Psychology 207: Exam 4 (Ch. 12-15)

We want to study the mean difference in autonomy between first-born and second-born children. Instead of taking a random sample of children we take a random sample of families and sort the children into first- and second-born. The dependent variable is a measure of autonomy. This experiment would most likely employ: a. A repeated measures analysis. b. An independent measures analysis. c. A scatterplot. d. A correlation coefficient.

*a. A repeated measures analysis.*

When we speak about pooling variances we are referring to: a. Averaging the two groups' variances. b. Using the variance of the larger group. c. Adding up the variances in each of two groups and dividing by 2. d. Taking the variance of both sets of data combined.

*a. Averaging the two groups' variances.*

Which of the following would most likely NOT be independent samples? a. Group 1 contains wives and Group 2 contains their husbands. b. Group 1 contains women and Group 2 contains men. c. Group 1 contains dyslexic children and Group 2 contains non-dyslexic children. d. Group 1 contains smokers and Group 2 contains nonsmokers.

*a. Group 1 contains wives and Group 2 contains their husbands.*

If we have two experiments in which the groups are equally different, the one with the larger power will be the one that: a. Had the larger sample size. b. Had the smaller level of alpha. c. Had the larger beta. d. Assigned subjects randomly.

*a. Had the larger sample size.*

If we have two related samples, the power of our test will: a. Increase as the correlation between the two samples increases. b. Increase as the correlation between the two samples decreases. c. Be at its maximum when the correlation is zero. d. The correlation is irrelevant.

*a. Increase as the correlation between the two samples increases.*

When testing the means of independent samples, the null hypothesis is best thought of as: a. The mean of population 1 is equal to the mean of population 2. b. The mean of population 1 is unequal to the mean of population 2. c. The mean of a set of difference scores is equal to 0. d. The two population means don't differ by more than 10 points.

*a. The mean of population 1 is equal to the mean of population 2.*

If we use the standard approach to solve for the necessary sample size so that we have power = .75 in a two-sample t test, the value of N that we will obtain from the tables is: a. The number of subjects we will need in each group. b. The number of subjects we will need overall. c. The number of groups we will need to run. d. The number of subjects that we actually have available.

*a. The number of subjects we will need in each group.*

If the standard deviation of the population is 15 and we repeatedly draw samples of 25 observations each, the resulting sample means will have a standard error of: a. 2 b. 3 c. 15 d. 0.60

*b. 3*

When we are worrying about power we are most concerned with, a. A Type I error. b. A Type II error. c. The right answer to the wrong question. d. None of these.

*b. A Type II error.*

A weighted average is: a. The average of two weights. b. An average that gives greater influence to the larger sample. c. An average that reduces the influence of the larger sample. d. The balance between the sample size and the mean.

*b. An average that gives greater influence to the larger sample.*

The point of calculating effect size measures is to: a. Decide if something is statistically significant. b. Convey useful information to the reader about what you found. c. Reject the null hypothesis. d. Prove causality.

*b. Convey useful information to the reader about what you found.*

A t test, in general, involves: a. Dividing the difference between means by the standard deviation of the population. b. Dividing the difference between means by the standard error of a distribution of differences between means. c. Dividing the difference between means by the standard deviation of the sample. d. Dividing the difference in standard deviations by the size of the larger mean.

*b. Dividing the difference between means by the standard error of a distribution of differences between means.*

Which of the following was NOT an advantage of repeated measures designs discussed in the text? a. It helps to control for extraneous variables. b. It is easier to calculate the statistics. c. It allows us to avoid problems associated with variability from subject to subject. d. It requires fewer subjects that other designs.

*b. It is easier to calculate the statistics.*

For a t test with one sample we: a. Lose one degree of freedom because we have a sample. b. Lose one degree of freedom because we estimate the population mean. c. Lose two degrees of freedom because of the mean and the standard deviation. d. Have N degrees of freedom.

*b. Lose one degree of freedom because we estimate the population mean.*

The variance of an individual sample is more likely than not to be: a. Larger than the corresponding population variance. b. Smaller than the corresponding population variance. c. The same as the population variance. d. Less than the population mean.

*b. Smaller than the corresponding population variance*

A repeated measures t test is more likely to lead to rejection of the null hypothesis if: a. Subjects show considerable variability in their change scores. b. The degree of change is consistent across subjects. c. Many subjects show no change. d. Some subjects change a lot more than others.

*b. The degree of change is consistent across subjects.*

Cohen's d refers to: a. The probability associated with t. b. The difference between the means before and after treatment divided by a standard deviation. c. The difference between the means before and after treatment. d. The "diagnostic" statistic.

