chapter 1

Chapter 9: Inferences Based on Two Samples35. There are two changes that must be made to the procedure we currently use. First, theequation used to compute the value of the t test statistic is:t( x − y ) − ( )= where s p issp∆1 1 +m ndefined as in Exercise 34 above. Second, the degrees of freedom = m + n – 2. Assumingequal variances in the situation from Exercise 33, we calculate s p as follows:sp⎛ 7 ⎞ ⎛ 9 ⎞= ⎜ ⎟2⎝16⎠⎝16⎠( 32.8 − 40.5) − ( −5)== −2.24≈ −2.21 12.544 +8 102( 2.6) + ⎜ ⎟( 2.5) = 2. 544. The value of the test statistic is, then,t . The degrees of freedom = 16, and the p-value is P ( t < -2.2) = .021. Since .021 > .01, we fail to reject H o . This is the sameconclusion reached in Exercise 33.Section 9.3d , s =11. 862836. = 7. 25D1 Parameter of Interest: µD= true average difference of breaking load for fabric inunabraded or abraded condition.2 H : µ 00 D=3 H : µ > 04at =dsDD− µ/Dn=sdD− 05 RR: t ≥ t .= 2. 99801,77.25 − 06 t == 1. 7311.8628/ 8/n7 Fail to reject H o . The data does not indicate a difference in breaking load for the twofabric load conditions.274