Deck 9: Choosing the Right Research Design

Question 1

If you overheard a researcher say that "each of the three possible main effects in my study was highly significant," you could conclude that the analysis the researcher had conducted was:

A)a one-way,three-groups ANOVA
B)a 2-way ANOVA
C)a 3-way ANOVA
D)a test that had ruled out any statistical interactions

Answer 1

a 3-way ANOVA

Question 2

Cathryn conducted an experiment that involved administering an anti-anxiety drug to people with a history of Generalized Anxiety Disorder.Participants received 0 mg,50 mg,or 100 mg of the drug.Which participants received a placebo?

A)The 50 mg condition
B)The 100 mg condition
C)The 0 mg condition
D)There was no placebo

Answer 2

The 0 mg condition

Question 3

To analyze the data from a 2 x 2 factorial experiment,you should conduct:

A)a t-test
B)a two-way analysis of variance (ANOVA)
C)a four-way analysis of variance (ANOVA)
D)a factorial analysis contingency test (FACT)

Answer 3

a two-way analysis of variance (ANOVA)

Question 4

According to the text,carry-over effects are best controlled by:

A)reducing the number of conditions
B)making experimental tasks as brief as possible
C)varying the order of presentation of conditions
D)avoiding tasks that interfere with performance on a subsequent task

Question 5

If you plotted the results of a factorial study using lines whose slopes represents mean differences across different experimental conditions,the presence of a statistical interaction would show up in the form of:

A)lines that are separated by a statistically significant distance
B)lines that generally slope upwards
C)lines that generally slope downward
D)nonparallel lines

Question 6

In the Gilbert et al.(1988)study of causal attribution discussed in the text,the researchers independently manipulated cognitive load (low vs.high)and discussion topics (anxiety-provoking vs.relaxing).Thus,the design of this study was:

A)a one-way,four-groups design
B)a Latin square factorial
C)a mixed-model design
D)a 2 x 2 factorial design

Question 7

The major limitation of one-way experimental designs is that they only allow researchers to look at o_____ at a time.

A)control group
B)dependent variable
C)independent variable
D)type of participant

Question 8

Chloe conducted a 2 x 2 factorial study in which her independent variables were gender and frustration.Her dependent variable was verbal aggression.She found that frustration significantly increased verbal aggression in both women and men (to an equal degree).She also found that,on average,women engaged in significantly more verbal aggression than men did (regardless of frustration level).Chloe seems to have observed:

A)a gender x frustration interaction
B)a main effect of gender and a main effect of frustration
C)a main effect of frustration and a gender x frustration interaction
D)a gender interaction and a frustration interaction

Question 9

The simplest possible kind of true experiment is:

A)a one-group design
B)a two-groups design
C)a one-way design
D)a 2 x 2 factorial design

Question 10

When an effect exists at one level of a second independent variable but is weaker or nonexistent at a different level of the second independent variable,the observed pattern is referred to as a:

A)cross-over interaction
B)spreading interaction
C)simple main effect
D)factorial

Question 11

How many cells (i.e. ,how many unique experimental conditions)are there in a 2 X 2 X 3 completely between-subjects factorial experiment?

A)6
B)7
C)12
D)There is no way to tell without knowing more about the design.

Question 12

In a 2 x 2 factorial study,what is the maximum number of main effects that you could possibly observe?

A)two
B)one
C)three
D)four

Question 13

A disordinal interaction is the same as:

A)a cross-over interaction
B)a spreading interaction
C)a two-way interaction
D)an interaction and two significant main effects

Question 14

In her research on stereotyping,Efrat conducted a very large 2 x 2 (Target Gender x Judgmental Time Pressure)factorial study.She observed an interesting main effect of target gender.However,she did not observe a main effect of time pressure,and she did not observe any evidence for a significant interaction.Efrat's failure to observe a significant interaction could be thought of as evidence for:

A)the internal validity of her target gender effect
B)the external validity of her target gender effect
C)the lack of internal validity of her target gender effect
D)the lack of external validity of her target gender effect

Question 15

Simple effects tests are conducted to:

A)figure out the exact nature of a significant interaction
B)determine whether an apparent interaction is statistically significant
C)find out if an interaction replicates in a subset of a larger sample
D)determine if a lack of an interaction is due to a lack of statistical power

Question 16

To find out whether participants figured out what was being manipulated in an experiment,an experimenter is most likely to use_________ immediately after participants have completed the study.

A)post-manipulation interview
B)feedback protocol
C)manipulation check
D)structured debriefing

Question 17

In comparison with purely between-subjects experimental designs,one drawback of purely within-subjects experimental designs is that:

A)they usually do not do as good a job of controlling for individual differences
B)they require the use of matching,which is very difficult to perform effectively
C)they open the door to problems such as sequence effects or carryover effects
D)they are usually more costly and time-consuming than between-subjects designs

Question 18

A Latin square is one specific example of:

A)ideal counterbalancing
B)partial counterbalancing
C)complete counterbalancing
D)reverse counterbalancing

Question 19

Is it possible for a statistical analysis to yield three main effects in a 2 x 2 factorial study?

A)No,because there are only two independent variables.
B)No,because there are only two cells in such a design.
C)Yes,but only if the test for an interaction is not significant.
D)Yes,but only if the test for an interaction is also significant.

Question 20

Maurice conducted a 2 x 2 completely within-subjects experiment.He counterbalanced all of his within-subjects conditions by making use of 24 different orders.If he used 24 different orders,which form of counterbalancing must he have used?

A)reverse counterbalancing
B)an ideal Latin square
C)randomized comprehensive counterbalancing
D)complete counterbalancing

Question 21

Relative to within-subjects designs,mixed-model designs make it easier:

A)to prevent participants from becoming aware of your manipulation
B)for participants to figure out the researcher's hypothesis
C)to gain control over individual differences
D)to conduct the appropriate statistical analysis of one's data

Question 22

A design that incorporates both within-subjects and between-subjects measures of independent variables is known as a:

A)Latin square
B)mixed-model design
C)pseudo-experiment
D)counterbalancing technique

Question 23

A counterbalancing technique that guarantees that the average serial position of any given condition in a study is exactly the same for all the unique conditions is known as:

A)reverse counterbalancing
B)partial counterbalancing
C)structured debriefing
D)serial position counterbalancing