How To (and How Not To) Assess the Integrity of Managers (19 pages)

Document rating:
  5.0 out of 5
Vote this doc:
background image

approach was the best predictor of leadership in the second study; the competency approach was the
poorest predictive measure in the first study. Nonetheless, it is possible that our results would have
been different had we used subordinate job attitudes and overall ratings of their bosses in the first
study.

Perhaps the biggest difficulty in comparing the two methods concerns how the measures of

integrity were computed and intended to be used. In the competency study, we used the average
rating across several subordinates. This aggregation reduced the variability and increased the
reliability in the integrity scores relative to the dubious reputation study—which used ratings from
only one subordinate per manager. This may have accentuated the differences between the two
samples in the proportion of managers who scored low on integrity. It is unlikely that this accounts
for all of the differences, but it is impossible to quantify the effect with our current dataset. Finally,
in the competency study, raters knew their ratings would be provided to their managers as
developmental feedback. Although raters were assured of anonymity and the organization had a
good track record for maintaining confidentiality, it is possible that they felt less comfortable
providing low ratings of integrity. In the dubious reputation study, raters knew that their bosses
would not see their ratings. It is well documented that raters give more favorable ratings when they
know they will be accountable for them (Curtis, Harvey, & Ravden, 2005), and this may have
affected our results to some degree.

Despite these limitations, the results presented here suggest that the dubious reputation approach

offers some advantages over the competency approach for identifying managers with potential
integrity problems. We regard the present effort as a proof of concept for the dubious reputation
approach. Further research is needed in applied settings under typical conditions to confirm the
results that we observed in study two. In particular, we need to know if subordinates will provide
low ratings of their bosses’ integrity when they know the survey results will be presented in some
form to those bosses. Further, longitudinal research is needed that compares how well the two
methods identify managers who eventually get into ethical trouble.

Implications

Our results have implications for human resource and talent management professionals and con-
sulting psychologists. First, they raise doubts about the ability of typical competency ratings to
identify managers who may lack integrity. We suggest that organizations using competency ratings
to conduct “integrity audits” consider alternative approaches. We also recommend that coaches and
consultants who focus on issues of character and integrity in development rely on more than just
competency ratings of integrity.

One possibility is to replace competency ratings with ratings based on the dubious reputation

method. For instance, the PLIS scale is in the public domain and available for free at www
.sbcraig.com/plis. Another option would be to construct similar scales based on the dubious
reputation method by focusing on subordinates’ expectations that their boss would behave uneth-
ically if given the chance. A third option would be to create hybrid scales that combine the positive
items of competency scales with more pointed items from the dubious reputation perspective.

3

Of course, asking subordinates to speculate on the likelihood that their manager would behave

unethically is a process that needs to be implemented with great care. Key issues include anonymity
and confidentiality as well as safety mechanisms that encourage candid ratings by protecting
subordinates from retribution, a clearly communicated policy on how the data will be used, and
visible support from senior leadership and their commitment to a culture of integrity. Finally,
collecting such information should be accompanied by a commitment to take action when results
suggest that there may be a problem with an unethical manager. And this is where things get
complicated. Obviously, it is ill-advised to remove a manager because subordinates believe that the
manager might behave unethically. We suggest using this kind of data as a red flag—an indication

3

We thank an anonymous reviewer for this suggestion.

230

KAISER AND HOGAN

This

document

is

copyrighted

by

the

American

Psychological

Association

or

one

of

its

allied

publishers.

This

article

is

intended

solely

for

the

personal

use

of

the

individual

user

and

is

not

to

be

disseminated

broadly.

Document rating:
  5.0 out of 5
Vote this doc: