Textbook Chapter 10 (Lectures 8-12)

What is a between-subjects design?

A design where different participants are assigned to each condition (e.g., experimental vs. control group).

What is a within-subjects design?

A design where the same participants experience all conditions of the independent variable.

Why don’t within-subjects designs need random assignment to create equivalent groups?

Because the same people serve in all conditions, so groups are inherently equivalent.

What are the main advantages of within-subjects designs?

(1) Fewer participants needed, (2) greater sensitivity—removes variability due to individual differences.

What is the main disadvantage of within-subjects designs?

Risk of order effects (practice, fatigue, contrast).

What are the three basic steps in planning a between-subjects experiment?

(1) Gather equivalent groups of participants, (2) manipulate the independent variable, (3) measure the dependent variable.

What does it mean to operationalize a variable?

Defining the independent or dependent variable in measurable terms (e.g., “hours of sleep” or “memory test score”).

What is a confounding variable?

A variable not of interest that changes along with the independent variable and could explain results.

What is internal validity?

The degree to which results can be confidently attributed to the independent variable rather than alternative explanations.

How can researchers control for confounds?

By using experimental control, random assignment, or careful design adjustments (e.g., keeping group size constant when testing “crowding”).

What is a practice effect?

Performance improves over time due to repeated practice with a task.

What is a fatigue effect?

Performance worsens as participants get tired, bored, or distracted.

What is a contrast effect?

A response to a later condition is influenced by the earlier condition (e.g., theft seems milder after murder).

What are two main ways to reduce order effects?

(1) Counterbalancing (complete or partial/Latin square), (2) adding time intervals or rest between conditions.

What is complete counterbalancing?

Including all possible orders of conditions (e.g., AB and BA for two conditions).

What is a Latin square design?

A partial counterbalancing method ensuring each condition appears in each ordinal position and before/after every other condition exactly once.

What is a pretest-posttest design?

Participants are measured on a variable before and after manipulation to check group equivalence and measure change.

What is a posttest-only design?

A design without a pretest, only measuring the dependent variable after manipulation.

What are the main advantages of adding a pretest?

(1) Helps with small samples, (2) allows participant selection, (3) checks for selective attrition.

What is the main disadvantage of adding a pretest?

It may sensitize participants to the hypothesis (demand characteristics).

What is selective attrition?

When dropout rates differ across conditions in a way that biases results (e.g., heavier smokers dropping out of treatment group).

What is a matched pairs design?

A between-subjects method where participants are paired on a relevant variable, and each member of the pair is randomly assigned to different conditions

When is a matched pairs design most useful?

When sample sizes are small, or individual differences strongly influence the dependent variable.

What is a disadvantage of matched pairs designs?

Matching can be costly and time-consuming, especially if many variables are used.

What are selection differences in between-subjects experiments?

When participants in conditions differ systematically (e.g., mostly high-income in one group, low-income in another), creating a confound.