Ch11 - Within-Subjects Designs

B. Assigned to more than one treatment condition

What defines a within-subjects design?

A. Subjects are assigned to only one condition

B. Subjects are assigned to more than one treatment condition

C. Subjects are randomly assigned to groups

D. Subjects serve as a control group only

C. Repeated-measures design

Why is a within-subjects design also called a repeated-measures design?

A. Subjects are measured once

B. Subjects repeat the same task once

C. Subjects are measured on the DV after each treatment

D. Subjects repeat the experiment on different days

D. Ability to detect an IV's effect on the DV

What is statistical power?

A. Strength of the independent variable

B. Sample size of the experiment

C. Level of significance

D. Ability to detect an IV's effect on the DV

A. It allows detection of practically meaningful differences

Why is statistical power desirable?

A. It allows detection of practically meaningful differences

B. It eliminates random error

C. It removes the need for counterbalancing

D. It guarantees significant results

C. Statistically significant but practically meaningless differences

What is a potential problem with excessive statistical power?

A. Failure to detect differences

B. Increased sampling error

C. Statistically significant but practically meaningless differences

D. Reduced external validity

B. Subjects serve as their own control

What is a basic principle of within-subjects designs?

A. Subjects are compared to others

B. Subjects serve as their own control

C. Subjects are randomly matched

D. Subjects are exposed to one condition

D. Comparing performance across conditions

How do researchers determine treatment effects in a within-subjects design?

A. Comparing different groups

B. Comparing pretest scores

C. Comparing population means

D. Comparing performance across conditions

A. Assigned to all levels of two or more IVs

What is a within-subjects factorial design?

A. Subjects are assigned to all levels of two or more IVs

B. Subjects are assigned to one level of each IV

C. Subjects are randomly grouped

D. Subjects are matched across conditions

C. At least one within- and one between-subjects variable

What is a mixed design?

A. All variables are within-subjects

B. All variables are between-subjects

C. At least one within- and one between-subjects variable

D. A correlational design

B. Uses fewer subjects

Which is an advantage of within-subjects designs?

A. Eliminates order effects

B. Uses fewer subjects

C. Requires fewer measurements

D. Avoids fatigue

D. Greater statistical power

Why do within-subjects designs typically have greater statistical power?

A. They increase sample size

B. They reduce treatment levels

C. They eliminate confounds

D. They reduce individual differences

A. Subjects participate longer

Which is a disadvantage of within-subjects designs?

A. Subjects participate longer

B. Requires more subjects

C. Lower statistical power

D. Less data collected

C. Treatment conditions may interfere

Why can within-subjects designs be problematic?

A. Treatments are independent

B. Subjects forget instructions

C. Treatment conditions may interfere

D. No control group exists

B. When one treatment precludes another

When can't a within-subjects design be used?

A. When sample size is large

B. When one treatment precludes another

C. When IVs have many levels

D. When the DV is continuous

D. Performance changes due to treatment order

What are order effects?

A. Random measurement error

B. Differences between groups

C. Treatment manipulation errors

D. Performance changes due to treatment order

A. Progressive error

What term includes both practice and fatigue effects?

A. Progressive error

B. Random error

C. Sampling error

D. Measurement bias

C. Fatigue effect

What type of order effect involves declining performance?

A. Practice effect

B. Carryover effect

C. Fatigue effect

D. Learning effect

B. Practice effects

What are positive performance changes across conditions called?

A. Fatigue effects

B. Practice effects

C. Random effects

D. Ceiling effects

D. Increased familiarity with the task

Which factor may contribute to practice effects?

A. Equipment failure

B. Fatigue

C. Irritation

D. Increased familiarity with the task

A. Order exists once multiple treatments are presented

Why can't order effects be eliminated in within-subjects designs?

A. Order exists once multiple treatments are presented

B. Subjects forget earlier conditions

C. Treatments are random

D. IVs cannot be controlled

C. Counterbalancing

What method is used to control order effects in within-subjects designs?

A. Matching

B. Random assignment

C. Counterbalancing

D. Blocking

B. Subject-by-subject and across-subjects

What are the two major counterbalancing strategies?

A. Complete and partial

B. Subject-by-subject and across-subjects

C. Random and fixed

D. Linear and nonlinear

D. Presenting treatments multiple times per subject

What characterizes subject-by-subject counterbalancing?

A. One sequence per subject

B. Random assignment only

C. One treatment per subject

D. Presenting treatments multiple times per subject

A. Reverse counterbalancing

Which subject-by-subject technique uses a mirror-image sequence like ABBA?

