Week 10: Factorial Designs

A set of vocabulary flashcards covering the concepts, designs, and interpretations of factorial experiments including main effects, interactions, and design types.

Factorial Design

A study with $2$ or more independent variables (factors) being tested simultaneously across all their combinations.

Main Effect

The overall effect of one factor, averaged (collapsed) across all levels of the other factor(s).

Marginal Means

The average DV score for all participants in one level of a factor, calculated by ignoring which level of the other factor they were in.

Interaction

Occurs when the effect of one independent variable depends on the level of another independent variable; technically detected as a 'difference in differences'.

Simple Effect

The effect of one factor at one specific level of the other factor.

Crossover Interaction

A disordinal interaction where the lines on a graph cross and the effect of one variable flips direction completely depending on the level of the other variable.

Spreading Interaction

An ordinal interaction where the lines on a graph diverge (do not cross) and are not parallel, showing the effect is stronger at one level.

Independent-Groups Factorial Design

A design where both IVs are between-subjects, meaning each participant experiences only one condition; this requires the most participants.

Within-Groups Factorial Design

A design where both IVs are within-subjects, meaning every participant experiences all conditions; this requires the fewest participants.

Mixed Factorial Design

A design where one independent variable is between-subjects and another independent variable is within-subjects.

Three-way Interaction

An interaction that exists when the two-way interaction between two variables changes depending on the level of a third variable.

Cells

The number of conditions in a factorial design, calculated by multiplying the levels of the factors (e.g., $2 \times 2 = 4$ cells).

Factorial Design Notation ($A \times B$)

A notation where the number of values indicates the number of IVs and the numbers themselves indicate how many levels each IV has (e.g., $2 \times 2 \times 3$ has $3$ IVs).

Statistical Significance

An interpretation of how large or precise an estimate is to determine if it matters in the real world, distinct from statistical significance ($p$-value).

Efficiency

An advantage of factorial designs allowing researchers to test multiple questions in one study instead of running separate experiments.

Realism

An advantage of factorial designs that allows for the study of real psychological phenomena which typically involve multiple interacting variables.

Overall Effect

Another name for a main effect, as it represents the overall pattern rather than what happens in every individual condition.