Inferential Statistical Tests Flashcards
The numerator of the F-ratio produced by a one-way ANOVA is referred to as ______ and is a measure of __________.
A. MSW; treatment effects plus error
B. MSB; treatment effects plus error
C. MSW; error only
D. MSB; error only
Answer B is correct. The F-ratio is calculated by dividing MSB (mean square between) by MSW (mean square within). MSB is a measure of variability in dependent variable scores that’s due to treatment effects plus error, while MSW is a measure of variability due to error only.
To compare the mean achievement test score obtained by a sample of sixth-grade students in one school district in California to the mean achievement test score of all sixth-grade students in California, you would use which of the following?
A. single-sample chi-square test
B. factorial ANOVA
C. t-test for related samples
D. t-test for a single sample
Answer D is correct. The t-test for a single sample is used to compare a sample mean to a known population mean.
Which of the following statistical tests is used to control the effects of an extraneous variable on a dependent variable by treating the extraneous variable as an independent variable?
A. split-plot ANOVA
B. factorial ANOVA
C. ANCOVA
D. randomized block ANOVA
Answer D is correct. The randomized block ANOVA is used to control the effects of an extraneous variable on a dependent variable by including it as independent variable so that its main and interaction effects on the dependent variable can be determined.
A Cohen’s d of .6 suggests that a treatment has had:
A. no effect.
B. a small effect.
C. a medium effect.
D. a large effect.
Answer C is correct. As described by Cohen (1969), a d less than .2 indicates the treatment had a small effect, a d between .2 and .8 indicates the treatment had a medium effect, and a d larger than .8 indicates the treatment had a large effect.
An advantage of using the Bonferroni procedure and the Tukey HSD test and other post hoc tests is that they:
A. increase statistical power.
B. reduce statistical power.
C. reduce the probability of making a Type I error.
D. reduce the probability of making Type I and Type II errors.
Answer C is correct. Knowing that the Bonferroni procedure and the Tukey HSD test and other post hoc tests reduce the experimentwise error rate and that the experimentwise error rate refers to a Type I error would have helped you identify answer C as the correct answer. Answer A is incorrect because post hoc tests reduce the experimentwise error rate by reducing alpha for each test, and reducing alpha reduces statistical power (i.e., a smaller alpha makes it harder to reject the null hypothesis whether it is true or false). Answer B is incorrect because, although these tests reduce statistical power, this question is asking for an advantage of these tests and reducing power is a disadvantage – not an advantage – of using them. Answer D is incorrect because changing the size of alpha has opposite effects on the chances of making Type I and Type II errors. In other words, reducing alpha decreases the probability of making a Type I error and increases the probability of making a Type II error, and vice versa.
A psychologist wants to compare the number of therapy clients who have just received a diagnosis of specific phobia who say they prefer in vivo exposure, imaginal exposure, or virtual reality exposure as part of their treatment. The psychologist will use which of the following to make this comparison?
A. single-sample chi-square test
B. multiple-sample chi-square test
C. one-way ANOVA
D. three-way ANOVA
Answer A is correct. This study includes only one variable (type of exposure) which is measured on a nominal scale. Consequently, the appropriate statistical test is the single-sample chi-square test.
Dr. Bashir conducts a study to compare the effects of in vivo exposure, imaginal exposure, and virtual reality exposure for clinic clients with claustrophobia on three physiological measures of arousal and a self-report measure of anxiety. Which of the following statistical tests is Dr. Bashir most likely to use to analyze the data she obtains?
A. randomized block ANOVA
B. factorial ANOVA
C. MANOVA
D. mixed ANOVA
Answer C is correct. The MANOVA (multivariate analysis of variance) is the appropriate statistical test when a study includes one or more independent variables and two or more dependent variables.
An investigator would use Jacobson and Truax’s reliable change index (RCI) to determine if the difference between an individual’s pre- and posttreatment scores on a measure of symptom severity is:
A. due to measurement error.
B. due to sampling error.
C. a significant effect size.
D. practically significant.
Answer A is correct. A reliable change index is used to determine if a difference in an individual’s pretreatment and posttreatment test scores is statistically reliable – i.e., the extent to which the change is due to actual change rather than measurement error. It is part of the Jacobson-Truax method which is used to determine if an intervention has clinical significance.
