Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Slide 21Slide 22Slide 23Slide 24Slide 25Slide 26Slide 27Slide 28Slide 29Slide 3001/14/19 Slide 1The introductory statement in the question indicates:•The data set to use: GSS2000R.SAV•The task to accomplish: a one-sample test of a population proportion•The variable to use in the analysis: favor or oppose death penalty for murder [cappun]•The population proportion to use in the comparison with our sample: 22%•The alpha level of significance for the hypothesis test: 0.0501/14/19 Slide 2The first statement asks about the level of measurement. The SPSS Binomial test only works if the variable is dichotomous.01/14/19 Slide 3To determine the level of measurement, we examine the information about the variable. Select the Variables command from the Utilities menu.01/14/19 Slide 4Scroll down the list of variables to locate cappun."Favor or oppose death penalty for murder" [cappun] contains two categories that do not represent missing data. The variable is dichotomous, satisfying the requirement to use the SPSS Binomial Test.Close the dialog box when we are finished.01/14/19 Slide 5"Favor or oppose death penalty for murder" [cappun] is dichotomous, satisfying the requirement to use the SPSS Binomial Test.Mark the check box for a correct answer.01/14/19 Slide 6If we had specified a variable that is not dichotomous, SPSS prints this message in the output. SPSS can compute the Binomial test for variables that have more than two categories if we recode the variable, or if we use the syntax version of the SPSS commands.01/14/19 Slide 7The next statement asks about the sample size. If we have 10 or more subjects in both the group that has the characteristic identified in the proportion and the group that does not have it, we can rely on the Central Limit Theorem to justify using probabilities based on the normal distribution.To answer this question, we compute the Binomial Test in SPSS.01/14/19 Slide 8To compute the one-sample test of a proportion in SPSS, select the Nonparametric Tests > Binomial test from the Analyze menu.01/14/19 Slide 9First, move the variable cappun to the Test Variable List.Second, enter the test proportion stated in the problem (from previous research).Third, click on the OK button to produce the output.01/14/19 Slide 10The sample size is large enough to make the sampling model for the sample proportions approximately normal based on the Central Limit Theorem if there are 10 or more cases in each of the categories compared in the test. There were 67 in the category "survey respondents who opposed the death penalty for persons convicted of murder" and 171 in the other category "survey respondents who favored the death penalty for persons convicted of murder". Both groups used to test the hypothesis had the required minimum of 10 cases. The sample size requirement is satisfied.01/14/19 Slide 11Both groups used to test the hypothesis had the required minimum of 10 cases. The sample size requirement is satisfied.Mark the check box for a correct answer.01/14/19 Slide 12The next questions asks us about the size of the proportion of cases in the target category, “opposed.”01/14/19 Slide 13The correct percentage of survey respondents in the category opposed in our sample was 28.2%, which SPSS rounds off to .28.01/14/19 Slide 14The correct percentage of survey respondents in the category opposed in our sample was 28.2%.Mark the check box for a correct answer.01/14/19 Slide 15The next statement asks us what the null hypothesis for the test states.01/14/19 Slide 16The population proportion is given in the statement of the problem.01/14/19 Slide 17This is the value entered in the Test Proportion text box, i.e. 0.22.This is the also the value that SPSS prints in the output table.01/14/19 Slide 18The null hypothesis for the test states that the true proportion of survey respondents who opposed the death penalty for persons convicted of murder is equal to the population proportion, which in this problem is the proportion cited in previous research, 0.220.Mark the check box for a correct answer.01/14/19 Slide 19The next statement asks us about the probability (p-value or sig. value) for the test of a population proportion.01/14/19 Slide 20The probability that the proportion in our sample was different from the proportion reported in previous research was p = .031. The probability for the two-tailed test is computed by doubling the probability of the one-tailed test reported in the SPSS output (2 x .015 = .031).SPSS computes the one-tailed probability for the test. To get the two-tailed probability, we double this number.01/14/19 Slide 21The probability that the proportion in our sample was different from the proportion reported in previous research was p = .031. Mark the check box for a correct answer.01/14/19 Slide 22The next statement asks about the statistical decision or conclusion that we would make based on the p-value.01/14/19 Slide 23When the p-value for the statistical test is less than or equal to alpha, we reject the null hypothesis and interpret the results of the test. If the p-value is greater than alpha, we fail to reject the null hypothesis and do not interpret hypothesis. The p-value for this test (p = .031) is less than or equal to the alpha level of significance ( p = .050) supporting the conclusion that we reject the null hypothesis. Mark the check box for a correct answer.01/14/19 Slide 24The final statement asks us to interpret the results of the statistical test.01/14/19 Slide 25Since we rejected the null hypothesis and since the proportion of survey respondents who opposed the death penalty for persons convicted of murder in our sample is actually larger than the proportion reported in research, it is reasonable to suggest that actual proportion of survey respondents who opposed the death penalty for persons convicted of murder in the population has increased. We mark the check box for a correct answer.When we do not reject the null hypothesis, we do not interpret the results.01/14/19 Slide 26If we did not satisfy the level of measurement or the sample size, we should not use the test or its results.We would not mark any of the check boxes and None of the above would be the correct answer.01/14/19 Slide 27YesTarget variable is dichotomous?YesDo not mark check box.Mark statement check box.No10+