Deck 10: Approaches to Instruction

Question 1

The reason for statistically less error variance in within-subjects designs is:

A) less variability within each of the two groups.
B) smaller treatment differences.
C) individual differences are removed from the error variance.
D) all of these

Answer 1

C

Question 2

In a within-subjects design, who receives all levels of the independent variable?

A) every participant
B) those participants randomly assigned to groups
C) only those assigned to the control group
D) only those assigned to the experimental group

Answer 2

A

Question 3

Why can we be certain that the groups are exactly the same before the treatment has begun?

A) the independent variable is operationally defined
B) the same participants are used in each group
C) random assignment is used to assign participants
D) random selection is used to assigned participants

Answer 3

B

Question 4

In a within-subjects design:

A) each participant serves as his or her own control.
B) participants receive only one level of the independent variable.
C) between-group variance does not equal error variance at the start of an experiment.
D) groups are not equated at the start of the experiment.

Question 5

Another term for error variance is:

A) within-subjects variance
B) between-groups variance.
C) systematic variance.
D) within-groups variance.

Question 6

The design below best illustrates which of the following designs?

A) between-subjects design
B) repeated measures design
C) quasi-experimental design
D) within-subjects design

Question 7

The chapter describes an example of a within-subjects study examining miscalculation of space perception in airline pilots. What was the dependent variable?

A) amount of landing error
B) type of aircraft simulated
C) years of experience as a pilot
D) number of lights on the runway in the simulation

Question 8

Under what condition would a within-subjects design not be appropriate?

A) when participants can serve as their own comparison
B) when the treatment is expected to have a lasting effect on the participants
C) when there are few potential participants
D) when treatment effects are likely to be large

Question 9

Which of the following best illustrates why would we expect a reduction of error variance in a within-subjects design?

A) Stephen's behavior in situations A, B, and C is more similar than Mary's, Kevin's, and Stuart's behaviors in situation X, Y, and Z.
B) Stephen's behavior in situations A, B, and C is less similar than Mary's, Kevin's, and Stuart behaviors in situations X, Y, and Z.
C) Stephen's behavior in situation A is similar to Mary's, Kevin's, and Stuart's behaviors in situation A.
D) Stephen's behavior in situation A is not similar to Mary's, Kevin's, and Stuart behaviors in situation A.

Question 10

What effect do within-subjects designs have on the number of research participants?

A) It increases the number needed since the error variance increases.
B) It increases the number needed because the F ratio tends to be larger.
C) It reduces the number needed since each participant serves as her own control.
D) It reduces the number needed since most studies of this type are concerned with populations that are small.

Question 11

What effect do within-subjects designs have on F ratios?

A) F ratios are more sensitive to changes in the treatments
B) F ratios are less sensitive to changes in the treatments
C) F ratios tend to be smaller
D) There is no effect on F ratios.

Question 12

In a within-subjects design:

A) larger effect sizes are needed to reject the null hypothesis.
B) smaller treatment differences are needed to reject the null hypothesis.
C) a smaller F is needed to reject the null hypothesis.
D) the variability of scores in each condition is reduced.

Question 13

The design that allows for the greatest reduction of error variance is:

A) within-subjects.
B) between-subjects.
C) completely randomized.
D) factorial.

Question 14

Within-subjects designs are good for:

A) equating groups
B) reducing error variance
C) increasing the sensitivity of a study
D) all of these

Question 15

Which of the following methods will create groups that are the most equivalent at the start of an experiment?

A) random assignment
B) matching
C) within-subjects design
D) counterbalancing

Question 16

The design below best illustrates which of the following designs?

A) between-subjects design
B) correlational design
C) quasi-experimental design
D) within-subjects design

Question 17

A within-subjects design is more sensitive than a between-subjects design because:

A) there are fewer participants.
B) there is less error variance.
C) there are fewer degrees of freedom.
D) all of these

Question 18

In a within-subjects design, as compared to a between-subjects design, you are:

A) less likely to reject the null hypothesis.
B) more likely to make a Type II error.
C) more likely to obtain a small F ratio.
D) more likely to obtain a large F ratio.

Question 19

A researcher wants to conduct an experiment that uses a 2 X 3 X 3 design and requires ten participants in each cell. ____ participants would be needed for a between-subjects design but only ____ would be needed if a within-subjects design is used.

A) Eight, five
B) Eighteen, ten
C) Eight, three
D) Twenty-four, eight

Question 20

Which of the following is not an advantage of the within-subjects design?

A) It has greater statistical power
B) It uses fewer participants
C) It increases the size of the F ratio
D) It is most appropriate when the treatment has a lasting effect

Question 21

For complete counterbalancing:

A) order and time effects must be eliminated.
B) each condition must occur equally often.
C) each condition must precede and follow all other conditions an equal number of times.
D) each condition must occur equally often and precede and follow all other conditions an equal number of times.

Question 22

In the within-subjects design, the effects of treatment order can be controlled by:

A) matching.
B) counterbalancing.
C) blocking.
D) repeated measures.

Question 23

If practice of experimental tasks leads to an increase in performance by participants, ____ are said to have occurred.

A) carryover effects
B) order effects
C) time-related effects
D) counterbalancing effects

Question 24

The design below best illustrates which of the following designs?

