Key Concepts in Experimental Design, Statistics, and Causality in Psychology

What are the three conditions for causality?

Association; Temporal Precedence; Nonspuriousness

What is association?

IV and DV are statistically related

What is temporal precedence?

The cause happens before the effect

What is nonspuriousness?

No third variable explains the relationship

What is an independent variable (IV)?

The variable manipulated by the researcher

What is a dependent variable (DV)?

The outcome being measured

What is a confound?

An uncontrolled variable that systematically varies with the IV

What are demand characteristics?

When participants change behavior because they guess the study purpose

What is randomization?

Equal chance of assignment; reduces confounds

What is internal validity?

Confidence that the IV caused the DV

What is external validity?

Ability to generalize results to the real world

Lab experiment: pros and cons?

High internal validity; low external validity

Field experiment: pros and cons?

Low internal validity; high external validity

What is a between-subjects design?

Different participants in each condition

What is a within-subjects design?

Same participants in all conditions

What is a factorial design?

Multiple IVs tested at once; allows interaction effects

What do t-tests compare?

Means

What type of DV is required for t-tests?

Continuous (interval or ratio)

What is a one-sample t-test?

Compares one group mean to a known value

What is an independent samples t-test?

Compares two unrelated groups

What is a paired samples t-test?

Compares same participants twice

What does the p-value indicate?

If p < .05, results are statistically significant

What is the null hypothesis?

No difference between group means

What is ANOVA used for?

Comparing 3 or more group means

What is a one-way ANOVA?

One IV with 3+ groups

What is a repeated measures ANOVA?

Same participants across conditions or time points

What is a two-way ANOVA?

Two IVs and their interaction effects

What is the F-statistic?

Ratio of between-group variance to within-group variance

When is an F-statistic significant?

When p < .05

What does a significant ANOVA NOT tell you?

Which specific groups differ

What must you do after a significant ANOVA?

Run post hoc tests

What is the relationship between t and F?

F = t² (in one-way ANOVA with two groups)

What are post hoc tests?

Tests to determine which groups differ after ANOVA

What are planned contrasts?

Pre-planned comparisons between groups

What is sentiment analysis?

Classifies text as positive, negative, or neutral

What are synthetic consumers?

AI-generated personas simulating consumer segments

What is topic modeling?

Identifies themes in large text data

What is a limitation of AI in research?

Cannot establish causality

Another limitation of AI?

Can amplify bias or generate hallucinations

When do you use a one-sample t-test?

One group vs a known value

When do you use an independent samples t-test?

Two unrelated groups

When do you use a paired samples t-test?

Same participants measured twice

When do you use a one-way ANOVA?

3+ independent groups

When do you use a repeated measures ANOVA?

Same participants across 3+ conditions

When do you use a two-way ANOVA?

Two independent variables

If p > .05, what do you do?

Fail to reject the null hypothesis

If ANOVA is not significant, should you run post hoc tests?

No

What type of DV is required for these tests?

Continuous