does Drug Use Cause Crime?

Temporal order: which comes first?

A statistical relationship exists, but underlying causes affect both drug use and crime (IV threat)

Relationship between crime and drug use varies by type of drug

How will policy affect drug use and crime? A crackdown on all drugs among all populations will do little to reduce serious crime.

Scientific Realism

Scientific realism: bridges/links idiographic and nomothetic approaches to explanation by seeking to understand how casual mechanisms operate in specific contexts

Scientific realism studies how other possible influences are involved in cause-and-effect relationships

Contains elements of both idiographic and monothetic modes of explanation
Traditional approaches try to isolate causal mechanisms from other possible influences
But scientific realist approach views other possible influences as context in which causal mechanisms operate