Join StudySoup for FREE

Get Full Access to
UMB - CCJS 300 - Study Guide

Description

Reviews

C r i m i n a l J u s t i c e

RESEARCH METHODS

Purposes Inaccurate

Features of Scientific Method

Objective

Selective

Observations partial picture

Observations circumstance, feeling,

environment

Illogical

Reasoning (Gambler's Fallacy)

Resistance to Change

Ideology and Politics

funding

Empirical

Self

Correcting errors can be fixed

Verifiable

proveCumulative

Neutrality

value free

Researchers

Types of Questions

Professors/ Academics

Government Employees

Practitioners (non

Academics)

Measurement operationalize

Descriptive baseline

information

Exploratory come up with questions and validate it

Journalists Students

?

Causal

establish causality

Evaluative

impact of research

RESEARCH paradigms change over time

Logical Thinking

Approaches

Deductive

Logic

hypothesis from theory

Inductive Logic generalizations from

observations Don't forget about the age old question of What if new positive mutation (a) is additive?

Positivist

scientific

approaches

ex. sociological theorists

Interpretive understand

everyday things

ex. how ethics influences

behavior

Critical

liberating knowledge

ex. feminism

Time

Cross-Sectional

Longitudinal

Exploration

Purposes

Description

snapshot of one time

time series panel

cohort

case study

explore what is happening

Explanation

understand relationship causal process

describe what is going on

Evaluation/Application apply criteria

Research Process

Data Gathering

Finding

20%

Induction

20%

is

s

ly

a

n

A

Finding

TheoryDed

u

cWe also discuss several other topics like What is nuer concept of structural space?

t

i

g

n

i r

H

y

p

o

o n

a t

e

th

Theory

Research Design ment

a

D

h

t

a

GWe also discuss several other topics like What is herodotus known for?

ngi seD

e

s

isOpe ar t oi

20% We also discuss several other topics like Who wrote dido aeneas?

20%

Measu er

hcr aeseR

an We also discuss several other topics like What is collateral sprouting?

Steps

il az t oi n

Hypothesis 20%

Problem

Formulation

Research Design

Data

Collection

Analysis and Presentation

Conclusions, Interpretations, and Limitations

E T H I C S I N

R E S E A R C H

Legal and Moral

Ethical Horror Stories Belmont Report (1974)

Biomedical

Nazi Experiments

Tuskegee Experiments

LSD

Social Sciences

Milgram Study - obedience

Zimbardo Simulated Prison Study - role play Humphrey's Tearoom Trade Study - male homosexual behavior

Project Camelot - CIA

Researcher Fraud

Fraud - make something up

Plagiarism - steal someone's work

E T H I C S

Principle of Respect for Persons - treat

respondents well

Principle of Beneficence - do no harm

Principle of Justice - distribute benefits and burden Don't forget about the age old question of What are the advantages of hedonism?

No Harm to Participants

Transparency

Physical and

psychological

harm

Waivers

Avoid

deception

Establish trust

Voluntary

Participation

Anonymity and

Confidentiality

Reduce biases Can offer

benefits

Aggregate

data

Confidential - won't be

traced back

How research is conducted How data is collected

Maintain

ethical

standards

Oversight

Informed

Consent - let people know Special

Population

Code of

Professional Ethics

Institutional Review

Boards

Ethical Problems Avoid Problems

Brajuha Case - subpoena field notes

Ofshe Case - lawsuits Hutchinson Case - attack on research

Scarce Case - jailed for refusing to give up notes

Careful research design Alternative

methodologies (ex. observation)

The Experimental Model

Gold Standard of Research Design Positivist Approach

Developed from Natural Sciences

Psychology

* Perceptual Psychology * Behaviorism

* Mental Measurement * Human Factors

Criminology Topics * Arrest for Domestic Violence

* Drug Education

* Correction Boot Camps

Purposes

Explanation - explain behavior

Evaluation - what happens when laws change

Establishing

Causality

1. Demonstrate relationship exists 2. Time order

3. Eliminate rival causal factors

Internal Validity External Validity did X change Y generalization

Internal Factors External Factors

History

events

between

experiments

Maturation time as

variable

Statistical

Regression

regress

towards

mean

Instrumentation

Selection Bias differential

respondents * random

assignment

Experimental Mortality

loss of

respondents

Reactive Effect of

Experimental

Arrangements

(Hawthorne Effect)

react to something

else

Multiple

Treatment

Inference

bleed over

effect

Interactive

Testing Effects pretest decrease responsiveness

Interaction of Selection Bias and

Experimental Variable

change the way

you measure

Testing

effects of

Selection

Other Factors

Post Hoc Error

Placebo

time order not

test on

second test

Maturation Interaction

enough to establish causality

Effect

mind over matter

Experimental Designs

True Experimental Design

* Group Equivalence - randomization

* Pretests and Posttests * Experimental and Control Group

attribute differences to experimental manipulation

Quasi

Experimental Design

no random

assignment

some matching

Pre

Experimental Design

no group

equivalence

weakest design typically exploratory

Control - helps determine causality

Quick and Inexpensive - compared to field

observations

Manageable - replication

Artificiality - Hawthorne Effect

Difficulty Recruiting

Experimenter Effects - bias use double-blind

experiment

STATISTICS AND DATA ANALYSIS

LEVELS OF MEASUREMENT

Nominal

categorical ex. gender

Ordinal

rank order values ex. attitude

questions

Interval

equal intervals ex. temperature

Ratio

meaningful zero ex. income

RESEARCH

Descriptive Research describe phenomenon, baseline data

Inferential Research infer whether

relationship exists, hypothesis testing

VARIABLE

Independent Variable predictor variables; influence dependent variables

Dependent Variables outcome variable

PROBABILITY

Bounding Rules - always between 0 (impossible) and 1 (already happened)

Addition and

Multiplication Rules

MEASURES OF

CENTRAL TENDENCY Middle of Data

MEASURES OF DISPERSION Spreadoutness of Data

Mode

most frequent number

any level of

measurement

not sensitive to outliers

Median

middle number (50th

percentile)

ordinal,

interval, ratio

not sensitive to outliers

Mean

arithmetic average

interval, ratio

sensitive to outliers

Variation Ratio nominal

VR = 1 -

(frequency/n)

Variance

s^2 = [∑(X1 - X̄ )^2]/(n-1)

Range

range = max - min

Standard

Deviation s = √s2

Interquartile Range (IQR) middle 50%

PROBABILITY DISTRIBUTION

Ex. z stat, t stat, chi-square Normal Distribution

SAMPLING

DISTRIBUTION

Central Limit Theorem (CLT) - if sample is large enough, don't need

UnimodalBilateral Symmetry

Theoretical

CurveAsymptotic

Bell

Shaped

mean, median, mode at same place

to worry about shape of underlying population

Rule of Thumb: 30

HYPOTHESIS TESTING STATISTICAL TESTS 1. Hypothesis (Null and Alternative)

null - no difference/relationship 2. Probability Distribution and Statistics

ex. z-test

z Test

t Test

Correlation

ANOVA

3. Alpha (α) Level

criteria

4. Calculations

5. Decision and Interpretation reject or fail to reject null

Chi-Square Regression Coefficient

Canonical

Correlation