Chapter 10: Statistical Inferences About Two Populations Chapter 10 Statistical Inferences about Two Populations LEARNING OBJECTIVES The general focus of Chapter 10 is on testing hypotheses and constructing confidence intervals about parameters from two populations, thereby enabling you to 1. Test hypotheses and construct confidence intervals about the difference in two population means using the z statistic. 2. Test hypotheses and establish confidence intervals about the difference in two population means using the t statistic. 3. Test hypotheses and construct confidence intervals about the difference in two related populations or in matched-pairs experiments from one population 4. Test hypotheses and construct confidence intervals about the difference in two population proportions. 5. Test hypotheses and construct confidence intervals about two population variances. CHAPTER TEACHING STRATEGY The major emphasis of chapter 10 is on analyzing data from two samples. The student should be ready to deal with this topic given that he/she has tested hypotheses and computed confidence intervals in previous chapters on single sample data. In this chapter, the approach as to whether to use a z statistic or a t statistic for analyzing the differences in two sample means is the same as that used in chapters 8 and 9. When the population variances are known, the z statistic can be used. However, if the population variances are unknown and sample variances are being used, then the t test is the appropriate statistic for the analysis. It is always an assumption underlying the use of the t statistic that the populations are normally distributed. If sample sizes are small and the population variances are known, the z statistic can be used if the populations are normally distributed. © 2010 John Wiley & Sons Canada, Ltd. 319 Chapter 10: Statistical Inferences About Two Populations In conducting a t test for the difference of two means from independent populations, there are two different formulas given in the chapter. One version of this test uses a "pooled" estimate of the population variance and assumes that the population variances are equal. The other version does not assume equal population variances and is simpler to compute. In doing hand calculations, it is generally easier to use the “pooled” variance formula because the degrees of freedom formula for the unequal variance formula is quite complex. However, it is good to expose students to both formulas since computer software packages often give you the option of using the “pooled” that assumes equal population variances or the formula for unequal variances. A t test is also included for related (non independent) samples. It is important that the student be able to recognize when two samples are related and when they are independent. The first portion of section 10.3 addresses this issue. To underscore the potential difference in the outcome of the two techniques, it is sometimes valuable to analyze some related measures data with both techniques and demonstrate that the results and conclusions are usually quite different. You can have your students work problems like this using both techniques to help them understand the differences between the two tests (independent and dependent t tests) and the different outcomes they will obtain. A z test of proportions for two samples is presented here along with an F test for two population variances. This is a good place to introduce the student to the F distribution in preparation for analysis of variance in Chapter 11. The student will begin to understand that the F values have two different degrees of freedom. The F distribution tables are upper tailed only. For this reason, formula 10.14 is given in the chapter to be used to compute lower tailed F values for two-tailed tests. © 2010 John Wiley & Sons Canada, Ltd. 320 Chapter 10: Statistical Inferences About Two Populations CHAPTER OUTLINE 10.1 Hypothesis Testing and Confidence Intervals about the Difference in Two Means using the z Statistic (Population Variances Known) Hypothesis Testing Confidence Intervals Using the Computer to Test Hypotheses about the Difference in Two Population Means Using the z Test 10.2 Hypothesis Testing and Confidence Intervals about the Difference in Two Means: Independent Samples and Population Variances Unknown Hypothesis Testing Using the Computer to Test Hypotheses and Construct Confidence Intervals about the Difference in Two Population Means Using the t Test Confidence Intervals 10.3 Statistical Inferences for Two Related Populations Hypothesis Testing Using the Computer to Make Statistical Inferences about Two Related Populations Confidence Intervals 10.4 Statistical Inferences about Two Population Proportions, p1- p2 Hypothesis Testing Confidence Intervals 10.5 Testing Hypotheses about Two Population Variances Using the Computer to Test Hypotheses about Two Population Variances KEY TERMS Correlated t test Dependent Samples F Distribution F Value