Causation & Research Design
What three criteria are required to establish causation?
Association, time order, and nonspuriousness.
What are the three core components of a true experiment?
Random assignment, control group, treatment, pre/post measurement.
What is a strength of survey research?
Low cost, large samples, generalizability.
What is reliability?
Consistency of measurement.
What is the difference between descriptive and inferential statistics?
Descriptive statistics summarize the data you have (they stay home with the sample), while inferential statistics test hypotheses and make predictions or generalizations beyond the sample.
What is a spurious relationship?
A false association caused by a third variable.
Why is random assignment essential?
It ensures that the treatment and control groups are equivalent at the start.
What is a major weakness of survey research?
Measurement error/low validity, low depth, self-report bias.
What is validity?
Accuracy — measuring what it's intended to measure.
When is the median preferred over the mode?
When data are skewed or contain outliers, because the median better represents the center.
Which longitudinal design follows the exact same participants over time, unlike cohort or repeated cross-sectional designs?
Panel design
What is the difference between a treatment and a control group?
Treatment receives the intervention; control does not.
What is an omnibus survey?
A survey covering many unrelated topics.
What is the difference between primary and secondary data?
Primary data is collected firsthand by the researcher; secondary data was collected by someone else for another purpose.
What does it mean when a distribution is skewed? What does the curve look like?
Values cluster on one side with a long tail.
Which explanation aims for broad, generalizable causal laws, and which focuses on unique individual circumstances?
Nomothetic = broad/generalizable; Idiographic = case-specific/deep detail.
What type of experiment was the domestic violence arrest study in Bachman & Schutt?
A quasi-experiment or field experiment because it not have all of the features of a "true" experiment.
What is the difference between a double-barreled and double-negative question?
Double-barreled asks two things at once; double-negative confuses through wording.
What advantage does secondary data offer compared to collecting primary data yourself?
It is less expensive and faster because the data already exists.
What is standard deviation used to measure?
Spread or variability around the mean. Shows how different your data points are.
What advantage does longitudinal research offer that cross-sectional research cannot?
It allows researchers to see how individuals or groups change and how earlier events influence later outcomes.
List two common threats to internal validity in an experiment. In other words, what can ruin the researchers' ability to prove causality?
Selection bias, history, maturation, attrition
What is an index on a survey?
When researchers add together responses from several related survey items to create one score.
Name two common sources criminologists use for secondary data analysis.
UCR and the NCVS
What p-value level is typically used to determine statistical significance in criminal justice research?
p ≤ .05 (less than or equal to)