1. A textile company is interested in knowing if there is a difference in the breaking strength of four different kinds of thread. A test was conducted where five measurements (in ounces) were taken for each kind of thread. Dummy variables were created named threada, threadb, and threadc, which indicated if a strength reading was obtained on thread a, b, or c, respectively. A regression analysis was performed with breaking strength as the dependent variable and threada, threadb, and threadc as independent variables, the results are attached at the end of the sheet. a) Write the linear model for the regression analysis, give the F statistic and P value for the test of equality of breaking strength means for the four threads, and state your conclusion for the test using alpha = .05. b) Use the printout from the regression analysis to give estimates of the mean breaking strength for each of the four thread groups. c) On the regression analysis printout on the row with information about threadb is a t statistic ( t = 0.97, P = 0.3477). Express the null hypothesis that is being tested by this t statistic in terms of the population means i . If we have a single group (such as a control group) that we wish to test against each other group, how can we arrange it so that the regression analysis shows these pairwise tests?
1. A textile company is interested in knowing if there is a difference in the breaking strength of
four different kinds of thread. A test was conducted where five measurements (in ounces) were
taken for each kind of thread. Dummy variables were created named threada, threadb, and
threadc, which indicated if a strength reading was obtained on thread a, b, or c, respectively. A
threadb, and threadc as independent variables, the results are attached at the end of the sheet.
a) Write the linear model for the regression analysis, give the F statistic and P value
for the test of equality of breaking strength means for the four threads, and state your conclusion
for the test using alpha = .05.
b) Use the printout from the regression analysis to give estimates of the
strength for each of the four thread groups.
c) On the regression analysis printout on the row with information about threadb is a t
statistic ( t = 0.97, P = 0.3477). Express the null hypothesis that is being tested by this t statistic in
terms of the population means i . If we have a single group (such as a control group) that we
wish to test against each other group, how can we arrange it so that the regression analysis shows
these pairwise tests?
Trending now
This is a popular solution!
Step by step
Solved in 5 steps with 22 images