Independent Samples Matched-Pairs Test t test for related Measures © 2010 John Wiley & Sons Canada, Ltd. 321 Chapter 10: Statistical Inferences About Two Populations SOLUTIONS TO PROBLEMS IN CHAPTER 10 10.1 Sample 1 x 1 = 51.3 σ12 = 52 n1 = 31 a) Ho: Ha: Sample 2 x 2 = 53.2 σ22 = 60 n2 = 32 µ1 - µ2 = 0 µ1 - µ2 < 0 For one-tail test, = .10 z = z.10 = -1.28 ( x1 x 2 ) ( 1 2 ) 1 2 n1 2 2 (51.3 53.2) (0) n2 52 60 31 32 = -1.01 Since the observed z = -1.01 > zc = -1.28, the decision is to fail to reject the null hypothesis. b) Critical value method: zc = ( x 1 x 2 ) c ( 1 2 ) 12 n1 -1.28 = 22 n2 ( x1 x 2 ) c (0) 52 60 31 32 ( x 1 - x 2)c = -2.41 c) The area for z = -1.01 using Table A.5 is .3438. The p-value is .5000 - .3438 = .1562 © 2010 John Wiley & Sons Canada, Ltd. 322 Chapter 10: Statistical Inferences About Two Populations 10.2 Sample 1 n1 = 32 x 1 = 70.4 1 = 5.76 Sample 2 n2 = 31 x 2 = 68.7 2 = 6.1 For a 90% C.I., z.05 = 1.645 ( x1 x 2 ) z 12 n1 22 n2 (70.4) – 68.7) + 1.645 5.762 6.12 32 31 1.7 ± 2.46 -.76 < µ1 - µ2 < 4.16 10.3 a) Sample 1 Sample 2 x 1 = 88.23 12 = 22.74 n1 = 30 x 2 = 81.2 22 = 26.65 n2 = 30 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 For two-tail test, use /2 = .01 z = ( x 1 x 2 ) ( 1 2 ) 1 2 n1 2 2 n2 z.01 = + 2.33 (88.23 81.2) (0) 22.74 26.65 30 30 = 5.48 Since the observed z = 5.48 > z.01 = 2.33, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 323 Chapter 10: Statistical Inferences About Two Populations 12 b) ( x 1 x 2 ) z n1 22 (88.23 – 81.2) + 2.33 n2 22.74 26.65 30 30 7.03 + 2.99 4.04 < < 10.02 This supports the decision made in a) to reject the null hypothesis because zero is not in the interval. 10.4 Computers/electronics Food/Beverage x 1 = 1.96 12 = 1.0188 n1 = 50 Ho: Ha: x 2 = 3.02 22 = .9180 n2 = 50 µ1 - µ2 = 0 µ1 - µ2 0 For two-tail test, /2 = .005 z = ( x1 x 2 ) ( 1 2 ) 1 2 n1 2 2 n2 z.005 = ±2.575 (1.96 3.02) (0) 1.0188 0.9180 50 50 = -5.39 Since the observed z = -5.39 < zc = -2.575, the decision is to reject the null hypothesis. 10.5 A n1 = 40 x 1 = 5.3 12 = 1.99 For a 95% C.I., B n2 = 37 x 2 = 6.5 22 = 2.36 z.025 = 1.96 © 2010 John Wiley & Sons Canada, Ltd. 324 Chapter 10: Statistical Inferences About Two Populations ( x1 x 2 ) z 12 n1 (5.3 – 6.5) + 1.96 22 n2 1.99 2.36 40 37 -1.86 < < -.54 -1.2 ± .66 The results indicate that we are 95% confident that, on average, Plumber B does between 0.54 and 1.86 more jobs per day than Plumber A. Since zero does not lie in this interval, we are confident that there is a difference between Plumber A and Plumber B. 10.6 Managers n1 = 35 x 1 = 1.84 1 = .38 Specialty n2 = 41 x 2 = 1.99 2 = .51 for a 98% C.I., ( x1 x 2 ) z 12 n1 z.01 = 2.33 22 n2 .382 .512 (1.84 - 1.99) ± 2.33 35 41 -.15 ± .2384 -.3884 < µ1 - µ2 < .0884 Point Estimate = -.15 Hypothesis Test: 1) Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 2) z = ( x 1 x 2 ) ( 1 2 ) 12 n1 22 n2 © 2010 John Wiley & Sons Canada, Ltd. 325 Chapter 10: Statistical Inferences About Two Populations 3) = .02 4) For a two-tailed test, z.01 = + 2.33. If the observed z value is greater than 2.33 or less than -2.33, then the decision will be to reject the null hypothesis. 5) Data given above 6) z = (1.84 1.99) (0) (.38) 2 (.51) 2 35 41 = -1.47 7) Since z = -1.47 > z.01 = -2.33, the decision is to fail to reject the null hypothesis. 8) There is no significant difference in the hourly rates of the two groups. 10.7 1999 2009 x 1 = 190 1 = 18.50 n1 = 51 x 2 = 198 2 = 15.60 n2 = 47 = .01 H0: 1 - 2 = 0 Ha: 1 - 2 < 0 For a one-tailed test, z = z.01 = -2.33 ( x 1 x 2 ) ( 1 2 ) 1 2 n1 2 2 n2 (190 198) (0) (18.50) 2 (15.60) 2 51 47 = -2.32 Since the observed z = -2.32 > z.01 = -2.33, the decision is to fail to reject the null hypothesis. 10.8 Vancouver Montreal n1 = 31 x 1 = 3.88 12 = .12 n2 = 31 x 2 = 5.59 22 = .06 For a 99% C.I., z.005 = 2.575 © 2010 John Wiley & Sons Canada, Ltd. 326 Chapter 10: Statistical Inferences About Two Populations 12 ( x1 x 2 ) z n1 22 n2 (3.88 – 5.59) ± 2.575 .12 .06 31 31 −1.91 < < −1.51 − 1.71 ± .20 Between $ 1.51 and $ 1.91 difference with Montreal being more expensive. 10.9 Manan x 1 = 5.8 1 = 1.7 n1 = 36 Ho: Ha: Prairie x 2 = 5.0 2 = 1.4 n2 = 45 µ1 - µ2 = 0 µ1 - µ2 0 For two-tail test, /2 = .025 z = ( x 1 x 2 ) ( 1 2 ) 12 n1 22 n2 z.025 = ±1.96 (5.8 5.0) (0) (1.7) 2 (1.4) 2 36 45 = 2.27 Since the observed z = 2.27 > zc = 1.96, the decision is to reject the null hypothesis. 10.10 A x 1 = 8.05 1 = 1.36 n1 = 50 Ho: Ha: B x 2 = 7.26 2 = 1.06 n2 = 38 µ1 - µ2 = 0 µ1 - µ2 > 0 For one-tail test, = .10 z = ( x 1 x 2 ) ( 1 2 ) 1 2 n1 2 2 n2 z.10 = 1.28 (8.05 7.26) (0) (1.36) 2 (1.06) 2 50 38 = 3.06 © 2010 John Wiley & Sons Canada, Ltd. 327 Chapter 10: Statistical Inferences About Two Populations Since the observed z = 3.06 > zc = 1.28, the decision is to reject the null hypothesis. = .01 10.11 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 < 0 df = 8 + 11 - 2 = 17 Sample 1 n1 = 8 x 1 = 24.56 s12 = 12.4 Sample 2 n2 = 11 x 2 = 26.42 s22 = 15.8 For one-tail test, = .01 Critical t.01,17 = -2.567 t= ( x1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 = (24.56 26.42) (0) = -1.05 12.4(7) 15.8(10) 1 1 8 11 2 8 11 Since the observed t = -1.05 > t.01,19 = -2.567, the decision is to fail to reject the null hypothesis. 10.12 a) Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 =.10 Sample 1 n1 = 20 x 1 = 118 s1 = 23.9 Sample 2 n2 = 20 x 2 = 113 s2 = 21.6 df = 20 + 20 - 2 = 38 For two-tail test, t = t = /2 = .05 Critical t.05,38 = 1.697 (used df=30) ( x1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (118 113) (0) (23.9) (19) (21.6) 2 (19) 1 1 20 20 2 20 20 2 = = 0.69 © 2010 John Wiley & Sons Canada, Ltd. 328 Chapter 10: Statistical Inferences About Two Populations Since the observed t = 0.69 < t.05,38 = 1.697, the decision is to fail to reject the null hypothesis. s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 b) ( x1 x 2 ) t (118 – 113) + 1.697 2 = (23.9) 2 (19) (21.6) 2 (19) 1 1 20 20 2 20 20 5 + 12.224 -7.224 < 1 - 2 < 17.224 10.13 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 > 0 = .05 Sample 1 n1 = 10 x 1 = 45.38 s1 = 2.357 Sample 2 n2 = 10 x 2 = 40.49 s2 = 2.355 For one-tail test, = .05 Critical t.05,18 = 1.734 t = t = df = n1 + n2 - 2 = 10 + 10 - 2 = 18 ( x1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (45.38 40.49) (0) (2.357) 2 (9) (2.355) 2 (9) 1 1 10 10 2 10 10 = = 4.64 Since the observed t = 4.64 > t.05,18 = 1.734, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 329 Chapter 10: Statistical Inferences About Two Populations 10.14 =.01 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 df = 18 + 18 - 2 = 34 Sample 1 n1 = 18 x 1 = 5.333 s12 = 12 Sample 2 n2 = 18 x 2 = 9.444 s22 = 2.026 For two-tail test, /2 = .005 t = t = Critical t.005,34 = ±2.75 (used df=30) ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 = (5.333 9.444) (0) = -4.66 12(17) (2.026)17 1 1 18 18 2 18 18 Since the observed t = -4.66 < t.005,34 = -2.75, reject the null hypothesis. b) For 98% confidence, t.01, 30 = 2.457 s (n 1) s 2 (n2 1) 1 1 ( x1 x 2 ) t 1 1 n1 n2 2 n1 n2 2 2 (5.333 – 9.444) + 2.457 = (12)(17) (2.026)(17) 1 1 18 18 2 18 18 -4.111 + 2.169 -6.280 < 1 - 2 < -1.942 10.15 Montreal n1 = 21 x1 = 252,000 s1 = 4,900 Halifax n2 = 26 x 2 = 243,000 s2 = 3,700 90% level of confidence, /2 = .05 df = 21 + 26 – 2 = 45 t .05,45 = 1.684 (used df = 40) © 2010 John Wiley & Sons Canada, Ltd. 330 Chapter 10: Statistical Inferences About Two Populations s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 ( x1 x 2 ) t 2 = (4,900) 2 (20) (3,700) 2 (25) 1 1 (252,000 – 243,000) + 1.684 = 21 26 2 21 26 9,000 + 2112.24 6887.76 < 1 - 2 < 11,112.24 Since zero does not fall into the 90% confidence interval, we can state at = 0.10 that there is a difference in the mean prices. = .10 df = 12 + 12 - 2 = 22 10.16 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 Co-op n1 = 12 x 1 = $15.645 s1 = $1.093 Interns n2 = 12 x 2 = $15.439 s2 = $0.958 For two-tail test, /2 = .05 Critical t.05,22 = ± 1.717 t = t = ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 = (15.645 15.439) (0) (1.093) 2 (11) (0.958) 2 (11) 1 1 12 12 2 12 12 = 0.49 Since the observed t = 0.49 < t.05,22 = 1.717, the decision is to fail to reject the null hypothesis. 90% Confidence Interval: s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 ( x1 x 2 ) t t.05,22 = ± 1.717 2 = © 2010 John Wiley & Sons Canada, Ltd. 331 Chapter 10: Statistical Inferences About Two Populations (1.093) 2 (11) (0.958) 2 (11) 1 1 = 12 12 2 12 12 (15.645 – 15.439) + 1.717 0.206 + 0.7204 -0.5144 < 1 - 2 < 0.9264 10.17 Let Toronto be group 1 1) Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 > 0 2) t = ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 3) = .05 4) For a one-tailed test and df = 8 + 9 - 2 = 15, t.05,15 = 1.753. If the observed value of t is greater than 1.753, the decision is to reject the null hypothesis. 5) Toronto n1 = 8 x 1 = 47 s1 = 3 6) t = Montreal n2 = 9 x 2 = 44 s2 = 3 (47 44) (0) 7(3) 2 8(3) 2 15 = 2.06 1 1 8 9 7) Since t = 2.06 > t.05,15 = 1.753, the decision is to reject the null hypothesis. 8) The rental rates in Toronto are significantly higher than in Montreal. 10.18 nE = 22 x E = 112 sE = 11 nQ = 20 x Q = 122 sQ = 12 df = nE + nQ - 2 = 22 + 20 - 2 = 40 For a 98% Confidence Interval, /2 = .01 and t.01,40 = 2.423 © 2010 John Wiley & Sons Canada, Ltd. 332 Chapter 10: Statistical Inferences About Two Populations s E (n E 1) sQ (nQ 1) 2 (x E xQ ) t 2 n E 1 nQ 2 (112 – 122) + 2.423 1 1 n E nQ = (11)2 (21) (12) 2 (19) 1 1 22 20 2 22 20 -10 ± 8.60 -$18.60 < µ1 - µ2 < -$1.40 Point Estimate = -$10 10.19 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 df = n1 + n2 - 2 = 11 + 11 - 2 = 20 Perth n1 = 11 x 1 = $67,381.82 s1 = $2,067.28 Mexico City n2 = 11 x 2 = $63,390.91 s2 = $1,526.08 For a two-tail test, /2 = .025 t = t = Critical t.025,20 = ±2.086 ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 = (67,381.82 63,390.91) (0) (2,067.28) 2 (10) (1,526.08) 2 (10) 1 1 11 11 2 11 11 = 5.15 Since the observed t = 5.15 > t.025,20 = 2.086, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 333 Chapter 10: Statistical Inferences About Two Populations 10.20 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 > 0 df = n1 + n2 - 2 = 9 + 10 - 2 = 17 Men n1 = 9 x 1 = $110.92 s1 = $28.79 Women n2 = 10 x 2 = $75.48 s2 = $30.51 This is a one-tail test, = .01 t = t = Critical t.01,17 = 2.567 ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (110.92 75.48) (0) (28.79) 2 (8) (30.51) 2 (9) 1 1 9 10 2 9 10 = = 2.60 Since the observed t = 2.60 > t.01,17 = 2.567, the decision is to reject the null hypothesis. 10.21 Ho: D = 0 Ha: D > 0 Sample 1 38 27 30 41 36 38 33 35 44 n=9 Sample 2 22 28 21 38 38 26 19 31 35 d =7.11 d 16 -1 9 3 -2 12 14 4 9 sd=6.45 = .01 df = n - 1 = 9 - 1 = 8 For one-tail test and = .01, the critical t.01,8 = ±2.896 © 2010 John Wiley & Sons Canada, Ltd. 334 Chapter 10: Statistical Inferences About Two Populations t = d D 7.11 0 = 3.31 sd 6.45 n 9 Since the observed t = 3.31 > t.01,8 = 2.896, the decision is to reject the null hypothesis. 10.22 Ho: Ha: D=0 D0 Before 107 99 110 113 96 98 100 102 107 109 104 99 101 After 102 98 100 108 89 101 99 102 105 110 102 96 100 n = 13 d = 2.5385 df = n - 1 = 13 - 1 = 12 d 5 1 10 5 7 -3 1 0 2 -1 2 3 1 sd=3.4789 For a two-tail test and /2 = .025 t = = .05 Critical t.025,12 = ±2.179 d D 2.5385 0 = 2.63 sd 3.4789 n 13 Since the observed t = 2.63 > t.025,12 = 2.179, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 335 Chapter 10: Statistical Inferences About Two Populations d = 40.56 10.23 n = 22 sd = 26.58 For a 98% Level of Confidence, /2 = .01, and df = n - 1 = 22 - 1 = 21 t.01,21 = 2.518 d t sd n 40.56 ± (2.518) 26.58 22 40.56 ± 14.27 26.29 < D < 54.83 10.24 Before 32 28 35 32 26 25 37 16 35 After 40 25 36 32 29 31 39 30 31 n=9 d = -3 df = n - 1 = 9 - 1 = 8 d -8 3 -1 0 -3 -6 -2 -14 4 sd = 5.6347 For 90% level of confidence and /2 = .05, t = d t = .025 t.05,8 = 1.86 sd n t = -3 + (1.86) 5.6347 = -3 ± 3.49 9 -6.49 < D < 0.49 © 2010 John Wiley & Sons Canada, Ltd. 336 Chapter 10: Statistical Inferences About Two Populations 10.25 City Cost Resale d Calgary Edmonton Fredericton Halifax London, ON Montreal Regina Toronto Vancouver Victoria Winnipeg 20427 27255 22115 23256 21887 24255 19852 23624 25885 28999 20836 25163 24625 12600 24588 19267 20150 22500 16667 26875 35333 16292 -4736 2630 9515 -1332 2620 4105 -2648 6957 - 990 -6334 4544 d = 1302.82 sd = 4938.22 n = 11, df = 10 = .01 d t /2 = .005 sd n t.005,10= 3.169 = 1302.82 + 3.169 4938.22 = 1302.82 + 4718.42 11 -3415.6 < D < 6021.2 10.26 Ho: Ha: D=0 D<0 Before 2 4 1 3 4 2 2 3 1 After 4 5 3 3 3 5 6 4 5 d =-1.778 n=9 d -2 -1 -2 0 1 -3 -4 -1 -4 sd=1.716 = .05 df = n - 1 = 9 - 1 = 8 For a one-tail test and = .05, the critical t.05,8 = -1.86 t = d D 1.778 0 = -3.11 sd 1.716 n 9 © 2010 John Wiley & Sons Canada, Ltd. 337 Chapter 10: Statistical Inferences About Two Populations Since the observed t = -3.11 < t.05,8 = -1.86, the decision is to reject the null hypothesis. 10.27 Before 6.59 5.95 7.50 6.26 7.76 6.46 5.56 5.95 5.82 5.66 6.10 After 5.09 5.82 5.56 5.56 6.21 6.08 4.91 6.21 5.17 5.25 5.77 d = 0.7255 n = 11 d 1.50 0.13 1.94 0.70 1.55 0.38 0.65 -0.26 0.65 0.41 0.33 sd=0.6684 df = n - 1 = 11 - 1 = 10 For a 98% level of confidence and /2=.01, t.01,10 = 2.764 d t sd n 0.7255 ± (2.764) 0.6684 11 = 0.7255 ± 0.5570 0.1685 < D < 1.2825 10.28 H0: D = 0 Ha: D > 0 n = 27 df = 27 – 1 = 26 d = 3.17 sd = 5 Since = .01, the critical t.01,26 = 2.479 t = d D 3.71 0 = 3.86 sd 5 n 27 Since the observed t = 3.86 > t.01,26 = 2.479, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 338 Chapter 10: Statistical Inferences About Two Populations d = 75 10.29 n = 21 sd = 30 df = 21 - 1 = 20 For a 90% confidence level, /2=.05 and t.05,20 = 1.725 d t sd n 75 + 1.725 30 = 75 ± 11.29 21 63.71 < D < 86.29 10.30 Ho: Ha: D=0 D0 d = 1.32 n = 15 sd = 2.08 = .01 df = 15 - 1 = 14 For a two-tail test, /2 = .005 and the critical t.005,14 = + 2.977 d D 1.32 0 = 2.46 sd 2.08 t = n 15 Since the observed t = 2.46 < t.005,14 = 2.977, we fail to reject the null hypothesis. 10.31 a) Sample 1 n1 = 368 x1 = 175 pˆ 1 p Sample 2 n2 = 405 x2 = 182 x1 175 = .476 n1 368 pˆ 2 x2 182 = .449 n2 405 x1 x2 175 182 357 = .462 n1 n2 368 405 773 Ho: p1 - p2 = 0 Ha: p1 - p2 0 For two-tail, /2 = .025 and z.025 = ±1.96 © 2010 John Wiley & Sons Canada, Ltd. 339 Chapter 10: Statistical Inferences About Two Populations z ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n (.476 .449) (0) 1 1 (.462)(. 538) 368 405 = 0.75 Since the observed z = 0.75 < zc = 1.96, the decision is to fail to reject the null hypothesis. b) p Ho: Ha: Sample 1 p̂ 1 = .38 n1 = 649 Sample 2 p̂ 2 = .25 n2 = 558 n1 pˆ 1 n2 pˆ 2 649(.38) 558(.25) = .32 n1 n2 649 558 p1 - p2 = 0 p1 - p2 > 0 For a one-tail test and = .10, z ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n2 z.10 = 1.28 (.38 .25) (0) 1 1 (.32)(. 68) 649 558 = 4.83 Since the observed z = 4.83 > zc = 1.28, the decision is to reject the null hypothesis. 10.32 a) n1 = 85 n2 = 90 p̂ 1 = .75 p̂ 2 = .67 For a 90% Confidence Level, z.05 = 1.645 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.75 - .67) ± 1.645 (.75)(.25) (.67)(.33) = .08 ± .11 85 90 -.03 < p1 - p2 < .19 © 2010 John Wiley & Sons Canada, Ltd. 340 Chapter 10: Statistical Inferences About Two Populations b) n1 = 1100 n2 = 1300 p̂ 1 = .19 p̂ 2 = .17 For a 95% Confidence Level, /2 = .025 and z.025 = 1.96 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.19 - .17) + 1.96 (.19)(.81) (.17)(.83) = .02 ± .03 1100 1300 -.01 < p1 - p2 < .05 c) n1 = 430 pˆ 1 n2 = 399 x1 275 = .64 n1 430 x1 = 275 pˆ 2 x2 = 275 x2 275 = .69 n2 399 For an 85% Confidence Level, /2 = .075 and ( pˆ 1 pˆ 2 ) z z.075 = 1.44 pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.64 - .69) + 1.44 (.64)(.36) (.69)(.31) = -.05 ± .047 430 399 -.097 < p1 - p2 < -.003 d) n1 = 1500 pˆ 1 n2 = 1500 x1 = 1050 x1 1050 = .70 n1 1500 x2 = 1100 pˆ 2 x2 1100 = .733 n2 1500 For an 80% Confidence Level, /2 = .10 and z.10 = 1.28 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 © 2010 John Wiley & Sons Canada, Ltd. 341 Chapter 10: Statistical Inferences About Two Populations (.70 - .733) ± 1.28 (.70)(.30) (.733)(.267) = -.033 ± .021 1500 1500 -.054 < p1 - p2 < -.012 10.33 H0: pm - pw = 0 Ha: pm - pw < 0 nm = 374 For a one-tailed test and = .05, p z nw = 481 p̂ m = .59 p̂ w = .70 z.05 = -1.645 nm pˆ m nw pˆ w 374(.59) 481(.70) = .652 nm n w 374 481 ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n2 (.59 .70) (0) 1 1 (.652)(. 348) 374 481 = -3.35 Since the observed z = -3.35 < z.05 = -1.645, the decision is to reject the null hypothesis. 10.34 n1 = 210 n2 = 176 p̂1 = .24 For a 90% Confidence Level, /2 = .05 and ( pˆ 1 pˆ 2 ) z p̂2 = .35 z.05 = + 1.645 pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.24 - .35) + 1.645 (.24)(.76) (.35)(.65) = -.11 + .0765 210 176 -.1865 < p1 – p2 < -.0335 © 2010 John Wiley & Sons Canada, Ltd. 342 Chapter 10: Statistical Inferences About Two Populations 10.35 Computer Firms p̂ 1 = .48 n1 = 56 Banks p̂ 2 = .56 n2 = 89 n1 pˆ 1 n2 pˆ 2 56(.48) 89(.56) = .529 n1 n2 56 89 p Ho: Ha: p1 - p2 = 0 p1 - p2 0 For two-tail test, /2 = .10 and zc = ±1.28 z ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n (.48 .56) (0) 1 1 (.529)(. 471) 56 89 = -0.94 Since the observed z = -0.94 > zc = -1.28, the decision is to fail to reject the null hypothesis. 10.36 A n1 = 35 x1 = 5 pˆ 1 B n2 = 35 x2 = 7 x1 5 = .14 n1 35 pˆ 2 x2 7 = .20 n2 35 For a 98% Confidence Level, /2 = .01 and z.01 = 2.33 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.14 - .20) ± 2.33 (.14)(.86) (.20)(.80) 35 35 = -.06 ± .21 -.27 < p1 - p2 < .15 © 2010 John Wiley & Sons Canada, Ltd. 343 Chapter 10: Statistical Inferences About Two Populations 10.37 H0: p1 – p2 = 0 Ha: p1 – p2 0 = .10 p̂ = .09 p̂ = .06 1 For a two-tailed test, /2 = .05 and p Z n1 = 780 2 n2 = 915 z.05 = + 1.645 n1 pˆ 1 n2 pˆ 2 780(.09) 915(.06) = .0738 n1 n2 780 915 ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n (.09 .06) (0) 1 1 (.0738)(. 9262) 780 915 = 2.35 Since the observed z = 2.35 > z.05 = 1.645, the decision is to reject the null hypothesis. 10.38 n1 = 850 n2 = 910 p̂ = .60 p̂ 2 = .52 1 For a 95% Confidence Level, /2 = .025 and z.025 = + 1.96 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.60 - .52) + 1.96 (.60)(.40) (.52)(.48) = .08 + .046 850 910 .034 < p1 – p2 < .126 10.39 H0: 12 = 22 Ha: 12 < 22 = .01 dfnum = 10 - 1 = 9 Table F.99, 9,11 n1 = 10 n2 = 12 s12 = 562 s22 = 1013 dfdenom = 12 - 1 = 11 1 F.01,11, 9 1 F.01,10, 9 1 .190 5.26 © 2010 John Wiley & Sons Canada, Ltd. 344 Chapter 10: Statistical Inferences About Two Populations F = s1 2 s2 2 562 = .5548 1013 Since the observed F = .5548 >F.99,9,11 = .190, the decision is to fail to reject the null hypothesis. 10.40 H0: 12 = 22 Ha: 12 22 = .05 dfnum = 5 - 1 = 4 F = 2 s2 2 s1 = 4.68 s2 = 2.78 dfdenom = 19 - 1 = 18 The critical table F values are: s1 n1 = 5 n2 = 19 F.025,4,18 = 3.61 F..975,18,4 = .277 (4.68) 2 = 2.83 (2.78) 2 Since the observed F = 2.83 < F.025,4,18 = 3.61, the decision is to fail to reject the null hypothesis. 10.41 City 1 City 2 1.029 1.020 1.017 1.002 0.996 1.017 1.014 1.014 1.014 0.983 n1 = 10 0.999 1.026 1.017 1.008 0.990 1.038 1.017 1.032 1.011 1.014 df1 = 9 n2 = 10 df2 = 9 s12 = 0.0001769 s22 = 0.0002104 H0: 12 = 22 Ha: 12 22 = .01 /2 = .005 © 2010 John Wiley & Sons Canada, Ltd. 345 Chapter 10: Statistical Inferences About Two Populations Upper tail critical F value = F.005,9,9 = 6.54 Lower tail critical F value = F.995,9,9 = 0.153 2 s .0001769 F = 12 = 0.8408 .0002104 s2 Since the observed F = 0.8408 is greater than the lower tail critical value of 0.153 and less than the upper tail critical value of 6.54, the decision is to fail to reject the null hypothesis. 10.42 Let Thunder Bay = group 1 and Moncton = group 2 1) H0: 12 = 22 Ha: 12 22 2 s1 2) F = 2 s2 3) = .01 4) df1 = 12 df2 = 10 This is a two-tailed test The critical table F values are: F.005,12,10 = 5.66 F.995,10,12 = .177 If the observed value is greater than 5.66 or less than .177, the decision will be to reject the null hypothesis. 