In: Statistics and Probability
A technician compares repair costs for two types of microwave ovens (type I and type II). He believes that the repair cost for type I ovens is greater than the repair cost for type II ovens. A sample of 56 type I ovens has a mean repair cost of $76.66, with a standard deviation of $18.63. A sample of 75 type II ovens has a mean repair cost of $72.66, with a standard deviation of $22.09. Conduct a hypothesis test of the technician's claim at the 0.1 level of significance. Let μ1 be the true mean repair cost for type I ovens and μ2 be the true mean repair cost for type II ovens.
Step 1: State the null and alternative hypotheses for the test.
Step 2: Compute the value of the test statistic. Round your answer to two decimal places.
Step 3: Determine the decision rule for rejecting the null hypothesis H0. Round the numerical portion of your answer to two decimal places.
Step 4:Make the decision for the hypothesis test. (Reject null hypothesis or fail to reject null hypothesis)
As given, let the true means for repair costs for Type 1 and Type 2 ovens be μ1 and μ2 respectively. Here, we carry out 2-sample unpaired T-test as the 2 samples are not paired.
1) The null and alternative hypotheses are:
H0: μ1 - μ2 <= 0
Ha: μ1 - μ2 > 0
2) Value of the test statistic is given by:
T = [ () ] / sqrt[(s1^2/n1) + (s2^2/n2)], where and are the sample means, s1 and s2 are the sample std deviations, and n1 and n2 are the sample sizes.
Plugging in the given values, we get
T = 1.22
3) The critical T-score for the right-tailed T-test at significance level of 0.1 and d.o.f = 56+75-2 = 129 is given by,
T-critical = T0.1, 129 = 1.29
Decision rule: We reject the null hypothesis if the T-statistic is greater than the critical T-score at the given significance level.
4) Decision: Since T = 1.22 < 1.29, we fail to reject the null hypothesis at the significance level of 0.1