A. Reverse counterbalancing

B. Block randomization

C. Latin square

D. Complete counterbalancing

C. Linear progressive error

Reverse counterbalancing is effective for controlling what type of progressive error?

A. Nonlinear

B. Random

C. Linear

D. Cyclical

B. It cannot control nonlinear progressive error

Why is reverse counterbalancing limited?

A. It increases fatigue

B. It cannot control nonlinear progressive error

C. It requires too many subjects

D. It removes treatment effects

D. Random sequence of all treatments

What is block randomization?

A. Fixed treatment order

B. Partial treatment exposure

C. One block per experiment

D. Random sequence of all treatments

A. Long and expensive procedures

What is a problem with subject-by-subject counterbalancing?

A. Long and expensive procedures

B. Poor statistical power

C. No control of order effects

D. Limited treatment exposure

C. Across-subjects counterbalancing

What alternative controls progressive error by distributing it across subjects?

A. Subject-by-subject counterbalancing

B. Reverse counterbalancing

C. Across-subjects counterbalancing

D. Blocking

D. Uses all possible treatment sequences

What is complete counterbalancing?

A. Uses one random order

B. Uses partial sequences

C. Uses mirror-image orders

D. Uses all possible treatment sequences

B. Partial counterbalancing

What type of counterbalancing uses only some of the possible orders?

A. Complete counterbalancing

B. Partial counterbalancing

C. Reverse counterbalancing

D. Block randomization

A. Latin square counterbalancing

Which is a partial counterbalancing technique?

A. Latin square counterbalancing

B. Reverse counterbalancing

C. Subject repetition

D. Matching

C. When individual differences are large

When is a within-subjects design usually preferable?

A. When treatments interfere

B. When experiments are long

C. When individual differences are large

D. When order effects are impossible

D. Risk of asymmetrical carryover

Why might a within-subjects design not be feasible?

A. Low statistical power

B. Too many subjects

C. Lack of control

D. Risk of asymmetrical carryover

B. Within-subjects design

A researcher exposes the same participants to multiple treatment conditions and compares their performance across those conditions. What design is being used?

A. Between-subjects design

B. Within-subjects design

C. Matched-groups design

D. Mixed design

D. Repeated-measures design

Participants' reaction times are measured after each treatment condition in an experiment. Why is this design called repeated-measures?

A. Because treatments are repeated

B. Because subjects are replaced

C. Because the IV is repeated

D. Because the DV is measured after each treatment

A. Statistical power

An experiment is able to detect small but meaningful differences between conditions. What concept explains this ability?

A. Statistical power

B. External validity

C. Reliability

D. Randomization

C. Diminishing returns

A study finds a statistically significant difference that has no real-world importance (e.g., 0.1 mm Hg blood pressure change). What concept does this illustrate?

A. Low power

B. Order effects

C. Diminishing returns of power

D. Confounding

B. Subjects as their own control

Why do within-subjects designs typically have greater statistical power than between-subjects designs?

A. They use random sampling

B. Subjects serve as their own control

C. They eliminate order effects

D. They use more subjects

A. Within-subjects factorial design

Participants experience all levels of two independent variables in a single experiment. What design is this?

A. Within-subjects factorial design

B. Between-subjects factorial design

C. Matched-groups design

D. Mixed design

D. Mixed design

An experiment includes one variable manipulated between subjects and another manipulated within subjects. What design is used?

A. Between-subjects design

B. Within-subjects design

C. Factorial design

D. Mixed design

C. Advantage of fewer subjects

A researcher has a very small sample size but still wants to test multiple treatments. Why is a within-subjects design useful here?

A. It eliminates confounds

B. It reduces training effects

C. It requires fewer subjects

D. It prevents order effects

B. Disadvantage of interference

A drug administered in the first condition permanently alters participants' physiology, making later conditions impossible. Why can't a within-subjects design be used?

A. Low power

B. One condition precludes another

C. Sample size limitations

D. Lack of random assignment

A. Order effect

Participants improve simply because they become more familiar with the task as the experiment progresses. What is this called?

A. Order effect

B. Interaction

C. Confounding

D. Blocking

D. Progressive error

Both fatigue and practice effects are collectively referred to as what?

A. Random error

B. Carryover

C. Confounds

D. Progressive error

C. Practice effect

Participants' scores improve because they develop strategies or become familiar with the equipment. What effect is occurring?

A. Fatigue effect

B. Order confound

C. Practice effect

D. Sampling bias

B. Fatigue effect

Participants' performance declines due to boredom or tiredness later in the experiment. What type of error is this?

