The SAS System 1 09:56 Thursday, October 7, 2010 The GLM Procedure Class Level Information Class Levels Values twistlevel 3 1 2 3 testspeed 3 1 2 3 bobbin 6 1 2 3 4 5 6 Number of Observations Read 54 Number of Observations Used 54 The SAS System 2 09:56 Thursday, October 7, 2010 The GLM Procedure Dependent Variable: hairiness Sum of Source DF Squares Mean Square F Value Pr > F Model 13 621669.1296 47820.7023 207.09 <.0001 Error 40 9236.7407 230.9185 Corrected Total 53 630905.8704 R-Square Coeff Var Root MSE hairiness Mean 0.985360 2.475890 15.19600 613.7593 Source DF Type I SS Mean Square F Value Pr > F twistlevel 2 611791.8148 305895.9074 1324.69 <.0001 testspeed 2 4636.9259 2318.4630 10.04 0.0003 twistlevel*testspeed 4 826.2963 206.5741 0.89 0.4762 bobbin 5 4414.0926 882.8185 3.82 0.0063 Source DF Type III SS Mean Square F Value Pr > F twistlevel 2 611791.8148 305895.9074 1324.69 <.0001 testspeed 2 4636.9259 2318.4630 10.04 0.0003 twistlevel*testspeed 4 826.2963 206.5741 0.89 0.4762 bobbin 5 4414.0926 882.8185 3.82 0.0063 The SAS System 3 09:56 Thursday, October 7, 2010 The GLM Procedure Tukey's Studentized Range (HSD) Test for hairiness NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 40 Error Mean Square 230.9185 Critical Value of Studentized Range 3.44208 Minimum Significant Difference 12.329 Means with the same letter are not significantly different. Tukey Grouping Mean N twistlevel A 755.000 18 1 B 588.222 18 2 C 498.056 18 3 The SAS System 4 09:56 Thursday, October 7, 2010 The GLM Procedure Bonferroni (Dunn) t Tests for hairiness NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 40 Error Mean Square 230.9185 Critical Value of t 2.49886 Minimum Significant Difference 12.658 Means with the same letter are not significantly different. Bon Grouping Mean N twistlevel A 755.000 18 1 B 588.222 18 2 C 498.056 18 3 The SAS System 5 09:56 Thursday, October 7, 2010 The GLM Procedure Tukey's Studentized Range (HSD) Test for hairiness NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 40 Error Mean Square 230.9185 Critical Value of Studentized Range 3.44208 Minimum Significant Difference 12.329 Means with the same letter are not significantly different. Tukey Grouping Mean N testspeed A 626.833 18 3 B 608.000 18 2 B B 606.444 18 1 The SAS System 6 09:56 Thursday, October 7, 2010 The GLM Procedure Bonferroni (Dunn) t Tests for hairiness NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 40 Error Mean Square 230.9185 Critical Value of t 2.49886 Minimum Significant Difference 12.658 Means with the same letter are not significantly different. Bon Grouping Mean N testspeed A 626.833 18 3 B 608.000 18 2 B B 606.444 18 1 The SAS System 7 09:56 Thursday, October 7, 2010 The GLM Procedure Level of Level of ----------hairiness---------- twistlevel testspeed N Mean Std Dev 1 1 6 744.666667 32.1662349 1 2 6 744.500000 22.6870007 1 3 6 775.833333 18.9252917 2 1 6 582.333333 18.6404578 2 2 6 584.833333 17.0577451 2 3 6 597.500000 12.5339539 3 1 6 492.333333 3.1411251 3 2 6 494.666667 2.6583203 3 3 6 507.166667 3.1885211