252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . D. COMPARISON OF TWO SAMPLES 1. Two Means, Two Independent Samples, Large Samples. Text 10.1-10.3, 10.7 [10.1 – 10.3, 10.5] (10.1 – 10.3, 10.5) 2. Two Means, Two Independent Samples, Populations Normally Distributed, Population Variances Assumed Equal. Text 10.4, 10.13a, 10.20a, b, e [10.4, 10.15a, 10.13a,b,e]. For the last problem: x1 17 .5571 , s1 1.9333 , x 2 19.8905 , s 2 4.5767 (10.4, 10.14, 10.12a,b,e) 3. Two Means, Two independent Samples, Populations Normally Distributed, Population Variances not Assumed Equal. Optional Text 10.20[10.13c,d] (10.12c,d) See data above. D3, D4 4. Two Means, Paired Samples (If samples are small, populations should be normally distributed). Text 10.26, 10.29[10.36, 10.37], D1, D2 (10.32*(in 252hwkadd.), [10.34] (different numbers), 10.25[10.35], D1, D2) 5. Rank Tests. a. The Wilcoxon-Mann-Whitney Test for Two Independent Samples. Text 12.65[10.48] (10.46) b. Wilcoxon Signed Rank Test for Paired Samples. Text 12.74-12.76[10.57-59] (10.80-82 on CD), Downing & Clark 18-15, 18-9 (in chapter 17 in D&C 3rd edition), D5 6. Proportions. Text 10.32, 10.38, 10.39, 12.32** [12.2, 12.7*, 12.8*] (12.2) 7. Variances. Text 10.40, 10.43-10.48 [10.16, 10.19 - 10.24, 10.25] (10.15, 10.18 - 10.23, 10.24) D6a (below), D6, D7 (A summary problem), D8 (A summary problem) Graded assignment 3 will be posted. Solutions to outline points 1 through 3 are in this document. From the formula table Interval for Confidence Hypotheses Test Ratio Interval Difference H 0 : D D0 * d z 2 d d D0 z between Two H : D D , d 1 0 Means ( 12 22 D 1 2 d known) n n 1 Critical Value d cv D0 z 2 d 2 d x1 x2 Difference between Two Means ( unknown, variances assumed equal) Difference between Two Means( unknown, variances assumed unequal) H 1 : D D0 , 1 1 n1 n 2 sd s p D 1 2 DF n1 n2 2 H 0 : D D0 * D d t 2 s d DF H 1 : D D0 , s12 s22 n1 n2 sd s12 s22 n 1 n2 D 1 2 t sˆ 2p t d D0 sd d cv D0 t 2 s d n1 1s12 n2 1s22 n1 n2 2 d D0 sd d cv D0 t 2 s d 2 s12 2 n1 n1 1 * Same as H 0 : D D0 * D d t 2 s d s 22 2 n2 n2 1 H 0 : 1 2 H 1 : 1 2 if D0 0. 1 (Open this document in 'Page Layout' view!)Edited to replace or with D . 252solnD1 10/18/06 Problems with 2 means and known population variances or 2 large independent samples 1 20 2 10 Exercise 10.1: If for sample 1 n1 40 and for sample 2 n2 50 , what is the value of the z-test x 72 x 66 1 2 statistic for testing differences between the population means. Solution: From the outline if D 1 2 and d x1 x 2 , the formula for a test ratio is t x x 2 10 20 d D0 or t 1 where, for large samples we use s d sd sd with z . If the population variances are known, replace sd with d 1 20 the value of z when for sample 1 n1 40 and for sample 2 x 72 1 d x1 x2 72 66 6 , D0 10 20 0 , and d 3.4641 . So z d D0 d 12 n1 22 n2 s12 s 22 and replace t n1 n 2 . The problem asks for 2 10 n2 50 . Here x 66 2 12 n1 22 n2 20 2 40 10 2 50 10 2 x x 10 20 72 66 0 1.732 6-0 1.732 or z 1 2 3.4641 12 22 20 2 10 2 n1 n2 40 50 Exercise 10.2: What is your decision in 10.1 if you are testing H0: 1 2 and H1: 1 2 and .01 . Solution: .01 . .01 , so z z.005 2.576 . Make a diagram: since this is a 2-sided test, your 2 rejection regions are below -2.576 and above 2.576. So we have H0: Decision rule: If z 2.576 or z 2.576 , reject H0. Test statistic: z d D0 d 1 2 and H1: 1 2 . 