STATS CH 16

Cohen's d requires that a mean difference be divided by the square root of the a) b) c) n d)

A

If a treatment effect exists, a) variability between groups will tend to exceed variability within groups. b) variability within groups will tend to exceed variability between groups. c) variability between groups and within groups will tend to be about the same. d) total variability will tend to exceed variability between groups and within groups.

A

In an experiment involving four different groups, each consisting of 5 subjects, the degrees of freedom for between groups equals a) 3 b) 4 c) 5 d) 20

A

In an experiment involving four different groups, each consisting of 8 subjects, the critical F occupies the cell intersected by column and row degrees of freedom, respectively, of a) 3 and 28. b) 3 and 32. c) 4 and 16. d) 4 and 32.

A

In an experiment involving three different groups, each consisting of 10 subjects, the degrees of freedom for within groups equals a) 27 b) 28 c) 29 d) 30

A

In analysis of variance, observed mean differences appear, somewhat disguised, as variability a) between groups. b) between subjects. c) within groups. d) within subjects.

A

Rejection of the overall null hypothesis indicates that a) at least one population mean differs from all others. b) at least several population means differ from each other. c) most population means are different. d) all population means are different.

A

Cohen's guidelines for values of d associated with small, medium, and large effects are a) the same for two independent samples, but not for two related samples. b) the same as for both two independent or two related samples. c) less stringent for ANOVA. d) more stringent for ANOVA.

B

If a published report of an F test specified that p <.01, you could conclude that the test result is a) rare, supporting the null hypothesis. b) rare, supporting the research hypothesis. c) common, supporting the null hypothesis. d) common, supporting the research hypothesis.

B

Use Cohen's d to estimate the effect size of a) any difference between pairs of means. b) any significant difference between pairs of means. c) the largest difference between pairs of means. d) the mean difference between all pairs of means.

B

12 In analysis of variance, variance estimates consist of a) differences between sample means divided by degrees of freedom. b) differences between sample means divided by sample size. c) sum of squares divided by degrees of freedom. d) sum of squares divided by sample size.

C

Any sum of squares term always equals the a) square of the sum of all scores. b) square of the deviations of all scores about their mean. c) sum of the squared deviations of all scores about their mean. d) sum of the deviations of squared scores about their mean.

C

The squared curvilinear correlation is obtained by dividing the a) between sum of squares by the within sum of squares. b) within sum of squares by the total sum of squares. c) between sum of squares by the total sum of squares. d) within sum of squares by the between sum of squares.

C

Tukey's HSD controls the a) probability of a type I error. b) probability of a type II error. c) cumulative probability of a type I error. d) cumulative probability of a type II error.

C

Variability between groups is based on the variation among the scores of a) equal numbers of subjects. b) randomly selected subjects. c) subjects treated differently. d) subjects treated similarly.

C

If the value of the observed F is less than one -- say, it equals 0.83 -- you can conclude that the null hypothesis a) is false. b) is true. c) is probably true. d) could be true.

D

In those cases where the status of a particular comparison is ambiguous, being designated as significant by some multiple comparison tests and as nonsignificant by the remaining tests, it is suggested that this comparison be reported as having a) borderline significance. b) questionable significance. c) latent significance. d) no significance.

A

It is recommended that reports in the literature describe not only the results of statistical tests and estimates of effect size, but also a) descriptions of group means and standard deviations. b) an ANOVA summary table. c) descriptions of detailed calculations. d) all of these.

A

An F test of the null hypothesis is based on the notion that if the null hypothesis is true, the numerator of the F ratio tends to be a) substantially larger than its denominator. b) about the same as its denominator. c) larger than its denominator. d) smaller than its denominator.

B

Given a significant F test, the size of the overall effect can be estimated by using a measure equal to the squared a) correlation between independent and dependent variables. b) curvilinear correlation between independent and dependent variables. c) t ratio. d) treatment effect.

B

In analysis of variance, variance estimates often are referred to as a) variance means. b) mean squares. c) sum of squares. d) degrees of freedom.

B

Rather than being concerned about the single multiple comparison test that is most appropriate in a particular application, you could initiate with the aid of a computer program a) several multiple comparison tests. b) an entire series of multiple comparison tests. c) a protected t test. d) a more complete description of all mean differences.

