12-1 Chapter 12 Chi-Square and Analysis of Variance (ANOVA) © The McGraw-Hill Companies, Inc., 2000 12-2 Outline 12-1 Introduction 12-2 Test for Goodness of Fit 12-3 Tests Using Contingency Tables 12-4 Analysis of Variance (ANOVA) © The McGraw-Hill Companies, Inc., 2000 12-3 Objectives Test a distribution for goodness of fit using chi-square. Test two variables for independence using chi-square. Test proportions for homogeneity using chi-square. Use ANOVA technique to determine a difference among three or more means. © The McGraw-Hill Companies, Inc., 2000 12-4 12-2 Test for Goodness of Fit When one is testing to see whether a frequency distribution fits a specific pattern, the chi-square goodness-of-fit test is used. © The McGraw-Hill Companies, Inc., 2000 12-5 12-2 Test for Goodness of Fit Example Suppose a market analyst wished to see whether consumers have any preference among five flavors of a new fruit soda. A sample of 100 people provided the following data: Cherry Straw- Orange berry 32 28 16 Lime Grape 14 10 © The McGraw-Hill Companies, Inc., 2000 12-6 12-2 Test for Goodness of Fit Example If there were no preference, one would expect that each flavor would be selected with equal frequency. In this case, the equal frequency is 100/5 = 20. That is, approximately 20 people would select each flavor. © The McGraw-Hill Companies, Inc., 2000 12-7 12-2 Test for Goodness of Fit Example The frequencies obtained from the sample are called observed frequencies. The frequencies obtained from calculations are called expected frequencies. Table for the test is shown next. © The McGraw-Hill Companies, Inc., 2000 12-8 12-2 Test for Goodness of Fit Example Freq. Cherry Straw- Orange Lime Grape berry Observed 32 28 16 14 10 Expected 20 20 20 20 20 © The McGraw-Hill Companies, Inc., 2000 12-9 12-2 Test for Goodness of Fit Example The observed frequencies will almost always differ from the expected frequencies due to sampling error. Question: Are these differences significant, or are they due to chance? The chi-square goodness-of-fit test will enable one to answer this question. © The McGraw-Hill Companies, Inc., 2000 12-10 12-2 Test for Goodness of Fit Example The appropriate hypotheses for this example are: H0: Consumers show no preference for flavors of the fruit soda. H1: Consumers show a preference. The d. f. for this test is equal to the number of categories minus 1. © The McGraw-Hill Companies, Inc., 2000 12-11 12-2 Test for Goodness of Fit Formula O E 2 2 E d . f . number of categories 1 O observed frequency E expected frequency © The McGraw-Hill Companies, Inc., 2000 12-12 12-2 Test for Goodness of Fit Example Is there enough evidence to reject the claim that there is no preference in the selection of fruit soda flavors? Let = 0.05. Step 1: State the hypotheses and identify the claim. © The McGraw-Hill Companies, Inc., 2000 12-13 12-2 Test for Goodness of Fit Example H0: Consumers show no preference for flavors (claim). H1: Consumers show a preference. Step 2: Find the critical value. The d. f. are 5 – 1 = 4 and = 0.05. Hence, the critical value = 9.488. © The McGraw-Hill Companies, Inc., 2000 12-14 12-2 Test for Goodness of Fit Example Step 3: Compute the test value. = (32 – 20)2/20 + (28 is – 20)2/20 + … + (10 – 20)2/20 = 18.0. Step 4: Make the decision. The decision is to reject the null hypothesis, since 18.0 > 9.488. © The McGraw-Hill Companies, Inc., 2000 12-15 12-2 Test for Goodness of Fit Example Step 5: Summarize the results. There is enough evidence to reject the claim that consumers show no preference for the flavors. © The McGraw-Hill Companies, Inc., 2000 12-16 12-2 Test for Goodness of Fit Example 9.488 © The McGraw-Hill Companies, Inc., 2000 12-17 12-2 Test for Goodness of Fit Example The advisor of an ecology club at a large college believes that the group consists of 10% freshmen, 20% sophomores, 40% juniors, and 30% seniors. The membership for the club this year consisted of 14 freshmen, 19 sophomores, 51 juniors, and 16 seniors. At = 0.10, test the advisor’s conjecture. © The McGraw-Hill Companies, Inc., 2000 12-18 12-2 Test for Goodness of Fit Example Step 1: State the hypotheses and identify the claim. H0: The club consists of 10% freshmen, 20% sophomores, 40% juniors, and 30% seniors (claim) H1: The distribution is not the same as stated in the null hypothesis. © The McGraw-Hill Companies, Inc., 2000 12-19 12-2 Test for Goodness of Fit Example Step 2: Find the critical value. The d. f. are 4 – 1 = 3 and = 0.10. Hence, the critical value = 6.251. Step 3: Compute the test value. = (14 – 10)2/10 + (19 – 20)2/20 + … + (16 – 30)2/30 = 11.208. © The McGraw-Hill Companies, Inc., 2000 12-20 12-2 Test for Goodness of Fit Example Step 4: Make the decision. The decision is to reject the null hypothesis, since 11.208 > 6.251. Step 5: Summarize the results. There is enough evidence to reject the advisor’s claim. © The McGraw-Hill Companies, Inc., 2000 12-21 12-3 Tests Using Contingency Tables When data can be tabulated in table form in terms of frequencies, several types of hypotheses can be tested using the chi-square test. Two such tests are the independence of variables test and the homogeneity of proportions test. © The McGraw-Hill Companies, Inc., 2000 12-22 12-3 Tests Using Contingency Tables The test of independence of variables is used to determine whether two variables are independent when a single sample is selected. The test of homogeneity of proportions is used to determine whether the proportions for a variable are equal when several samples are selected from different populations. © The McGraw-Hill Companies, Inc., 2000 12-23 12-3 Test for Independence Example Suppose a new postoperative procedure is administered to a number of patients in a large hospital. Question: Do the doctors feel differently about