Virginia's Study
Alcohol & Driving Ability Study
Chapter 13!
Chapter 14!
Dr. Drew's Study
100
Which of the following is true?

a. This is a within-subjects factorial design
b. This is a mixed subjects factorial design
c. This is a between subjects factorial design
d. This is a 2 x 10 factorial design
e. There is not enough information given to determine the type of design
This is a between subjects factorial design
100
In this study, the independent variable was ____; the dependent variable was measured by ____; and an extraneous variable held constant (to avoid confounding) was ____.

a. red light; braking speed; time since alcohol consumption
b. timing of the red light; amount of alcohol consumed; rate of alcohol consumption
c. amount of alcohol consumed; degree of intoxication; time since alcohol consumption
d. alcohol consumption; braking speed; time since alcohol consumption
e. braking speed; degree of intoxication; alcohol consumption
d. alcohol consumption; braking speed; time since alcohol consumption
100
pp. 320-321 – What is a nonequivalent-control group design?
If both an experimental and a control group are part of an experiment but subjects have not been allocated randomly to the two groups, we have a nonequivalent-control-group design. This is the most typical quasi experimental design.
100
pp. 351 - 352 – What are the definitions of mean/median/mode?
Mode: Most common score in a frequency distribution Median: The middle number in a set of ordered data. Mean: The common average
100
The design of the study is a(n)

a. 2 X 2 quasi-experimental factorial design
b. one way design with two independent variables
c. 2 X 2 between subjects factorial design
d. 2 X 2 mixed-subjects factorial design
e. interrupted time series design
d. 2 X 2 mixed-subjects factorial design
200
How should the data in this type of study be analyzed?

a. With an independent-samples t-test
b. With a dependent-samples t-test
c. With a 2x2 between-subjects ANOVA, with relevant post-hoc tests, if necessary
d. With a 2x2 within-subjects ANOVA, with relevant post-hoc tests, if necessary
e. With a measure of the correlation between junkfood and depression, such as Pearson’s correlation coefficient
With a 2x2 between-subjects ANOVA, with relevant post-hoc tests, if necessary
200
The values or levels of the dependent variable lie on a ________ scale

a. nominal
b. interval
c. ratio
d. ordinal
e. None of the above
c. ratio!
200
p.324 – What do we call the design of an experiment when there is one between-subjects variable and one within-subjects variable?
A mixed-factorial design
200
p.354 – What is “variance?” What is “standard deviation?”
Variance: The average of the squared deviations from the mean Std Deviation: The square root of the variance; a measure of variability in the same units as the scores being described.
200
In an ANOVA, which means would be compared to test for a possible main effect of praise?

a. 15 and 6
b. 12 and 18
c. 10 and 11
d. 8 and 4
e. (b) and (d)
c. 10 and 11
300
If Virginia changed the design so that she used pre-existing groups for the two levels of the junk food variable (i.e., she found children that already were eating 2 or 10 servings of junk food per week), this investigation would then be classified as a . . . ?

a. An experimental study
b. A quasi-experimental study
c. A correlational study
d. None of the above
A quasi-experimental study
300
The appropriate statistical test to analyze the data would be

a. an independent or between-subjects t test
b. a dependent or within-subjects t test
c. a one way ANOVA followed by appropriate posttest comparisons
d. a two way ANOVA followed by the appropriate posttest comparisons
e. none of the above
c. a one way ANOVA followed by appropriate posttest comparisons
300
p.325 – What is an “interrupted time-series design?”
A research design that allows the same group to be compared over time by considering the trend of the data before and after experimental manipulation
300
pp.364 – 365 – What does “correlation” refer to? What is a “correlation coefficient?”
Correlation: A number between 1 and -1 that captures the strength of the relationship between two variables. Correlation Coefficient: A statistic indicating the strength of the relationship between two variables.
300
Suppose the ANOVA reveals a significant interaction but not a significant main effect of praise. Dr. Drew would be justified in concluding that

a. Praise had no effect on task performance
b. There is in fact a main effect of praise but it is hidden in the data
c. The effect of praise on performance depended on the level of task difficulty
d. There must also be a main effect of difficulty
e. (b) and (c)
c. The effect of praise on performance depended on the level of task difficulty
400
To prevent the type of junk food that participants consume from becoming a possible confounding variable in this study, which of the following would control for this issue?

a. Mandate that all participants eat potato chips for all junk food servings
b. Have participants in the high junk food condition eat potato chips and the participants in the low junk food condition eat chocolate chip cookies
c. Randomly assign half of the participants in each junk food condition to eat potato chips for all junk food servings and the other half of the participants in each junk food condition to eat chocolate chip cookies for all junk food servings.
d. A & C, but not B
e. A, B, & C
d. A & C, but not B
400
Unless proper control procedures were used in this study, a possible problem could arise from

a. carryover effects
b. practice and fatigue effects
c. all of the above
d. none of the above
d. none of the above!
400
p.327 – What is a “repeated-treatment design?”
It’s a design in which a treatment is withdrawn and then presented a second time
400
p.354 – What is a “percentile?”
A score below which a certain percentage of the cases in a distribution fall; a percentile is a score, not a percentage.
400
If the interaction is significant and we want to know whether type of praise makes a difference when the task is easy, which means would we compare

a. 10 and 11
b. 12 and 18
c. 18 and 4
d. 15 and 6
e. 12 and 8
b. 12 and 18
500
Virginia finds that for children who eat more junk food, exercise causes lower levels of depression. For children who eat less junk food, there is no effect of exercise. Virginia has found which type of effect?

a. A main effect of junk food
b. A main effect of exercise
c. An interaction between junk food and exercise
d. Both A & C are true
e. Both B & C are true
c. An interaction between junk food and exercise
500
What type of post-hoc tests should be used in this experiment, if the initial analysis (I'm not going to state the initial analysis in case someone picks this question first!) yields significant results?
Indepedent-samples t-tests!
500
p.328 – What is a “cross-sectional study?”
In developmental research, it’s a study that tests different age groups at the same time.
500
p. 361 – How are the mean and mode affected by skewness? Why?
Mode is not affected at all by the skewness of the data because, as we have seen, it is not affected by any other scores. The mean, however, is lowest because it is affected by the distance of the low scores from the middle, as well as by the number of scores.
500
Finally, how would we analyze data from this experiment?
With a mixed-subjects ANOVA with appropriate post-hoc tests
M
e
n
u