B

Regardless of whether the null hypothesis is true or false, variability within groups reflects a) only the treatment effect. b) only random error. c) random error and possibly the treatment effect. d) both random error and the treatment effect.

B

To pinpoint specific differences between pairs of population means, a) do another study. b) use a multiple comparison test. c) use an F test. d) inspect the observed differences between sample means.

B

Tukey's test should be used only if a) the overall null hypothesis is retained. b) the overall null hypothesis is rejected. c) analysis of variance is used. d) an F test is used.

B

Use analysis of variance rather than a t test whenever the null hypothesis makes a claim about a) more than one population mean. b) more than two population means. c) the population variance. d) the population shape.

B

You needn't be too concerned about violating the assumptions of the F test so long as a) sample sizes are equal and very large. b) sample sizes are equal and larger than 10. c) sample variances are equal and very large. d) sample variances are equal and larger than 10.

B

Essentially, the F ratio reflects the ratio of the a) true null hypothesis to the false null hypothesis. b) treatment effect to the null hypothesis. c) observed sample mean differences to the estimated error term. d) observed sample mean differences to the null hypothesis.

C

Given the following ANOVA summary table, SOURCE SS df MS F ________________________________ Between 60 4 15 0.75 Within 1000 50 20 Total 1060 54 the total number of subjects equals a) 50 b) 54 c) 55 d) 108

C

Rejection of the overall null hypothesis usually raises some additional questions regarding a) the estimated size of the overall effect. b) which differences between population means cause rejection of the overall null hypothesis. c) both a and b d) neither a nor b

C

You might choose the most appropriate multiple comparison test (from among the many possible tests) depending on a) computational simplicity. b) the level of significance. c) the relative seriousness of type I and II errors. d) any of the above considerations.

C

A treatment effect exists if differences exist between a) all sample means. b) all population means. c) at least one pair of sample means. d) at least one pair of population means.

D

Given absolute differences of 5, 3, 6, 1, 5, and 8 for six pairs of means, and a critical HSD value of 4.23, it would be correct to conclude that the number of significant differences equals a) one. b) two. c) three. d) four.

D

If the null hypothesis is false because of a sizable treatment effect, the value of F tends to be a) considerably smaller than one. b) smaller than one. c) larger than one. d) considerably larger than one.

D

If the null hypothesis is false, variability between groups reflects a) only the treatment effect. b) only random error. c) random error and possibly the treatment effect. d) both random error and the treatment effect.

D

If the observed F equals or exceeds the critical F, the experimental outcome is a) common, and the null hypothesis is retained. b) common, and the null hypothesis is rejected. c) rare, and the null hypothesis is retained. d) rare, and the null hypothesis is rejected.

D

If the sum of squares for between groups equals 50 and that for within groups equals 70, the sum of squares for total variability must a) lie between 50 and 70. b) equal 60. c) equal 20. d) equal 120.

D

In the Minitab output for one-factor ANOVA, each of the three sample means generates a 95 percent confidence interval based on the pooled standard deviation or a) between groups mean square. b) within groups or error mean square. c) square root of the between groups mean square. d) square root of the within groups or error mean square.

D

In the analysis of variance, the F test is equivalent to a a) directional test because only the upper tail is critical. b) directional test because all variations are squared. c) nondirectional test because only the upper tail is critical. d) nondirectional test because all variations are squared.

D

The squared curvilinear correlation indicates the proportion of variance in a) independent variable caused by the dependent variable. b) independent variable attributable to the dependent variable. c) dependent variable caused by the independent variable. d) dependent variable attributable to the independent variable.

D

When all possible differences between pairs of population means are evaluated not with an F test, but with a series of regular t tests, the probability of at least one a) type I error is larger than the specified level of significance. b) type I error is smaller than the specified level of significance. c) type II error is smaller than the specified level of significance. d) type II error is larger than the specified level of significance.

A

When there is very extensive overlap between scores for all groups, the value of the squared curvilinear correlation tends toward a) zero. b) one-half. c) one. d) any number between zero or one.

A

In a table that summarizes the results of an analysis of variance, the word "Between" might be replaced by a) "Within." b) "Error." c) "Treatment." d) "Total."

C