*b. The difference between the means before and after treatment divided by a standard deviation.*

In the test on the treatment of anorexia, the resulting t was -3.22, which was significant (p < .05). This tells us that: a. The family therapy condition had a mean that was reliably less than the mean of the control condition. b. The family therapy condition had a mean that was reliably more than the mean of the control condition. c. The family therapy condition had a mean that was unreliably less than the mean of the control condition. d. The family therapy condition had a mean that was unreliably more than the mean of the control condition.

*b. The family therapy condition had a mean that was reliably more than the mean of the control condition.*

The null hypothesis of a related scores t test is: a. Mean 1 equals (=) the mean of 2 which is equal to the mean of D + 1 b. The mean of D is equal to the difference of mean of 2 and mean 1 which is equal to 0 c. The mean of 1 equals 0 d. The mean of D equals the mean of 2

*b. The mean of D is equal to the difference of mean of 2 and mean 1 which is equal to 0*

To increase power, the easiest variable to control in designing an experiment is usually: a. The difference between the population means. b. The sample size. c. The sample standard deviation. d. The shape of the distribution.

*b. The sample size.*

The effect size can best be thought of as: a. The size of the largest mean. b. The size of the difference between means in standard deviation units. c. A mixture of the size of the difference and the sample size. d. The largest difference we would hope to find.

*b. The size of the difference between means in standard deviation units.*

It is important to think about power because: a. Everyone uses it. b. We don't want to run experiments that have little chance of finding something. c. It is likely to guarantee successful studies. d. None of these.

*b. We don't want to run experiments that have little chance of finding something.*

In one-sample tests of means we: a. compare one sample mean with another b. compare one sample mean against a population mean c. compare two sample means to each other d. compare a set of population means

*b. compare one sample mean against a population mean*

The text suggested that the highest value of power that we are likely to be able to afford, assuming that we don't have huge differences between our groups, is something like: a. .20 b. .50 c. .80 d. .95

*c. .80*

Mean scores in Tetris were recorded for both math majors and English majors. The mean score for math majors was 20,752 (s12 = 851, N1 = 17) and the mean score for English majors was 14,922 (s22 = 395, N2 = 56). The pooled variance would probably be closest to which value? a. 800 b. 623 c. 500 d. 900

*c. 500*

In calculating power we calculate a statistic called delta. This statistic is: a. The effect size. b. The sample size. c. A combination of the effect size and the sample size. d. The difference we are looking for.

*c. A combination of the effect size and the sample size.*

Cohen's d is an example of: a. A measure of correlation. b. A r-family measure. c. A d-family measure. d. A correlational measure.

*c. A d-family measure.*

In deciding whether or not to pool the variances in calculating a t score, one should consider: a. The differences in the sample variances. b. The differences in the sample sizes. c. Both the differences in the sample variances and the differences in the sample sizes. d. Neither the differences in the sample variances nor the differences in the sample sizes.

*c. Both the differences in the sample variances and the differences in the sample sizes.*

The effect size is generally denoted by: a. Beta. b. Alpha. c. D-hat. d. Delta

*c. D-hat.*

When people such as Cohen say that a medium size difference has an effect size of .50, they are saying that we want the power for: a. One mean being half as large as another. b. One mean being half a point larger than another. c. One mean being half a standard deviation larger than another. d. One mean being based on half as many subjects as the other.

*c. One mean being half a standard deviation larger than another.*

The term "effect size" refers to: a. How large the resulting t statistic is. b. The size of the p value, or probability associated with that t. c. The actual magnitude of the mean or difference between means. d. The value of the null hypothesis.

*c. The actual magnitude of the mean or difference between means.*

The example of the moon illusion discussed in the text illustrates the fact that: a. The moon grows as it rises in the sky. b. An effect size is best based on medians. c. The best estimate of the size of an effect need not use the standard deviation. d. Standardized effect sizes are clearly superior.

*c. The best estimate of the size of an effect need not use the standard deviation.*

The most important characteristic of two independent samples is: a. We measure the same subjects at two separate times. b. The subjects in one sample are part of the same family as subjects in the other sample. c. The set of scores for one sample is uncorrelated with the scores in the other sample. d. The subjects in a group are people who know each other.

*c. The set of scores for one sample is uncorrelated with the scores in the other sample.*

As the value of the mean difference score decreases: a. The t score increases. b. The t score stays the same. c. The t score decreases. d. You cannot predict how the t score will be affected.

*c. The t score decreases.*

For a very conservative test when we don't pool the variances we should: a. Work at alpha = .01 instead of alpha = .05. b. Insist that the critical value of t be twice the tabled value. c. Use a much reduced number of degrees of freedom. d. Throw up our hands and call the problem too difficult.

*c. Use a much reduced number of degrees of freedom.*

If we have run a t test with 35 observations and have found a t of 3.60, which is significant at the .05 level, we would write: a. t(35) = 3.60, p <.05 b. t(34) = 3.60, p >.05 c. t(34) = 3.60, p <.05 d. t(35) = 3.60, p <05

*c. t(34) = 3.60, p <.05*

All of the following increase the magnitude of the t statistic and/or the likelihood of rejecting H0 EXCEPT: a. A greater difference between the sample mean and the population mean. b. An increase in sample size c. A decrease in sample variance. d. A smaller significance level (alpha).

