Test Bank Docx Chapter.14 T Test For Two Independent Samples

MULTIPLE‑CHOICE TEST ITEMS

CHAPTER 14

t TEST FOR TWO INDEPENDENT SAMPLES

14.1 Subjects in an experiment are viewed as

a) a random sample from a small real population.

b) a random sample from a large real population.

c) if they were a random sample from a small hypothetical population.

d) if they were a random sample from a large hypothetical population.

14.2 Generalizations to hypothetical populations are viewed as

a) conclusive.

b) provisional.

c) valid.

d) worthless.

14.3 In a study with 30 grade‑school children per group, students in the experimental reading program do 10 points better, on the average, than do students in the regular program. Therefore you can conclude that this difference is

a) statistically significant.

b) important.

c) statistically significant and important.

d) undecipherable without additional information.

14.4 To determine whether it qualifies as a common or rare outcome under the null hypothesis, the one observed difference between sample means is viewed as originating from the

a) two samples.

b) two underlying populations.

c) sampling distribution for the sample mean.

d) sampling distribution for the differences between sample means.

14.5 The mean of all possible differences between sample means equals

a) the difference between population means.

b) any observed difference between sample means.

c) most observed differences between sample means.

d) the population mean.

14.6 If daily meditation has little or no effect on college grade point average, the difference between population means is

a) negative.

b) close to zero.

c) positive.

d) unknown.

14.7 The null hypothesis presumes that any observed difference between sample means

a) is a common outcome.

b) reflects an effect.

c) is due to variability.

d) occurs rarely.

14.8 The null hypothesis assumes that the difference between population means is

a) zero.

b) virtually zero.

c) some number reflecting the research hypothesis.

d) unknown.

14.9 The standard error of the differences between sample means serves as a rough indicator of the average amount by which

a) individual observations deviate from the observed difference between sample means.

b) individual observations deviate from the hypothesized difference between population means.

c) observed differences between sample means deviate from the difference between population means.

d) observed differences between sample means deviate from the population mean.

14.10 Generally speaking, the standard error serves as a yardstick to determine whether the observed difference between sample means can be reasonably attributed to

a) variability.

b) the sampling distribution.

c) the underlying populations.

d) the research hypothesis.

14.11 The standard error would be smallest if both sample sizes equal

a) 25

b) 20

c) 15

d) 10

14.12 The z test for two independent samples is hardly ever appropriate – and hence ignored in the textbook -- because it requires that both

a) sample sizes be very large.

b) sample sizes be equal.

c) population means be known.

d) population standard deviations be known.

14.13 The t test for two independent samples has degrees of freedom equal to the

a) two sample sizes combined.

b) two sample sizes combined minus one.

c) two sample sizes combined minus two.

d) two sample sizes combined minus three.

14.14 The pooled variance estimate (used to calculate the standard error) can be characterized as a type of __________________ for the two sample variances.

a) mean

b) standard deviation

c) variance

d) range

14.15 Given a concern only that meditation improves college grade point averages, the alternative hypothesis should be a

a) two‑tailed test.

b) one‑tailed test with the upper tail critical.

c) one‑tailed test with the lower tail critical.

d) test with the .05 level of significance.

14.16 The p‑value for a test result represents the probability

a) that the null hypothesis is true.

b) that the decision is correct.

c) of the observed result, whether the null hypothesis is true or false.

d) of the observed result, given that the null hypothesis is true.

14.17 Smaller p-values tend to discredit the

a) null hypothesis.

b) research hypothesis.

c) research.

d) researcher.

14.18 A p-value represents, most directly, the degree of

a) rarity of the test result.

b) suspicion about the null hypothesis.

c) truth of the null hypothesis.

d) credibility of the test result.

14.19 Two-tailed p-values can be described as being

a) twice as large as the corresponding one-tailed p-value.

b) appropriate when the investigator is concerned about deviations in either direction.

c) equivalent to shaded areas located in both tails of the sampling distribution.

d) all of the above

14.20 A researcher would prefer to report an approximate p‑value that is

a) less than .05.

b) more than .05.

c) less than .01.

d) more than .01.

14.21 Given tabular entries of 2.145, 2.977, and 4.140 that correspond to p‑values of .05, .01, and .001 respectively, the p‑value for an observed t of 3.50 could be expressed as

a) less than .001.

b) less than .01.

c) less than .05.

d) more than .05.

14.22 Given tabular entries of and that correspond to p‑values of .05, .01, and .001 respectively, the p‑ value for an observed t of could be expressed as

a) less than .001.

b) less than .01.

c) less than .05.

d) more than .05.

14.23 A researcher would prefer to report an exact p-value of

a) .005

b) .03

c) .09

d) .15

14.24 The less structured (p‑value) approach to hypothesis testing is very attractive when

a) test results are borderline.

b) a type I error is particularly serious.

c) sample size is small.

d) the null hypothesis is false because of a large effect.

14.25 One weakness of the less structured p-value approach is that it makes it difficult to deal with the important notions of

a) causality.

b) the null and research hypotheses.

c) common and rare outcomes.

d) the type I and type II errors.

14.26 Unlike a p‑value, a level of significance is a degree of rarity

a) based on the assumption that the null hypothesis is true.

b) derived from the tabled sampling distribution.

c) specified before the test result has been observed.

d) associated with a specific probability.