6-0 1.732 3.4641 Decision: Since zcalc 1.732 is between the critical bounds of z = 2.576, do not reject H0. There is inadequate evidence to conclude the two population means are different. Exercise 10.3: What is the p-value for Problem 10.1? Solution: pvalue 2Pz 1.732 2.5 .4582 .0836 2 252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . Exercise 10.7 [10.5 in 9th]: We wish to determine whether there is a difference between the life expectancy of light bulbs produced by two different machines. The problem says the following: Population Standard deviation Sample size Sample Mean Machine 1 110 hrs 25 375 Machine 2 125 hrs 25 362 a) Is there a difference between the mean life of the bulbs? b) Find and interpret a p-value. Solution: (a) H0: 1 2 Light bulbs produced by machine 1 have the same average life expectancy as light bulbs produced by machine 2. H1: 1 2 Light bulbs produced by machine 1 have a different average life expectancy as light bulbs produced by machine 2. 1 110 2 125 Given: .05, n1 25 and n2 25 . d x1 x2 375 362 13 , x 362 x 375 2 1 D0 10 20 0 , and d (i) Test Ratio: t 12 n1 22 n2 110 2 25 125 2 25 484 625 33 .307 .3 x x 2 10 20 d D0 or t 1 . If this test ratio lies between t z , do sd sd 2 2 not reject H 0 . z d D0 13 0 0.390 . Make a diagram with zero in the middle sd 33 .307 showing 'reject' regions below -1.960 and above z .025 1.960 . Since 0.390 does not fall in one of the 'reject' regions, do not reject H 0 . According to the Instructor’s Solution Manual (edited): Decision rule: If z < – 1.960 or z > 1.960, reject H0. ( X1 – X2 ) – ( 1 – 2 ) (375 – 362) – 0 = 0.390 12 2 2 1102 125 2 n1 n2 25 25 Decision: Since zcalc = 0.390 is between the critical bounds of 1.960, do not reject H0. Test statistic: Z (ii). Critical Value: d CV D0 t 2 s d or x1 x 2 CV 10 20 t 2 s d . If d x1 x 2 is between the two critical values, do not reject H 0 . d CV D0 z 2 s d 0 1.960 33.307 0 65 .28 , or -65.28 and +65.28. Make a diagram with 0 in the middle showing 'reject' regions below -65.28 and above +65.28. Since d 13 does not fall in a 'reject' region, do not reject H 0 . (iii) Confidence Interval: D d z 2 s d 13 1.960 33.307 13 65 .28 or -52.28 to 87.28. Since D0 0 falls in the confidence interval, do not reject H 0 . According to the Instructor’s Solution Manual, there is not enough evidence to conclude that light bulbs produced by machine 1 have a different average life expectancy from light bulbs produced by machine 2. (b) pval 2Pz 0.390 2.5 .1517 .6966 . (Since the p-value is above the significance level , do not reject H 0 .) According to the Instructor’s Solution Manual, the probability of obtaining samples whose means differ by 13 hours or more when the null hypothesis is true is 0.6966. 3 (Open this document in 'Page Layout' view!)Edited to replace or with D . 252solnD1 10/18/06 Problems with 2 means and 2 independent samples with variances assumed equal. Text 10.4, 10.15a, 10.13a,b,e x1 17.5571, s1 1.9333 , x2 19.8905 , s 2 4.5767 (10.4, 10.14, 10.12a,b,e) s1 4 s2 5 Exercise 10.4: Given: n1 8 and n2 15 . a) What is the value of the pooled-variance t statistic for x 42 x 34 1 2 testing the equality of the population means? b) If we need the critical value of the t-ratio, how many degrees of freedom are there? c) If we use a significance level of .01 what is the critical value for the test H : 2 ratio for a one tailed test of the hypotheses 0 1 ? d) What is your statistical decision? H1 : 1 2 s1 4 Solution: Given: n1 8 and x 42 1 s2 5 n2 15 . x 34 2 d x1 x2 42 34 8 . We are assuming 12 22 . n 1s12 n2 1s22 742 14 52 22 and s s 2 1 1 So s p2 1 p d n1 n2 2 21 n1 n2 1 1 22 22 0.125 0.06667 22 0.1916667 4.2166667 2.05345 8 15 The test ratio is t (a) t d D0 8-0 3.896 . The Instructor’s Solution Manual has sd 2.05345 ( X 1 X 2 ) ( 1 2 ) 1 1 Sp n1 n2 (42 34) 0 1 1 22 8 15 2 3.8959 (e) df = (n1 – 1) + (n2 – 1) = 7 + 14 = 21 H : 2 H 0 : 1 2 0 H 0 : D 0 .01, 0 1 or or D 1 2 . This is a H : H : 0 2 2 1 1 1 1 H1 : D 0 right-tail test. Make a diagram. Show a Normal curve with a mean at zero. Since 21 t t.01 2.518 , show a 1% rejection region above 2.518. Decision rule: If t > 2.518, reject H0. Decision: Since tcalc = 3.896 is above the critical bound of t = 2.518, reject H0. There is enough evidence to conclude that the first population mean is larger than the second population mean. We are sampling from two independent normal distributions having equal variances. (f) .05. A 2-sided confidence interval has the form D d t 2 sd (b) (c) (d) 21 1 2 x1 x2 t sd and t 21 t.025 2.080. . 8 2.080 2.05345 8 4.271 or 3.729 to 12.271. 2 2 or So D d t 2 sd The Instructor’s Solution Manual has 1 1 1 1 X 2 t S p2 42 34 2.0796 22 8 15 n1 n2 or 3.7296<1 2 12.2704 . X 1 4 252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . Exercise 10.13a [10.15a in 9th] (10.14a in 8th edition): We have the following numbers concerning the s1 0.012 two independent samples of fill weights of cans of pet food produced by two different lines; n1 11 x 8.005 1 s2 0.005 and n2 16 . If we assume that the population variances are equal is there evidence of a difference in x 7.997 2 the mean fill weight from the two lines? s1 0.012 s2 0.005 Solution: Given: n1 11 and n2 16 . d x1 x2 8.005 7.997 0.008 . x 7.997 x 8.005 2 1 df 11 1 16 1 25. We are assuming 12 22 . n 1s12 n2 1s22 100.012 2 15 0.005 2 .00144 .000375 0.0000726 7.26 10 5 So s p2 1 25 25 n1 n2 2 1 1 1 1 0.0000726 0.0000726 0.090909 0.0625000 and sd s p2 11 16 n n 2 1 H : 2 0.0000726 0.153409 0.0000111375 0.00333729 . We are testing 0 1 or H 1 : 1 2 H : D 0 H 0 : 1 2 0 or 0 where D 1 2 . .05 . H 1 : D 0 H 1 : 1 2 0 (i) Test Ratio: t x x 2 10 20 d D0 or t 1 . If this test ratio lies between sd sd 25 t t .025 2.060 , do not reject H 0 . 2 t d D0 0.008 0 2.397 . Make a sd 0.00333729 diagram with zero in the middle showing 'reject' regions below -2.060 and above 2.060 . Since 2.397 falls in one of the 'reject' regions, reject H 0 . Or you can say that, since 2.397 falls 25 25 2.060 and t .001 2.485 , for a one-sided test, .010 p value .025 . For a between t .025 2-sided test, we double this to get .020 p value .05 . If we want a more exact value, try my available Minitab macro tareaA as below. ————— 10/20/2005 6:04:19 PM ———————————————————— Welcome to Minitab, press F1 for help. Results for: notmuch.MTW #a nonsense worksheet used so that Mintab finds tareaA MTB > %tareaA Executing from file: tareaA.MAC Graphic display of t curve areas Finds and displays areas to the left or right of a given value or between two values. (This macro uses C100-C116 and K100-K120) Enter the degrees of freedom. DATA> 25 Do you want the area to the left of a value? (Y or N) n Do you want the area to the right of a value? (Y or N) y Enter the value for which you want the area to the right. DATA> 2.397 5 252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . ...working... t Curve Area Data Display mode 0 median 0 What this is saying is Pt 2.397 .9878 , so 2.397 t1.9878 t .0122 and p value 2.0122 .0244 . This is below .05 . According to the Instructor’s Solution Manual (edited): (a) H0: 1 2 where Populations: 1 = Line A, 2 = Line B H1: 1 2 Decision rule: df = 25. If |t| > 2.0595, reject H0. Test statistic: x x 2 1 2 8.005 7.997 0 2.3972 t 1 1 1 1 1 7.26 10 5 sˆ 2p 11 16 n1 n 2 Since t = 2.3972 > 2.0595 or p-value = 0.0243 < 0.05, reject H0. There is sufficient evidence of a difference in the average weight of cans filled on the two lines. (ii). Critical Value: d CV D0 t 2 s d or x1 x 2 CV 10 20 t 2 s d . If d x1 x 2 is between the two critical values, do not reject H 0 . d CV D0 t sd 0 2.0600.00333729 0 0.00687 , or 2 -0.00687 and +0.00687. Make a diagram with 0 in the middle showing 'reject' regions below -0.00867 and above +0.00687. Since d 0.008 falls in a 'reject' region, reject H 0 . (iii) Confidence Interval: D d t 2 s d 0.008 2.060 0.00333729 0.008 0.00687 or 0.0011 to 0.0149. Since D0 0 does not fall in the confidence interval, reject H 0 . Exercise 10.20a, b, e [10.13a, b, e in 9th edition] (10.12a, b, e in 8th edition): We are to compare two randomly selected groups of 21 employees each trained by two different methods. Data from the file training gives the following results for time necessary to assemble a part after training: x1 17.5571, s1 1.9333 , x2 19.8905 , s 2 4.5767 n1 21 and n2 21 a) Assume that the variances in the two populations of assembly times are equal. Is there evidence of a difference in the mean assembly times? Use .05 . b) What other assumption is necessary in a)? e) On the same assumptions construct and interpret a 95% confidence interval for the difference between the population means. Solution: (a) H0: 1 2 (where Populations: 1 = computer-assisted, 2 = team-based) Average assembly times in seconds are the same between employees trained in a computer-assisted, individual-based program and those trained in a team-based program. 6 (Open this document in 'Page Layout' view!)Edited to replace or with D . 252solnD1 10/18/06 H1: 1 2 Average assembly times in seconds are different between Employees trained in a computer-assisted, individual-based program and those trained in a team-based program. d x1 x 2 17.557119.8905 2.3334 . df 21 1 21 1 40 . We are assuming 12 22 . n 1s12 n2 1s22 201.9333 2 20 4.5767 2 74.7530 418 .9237 12.3419 So s p2 1 40 40 n1 n2 2 H : 2 1 1 1 1 12 .3419 1.175419 1.08417 . We are testing 0 1 and sd s p2 21 21 H 1 : 1 2 n1 n2 H : D 0 H : 2 0 or 0 1 or 0 where D 1 2 . .05 . H 1 : D 0 H 1 : 1 2 0 (i) Test Ratio: t x x 2 10 20 d D0 or t 1 . If this test ratio lies between sd sd 40 t t .025 2.021 , do not reject H 0 . 2 t d D0 2.3334 0 2.152 . Make a diagram with sd 1.08417 zero in the middle showing 'reject' regions below -2.021 and above +2.021. Since -2.152 falls in one of the 40 40 'reject' regions, reject H 0 . Or you can say that, since 2.152 falls between t .025 2.020 and t .01 2.421 , for a one-sided test, .010 p value .025 . For a 2-sided test, we double this to get .020 p value .05 . According to the Instructor’s Solution Manual (edited): Decision rule: df = 40. If t < – 2.0211 or t > 2.0211, reject H0. Test statistic: (n1 – 1) S12 (n2 –1) S2 2 (20) 1.93332 (20) 4.57672 = 12.3417 (n1 –1) (n2 – 1) 20 20 (X – X2 ) – (1 – 2 ) (17.5571– 19.8905) – 0 = – 2.1522 t 1 1 1 1 1 2 12.3417 S p 21 21 n n 1 2 Sp 2 Decision: Since tcalc = – 2.1522 is below the lower critical bound of – 2.0211, reject H0. There is enough evidence to conclude that the average assembly times in seconds are different between employees trained in a computer-assisted, individual-based program and those trained in a team-based program. (ii). Critical Value: d CV D0 t 2 s d or x1 x 2 CV 10 20 t 2 s d . If d x1 x 2 is between the two critical values, do not reject H 0 . d CV D0 t 2 s d 0 2.021 1.08417 0 2.1911 , or -2.1911 and 2.1911. Make a diagram with 0 in the middle showing 'reject' regions below -2.1911 and above +2.19117. Since d 2.3334 falls in a 'reject' region, reject H 0 . (iii) Confidence Interval: D d t 2 s d -2.3334 2.021 1.08417 -2.3334 2.1911 or -4.5245 to -0.1423. Since D0 0 does not fall in the confidence interval, reject H 0 . (b) One must assume that each of the two independent populations is normally distributed. The following sections (c) and (d) were not required, and were copied from the Instructor’s Solution 7 (Open this document in 'Page Layout' view!)Edited to replace or with D . 252solnD1 10/18/06 Manual to allow you to see what output looks like in the unequal variances case. 10.13 (c) (a) H0: 1 2 where Populations: 1 = computer-assisted, 2 = team-based cont. Average assembly times in seconds are the same between employees trained in a computer-assisted, individual-based program and those trained in a team-based program. H1: 1 2 Average assembly times in seconds are different between employees trained in a computer-assisted, individual-based program and those trained in a team-based program. Decision rule: df = 26. If p-value < 0.05, reject H0. Excel output: t-Test: Two-Sample Assuming Unequal Variances Computer-Assisted Program Team-based Program Mean 17.55714286 19.89047619 Variance 3.737571429 20.94590476 Observations 21 21 Hypothesized Mean Difference 0 df 27 t Stat -2.152203195 P(T<=t) one-tail 0.020240852 t Critical one-tail 1.703288035 P(T<=t) two-tail 0.040481703 t Critical two-tail 2.051829142 Since p-value = 0.041 < 0.05, reject H0. There is enough evidence to conclude that the average assembly times in seconds are different between employees trained in a computer-assisted, individual-based program and those trained in a team-based program. My solution: x1 17.5571, s1 1.9333 , x 2 19.8905 , s 2 4.5767 n1 21 , n2 21 and .05 . d x1 x 2 17.557119.8905 2.3334 . s12 n1 s 22 4.5767 2 0.9974372 , n2 21 1.9333 2 0.1779832 , 21 s12 s 22 0.1779832 0.9974372 1.1754205 , s d n1 n 2 DF s12 s 22 n1 n 2 2 so s12 s 22 1.1754205 1.0842 , and n1 n 2 2 2 1.1754205 2 0.1779832 2 0.9974372 2 1.3816134 (21) 29 .01388 28 .263 0.031678 0.9948809 1.0265679 21 21 n1 n2 n1 1 n2 1 Round this down to 28 degrees of freedom. Remember d 2.3334 , df 28, D 0 0 and s d 1.0842 . s12 (i) Test Ratio: t s 22 x x 2 10 20 d D0 or t 1 . If this test ratio lies between sd sd 28 t t .025 2.048 , do not reject H 0 . 2 t d D0 2.3334 0 2.152 . Make a diagram with sd 1.08417 zero in the middle showing 'reject' regions below -2.048 and above +2.048. Since -2.152 falls in one of the 8 252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . 28 28 'reject' regions, reject H 0 . Or you can say that, since 2.152 falls between t .025 2.048 and t .01 2.467 , for a one-sided test, .010 p value .025 . For a 2-sided test, we double this to get .020 p value .05 . Since these are both below the significance level, reject H 0 . (ii). Critical Value: d CV D0 t 2 s d or x1 x 2 CV 10 20 t 2 s d . If d x1 x 2 is between the two critical values, do not reject H 0 . d CV D0 t 2 s d 0 2.048 1.08417 0 2.2203 , or -2.2203 and 2.2203. Make a diagram with 0 in the middle showing 'reject' regions below -2.2203 and above +2.2203. Since d 2.3334 falls in a 'reject' region, reject H 0 . (iii) Confidence Interval: D d t 2 s d -2.3334 2.048 1.08417 -2.3334 2.2203 or -4.5537 to -0.11131. Since D0 0 does not fall in the confidence interval, reject H 0 . (d) The results in (a) and (c) are the same. (e) Since my answer is above, this is copied from the Instructor’s Solution Manual. Assuming equal variances: 1 1 1 1 X 2 t S p2 17.557 19.891 2.0211 12.3417 21 21 n1 n2 4.52<1 2 0.14 X 1 We are 95% confidence that the difference between the population means of the two training methods is between –4.52 and –0.14. 9 (Open this document in 'Page Layout' view!)Edited to replace or with D . 252solnD1 10/18/06 Problems with 2 means and 2 independent samples with variances not assumed equal. Exercise 10.20 [10.13c,d in 9th] (10.12c, d in 8th edition): ): We are to compare two randomly selected groups of 21 employees each trained by two different methods. Data from the file training gives the following results for time necessary to assemble a part after training: x1 17.5571, s1 1.9333 , x2 19.8905 , s 2 4.5767 n1 21 and n2 21 c) Assume that the variances in the two populations of assembly times are not equal. Is there evidence of a difference in the mean assembly times? Use .05 . d) Compare the results of a) and c). Solution: See above. Problem D3 (Optional): A secretary types 16 pages on word processor 1 and 16 pages on word processor 2. Her times are: x1 8.20 s12 4.10 x2 7.10 s22 4.20 If 1 2 =1-2 test D 0 at the 90 per cent confidence level. Assume that these are independent samples and that 12 22 . ( .10 ) Solution: We have n1 16 and n2 16 . class. From the Syllabus supplement: Difference D d t 2 s d Between Two s2 s2 Means( sd 1 2 Unknown, n1 n2 Variances 2 s12 s22 Assumed n n2 1 DF 2 Unequal) 2 s s The two-sided confidence interval for this problem was done in H 0 : D D0 t H 1 : D D0 D 1 2 d D0 sd d cv D0 t 2 s d 2 1 n1 n1 1 2 2 n2 n2 1 We found the following in class: s12 4.1 s 2 4.2 s2 s2 0.25625 , 2 0.26250 , so 1 2 0.25625 0.26250 0.51875 , n1 16 n2 16 n1 n2 sd DF s12 s 22 0.51875 0.720 , d x1 x2 8.20 7.10 1.10 and n1 n 2 s12 s 22 n1 n 2 2 2 2 0.51875 2 0.25625 2 0.26250 2 0.26910 29 .9 . 0.00438 0.00459 15 15 n1 n2 n1 1 n2 1 I will follow my own advice this time and round the degrees of freedom down to 29. (If we had followed 29 this advice in class, we would have used t.025 2.045 and the two-sided confidence interval would have s12 s 22 been D d t 2 s d 1.10 2.045 0.720 1.10 1.47 .) H 0 : 1 2 H 0 : D 0 We are now testing or . Since our hypotheses are one-sided we use H1 : 1 2 H 1 : D 0 t 29 1.311 .10 10 252solnD1 10/18/06 (Open this document in 'Page Layout' view!)Edited to replace or with D . d D0 x1 x 2 1 2 1.10 0 1.527 . Make a diagram with zero in the sd sd 0.720 middle showing a 'reject' region above t 29 1.311 . Since 1.527 falls in the 'reject' region, reject H . (i) Test Ratio: t 0 .10 (ii) Critical Value: d CV D0 t s d 0 1.311 0.720 0.943 . Make a diagram with zero in the middle showing a 'reject' above 0.943. Since d 1.10 falls in the 'reject' region, reject H 0 . (iii) Confidence interval: D d t 2 s d becomes D d t s d 1.10 1.311 0.720 0.157 . D 0.157 contradicts the null hypothesis D 0 so reject H0 . Problem D4: (Old Minitab Manual - modified) In a study of tool life , two independent samples of wear are taken. The first of these represents volume loss in millionths of cubic inches from 10 untreated tools. The second represents loss in the same units from 10 tools that were treated by a new wear retardant process. Untreated .56 .50 .69 .59 .47 .42 .45 .47 .50 .50 Treated .13 .13 .18 .23 .18 .31 .35 .23 .31 .33 On the assumption that the parent populations are Normal, test the hypothesis that the means are equal and do a confidence interval for the difference between the means ) (a) assuming that the variances are equal and (b) assuming that the variances are not equal. Solution: a) x1 x s12 1 n`1 x 2 1 5.15 0.515 10 nx12 2.709 10 0.515 2 9 0.628 10 0.238 2 9 n1 1 0.00625 . x2 2.38 x2 0.238 n2` 10 s22 x 2 2 nx2 2 n2 1 0.00684 . x1 0.56 0.50 0.69 0.59 0.47 0.42 0.45 0.47 0.50 0.50 5.15 H 0 : D 0 H 0 : 1 2 H : 2 0 or 0 1 or H : H : 0 2 2 H 1 : D 0 1 1 1 1 x12 0.3136 0.2500 0.4761 0.3481 0.2209 0.1764 0.2025 0 .2209 0.2500 0.2500 2.709 x2 0.13 0.13 0.18 0.23 0.18 0.31 0.35 0.23 0.31 0.33 2.38 x22 0.0169 0.0169 0.0324 0.0529 0.0324 0.0961 0.1225 0.0529 0.0961 0.1089 0.628 d x1 x2 0.515 0.238 0.277 a) We assume that the variances are equal. ( 12 22 ). So we use the traditional method for this problem. n 1s12 n2 1s22 9 0.00625 9 0.00684 0.00625 0.00684 0.00655 and s p2 1 18 2 n1 n2 2 1 1 1 1 0.00625 0.00625 0.2 0.03618 sd s p2 10 10 n1 n2 df n1 n2 2 10 10 2 18 . 11 252solnD1 10/18/06 (i) Test Ratio: t (Open this document in 'Page Layout' view!)Edited to replace or with D . d D0 sd 0.277 0 7.66 . If this test ratio lies between t , do not reject H 0 . 0.03618 2 Make a diagram with zero in the middle showing 'reject' regions below t.18 025 2.101 and above t.18 025 2.101. Since 7.66 falls in a 'reject' region, reject H 0 . (ii). Critical Value: d CV D0 t 2 s d 0 2.1010.03618 0.076 . Make a diagram with zero in the middle showing 'reject' regions below -0.076 and above 0.076. Since d 0.277 falls in a 'reject' region, reject H 0 . (iii) Confidence Interval: D d t 2 s d 0.277 2.1010.03618 0.277 0.076 or 0.201 to 0.353. Since zero is not on this interval, reject H 0 . b) (Optional) We assume that the variances are not equal. ( 12 22 ). approximation. So use the Satterthwaite s12 0.00625 s 2 .00684 s2 s2 0.000625 , 2 0.000684 , so 1 2 0.000625 0.000684 0.001309 , n1 10 n2 10 n1 n2 sd DF s12 s22 0.001309 0.0362 , d x1 x2 8.20 7.10 1.10 and n1 n2 s12 s22 n 1 n2 2 0..001309 2 0.000625 2 0.0006846250 2 17 .96 . We probably should round down to 17 2 2 s12 s22 9 9 n 1 n2 n1 1 n2 1 degrees of freedom, but note that, if we use 18 degrees of freedom, our results with this method are the same as those with the traditional method. d D0 0.277 0 7.66 . If this test ratio lies between t , do not reject H0 . (i) Test Ratio: t sd 0.0362 2 Make a diagram with zero in the middle showing 'reject' regions below t.17 025 2.110 and above t.17 025 2.110 . Since 7.66 falls in a 'reject' region, reject H0 . (ii). Critical Value: d CV D0 t 2 s d 0 2.110 0.03618 0.076 . Make a diagram with zero in the middle showing 'reject' regions below -0.076 and above 0.076. Since d 0.277 falls in a 'reject' region, reject H 0 . (iii) Confidence Interval: D d t 2 s d 0.277 2.110 0.03618 0.277 0.076 or 0.201 to 0.353. Since zero is not on this interval, reject H 0 . Parts not copied ©2003 Roger Even Bove 12