Answer the following questions and submit for grading. 1. An ESP experiment is done in which a participant guesses which of 8 cards the researcher has randomly picked, where each card is equally likely to be selected. This is repeated for 200 trials. The null hypothesis is that the subject is guessing, while the alternative is that the subject has ESP and can guess at higher than the chance rate. Write out the type 1 and type 2 errors in terms of this problem. 2. For each of the following, write out the null and alternative hypotheses. . a. Do female students, on the average, have a higher GPA? b. Is there a linear relationship between height and weight? c. Is there a difference in the proportions of male and female college students who smoke? Two-sample for 1st_Grade vs 12th_Grade Mean StDev SE Mean 1st_Grade 50 111.2 88.9 13 12th_Grade 50 49.5 38.8 5.5 Difference = mu 1st_Grade mu 12th_Grade Estimate for difference: 61.7 95% CI for difference: (34.3, 89.1) -Test of difference = 0 (vs not =): -Value = 4.50 -Value = 0.000 DF = 66 Sample Sample < 40 yrs 600 1000 0.60 ? 40 yrs 450 1000 0.45 Estimate for (1) - (2): 0.15 95% CI for (1) - (2): (0.107, 0.193) Test for (1) - (2) = 0 (vs not = 0): = 6.72 -Value = 0.000 5. For patients with a particular disease, the population proportion of those successfully treated with a standard treatment that has been used for many years is .75. A medical research group invents a new treatment that they believe will be more successful, i.e., population proportion will exceed .75. A doctor plans a clinical trial he hopes will prove this claim. A sample of 100 patients with the disease is obtained. Each person is treated with the new treatment and eventually classified as having either been successfully or not successfully treated with the new treatment. Test of p = 0.75 vs > 0.75 Sample X Sample p -Value -Value 1 80 100 0.800000 1.15 0.124 6. Refer to the information found in thearticle entitled from the Penn State Pulse (January, 2001). This was previously used in Lesson 9. a. What is the majority of the type of data summarized on the first page of this article? Measurement or categorical b. What population value should be used with this data? population mean or population proportion c. At the bottom of the first page of the article you find the statement * statistically significant at the .05 level. This statement implies that the -value is ? .05. Find the *s on the ofthe article. Precisely what two results are statistically significant? State these results in terms of the appropriate population value ie: population mean or population proportion). : Penn State Pulse, 21 Birthday, January 2001 7. Refer to the following article located in the Library Reservesuse the Library Reserves link in Angel Kirchheimer, S. (May 17, 2003). In studies that compare never smokers married to smokers with never smokers married to never smokers, the explanatory variable is ______ a. whether or not the spouse smokes. b. whether or not the person was married. c. whether or not the person developed lung cancer. d. whether or not the smoke is secondhand. A study that compares never smokers married to smokers with never smokers married to never smokers is which of the following? a. randomized experiment b. observational study c. matched pairs study The number 30% in this article represents which of the following quantities? a. risk b. relative risk c. increased risk d. odds Enstroms study is which of the following? a. randomized experiment b. prospective study c. retrospective study This article identifies the funding source used by Enstrom. As a statistical sleuth, what should you conclude from Enstroms study after knowing his funding source? a. results are definitely biased b. must first evaluate scientific procedures used in study before interpreting results c. results are definitely unbiased Which of the following is a concern about the study that was conducted by Enstrom? a. extending conclusions to all people in the United States b. the existence of confounding variables c. smoking habits probably changed from 1972 to 1998 d. results are based on a very small sample size
