Section SixCASE 6-1 continued
constitute a separate observation in the analysis. Thus, in the
regression analysis, the dependent variable (Y) is the number
of programmers. The independent variables are the dollar
volume of annual software sales (X 1 ), average number of hours
worked each year by programmers including overtime (X 2 ),
average years of experience of programmers (X 3 ), a rating of
the quality of programming supervision (X 4 ), and a rating of the
complexity of the software produced by each company (X 5 ).
The rating of the quality of supervision is a five-interval scale
ranging from 1 (poor) to 5 (good). Similarly, the software
complexity rating scheme is a five-interval scale ranging from 1
(simple) to 5 (complex). The regression model is Y = α + β 1 X 1
- β 2 X 2 +β 3 X 3 +β 4 X 4 + β 5 X 5 + ε, where α is the intercept or
constant; β 1 , β 2 , β 3 , β 4 , and β 5 are the respective regression
coefficients derived in the model; and ε is the error term.
In order to derive the values for α, β 1 , β 2 , β 3 , β 4 , and β 5 ,
the consulting firm will run the regression analysis, with each
observation being composed of matched data from one of the
65 different software companies. (Under most conditions, the
value for ε is assumed to be 0.) If the model is adequately
specified, statistically significant, and a substantial amount of
variance in the dependent variable is accounted for, the