PART ONE. MULTIPLE CHOICE. Please identify the answer which best completes the statement or question. (4 points each) 1. In Chi-square goodness of fit and contingency table tests, it is sometimes necessary to reduce the number of classifications used in order to a. make the distribution more normal. satisfy the condition that the variances are equal. provide the table with larger observed frequencies. get a more accurate estimate of the differences in the means. none of the above. 2. Which of the following null hypotheses are not tested using a chi-square test? a. the proportions of women favoring four weight control methods are equal b the average weekly wages for employees of four firms in the same industry are the same. family size is independent of educational achievement level. the observed distribution of scores is normally distributed with mean = 75 and standard deviation = 10. e. all of the above are Chi-square tests 3. In doing a test of independence, we use all of the following EXCEPT: I an estimate of the population variance. contingency tables. observed and expected frequencies number of rows and columns. a critical value taken from a Chi-Square table. 4. The number of degrees of freedom in a chi-square goodness of fit test depends upon: (1) the number of classes into which the sample observations are classified; (2) the number of observations in the sample; (3) the number of population parameters estimated from the sample data. a 1 only b. 2 only 1 and 2 only 1 and 3 only none of the above 5. is the difference between a sample statistic and its corresponding population parameter. a. Sampling error b. A sampling distribution C. A measurement error d. Bias e. ZeroStyles Editing Voi 6. When a false null hypothesis is rejected a. we commit a Type | error. b. we make the correct decision. c. we chose the wrong alpha. d. we commit a Type II error. e. we rely on the p value to see the probability of the null hypothesis being correct. 7. The correct sample size does NOT depend upon which of these factors? a. desired level of confidence b. margin of error the researcher can tolerate c. variability in the population being studied. d. the mean of the sample e. sample size depends on all of the above answers 8. The size of the standard error is affected by a. mean and standard deviation. b. Z score and mean. c. population size and sample size. d. variability in the population and sample size. e. mean and mode. 9. The population standard deviation a. is always the same as the sample standard deviation. b. measures something different than the sample standard deviation. c. corresponds to the median of the population. d. is usually known. e. affects the dispersion in the sampling distribution of the sample mean. 10. If the expected frequency for any comparison is less than 5: a. you cannot use the Chi-Square distribution b. you must use the finite population correction factor c. you must combine comparisons (categories) so that each expected frequency is at least 5. d. your degrees of freedom for a Chi-Square test will be the number of comparisons minus 1 e. it means that we estimated another population parameter.11. For which of the following situations would the logic underlying a test of independence apply (1) a random sample of students is selected from a large university and each student is classified on the basis of gender and whether or not he/she is employed (2) independent random samples of male and female employees in a large firm are selected and each is classified on a 0-100 job satisfaction scale a. 1 only b. 2 only c. 1 and 2 d. neither 1 nor 2 e. insufficient information to identify if a test of independence applies 12. A survey of 100 students at each of 5 universities are asked to rate their university's library on a scale of 1 (excellent) to 5 (poor). The correct test of the null hypothesis that the population proportion of students giving each rating are the same between two schools is a: a. one way ANOVA b. chi-square test of independence c. chi-square goodness of fit d. randomized block ANOVA e. a test of one population proportion 13. The probability that a Chi-Square random variable with 3 degrees of freedom is greater than is 5%. What number belongs in the blank? a. 0.352 b. 1.96 C. 6.251 d. 7.815 e. none of the above PART TWO. PROBLEMS. Please do all parts of all problems. Points distributed as noted. Partial credit is awarded where applicable. 1. Sears has determined that 14 percent of all purchases made during the Christmas I season are returned. If a store sells 150 video games, what is the probability that at most 18 percent will be returned? (8 pts)2. Tender Lubing Care, an auto diagnostic service in Clearlake Park, California, wants to estimate the average time to service a customer. a. Assume that a sample of 36 customers yielded a mean time of 37 minutes, with s = 6.2 minutes. Construct a 99 percent confidence interval. (6 pts) b. Calculate a 99 percent confidence with the information from part (a) if the sample size s 100. (6 pts) c. Why did the larger sample give you the smaller interval? (4 pts)