Psychology Research Methods: Experiments, Designs, and Statistical Analysis

What must exist for a cause-and-effect relationship between an IV and DV?

Association (correlation), temporal priority, and control of common-causal variables.

What is random assignment used for in experiments?

To create equivalence between groups and control common-causal variables.

What happens if there is no random assignment but IV is manipulated?

A quasi-experiment is performed.

What tests are used in one-way between-subjects designs?

Independent t-tests and Between-Subjects ANOVA.

What is the main goal of a true experiment?

To establish cause-and-effect relationships between IV and DV.

What is a one-way experiment?

An experiment with one independent variable.

What are within-subjects designs also called?

Repeated-measures designs.

What test is used for two groups in a within-subjects design?

Dependent t-test.

What test is used for three or more groups in a within-subjects design?

Within-Subjects ANOVA (Repeated Measures ANOVA).

What are advantages of repeated-measures designs?

Increased statistical power and economy of participants.

What are disadvantages of repeated-measures designs?

Carryover effects, practice effects, fatigue, and testing threats to internal validity.

What happens when multiple alphas are used?

They add up, increasing the chance of Type I Error (familywise error).

When do we reject the null hypothesis?

When p ≤ alpha (e.g., p ≤ .05).

What does it mean when p > alpha?

We fail to reject the null hypothesis (no significant difference).

What do significant differences mean?

Differences unlikely due to chance (random error).

What must true experimental designs include?

Manipulated IV, association, temporal priority, control of confounds, and random assignment.

What can true experiments claim?

Cause-and-effect relationships.

What is missing in quasi-experimental designs?

Random assignment.

What can quasi-experiments claim?

Limited cause-and-effect relationships.

What do nonexperimental designs lack?

Manipulated IV and temporal priority.

What can nonexperimental designs claim?

Only relationships, not causation.

What are factorial experimental designs?

Experiments with two or more independent variables (factors).

What does a 2x2 design mean?

Two independent variables with two levels each.

How many conditions are in a 2x2 design?

Four conditions.

What is a main effect?

The independent effect of one factor on the DV.

What is an interaction effect?

When the effect of one IV depends on the level of another IV.

Why are post-hoc tests used?

To find where significant differences exist and avoid inflated Type I error (familywise error).

What designs can factorial ANOVA be used for?

True, quasi-, or nonexperimental designs (between-, within-, or mixed-groups).