Appendix 13.1 weof Chemiech WEBAG ‘ite ‘The reaments ave entered rahe Row factor bax and the blocs are entered i the Column factor bo. Chapter 13. Experimental Design ond Analysis of Variance Use analysis of variance to test for any significant differences due to experience. Use a.05 level of significance, and for now, ignore the eect of . At the 05 level of significance test for any significant differences duc ta, years of experience, and interaction Yeas Post, Posty, Analysis of Variance with Minitab Completely Randomized Design In Seetion 13.2 we showed how analysts of variance could be used t et forth equi of population means using data from a completely randomized design. Tilsen Minitab can be used for this type of experinental design, we show how to tet what ‘he meno numberof units produced per week isthe same for each assembly medial. the Chemtech experiment introduced in Section 13.1. The sample data ate ener) into the first three columns of a Mintah worksheet; column 1 i lbeled A, column ? Jabeled B, and column 3 is labeled C. The following steps produce the Minitab oupuh Figure 135, Step 1, Select the Stat menu Step 2. Choose ANOVA Step 3. Choose One-way (Unstacked) Step 4. When the One-way Analysis of Variance dialog box appears Enter C1-C3 in the Responses (in separate eolumns) box Click OK Randomized Block Design In Section 13.4 we showed how analysis of variance could be used to test for the equality ‘of & population means using the data from @ randomized block design. To illustrate how ‘Minitab can be used for this type of experimental design, we show how to test whether the mean stress levels for air traffie controllers are the same for three work stations using the data in Table 13.3. The blocks (controllers), treatments (system), and stress level seo shown in Table 13.5 are entered into columns C1, C2, and C3 of a Minitab worksheet, = spectively. The following steps produce the Minitab output corresponding to the ANOVA in Table 13.8. |. Select the Stat menu . Choose ANOVA . Choose Two-way J. When the Two-way Analysis of Variance dialog box appears: Enter C3 in the Response box Enter C2 in the Row factor box Enter C1 in the Column factor box Select Fit Additive Model Click OK Factorial Experiment In Seetion 13.5 we showed how analysis of variance could be used to test for the equal ity of & population means using dats from a factorial experiment. To illustrate bow ‘Minitab can be used for this type of experimental design, we show how to analyze the data for the two-factor GMAT experiment introduced in that section, The GMAT scorestest whether bly method ig = tt are entered = A, column ig nitab empty [Appendix 13.2 wesliits or the equality eet illustrate how st viether the ions using the 8 level scores worksheet, e- 0 the ANOVA for the equal Mustrate how o analyze the GMAT scares Appendix 13.2 Analysis of Variance with Bxcal 555 shown in Table 13,11 are entered into column 1 of a Minitab worksheet; column 1 is labeled Score, column 2 is labeled Program, and column 3 is labeled College. The fol- owing steps produce the Minitab output corresponding 0 the ANOVA table shown in Figure 136. Step 1, Select the Stat menu Step 2. Choose ANOVA Step 3. Choose Two-way Step 4. When the Two-way Analysis of Variance dialog box appears: Enver C7 in the Response box Enter C2 in the Row factor box Enter C3 in the Column factor box Click OK Analysis of Variance with Excel Completely Randomized Design In Section 13.2 we showed how analysis of variance could be used to test for the equal ity of & population means using data from a completely randomized design. To illustrate how Excel can be used to test for the equality of & population means for this type of experimental design, we show how fo test whether the mean number of units produced per week is the same for each assembly method in the Chemitech experiment introduced in Section 13.1, The sample data are entered into worksheet rows 2to 6 of columns A, B, und Cas shown in Figure 13.7. The following steps are used to obtain the output showin in cells A8:G22; the ANOVA portion of this ourput corresponds to the ANOVA table shown in Table 133. Step 1. Click the Data tab on the Ribbot Step 2. In the Analysis group, click Data Analysis Step 3. Choose Anova: Single Faetor from the list of Analysis Tools Click OK Step 4. When the Anova: Single Factor dialog box appears: Enter A1:C6 in Input Range box Select Columns Select Labels in First Row Select Output Range and enter AS in the box Click OK Randomized Block Design In Section 13.4 we showed how anal “population means using data from a randomized block design. To illustrate how Excel he used for this type of experimental design, we show how to fest whether the mean stress levels for air traffic controllers are the same for three work stations, The stress level scores shown in Table 13.5 ate entered into worksheet rows 2to 7 of columns B, C, and D as shown, sure 13,8. The cells in rows 2 fo 7 of column A contain the number of each