14.27 A p‑value less than .01 implies that, in a more structured hypothesis test, the null hypothesis would have been

a) retained at the .05 level.

b) retained at the .01 level.

c) rejected at the .01 level.

d) rejected at the .001 level.

14.28 Statistical significance indicates that the null hypothesis is

a) false.

b) mildly false because of a small effect.

c) seriously false because of a large effect.

d) probably false.

14.29 Statistical significance that lacks importance is often caused by the use of

a) unduly small sample sizes.

b) excessively large sample sizes.

c) loose research hypotheses.

d) vague null hypotheses.

14.30 The importance of a statistically significant result might be clarified with

a) a point estimate based on the observed difference between means.

b) a confidence interval for the difference between population means.

c) Cohen's standardized effect estimate, d.

d) all of the above.

14.31 If two independent samples each consists of 20 subjects, a 95 percent confidence interval will based on degrees of freedom equal to

a) 20

b) 38

c) 39

d) 40

14.32 The numerical limits of a confidence interval for differences between population means refer to

a) each sample mean.

b) differences between sample means.

c) each population mean.

d) differences between population means.

14.33 If dissimilar signs are associated with the limits for a 95 percent confidence interval for differences between population means, a consistent interpretation is

a) possible.

b) possible under special circumstances.

c) not possible.

d) usually not possible.

14.34 Prior to taking a written test of self‑esteem (scored from a low of 0 to a high of 100), shy volunteers are randomly assigned to participate in weekend workshops dealing with either assertive behavior or group recreation. After analyzing their subsequent performance on the test of self‑esteem, the investigator reports a 95 percent confidence interval of to 14 points, tilted in favor of the group that had the workshop on assertive behavior. The boundaries for the 95 percent confidence interval ( to 14) require the conclusion that there is

a) a consistent effect in favor of the workshop for assertive behavior.

b) a consistent effect in favor of the workshop for group recreation.

c) no consistent effect.

d) evidence that the null hypothesis is probably true.

14.35 Prior to taking a written test of self‑esteem (scored from a low of 0 to a high of 100), shy volunteers are randomly assigned to participate in weekend workshops dealing with either assertive behavior or group recreation. After analyzing their subsequent performance on the test of self‑esteem, the investigator reports a 95 percent confidence interval of to 14 points, tilted in favor of the group that had the workshop on assertive behavior. The boundaries for the 95 percent confidence interval ( to 14) indicate that

a) no subject lost more than 5 points or gained more than 14 points on the test of self‑esteem.

b) about 95 percent of the subjects lost less than 5 points or gained less than 14 points on the test of self‑esteem.

c) the true population mean difference is between to 14 points.

d) the true population mean difference is probably between to 14 points.

14.36 You hope to demonstrate that a workshop on assertive behavior increases self‑esteem relative to a control workshop, as indicated by positive differences in the limits of a 95 percent confidence interval. Which would be the most preferred interval among the following list of possible 95 percent confidence intervals?

a) to 14

b) 3 to 8

c) 1 to 5

d) to 11

14.37 Cohen's d is obtained by dividing the observed mean difference by

a) the estimated standard error.

b) the pooled variance.

c) the best estimate of the sample standard deviation.

d) none of the above.

14.38 Cohen's d is

a) unit free.

b) immune to sample size.

c) appropriate whenever the mean difference is statistically significant.

d) all of the above.

14.39 According to the guidelines for Cohen's d, a value in the vicinity of .2 suggests that the estimated effect is

a) small and important.

b) small and could lack importance.

c) large and important.

d) large and could lack importance.

14.40 The value of Cohen's d is not inflated by

a) small sample sizes.

b) moderate sample sizes.

c) large sample sizes.

d) unequal sample sizes.

14.41 Values of d can be visualized as the degree of separation between pairs of normal curves. As the value of d increases, the separation between normal curves

a) increases.

b) decreases.

c) varies.

d) all of the above.

14.42 A distinctive feature of a meta-analysis is its dependence on

a) expert judgment.

b) a traditional literature review.

c) an intensive review of all relevant studies.

d) published studies only.

14.43 “Publication bias” refers to only reporting

a) results favorable to the Investigator’s point of view.

b) statistically significance results.

c) replicated studies.

d) well-publicized studies.

14.44 Published reports often include parenthetical statements that summarize

a) the research hypothesis.

b) any assumptions associated with the statistical analysis.

c) the statistical analysis.

d) the statistical analysis, including a p‑value and an estimate of effect size.

14.45 Error bars for dots in data graphs could reflect the size of the

a) standard deviation.

b) standard error.

c) confidence interval.

d) all of the above.

14.46 The t test for two independent samples assumes that both underlying populations are normally distributed with equal variances. You needn't be too concerned about violations of the assumptions of t for two independent samples (normally distributed populations with equal variances) as long as both samples sizes are

a) equal and fairly large.

b) equal and fairly small.

c) unequal and fairly large.

d) unequal and fairly small.

14.47 Ordinarily, when reading a computer output for a t test for two independent samples, you would report the t test results for

a) unequal or separate variances.

b) equal or pooled variances.

c) neither of the above

d) both of the above