A. Practice effect

B. Fatigue effect

C. Interaction

D. Carryover

D. Counterbalancing

A researcher varies the order of treatment presentation to control order effects. What technique is being used?

A. Matching

B. Random sampling

C. Blocking

D. Counterbalancing

A. Subject-by-subject counterbalancing

Each participant receives all treatments more than once in different orders to control progressive error. What strategy is this?

A. Subject-by-subject counterbalancing

B. Across-subjects counterbalancing

C. Partial counterbalancing

D. Randomized blocking

C. Reverse counterbalancing

A participant receives treatments in the order ABBA. What counterbalancing method is used?

A. Block randomization

B. Latin square

C. Reverse counterbalancing

D. Complete counterbalancing

B. Linear progressive error

Reverse counterbalancing is effective only when progressive error changes in what way?

A. Randomly

B. Linearly

C. Curvilinearly

D. Nonmonotonically

D. Nonlinear progressive error

Performance increases, then decreases, then increases again across conditions. What type of progressive error is this?

A. Linear

B. Constant

C. Random

D. Nonlinear

A. Block randomization

Each subject completes several blocks, each containing all treatments in a random order. What method is this?

A. Block randomization

B. Reverse counterbalancing

C. Latin square

D. Complete counterbalancing

C. Subject burden

Why can subject-by-subject counterbalancing become impractical as treatments increase?

A. It reduces power

B. It increases confounds

C. It creates long and expensive procedures

D. It eliminates randomization

B. Across-subjects counterbalancing

Each subject experiences every treatment only once, and order effects are distributed across participants. What strategy is this?

A. Subject-by-subject counterbalancing

B. Across-subjects counterbalancing

C. Reverse counterbalancing

D. Blocking

D. Complete counterbalancing

All possible treatment orders are used an equal number of times across participants. What method is this?

A. Partial counterbalancing

B. Latin square

C. Reverse counterbalancing

D. Complete counterbalancing

A. Partial counterbalancing

Only some of the possible treatment orders are used to control order effects. What is this called?

A. Partial counterbalancing

B. Complete counterbalancing

C. Subject-by-subject counterbalancing

D. Reverse counterbalancing

C. Latin square

A researcher selects a structured subset of treatment orders so each condition appears equally often in each position. What technique is used?

A. Randomized partial counterbalancing

B. Block randomization

C. Latin square

D. Reverse counterbalancing

B. Design choice

A researcher wants to control individual differences but has concerns about carryover effects. What should guide the design choice?

A. Statistical power only

B. Feasibility of a within-subjects design

C. External validity

D. Sample size only

True

In a within-subjects design, the same participants take part in more than one treatment condition.

False

Within-subjects designs measure the dependent variable only once for each participant.

True

Within-subjects designs are also referred to as repeated-measures designs.

True

Statistical power refers to an experiment's ability to detect the effect of an independent variable on a dependent variable.

False

Higher statistical power is always desirable even if the detected differences are practically meaningless.

True

Within-subjects designs often have greater statistical power because participants serve as their own control.

True

A within-subjects factorial design assigns participants to all levels of two or more independent variables.

True

A mixed design includes at least one within-subjects variable and one between-subjects variable.

True

One advantage of within-subjects designs is that they typically require fewer participants than between-subjects designs.

False

Within-subjects designs eliminate the need to control for order effects.

True

Order effects are changes in performance caused by the position of a treatment condition in a sequence.

True

Progressive error includes both practice effects and fatigue effects.

False

Fatigue effects occur when participants improve due to increased familiarity with a task.

True

Practice effects can occur because participants develop strategies or become more familiar with the equipment.

False

Order effects can be eliminated by presenting treatments in the same order to all participants.

True

Counterbalancing controls order effects by distributing progressive error across treatment conditions.

True

Subject-by-subject counterbalancing presents each treatment condition more than once to each participant.

True

Reverse counterbalancing presents treatments in a mirror-image sequence such as ABBA.

False

Reverse counterbalancing can control nonlinear progressive error.

True

Block randomization assigns each participant to multiple blocks containing all treatments in random order.

True

Across-subjects counterbalancing presents each treatment once and distributes order effects across participants.

True

Complete counterbalancing uses all possible treatment sequences an equal number of times.

True

Partial counterbalancing uses only some of the possible treatment sequences.

True

Latin square counterbalancing is a form of partial counterbalancing.

True

A within-subjects design is usually preferred when individual differences are large or when the number of participants is small.

False

Within-subjects designs are always feasible regardless of experiment length or carryover effects.