*d. A smaller significance level (alpha).*

Which of the following is NOT a part of the Central Limit Theorem? a. The mean of the sampling distribution approaches the population mean. b. The variance of the sampling distribution approaches the population variance divided by the sample size. c. The sampling distribution will approach a normal distribution as the sample size increases. d. All of these are a part of the Central Limit Theorem

*d. All of these are a part of the Central Limit Theorem*

Which of the following is involved in the calculation of power? a. Alpha b. The standard deviation c. The difference between the means d. All of these are involved.

*d. All of these are involved.*

All other things being equal, one study will have more power than another study if: a. It has more subjects. b. It controls error variance better. c. It compares groups that are more extreme. d. All of these.

*d. All of these.*

Which of the following is sometimes a serious problem with repeated measures designs? a. They require more subjects than designs with independent samples. b. All of these choices. c. Small sample sizes can distort the results more than with other designs. d. Carryover effects can cloud the interpretation.

*d. Carryover effects can cloud the interpretation.*

When we are using a two-tailed hypothesis test, the null hypothesis is of the form: a. H1: mean is not equal to 50 b. H1: mean is less than 50 c. H1: the mean is greater than 50 d. H0: the mean equals 50

*d. H0: the mean equals 50*

A harmonic mean can be used to calculate the mean sample size for unequal sample sizes. For a(n) ________ the values of k would equal ________. a. Related means t test; 3 b. Related means t test; 2 c. Independent samples t test; 3 d. Independent samples t test; 2

*d. Independent samples t test; 2*

If we built on Doob and Gross's study of horn-honking behavior to compare the time it takes for people to honk at an older car versus the time to honk at an expensive newer car, we could set it up as: a. An independent sample design. b. A repeated measures design. c. Neither an independent sample design nor a repeated measures design. d. It could be designed either way, depending upon our needs.

*d. It could be designed either way, depending upon our needs.*

With a one-sample t test, the value of t is: a. Always positive. b. Positive if the sample mean is too small. c. Negative whenever the sample standard deviation is negative. d. Positive if the sample mean is larger than the hypothesized population mean.

*d. Positive if the sample mean is larger than the hypothesized population mean.*

A 2-tailed significance level of .672 tells us: a. The means of the pre- and post-scores are large. b. The difference is significant at the .05 level. c. We can't tell whether the difference is significant or not. d. The difference is not significant at the .05 level.

*d. The difference is not significant at the .05 level.*

The null hypothesis in a repeated measures t test is: a. The hypothesis that the variance of scores stays constant from pretest to posttest. b. The hypothesis that post scores are larger than pre-scores. c. The hypothesis that the mean difference score is different from 0 in either direction. d. The hypothesis that the mean difference score is equal to 0.

*d. The hypothesis that the mean difference score is equal to 0.*

If we knew the population mean and variance, we would expect: a. The sample mean would closely approximate the population mean. b. The sample mean would differ from the population mean by no less than 1.96 standard deviations only 5% of the time. c. The sample mean would differ from the population mean by no more than 1.64 standard deviations only 5% of the time. d. The sample mean would differ from the population mean by more than 1.96 standard errors only 5% of the time.

*d. The sample mean would differ from the population mean by more than 1.96 standard errors only 5% of the time.*

If we have calculated a confidence interval and we find that it does NOT include the population mean: a. We must have done something wrong in collecting data. b. Our interval was too wide c. We made a mistake in calculation. d. This will happen a fixed percentage of the time.

*d. This will happen a fixed percentage of the time.*

What is the power in an experiment with two independent groups when the null hypothesis is true? a. .05 b. .95 c. .05/2 d. Undefined

*d. Undefined*

Suppose that we know that the sample mean is 18 and the population standard deviation is 3. We want to test the null hypothesis that the population mean is 20. In this situation we would: a. Reject the null hypothesis at alpha = .05. b. Reject the null hypothesis at alpha = .01 c. Retain the null hypothesis. d. We cannot solve this problem without knowing the sample size.

*d. We cannot solve this problem without knowing the sample size.*

Which of the following statistics comparing a sample mean to a population mean is most likely to be significant if you used a two-tailed test? a. t = 10.6 b. t = 0.9 c. t = -10.6 d. both t = 10.6 and t = -10.6

*d. both t = 10.6 and t = -10.6*

A researcher was interested in examining the different levels of aggression shown by college students and prisoners with life sentences. Aggression was measured by surveys given during one session. Which analysis should the researcher perform on the mean aggression scores of the groups? a. Related means t test b. Independent means t test c. Correlation d. Regression

b. Independent means t test