Tham khảo tài liệu 'product and process comparisons_9', kỹ thuật - công nghệ, điện - điện tử phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | . How can we compare several populations with unknown distributions the Kruskal-Wallis test 8 1 18 SUM 65 66 Compute the test statistic The corresponding H test statistic is 12 652 . . . 662 19 20 4 -3 20 From the chi-square table in Chapter 1 the critical value for a .05 with df k-1 3 is . Since we reject the null hypothesis. Note that the rejection region for the KW procedure is one-sided since we only reject the null hypothesis when the H statistic is too large. The KW test is implemented in the Dataplot command KRUSKAL WALLIS TEST Y X . NIST SEMATECH HOME TOOLS AIDS SEARCH SACK NEXT http div898 handbook prc section4 3 of 3 5 1 2006 10 38 53 AM . Assuming the observations are normal do the processes have the same variance ENGINEERING STATISTICS HANDBOOK hW tools raids lỉEÂtCH BACK Nixf 7. Product and Process Comparisons . Comparisons based on data from more than two processes . Assuming the observations are normal do the processes have the same variance Before comparing means test whether the variances are equal Techniques for comparing means of normal populations generally assume the populations have the same variance. Before using these ANOVA techniques it is advisable to test whether this assumption of homogeneity of variance is reasonable. The following procedure is widely used for this purpose. Bartlett s Test for Homogeneity of Variances Null Bartlett s test is a commonly used test for equal variances. Let s examine hypothesis the null and alternative hypotheses. Ho rrj . against Hữ the ơị are not all equal Test statistic Assume we have samples of size ni from the i-th population i 1 2 . . . k and the usual variance estimates from each sample c 2_ 2 --- where tii Now introduce the following notation Vj nj - 1 the ỉ j are the degrees of freedom and v 52 Mi i l http div898 handbook prc section4 1 of 3 5 1 2006 10 38 54 AM . .
