(Page 9 of 24) - What Can Federal Employees Teach Us About Productivity Improvement? authored by Frank, Sue. and Lewis, Gregory.

What Can Federal Employees Teach Us About Productivity Improvement?

good job” (Question 43). Although the MPS listed 14 factors and allowed respondents
to list other motivators, six best capture monetary and public service motivation. For the
Extrinsic Motivators variable, we gave respondents one point each for listing monetary
awards, a good performance rating, and increased chances for promotion in their top
three. For the Intrinsic Motivators variable, we gave respondents one point each for
listing a personal desire to make a contribution, their duty as a public employee, and
personal pride or satisfaction in their work. Respondents rated the intrinsic motivators
much more highly, with 77% listing pride in work and 53% listing the desire to make a
contribution. Only 27% put monetary awards in their top three, and more listed duty
(23%) than promotions (18%). To the extent that these accurately reflect the priorities
of Federal civil servants, it suggests weaknesses in the Federal reward system are not
as problematic as once thought.

Five variables are included to evaluate the conditions of the work unit. To

account for the constraints placed by downsizing, we use questions on whether
respondent's work unit had been downsized in the past five years, whether downsizing
had affected the unit’s institutional memory, and whether the unit had sufficient
employees. This should allow us both to measure the overall impact of downsizing and
to determine whether it primarily has its effect through lower expertise or fewer
employees. We included two other questions which measure whether employees were
given more flexibility on the job and whether the work was meaningful.

For most of the remaining variables, the MPS 2000 asked several questions on

similar topics, allowing us to develop more reliable measures of our concepts than
would be possible with single questions. To develop our measures, we grouped the
questions that best captured our theoretical concepts, then used principal components
factor analysis to be sure all questions were measuring similar concepts. In no case do
we include a question in the combined measure unless it loads at least .60 on the
factor. Rather than using factor scores, we calculate the mean across all variables in
the factor, so the final measures all have theoretical ranges from 1 (strongly disagreeing
on all items) to 5 (strongly agreeing with all). Question wording and Cronbach’s alpha
are listed in Table 1. In only one case do we keep a factor with an alpha below .70, the
threshold suggested by Carmines and Stimson (1979). Four have alphas greater than
.85.

Supervisors play a key role in motivating a productive workforce. Both the

importance of dealing with poor performers, and the reputed weakness of Federal
supervisors on this aspect of their jobs, caused us to split that out as a separate
variable. Satisfaction with Supervisor combines six questions about the supervisor's
management and technical skills, regard for subordinates' welfare, feedback on job
performance, career development support, and overall competence. As noted in
previous studies, employees were most impressed with their supervisors’ technical skills
(mean=3.5), and overall satisfaction with their supervisors (mean=3.4) were notably
higher than employee ratings of their management skills, feedback, or career support
(means of 3.0-3.1).

Employees were even less happy with how their supervisors handled poor

performers. They gave supervisors mean scores of 2.5 to 3.1 on dealing effectively with