A) between-subjects design
B) repeated measures design
C) quasi-experimental design
D) matched-subjects design

Question 25

A Latin square design is a special case of:

A) complete counterbalancing.
B) incomplete counterbalancing.
C) a between-subjects design.
D) a mixed design.

Question 26

Matched subjects procedures are:

A) more sensitive than within-subjects designs.
B) equally as sensitive as within-subjects designs.
C) more sensitive than between-subjects designs.
D) less sensitive than between-subjects designs.

Question 27

If matching is used as an experimental procedure, then the:

A) matching variable is analyzed.
B) matching variable is not analyzed.
C) procedure is called a randomized block design.
D) matching variable is analyzed and the procedure is called a randomized block design.

Question 28

In a mixed design:

A) one variable is manipulated between subjects and one variable is manipulated within subjects.
B) all variables are counterbalanced.
C) participants are matched on one variable only.
D) matching is used as a control procedure.

Question 29

If a treatment has a permanent effect on people, then it is inappropriate to use a ____ design.

A) between-subjects
B) within-subjects
C) naturalistic observation
D) completely randomized

Question 30

Likely confounds in a within-subjects design are:

A) order effects.
B) carryover effects.
C) time-related effects.
D) all of these

Question 31

In a matched-subjects procedure, participants are matched on a variable that:

A) is highly correlated with the independent variable.
B) is highly correlated with the dependent variable.
C) is uncorrelated with the dependent variable.
D) is not confounding.

Question 32

When matching is used as a control procedure, assignment of participants to groups is:

A) not random.
B) random.
C) counterbalanced.
D) controlled.

Question 33

When matching is used as an experimental procedure, the effect of the matching variable:

A) is randomized.
B) is eliminated.
C) cannot be interpreted causally.
D) is counterbalanced.

Question 34

If complete counterbalancing is used for a three condition experiment, there will be ____ possible orders.

A) three
B) four
C) five
D) six

Question 35

The procedure that eliminates order and time effects is:

A) intrasubject counterbalancing.
B) intragroup counterbalancing.
C) complete counterbalancing.
D) none of these

Question 36

When is incomplete counterbalancing most appropriate?

A) when there is a small number of conditions
B) when there is a large number of conditions
C) when there is a small sample size
D) when there is a large sample size

Question 37

Matched-subjects procedures reduce error variance:

A) more than between-subjects designs.
B) more than within-subjects designs.
C) less than between-subjects designs.
D) as much as within-subjects designs.

Question 38

____ effects refer to any increases or decreases in performance due to continued repetition of tasks.

A) Order
B) Counterbalanced
C) Matched
D) Mixed

Question 39

In a mixed design:

A) you do not need to counterbalance.
B) you do not have to worry about order effects.
C) error variances of the between and within comparisons are calculated differently.
D) all participants are not randomly assigned to conditions.

Question 40

The procedure that eliminates differential order effects is:

A) intrasubject counterbalancing.
B) intragroup counterbalancing.
C) complete counterbalancing.
D) none of these

Question 41

Under what conditions would you use a within-subjects design rather than a between-subjects design? Why?

Question 42

Describe the mixed design and its implication for the calculation of the F-ratio.

Question 43

An experimenter was interested in the effects of reward on performance in basketball shooting. Ten male and ten female individuals who had not played basketball before were selected for participation in the study. For the first 10 shots, they earned 5 cents for each basket; for the next 10 shots they earned 1 dollar per basket; for the final 10 shots, they were awarded 5 dollars per basket. Results indicated that females scored more baskets than males that all participants scored more baskets when given greater rewards, and that males' performance improved to a greater degree than females' as reward level increased.
What is (are) the independent variable(s) in this study?
What type of experimental design was used in this study?
What is the most likely confound present in this study?
How would you control for that confound?
Graph the results and interpret them in words.

Question 44

Discuss the logic of experimentation with respect to within-subjects designs and the statistical decision process.

Question 45

Matched-subjects designs have some advantages over both within-subjects designs and between-subjects designs.
Under what conditions would you choose a matched-subjects design rather than a within-subjects design?
What are the advantages of the matched-subjects design compared to a between-subjects design?

Question 46

An experimenter was interested in discovering which one of two reading programs was more effective for dyslexic children. What type of design would not be appropriate for this study and why not?

Question 47

In a randomized block design, the term ____ is usually used when matching is on interval or ratio scales.

A) blocking
B) leveling
C) stratifying
D) equating

Question 48

Compare and contrast within-subject designs and between-subjects designs in the context of a study examining the effects of computer screen size on eye fatigue.

Question 49

Discuss the types of counterbalancing and how they control order and time effects.

Question 50

One advantage of a matched subjects design is:

A) individuals do not have to be randomly assigned.
B) everything is automatically counterbalanced.
C) that error variance is decreased.
D) there are no individual differences.

Question 51

In a randomized block design, the term blocking is usually used when matching is on a(n):

A) nominal scale.
B) ordinal scale.
C) interval scale.
D) ratio scale.

Question 52

How are matched-subjects designs similar to both within-subjects and between-subjects designs.

Question 53

In a randomized block design, subjects are matched on gender. Since gender is measured on a nominal scale, ____ was used in matching participants.

A) leveling
B) stratifying
C) counterbalancing
D) blocking

Question 54

Describe how and why you would use counterbalancing in examining the effects of color on heart rate.

Question 55

Explain why it takes smaller treatment differences to reject the null hypothesis. Make sure you include a reference to the error variance.