Psych 213 Spring 2003 Practice exam



Psych 213 Spring 2003 Practice exam

1. The chi-square distribution is like the t distribution

in which of the following ways:

a. rejection regions/critical values are in the both tails

b. both depend on the df of the study

c. neither is symmetrical

d. both have a mean of 0

2. If your Chi-square result is significant, which of the following is true:

a. the obtained proportions would rarely be seen if Ho is true

b. the obtained proportions are likely if Ho is true

c. the obtained sample mean would rarely be seen if Ho is true

d. the obtained sample mean is likely if Ho is true

3. The basic logic of the Chi-square test of independence is that

a. a comparison of the means within each group provides information as to whether there is an effect of the independent variable

b. a comparison of obtained frequencies to expected frequencies provides information as to whether there is a relationship between the two variables

c. a comparison of the estimated population variance using the variance among group means to the estimated population variance using the variance within groups provides information as to whether there is an effect of the independent variable.

d. none of the above

5. One way to make a confidence interval for a population mean smaller (narrower) is to:

a. use a smaller sample mean

b. decrease the sample size

c. increase the level of confidence (i.e., from 95% to 99%)

d. none of the above

Computational problems

1. In a study of reading speed (measured in words per minute), researchers were interested in the effect of the size of type. Data were collected from 20 participants who were randomly assigned to conditions, and they are summarized below. Answer questions a through f using this information.

Groups Means Variance

small type 382 795.24

large type 436 686.44

a. State the null and alternative statistical hypotheses for this study.

b. What type of statistical test is appropriate to test the null hypothesis you gave for question 6?

c. The estimated standard error for this study would be computed using which formula? (although you will not be expected to carry out this computation on our exam) Please compute the estimated standard error here:

d. What is the value you obtain for your test statistic?

e. Make a statistical decision about your results. Use an alpha level of .05.

f. Draw a research conclusion from your results.

g. If the effect of type size is really .4, how much power did the experimenter have in this study?

2. A researcher is interested in whether the gender mix of developmental psychology graduate school applicants (nationally) in 2002 is the same as that for past years. In previous years the proportion of male applicants has been 50%, and the proportion of female applicants has been 50%. Several psychology departments keeping records of graduate school applicants’ race were contacted, and altogether their total numbers are believed to constitute a random sample of all graduate school applicants. The data obtained follow:

male applicants 136

female applicants 188

Total 324

a. What type of statistical analysis is appropriate to test the hypothesis that gender mix was the same in 2002 as in previous years.

b. Test the hypothesis that the racial mix was the same in 2002 as in previous years. What is your obtained value of the test statistic?

c. Make a statistical decision. Use an alpha level of .01.

3. A psychologist has obtained a random sample of 200 high school students, and each student has been classified in terms of gender and use or non-use of marijuana. The data are as follows:

| |user |non-user |

|M |8 |32 |

|F |64 |96 |

a. What type of statistical analysis is appropriate to test the hypothesis that gender and marijuana use are independent?

b. State the null and alternative hypotheses.

c. What is the value of the obtained test statistic?

d. Make a statistical decision. Use an alpha level of .01.

e. Draw a research conclusion

4. A psychologist is interested in the effectiveness of a new reading program. Students in a local kindergarden are randomly assigned to a control group (usual reading method) or to participate in the new reading course for six weeks. Because the researcher is concerned that children vary widely in their reading ability, she decides to compare the new program with the control group by matching children between the groups on reading achievement scores (measured before the six week program). The data are summarized below:

Pair control program

1 14 22

2 5 9

3 10 8

4 6 13

5 8 7

6 12 19

a. State the null and alternative hypotheses for this study.

b. What is the appropriate statistical method to test the null you gave above?

c. What is the value of the obtained test statistic. [Please note that for this practice test I am asking you to calculate the appropriate standard error. I may not ask you to do this on the final exam (I am more likely to give you intermediate values or standard errors to choose from).]

d. Make a statistical decision. Use an alpha level of .05.

e. Draw a research conclusion.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download