5) s12 = 393.4 6) F = s22 = 702.7 393.4 = 0.56 702.7 7) Since F = 0.56 is greater than .177 and less than 5.66, the decision is to fail to reject the null hypothesis. 8) There is no significant difference in the variances of number of days between Thunder Bay and Moncton. © 2010 John Wiley & Sons Canada, Ltd. 346 Chapter 10: Statistical Inferences About Two Populations 10.43 H0: 12 = 22 Ha: 12 > 22 dfnum = 12 - 1 = 11 = .05 n1 = 12 n2 = 15 s1 = 7.52 s2 = 6.08 dfdenom = 15 - 1 = 14 The critical table F value is F.05,10,14 = 2.60 F= s1 2 s2 2 (7.52) 2 = 1.53 (6.08) 2 Since the observed F = 1.53 < F.05,10,14 = 2.60, the decision is to fail to reject the null hypothesis. 10.44 H0: 12 = 22 Ha: 12 22 = .01 dfnum = 15 - 1 = 14 n1 = 15 n2 = 15 dfdenom = 15 - 1 = 14 The critical table F values are: F = s1 2 s2 2 s12 = 91.5 s22 = 67.3 F.005,12,14 = 4.43 F.995,14,12 = .226 91.5 = 1.36 67.3 Since the observed F = 1.36 < F.005,12,14 = 4.43 and > F.995,14,12 = .226, the decision is to fail to reject the null hypothesis. 10.45 Ho: Ha: µ1 - µ2 = 0 µ1 - µ2 0 For = .10 and a two-tailed test, /2 = .05 and z.05 = + 1.645 Sample 1 x 1 = 138.4 1 = 6.71 n1 = 48 Sample 2 x 2 = 142.5 2 = 8.92 n2 = 39 © 2010 John Wiley & Sons Canada, Ltd. 347 Chapter 10: Statistical Inferences About Two Populations z = ( x 1 x 2 ) ( 1 2 ) 12 n1 22 n2 (138.4 142.5) (0) (6.71) 2 (8.92) 48 39 = -2.38 Since the observed value of z = -2.38 is less than the critical value of z = -1.645, the decision is to reject the null hypothesis. There is a significant difference in the means of the two populations. 10.46 Sample 1 x 1 = 34.9 12 = 2.97 n1 = 34 Sample 2 x 2 = 27.6 22 = 3.50 n2 = 31 For 98% Confidence Level, z.01 = 2.33 ( x1 x 2 ) z 12 n1 22 n2 2.97 3.50 34 31 (34.9 – 27.6) + 2.33 = 7.3 + 1.04 6.26 < 1 - 2 < 8.34 10.47 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 > 0 Sample 1 x 1= 2.06 s12 = .176 n1 = 12 Sample 2 x 2 = 1.93 s22 = .143 n2 = 15 = .05 This is a one-tailed test with df = 12 + 15 - 2 = 25. The critical value is t.05,25 = 1.708. If the observed value is greater than 1.708, the decision will be to reject the null hypothesis. t = ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 © 2010 John Wiley & Sons Canada, Ltd. 348 Chapter 10: Statistical Inferences About Two Populations (2.06 1.93) (0) t = (.176)(11) (.143)(14) 1 1 25 12 15 = 0.85 Since the observed value of t = 0.85 is less than the critical value of t = 1.708, the decision is to fail to reject the null hypothesis. The mean for population one is not significantly greater than the mean for population two. 10.48 Sample 1 x 1 = 74.6 s12 = 10.5 n1 = 18 Sample 2 x 2 = 70.9 s22 = 11.4 n2 = 19 For 95% confidence, /2 = .025. Using df = 18 + 19 - 2 = 35, t30,.025 = 2.042 s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 ( x1 x 2 ) t 2 (10.5)(17) (11.4)(18) 1 1 18 19 2 18 19 (74.6 – 70.9) + 2.042 3.7 + 2.22 1.48 < 1 - 2 < 5.92 = .01 10.49 Ho: D = 0 Ha: D < 0 n = 21 df = 20 d = -1.16 sd = 1.01 The critical t.01,20 = -2.528. If the observed t is less than -2.528, then the decision will be to reject the null hypothesis. t = d D 1.16 0 = -5.26 sd 1.01 n 21 Since the observed value of t = -5.26 is less than the critical t value of -2.528, the decision is to reject the null hypothesis. The population difference is less than zero. © 2010 John Wiley & Sons Canada, Ltd. 349 Chapter 10: Statistical Inferences About Two Populations 10.50 Respondent 1 2 3 4 5 6 7 8 9 Before 47 33 38 50 39 27 35 46 41 After 63 35 36 56 44 29 32 54 47 d -16 -2 2 -6 -5 -2 3 -8 -6 sd = 5.703 df = 8 d = -4.44 For a 99% Confidence Level, /2 = .005 and t8,.005 = 3.355 d t sd = -4.44 + 3.355 n 5.703 = -4.44 + 6.38 9 -10.82 < D < 1.94 = .05 10.51 Ho: p1 - p2 = 0 Ha: p1 - p2 0 /2 = .025 z.025 = + 1.96 If the observed value of z is greater than 1.96 or less than -1.96, then the decision will be to reject the null hypothesis. Sample 1 x1 = 345 n1 = 783 x1 x2 345 421 = .4562 n1 n2 783 896 p pˆ 1 z Sample 2 x2 = 421 n2 = 896 x1 345 = .4406 n1 783 ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n pˆ 2 x2 421 = .4699 n2 896 (.4406 .4699) (0) 1 1 (.4562)(. 5438) 783 896 = -1.20 Since the observed value of z = -1.20 is greater than -1.96, the decision is to fail to reject the null hypothesis. There is no significant difference. © 2010 John Wiley & Sons Canada, Ltd. 350 Chapter 10: Statistical Inferences About Two Populations 10.52 Sample 1 n1 = 409 p̂ 1 = .71 Sample 2 n2 = 378 p̂ 2 = .67 For a 99% Confidence Level, /2 = .005 and ( pˆ 1 pˆ 2 ) z z.005 = 2.575 pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.71 - .67) + 2.575 (.71)(.29) (.67)(.33) = .04 ± .085 409 378 -.045 < p1 - p2 < .125 10.53 H0: 12 = 22 Ha: 12 22 = .05 n1 = 8 n2 = 10 dfnum = 8 - 1 = 7 dfdenom = 10 - 1 = 9 The critical F values are: F.025,7,9 = 4.20 s12 = 46 s22 = 37 F.975,9,7 = .238 If the observed value of F is greater than 4.20 or less than .238, then the decision will be to reject the null hypothesis. F = s1 2 s2 2 46 = 1.24 37 Since the observed F = 1.24 is less than F.025,7,9 =4.20 and greater than F.975,9,7 = .238, the decision is to fail to reject the null hypothesis. There is no significant difference in the variances of the two populations. 10.54 Term x t = $75,000 st = $22,000 nt = 27 Whole Life x w = $45,000 sw = $15,500 nw = 29 df = 27 + 29 - 2 = 54 For a 95% Confidence Level, /2 = .025 and t.025,50 = 2.009 (used df=50) © 2010 John Wiley & Sons Canada, Ltd. 351 Chapter 10: Statistical Inferences About Two Populations s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 ( x1 x 2 ) t 2 (22,000) 2 (26) (15,500) 2 (28) 1 1 (75,000 – 45,000) + 2.009 27 29 2 27 29 30,000 ± 10,160.11 19,839.89 < µ1 - µ2 < 40,160.11 10.55 Morning 43 51 37 24 47 44 50 55 46 Afternoon 41 49 44 32 46 42 47 51 49 n=9 d = -0.444 For a 90% Confidence Level: d t d 2 2 -7 -8 1 2 3 4 -3 sd =4.447 df = 9 - 1 = 8 /2 = .05 and t.05,8 = 1.86 sd n -0.444 + (1.86) 4.447 = -0.444 ± 2.757 9 -3.201 < D < 2.313 10.56 Marketing n1 = 400 x1 = 220 Ho: p1 - p2 = 0 Ha: p1 - p2 > 0 Accountants n2 = 450 x2 = 216 = .01 The critical table z value is: z.01 = 2.33 © 2010 John Wiley & Sons Canada, Ltd. 352 Chapter 10: Statistical Inferences About Two Populations p̂1 = p z 220 = .55 400 p̂2 = 216 = .48 450 x1 x2 220 216 = .513 n1 n2 400 450 ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n (.55 .48) (0) 1 1 (.513)(.487) 400 450 = 2.04 Since the observed z = 2.04 is less than z.01 = 2.33, the decision is to fail to reject the null hypothesis. There is no significant difference between marketing managers and accountants in the proportion who keep track of obligations “in their head”. 10.57 Accounting n1 = 16 x 1 = 26,400 s1 = 1,200 Data Entry n2 = 14 x 2 = 25,800 s2 = 1,050 H0: 12 = 22 Ha: 12 22 = .05 dfnum = 16 – 1 = 15 dfdenom = 14 – 1 = 13 The critical F values are: F.025,15,13 = 3.05 F = s1 2 s2 2 and F.975,15,13 = 0.33 1,440,000 = 1.31 1,102,500 Since the observed F = 1.31 is less than F.025,15,13 = 3.05 and greater than F.975,15,13 = 0.33, the decision is to fail to reject the null hypothesis. 10.58 Men Women n1 = 60 x 1 = 631 1 = 100 n2 = 41 x 2 = 848 2 = 100 For a 95% Confidence Level, /2 = .025 and z.025 = 1.96 © 2010 John Wiley & Sons Canada, Ltd. 353 Chapter 10: Statistical Inferences About Two Populations ( x1 x 2 ) z 12 n1 22 n2 1002 1002 (631 – 848) + 1.96 = -217 ± 39.7 60 41 -256.7 < µ1 - µ2 < -177.3 10.59 Ho: Ha: = .01 df = 20 + 24 - 2 = 42 µ1 - µ2 = 0 µ1 - µ2 0 Guelph n1 = 20 x 1 = 7.97 s1 = 1.5 Sudbury n2 = 24 x 2 = 6.7 s2 = 1.2 For two-tail test, /2 = .005 and the critical t.005,42 = ±2.704 (used df=40) t = t = ( x1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (7.97 6.7) (0) (1.5) (19) (1.2) 2 (23) 1 1 42 20 24 2 = 3.12 Since the observed t = 3.12 > t.005,40 = 2.704, the decision is to reject the null hypothesis. 10.60 With Fertilizer x 1 = 97.5 1 = 24.9 n1 = 35 Ho: Ha: Without Fertilizer x 2 = 58.7 2 = 18.8 n2 = 35 µ1 - µ2 = 0 µ1 - µ2 > 0 For one-tail test, = .01 and z.01 = 2.33 © 2010 John Wiley & Sons Canada, Ltd. 354 Chapter 10: Statistical Inferences About Two Populations z = ( x 1 x 2 ) ( 1 2 ) 12 n1 22 n2 (97.5 58.7) (0) (24.9) 2 (18.8) 2 35 35 = 7.36 Since the observed z = 7.36 > z.01 = 2.33, the decision is to reject the null hypothesis. 10.61 Specialty n1 = 350 p̂ 1 = .75 Discount n2 = 500 p̂ 2 = .52 = 0.01, /2 = .05 and z.05 = ±1.645 p n1 pˆ 1 n2 pˆ 2 350(.75) 500(.52) .615 n1 n2 350 500 Ho: p1 - p2 = 0 Ha: p1 - p2 0 z ( pˆ 1 pˆ 2 ) ( p1 p 2 ) (.75 .52) (0) 6.78 1 1 1 1 (.615)(.385) p q 350 500 n1 n2 Since the observed z = 6.78 > zc = 1.645, the decision is to reject the null hypothesis. 10.62 H0: 12 = 22 Ha: 12 22 dfnum = 6 = .01 n1 = 8 n2 = 7 2 s1 = 72,909 s2 = 129,569 2 dfdenom = 7 The critical F values are: F = s1 2 s2 2 F.005,6,7 = 9.16 F.995,7,6 = .11 72,909 = .56 129,569 Since F = .56 < F.005,6,7 = 9.16 but also > F.995,7,6 = .11, the decision is to fail to reject the null hypothesis. There is no difference in the variances of the shifts. © 2010 John Wiley & Sons Canada, Ltd. 355 Chapter 10: Statistical Inferences About Two Populations 10.63 Name Brand 54 55 59 53 54 61 51 53 n=8 Store Brand 49 50 52 51 50 56 47 49 d = 4.5 d 5 5 7 2 4 5 4 4 sd=1.414 df = 8 - 1 = 7 For a 90% Confidence Level, /2 = .05 and d t t.05,7 = 1.895 sd n 4.5 + 1.895 1.414 = 4.5 ± .947 8 3.553 < D < 5.447 10.64 Ho: Ha: = .01 µ1 - µ2 = 0 µ1 - µ2 < 0 df = 23 + 19 - 2 = 40 Alberta n1 = 23 x 1 = 20.917 s12 = 3.104 Saskatchewan n2 = 19 x 2 = 22.074 s22 = 1.422 For one-tail test, = .01 and the critical t.01,40 = -2.423 t = t = ( x 1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (20.917 22.074) (0) (3.104)( 22) (1.422)(18) 1 1 40 23 19 = -2.44 © 2010 John Wiley & Sons Canada, Ltd. 356 Chapter 10: Statistical Inferences About Two Populations Since the observed t = -2.44 < t.01,40 = -2.423, the decision is to reject the null hypothesis. 