In Chapters 11 and 14, we visited Circuit Systems, Inc., a company that was concerned about the effectiveness of its new program for reducing the cost of absenteeism among hourly workers. Besides allowing employees to be paid for unused sick leave, the program also pays $200 toward membership in a local health club for those employees who wish to join. Thomas Nelson, who introduced the new program, has collected data describing the 233 employees who have been with the company for both the year preceding the new program and the year following its implementation. The information Mr. Nelson has collected is in the CIRCUIT data file, and the variables and description of his program are as described in Circuit Systems, Inc. (A), in Chapter 11.
1. Use a multiple regression model to estimate the number of days of sick leave this year as a function of two variables: days of sick leave taken last year and whether the employee is a participant in the exercise program. Interpret the partial regression coefficients in the equation. Specifically, comment on the sign and numerical value of the partial regression coefficient for the dummy variable representing participation in the exercise program. On this basis, does the exercise program seem worthy of continuation?
2. What percentage of the variation in days of sick leave this year is explained by the regression equation you generated for question 1? To the extent that there is unexplained variation in the number of sick days taken this year, what other variables —including possible variables not in the database —might influence the number of days of sick leave taken this year?