Speech recognition technology has advanced to the point that it is now possible to communicate with a computer through verbal commands. A study was conducted to evaluate the value of speech recognition in human interactions with computer systems (Special Interest Group on Computer-Human Interaction Bulletin, July 1993). A sample of 45 subjects was randomly divided into three groups (15 subjects per group), and each subject was asked to perform tasks on a basic voice mail system. A different interface was employed in each group: (1) touch-tone, (2) human operator, or (3) simulated speech recognition. One of the variables measured was overall time (in seconds) to perform the assigned tasks. An analysis was conducted to compare the mean overall performance times of the three groups.

a. Identify the experimental design employed in this study.

b. Propose a regression model that will allow you to compare the three means.

c. In terms of means, give the appropriate null hypothesis to be tested.

d. In terms of the β’s of the model, part b, give the appropriate null hypothesis to be tested.

e. The sample mean performance times for the three groups are given below. Despite differences among the sample means, the null hypothesis of part c could not be rejected at α = .05. Explain how this is possible.

Transcribed Image Text:

Group Mean Performance Time (seconds) Touch-tone 1,400 Human operator 1,030 Speech recognition 1,040