Question

Absenteeism is a major problem for some companies and in some industries. Suppose a study was conducted on absenteeism in the warehousing industry. Observations on several variables that might be related to absenteeism were collected on 35 major warehouses in the Pacific Northwest.

Absent: The average number of absences per employee for the year (does not include vacation days or confirmed sick days)

Wage: The average annual wage paid to the warehouse employees (does not include manager salaries)

Pct U: Union membership. The percentage of employees who belong to a union at the warehouse.

Good R: 1 if the employee group self-reported a “good” relationship with management; 0 if the employee group self-reported otherwise.

Perform a complete multiple regression analysis to find a model that might be useful for predicting the average number of absences per employee for the year. Perform ALL steps as outlined in class. Use Minitab and show all your work. Use alpha = .10 for any required tests (and show all steps for any required test). STAPLE MULTIPLE PAGES and include all required computer output.

Absent	Wage	Pct U	Good R
5.4	42000	57.1	1
4.1	39350	41.5	1
11.5	31000	52.6	0
2.1	28000	65.1	0
5.9	30000	68.8	1
12.9	28000	46.4	0
3.5	40000	38.9	1
2.6	35820	17.2	1
8.6	29500	12.9	0
2.7	29500	18.1	0
6.6	36500	64.4	1
2.1	39600	63.7	1
3.8	31200	12.2	1
4.3	32000	11.8	0
4.3	29600	25.8	0
2.2	37560	53.2	1
8.6	32000	22.8	0
10.8	22980	49.8	0
2.9	32000	39.1	0
5.3	42320	32.6	1
8.2	29500	67.7	0
2.8	36500	10.8	1
2.4	37970	25.5	1
2.8	35180	31.8	1
5	29630	35	0
9.5	39800	41.9	1
4.3	41000	52.9	1
8.9	32890	64.4	0
7.2	27500	69.7	0
5.6	27500	61.8	1
2.4	40826	52.1	1
2.7	31970	57.4	0
13.4	29990	15.2	0
14.8	31450	38.7	0
10.7	36900	69.4	1

Answer 1

Regression Analysis: Absent versus Wage, Pct U, Good R

Analysis of Variance

Source      DF   Adj SS Adj MS F-Value P-Value
Regression   3   91.837 30.612     2.61    0.069
Wage       1    2.264   2.264     0.19    0.664
Pct U      1    8.673   8.673     0.74    0.397
Good R     1   24.989 24.989     2.13    0.155
Error       31 363.750 11.734
Total       34 455.587

Model Summary

      S    R-sq R-sq(adj) R-sq(pred)
3.42547 20.16%     12.43%       3.16%

Coefficients

Term           Coef   SE Coef T-Value P-Value   VIF
Constant       8.82      5.55     1.59    0.122
Wage      -0.000077 0.000176    -0.44    0.664 2.25
Pct U        0.0260    0.0303     0.86    0.397 1.01
Good R        -2.55      1.75    -1.46    0.155 2.27

Regression Equation

Absent = 8.82 - 0.000077 Wage + 0.0260 Pct U - 2.55 Good R

Fits and Diagnostics for Unusual Observations

                            Std
Obs Absent   Fit Resid Resid
33   13.40 6.91   6.49   2.01 R
34   14.80 7.41   7.39   2.23 R

R Large residual

Regression Equation

Absent = 8.82 - 0.000077 Wage + 0.0260 Pct U - 2.55 Good R