Section SixCASE 6-1 continued
consulting firm will have confidence in the statistical
justification for using its parameters. The adequacy of
explanatory power is indicated by the R^2 and the overall
model’s F values. (For the purposes of this case, assume that
the model is statistically adequate on all of these indicators.)
To predict the number of programmers that Core-Line
should be using, which is the value for Ŷ, the consulting firm
will plug into the regression model Core-Line’s X values for
annual sales, average annual hours worked by programmers,
average experience of programmers, rating of the quality of
programmer supervision, and rating of software complexity.
Using this model or equation, the consulting firm will then
compute the number of programmers (Ŷ value). This predicted
value or industry standard will then be compared with Core-
Line’s actual utilization of programmers in order to determine if
it is overstaffed. Further, for each independent variable having
a statistically significant relationship with the dependent
variable, the positive or negative sign of the regression
coefficient will provide direction for potential solutions to
utilization problems.