10.65 Wednesday 71 56 75 68 74 n=5 Friday 53 47 52 55 58 d = 15.8 sd = 5.263 d 18 9 23 13 16 df = 5 - 1 = 4 = .05 Ho: D = 0 Ha: D > 0 For one-tail test, = .05 and the critical t.05,4 = 2.132 d D 15.8 0 = 6.71 sd 5.263 t = n 5 Since the observed t = 6.71 > t.05,4 = 2.132, the decision is to reject the null hypothesis. = .05 10.66 Ho: p1 - p2 = 0 Ha: p1 - p2 0 Machine 1 x1 = 38 n1 = 191 pˆ 1 p Machine 2 x2 = 21 n2 = 202 x1 38 = .199 n1 191 pˆ 2 x2 21 = .104 n2 202 n1 pˆ 1 n2 pˆ 2 (.199)(191) (.104)(202) = .15 n1 n2 191 202 For two-tail, /2 = .025 and the critical z values are: © 2010 John Wiley & Sons Canada, Ltd. z.025 = ±1.96 357 Chapter 10: Statistical Inferences About Two Populations z ( pˆ 1 pˆ 2 ) ( p1 p 2 ) 1 1 p q n1 n (.199 .104) (0) 1 1 (.15)(. 85) 191 202 = 2.64 Since the observed z = 2.64 > zc = 1.96, the decision is to reject the null hypothesis. 10.67 Construction n1 = 338 x1 = 297 pˆ 1 Telephone Repair n2 = 281 x2 = 192 x1 297 = .879 n1 338 pˆ 2 x2 192 = .683 n2 281 For a 90% Confidence Level, /2 = .05 and z.05 = 1.645 ( pˆ 1 pˆ 2 ) z pˆ 1qˆ1 pˆ 2 qˆ 2 n1 n2 (.879 - .683) + 1.645 (.879)(.121) (.683)(.317) = .196 ± .054 338 281 .142 < p1 - p2 < .250 10.68 Aerospace n1 = 33 x 1 = 12.4 1 = 2.9 Automobile n2 = 35 x 2 = 4.6 2 = 1.8 For a 99% Confidence Level, /2 = .005 and z.005 = 2.575 ( x1 x 2 ) z 12 n1 (12.4 – 4.6) + 2.575 22 n2 (2.9) 2 (1.8) 2 33 35 = 7.8 ± 1.52 6.28 < µ1 - µ2 < 9.32 © 2010 John Wiley & Sons Canada, Ltd. 358 Chapter 10: Statistical Inferences About Two Populations 10.69 Discount x 1 = $47.20 1 = $12.45 n1 = 60 Specialty x 2 = $27.40 2 = $9.82 n2 = 40 = .01 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 For two-tail test, /2 = .005 and zc = ±2.575 ( x 1 x 2 ) ( 1 2 ) z = 1 2 n1 2 2 (47.20 27.40) (0) (12.45) 2 (9.82) 2 60 40 n2 = 8.86 Since the observed z = 8.86 > zc = 2.575, the decision is to reject the null hypothesis. 10.70 Before 12 7 10 16 8 n=5 After 8 3 8 9 5 d = 4.0 sd = 1.8708 d 4 4 2 7 3 df = 5 - 1 = 4 = .01 Ho: D = 0 Ha: D > 0 For one-tail test, = .01 and the critical t.01,4 = 3.747 t = d D 4.0 0 = 4.78 sd 1.8708 n 5 Since the observed t = 4.78 > t.01,4 = 3.747, the decision is to reject the null hypothesis. © 2010 John Wiley & Sons Canada, Ltd. 359 Chapter 10: Statistical Inferences About Two Populations = .01 10.71 Ho: µ1 - µ2 = 0 Ha: µ1 - µ2 0 df = 10 + 6 - 2 = 14 A n1 = 10 x 1 = 18.3 s12 = 17.122 B___ n2 = 6 x 2 = 9.667 s22 = 7.467 For two-tail test, /2 = .005 and the critical t.005,14 = ±2.977 t = t = ( x1 x 2 ) ( 1 2 ) s1 (n1 1) s 2 (n2 1) 1 1 n1 n2 2 n1 n2 2 2 (18.3 9.667) (0) = (17.122)(9) (7.467)(5) 1 1 14 10 6 4.52 Since the observed t = 4.52 > t.005,14 = 2.977, the decision is to reject the null hypothesis. 10.72 A t test was used to test to determine if Hong Kong has significantly different rates than Mumbai. Let group 1 be Hong Kong. Ho: Ha: µ1 - µ2 = 0 µ1 - µ2 0 x 1 = 130.4 n1 = 19 n2 = 23 x 2 = 128.4 s12 = 166.41 s22 = 193.21 98% C.I. and /2 = .01 t = 0.48 with a p-value of .634 which is greater than = .02. There is not enough evidence in these data to declare that there is a difference in the average rental rates of the two cities. 10.73 H0: D = 0 Ha: D 0 This is a related measure before and after study. Fourteen people were involved in the study. Before the treatment, the sample mean was 4.357 and after the treatment, the mean was 5.214. The higher number after the treatment indicates that subjects were more likely to “blow the whistle” after having been through the treatment. The observed t value was –3.12 which was more extreme than twotailed table t value of + 2.16 and as a result, the researcher rejects the null © 2010 John Wiley & Sons Canada, Ltd. 360 Chapter 10: Statistical Inferences About Two Populations hypothesis. This is underscored by a p-value of .0081 which is less than = .05. The study concludes that there is a significantly higher likelihood of “blowing the whistle” after the treatment. 10.74 The point estimates from the sample data indicate that in the Alberta city the market share is .310782 and in the Ontario city the market share is .27013o. The point estimate for the difference in the two proportions of market share is .0406524. Since the 99% confidence interval ranges from -.0393623 to +.120667 and zero is in the interval, any hypothesis testing decision based on this interval would result in failure to reject the null hypothesis. Alpha is .01 with a two-tailed test. This is underscored by an observed z value of 1.31 which has an associated p-value of .191 which, of course, is not significant for any of the usual values of . 10.75 A test of differences of the variances of the populations of the two machines is being computed. The hypotheses are: H0: 12 = 22 Ha: 12 > 22 Twenty-six pipes were measured for sample one and twenty-eight pipes were measured for sample two. The observed F = 2.0575 is significant at = .05 for a one-tailed test since the associated p-value is .0348. The variance of pipe lengths for machine 1 is significantly greater than the variance of pipe lengths for machine 2. © 2010 John Wiley & Sons Canada, Ltd. 361