Download (introduced in Chapter 19) Before purchasing videoconferencing equipment, a company ran tests of its current internal computer network. The goal of the tests was to measure how rapidly data moved through the network given the current demand on the network. Eighty files ranging in size from 20 to 100 megabytes (MB) were transmitted over the network at various times of day, and the time to send the files (in seconds) recorded. Two types of software were used to transfer the files, identified by the column labeled Vendor in the data table. The two possible values are “MS” and “NP”; use a dummy variable coded as 1 when Vendor = “MS.”
(a) Would it be appropriate for management to compare the two vendors based on a two- sample comparison of the times needed to transfer the files, or would such a comparison be confounded by different sizes of the files that were sent?
(b) Perform the two-sample t-test to compare the performance of the software provided by the two vendors. Summarize this analysis, assuming that there are no lurking variables.
(c) Compare the download times produced by the two vendors using an analysis of covariance. Summarize the comparison of download times based on this analysis. Use a dummy variable coded as + for Vendor ‘MS’ and 0 otherwise. (Assume for the moment that the model meets the conditions for the MRM.)
(d) Compare the results from parts (b) and (c). Do they agree? Explain why they agree or differ. You should take into account the precision of the estimates and your answer to part (a).
(e) Does the estimated multiple regression used in the analysis of covariance meet the similar variances condition?