controller (1,2, 3,4,5, 6) The following stops produce the Excel output shown in cells 49:30. The ANOVA portion ofthis outpat corresponds 40 the Minitab output shown in Table 13.8. of variance could be used to test for the equality of ‘Step 1, Click the Data tab on the Ribbon ‘Step 2. in the Analysis group, click Data Analysis556 Chapter 13. Experimental Design and Analysis af Variance FIGURE 13.7 EXCEL SOLUTION FOR THE CHEMITECH EXPERIMENT & 2 Pee a ae 1] Method A [Method B Method C] ze Sis. 3 ét = st Eee oa [4 so 2 aster 5 6 a Eee Pele ane cme — ([Rnove Sage Bacto es a = ; ee [SunmaaRy he pee ae Grow Sion [average cece ay Mabe 310,62 i Method aa0l 66 [sane “Meticd C 260) 52 in| ANOS La 18 Source of Variation |S rae 19 Beiwees Gronps 520 2 50 20 Within Groups ET) a ECE : ja i SE - | [Toul | ia | faa! cee) I aT | Sanne nS couarstaay Step 3. Choose Anova: Two-Factor Without Replication from the list of AnalysicToos Click OK Step 4. When the Anova: Two Factor Without Replication dialog box appears: Enter A1:D7 in Input Range box Select Labels Select Output Range and enter A9 in the box Click OK Factorial Experiment In Section 13.5 we showed how analysis of variance could be used to test for the equality of k population means using data from a factorial experiment. To illustrate how Excel ccan be used for this type of experimental design, wwe show the two-factor GMAT experiment introduc hhow to analyze the data for ed in that section, The GMAT scores shown in Table 13.10 are entered into worksheet rows to 7 of columns B, C, and D as shown i Figure 13.9. The following steps are used to obvain the output shown in cells A9:GA4s the ANOVA portion of this output corresponds to the Minitab output in Figure 13.6, Step 1. Click the Data tab on the Ribbon Step 2. In the Analysis group, click Data Analysis, ‘Step 3. Choose Anova: Two-Factor With Replication from the list of Analysis Tovls Click OK Step 4. Wheu the Anova: Two-Factor With Replication dialog box appears: Enter A1:D7 in Input Range box Entcr 2 in Rows per sample box LRAppendi 13.3 Analysis of e Completely Randomized Design Using Staoals 557 EXCEL SOLUTION FOR THE AIR TRAFFIC CONTROLLER STRESS TEST Contr [Sytem Stem [Sten C t Se % ce ‘i fos t woot 5 aot ie 3 oe | Anova: Two-Ractor Without Replication a a Sinan | Cones ras ce emote Tepes de al naa aaa ar ee fl ye ZI ces a Sik oat ete Tae ee [Syien stats a OW sae ria Fractal sispoostay 33 : Seton} 410 Select Output Range snd enter A9 in the box Click OK forthe equaliy rate howe Exel x133 Analysis of a Completely Randomized Design 1D as shown it = Using StatTools Is A9:G44 te In this appendix we show how StatTools can be used to test forthe equality of & population sneans fora completely randomized design. We use the Chemitech dat in Table 13.1 t0 il lustre, Beyin by using the Data Set Manager to create a StatTools dataset for these data "using the procedure desribed in the appendix in Chapter 1. The following steps canbe used to test for the equality of the three population means. WEB file) Step 1. Click the StatTools tab on the Ribbon Step 2. In the Analyses group, click Statistical Inference chemitech Step 3. Choose the One-Way ANOVA optioni 558 Chapter 13 Experimental Design andl Analysis of Variance FIGURE 19.9 EXCEL SOLUTION FOR THE TWO-FACTOR GMAT EXPERIMENT e gi = i “ris and Sic 2 |Shoureview ‘a 1 = i oo : 4 [ay pga e x 286 a : ¢ [Tivek oman a 0 7 . 7 ea a0 . a = oS 9 Rana era Oa REST a wW © [7 : 1 | Sou Bisnest| Bioeng AG ad Sienes | Teal a Foire S C . on z 3 zee Sia io] 00 aa] 06 = acre ‘sof 300 id ESAT 6 [Nara [S200] a 00 esr J Tapa ZERO ovat zy 3 mae i [su woo} Tis 50, aa 1 s avenge iol 55 ESEEERI — ‘isle cm Shoo] 0 io 86 sep] —— - TTikoear oar | { a i A Che sce a a ee Sie feo iho | avenge 5) SC SE TET ance a0 3136 6667] —— 3 esse ee | | can Sea % 3 Blecaefecaey : 2 a0 20 = aS ee [S| se a5 sie 30) Sat 7310 laxona ee i oan of oa Sg a Prue Fe i : Siagle {ei i S080] Ry Oo | a8 i s Colum 530 i eco] Toot| ons [a36 i [at fics cara] ol 0 C501 Sat i itn ao a] asses s pea a0 i z i fea L $= =—=—————— sansa i - Step 4. When the SttTeos-One-Way ANOVA ditog box appears: | : Iu the Variables section i Click the Format button and select Unstacked | i Select Method A, : Select Method B | i Select Method © : Select 95% in the Confidence Level box Click OK‘Appendis 13.3. Analysis ofe Completely Rondomized Design Using Stafools 559 Note that in step 4 we selected the Unstacked option after clicking the Format button, The Unstacked option means that the dats for the three treatments appear in separate columns. ‘of the worksheet In a stacked format, only two columns would be used. For example, the data could have been organized as follows: 5 ‘etiod [Unis Produced Method A 38 Matbod A Method A 35 ‘Metbod A 66 6 | Method A 67 | Metiod TMetior Method 8 Method Method B shod © Method C Method Medios ‘Methoa C aia are frequently recorded in a stacked format, Forstacked date, simply select the Stacked option after clicking the Format button.