this procedure from the nurses, or do they feel basically the same way? Data is on the next slide. © The McGraw-Hill Companies, Inc., 2000 12-24 12-3 Test for Independence Example Group Group Prefer Prefer old old procedure procedure 80 80 No No procedure preference Nurses Nurses Prefer Prefer new new procedure procedure 100 100 Doctors Doctors 50 50 120 120 30 30 20 20 © The McGraw-Hill Companies, Inc., 2000 12-25 12-3 Test for Independence Example The null and the alternative hypotheses are as follows: H0: The opinion about the procedure is independent of the profession. H1: The opinion about the procedure is dependent on the profession. © The McGraw-Hill Companies, Inc., 2000 12-26 12-3 Test for Independence Example If the null hypothesis is not rejected, the test means that both professions feel basically the same way about the procedure, and the differences are due to chance. If the null hypothesis is rejected, the test means that one group feels differently about the procedure from the other. © The McGraw-Hill Companies, Inc., 2000 12-27 12-3 Test for Independence Example Note: The rejection of the null hypothesis does not mean that one group favors the procedure and the other does not. The test value is the 2 value (same as the goodness-of-fit test value). The expected values are computed from: (row sum)(column sum)/(grand total). © The McGraw-Hill Companies, Inc., 2000 12-28 12-3 Test for Independence Example © The McGraw-Hill Companies, Inc., 2000 12-29 12-3 Test for Independence Example From the MINITAB output, the Pvalue = 0. Hence, the null hypothesis will be rejected. If the critical value approach is used, the degrees of freedom for the chi-square critical value will be (number of columns –1)(number of rows – 1). d.f. = (3 –1)(2 – 1) = 2. © The McGraw-Hill Companies, Inc., 2000 12-30 12-3 Test for Homogeneity of Proportions Here, samples are selected from several different populations and one is interested in determining whether the proportions of elements that have a common characteristic are the same for each population. © The McGraw-Hill Companies, Inc., 2000 12-31 12-3 Test for Homogeneity of Proportions The sample sizes are specified in advance, making either the row totals or column totals in the contingency table known before the samples are selected. The hypotheses will be: H0: p1 = p2 = … = pk H1: At least one proportion is different from the others. © The McGraw-Hill Companies, Inc., 2000 12-32 12-3 Test for Homogeneity of Proportions The computations for this test are the same as that for the test of independence. © The McGraw-Hill Companies, Inc., 2000 12-33 12-4 Analysis of Variance (ANOVA) When an F test is used to test a hypothesis concerning the means of three or more populations, the technique is called analysis of variance (ANOVA). © The McGraw-Hill Companies, Inc., 2000 12-34 12-4 Assumptions for the F Test for Comparing Three or More Means The populations from which the samples were obtained must be normally or approximately normally distributed. The samples must be independent of each other. The variances of the populations must be equal. © The McGraw-Hill Companies, Inc., 2000 12-35 12-4 Analysis of Variance Although means are being compared in this F test, variances are used in the test instead of the means. Two different estimates of the population variance are made. © The McGraw-Hill Companies, Inc., 2000 12-36 12-4 Analysis of Variance Between-group variance - this involves computing the variance by using the means of the groups or between the groups. Within-group variance - this involves computing the variance by using all the data and is not affected by differences in the means. © The McGraw-Hill Companies, Inc., 2000 12-37 12-4 Analysis of Variance The following hypotheses should be used when testing for the difference between three or more means. H0: = = … = k H1: At least one mean is different from the others. © The McGraw-Hill Companies, Inc., 2000 12-38 12-4 Analysis of Variance d.f.N. = k – 1, where k is the number of groups. d.f.D. = N – k, where N is the sum of the sample sizes of the groups. Note: The formulas for this test are tedious to work through, so examples will be done in MINITAB. See text for formulas. © The McGraw-Hill Companies, Inc., 2000 12-39 12-4 Analysis of Variance -Example A marketing specialist wishes to see whether there is a difference in the average time a customer has to wait in a checkout line in three large self-service department stores. The times (in minutes) are shown on the next slide. Is there a significant difference in the mean waiting times of customers for each store using = 0.05? © The McGraw-Hill Companies, Inc., 2000 12-40 12-4 Analysis of Variance -Example Store StoreAA 33 Store StoreBB 55 Store StoreCC 11 22 55 88 99 33 44 66 33 66 22 22 77 11 55 33 © The McGraw-Hill Companies, Inc., 2000 12-41 12-4 Analysis of Variance -Example Step 1: State the hypotheses and identify the claim. H0: = H1: At least one mean is different from the others (claim). © The McGraw-Hill Companies, Inc., 2000 12-42 12-4 Analysis of Variance -Example Step 2: Find the critical value. Since k = 3, N = 18, and = 0.05, d.f.N. = k – 1 = 3 – 1= 2, d.f.D. = N – k = 18 – 3 = 15. The critical value is 3.68. Step 3: Compute the test value. From the MINITAB output, F = 2.70. (See your text for computations). © The McGraw-Hill Companies, Inc., 2000 12-43 12-4 Analysis of Variance -Example Step 4: Make a decision. Since 2.70 < 3.68, the decision is not to reject the null hypothesis. Step 5: Summarize the results. There is not enough evidence to support the claim that there is a difference among the means. The ANOVA summary table is given on the next slide. © The McGraw-Hill Companies, Inc., 2000 12-44 12-4 Analysis of Variance -Example © The McGraw-Hill Companies, Inc., 2000