Practice Network: Competency Modeling: Viagraized Job Analysis Or
Impotent Imposter?
Michael Harris
University of Missouri–St. Louis
Greetings from Practice Network (PN) once again. As I stated in
my first column, my goal is to focus on a timely theme that will be of interest to
practitioners, as well as everyone else, including academics. This edition of PN considers
a term that has been the recent focus of considerable discussion: competency modeling. The
term "competency modeling" contains two words that sound sophisticated and
fashionable. The word "competency" calls to mind the recently popularized phrase
"core competencies," which is associated with organizational strategy. What
phrase do you associate with the word "modeling?" Many of you will answer
"causal" or "confirmatory," terms which sound highly analytical and
precise (or depending on your graduate school training, esoteric or frustrating!). Because
I had practically no exposure to competency modeling, I decided that this would be a good
topic for me to learn more and write about for PN. I therefore contacted some people who
seemed knowledgeable about competency modeling. I purposely kept the number of people that
I contacted down to enable me to pay for the long distance calls out of my "end of
fiscal year" phone account. The basic questions I asked the respondents were:
- What is competency modeling?
- How does competency modeling differ from traditional job analysis?
- How do you conduct competency modeling?
- What are the strengths and weaknesses of competency modeling?
A summary of these discussions follows, along with my own observations.
What is Competency Modeling?
Based on my conversations, competency modeling is the process of
determining what competencies are necessary for a job or job family. After talking with my
respondents, it occurred to me that I had not asked anyone to specifically define what a
competency was (though I was given several examples), so I went to my 1,500+ page (yes, I
still have a paper copy) dictionary and found that a competency involves the possession of
a "required skill, knowledge, qualification, or capacity." Based on the
dictionary definition and from what my respondents said, a competency is generally
quite similar to our traditional Knowledge, Skill, Ability, or Other requirement (KSAO).
In my conversations with respondents, however, there were two major
differences mentioned between our traditional KSAOs and a competency. First, competencies
may be defined more broadly than our traditional KSAOs. For example, materials I received
from the Office of Personnel Management included "self-esteem" as a competency.
Quite frankly, I doubt that self-esteem would be included as a KSAO, as the latter term is
used. Second, one respondent remarked that he viewed competencies as being similar to
"roles," much like I suppose Mintzberg used the term in developing his classic
typology of managerial roles. For example, a competency under this approach might be
"information conduit." What are the implications of this expanded notion of
competencies? I’m not sure.
In short, I believe that my respondents would agree that competency
modeling is a systematic process of determining the competencies needed to be successful
in an organization. Beyond that general definition, I think there are some major
differences, as well as some common overlap, in how my respondents understand competency
modeling.
How Does Competency Modeling Differ From
Traditional Job Analysis?
After one or two initial conversations with respondents, I began to
wonder whether the saying I heard in graduate school, "old wine in new barrels,"
was going to be an apt description. This was followed by a fear that I wouldn’t have
enough material to fill my column for this issue, so I decided to dig deeper and hope that
I found more fecund, as opposed to feculent, material to work with. After contacting more
people, I found some diversity of opinion regarding the difference between competency
modeling and traditional job analysis. First, let me explain very briefly that by
traditional job analysis I mean a process in which both tasks and KSAOs are compiled and
quantitatively rated by subject matter experts (SMEs).
With this definition in mind, I found two basic opinions as to the
difference between traditional job analysis and competency modeling. One position was that
competency modeling is no different than a traditional job analysis. Respondents in this
camp suggested that the only real difference was that competencies were defined somewhat
more broadly than traditional KSAOs. When asked what advantage competency modeling had
then over the traditional job analysis, respondents had several different reactions. One
respondent, who works in the public sector, indicated that the selection process that had
been used in the past was very task specific. Using a task specific approach was
far less appropriate in the 1990s because jobs were changing so much. Thus, competency
modeling enabled him to focus on testing for more general skills and abilities. Another
respondent, employed as a consultant, argued that the primary advantage of competency
modeling was the term. In other words, clients were more impressed by the term
"competency modeling" than the term "job analysis." Similarly, the
term "competency" seemed more marketable than the awkward sounding expression
"Knowledge, Skills, Abilities, and Other requirements." Parenthetically, I must
admit that I have always felt that "job analysis" sounded cold and sterile, as
do the words "Knowledge, Skills, and Abilities." And, does it matter what we
call this process, as long as we are doing it properly and not pretending it is something
that it is not?
The other camp indicated that competency modeling was different than a
traditional job analysis. These respondents described a process that differed from
traditional job analysis in a number of potentially important ways. Generalizing across
several different opinions, I would summarize this process of conducting competency
modeling as follows. First, the organization’s and department’s mission and
objectives are discussed with SMEs. Second, based on this information, the competencies
that are required to successfully meet the mission and objectives are described. Third,
the general competencies are defined by SMEs in specific, behavioral terms. As one
respondent stated, this definition of competency modeling is similar to a strategic job
analysis. Gatewood and Feild, in their textbook "Human Resource Selection,"
indicated on p. 313 that a major step in strategic job analysis, as compared to a
traditional job analysis, involves SMEs discussing how various issues (e.g., new
technology) will change the job at issue. Information about future tasks and KSAOs is then
based on these anticipated changes.
Thus, one major difference between this approach to competency modeling
and traditional job analysis is that the former approach explicitly links skills and
abilities to the bigger picture. One thing that quickly came to mind as I thought about
this is that it reminds me of a training needs analysis, in which the first step is an
"organizational analysis," during which the mission and objectives are
considered. I think that this is particularly helpful given the present emphasis in both
the for-profit and nonprofit world on articulating and adhering to a mission. When I
pointed this out to one respondent, she did agree with me (of course, she may have been
saying that to get me off the phone!). Stated somewhat differently, perhaps the closer
linkage of competency modeling to the organizational mission and objectives will provide
I-O psychologists with greater credibility.
A second difference emphasized by several respondents was that
competency modeling aimed for a wider application. Specifically, competency modeling aims
at identifying skills and abilities that can be used for a variety of purposes and for a
variety of jobs in the same organization. For example, using competency modeling, one may
be able to create a list of competencies that might be used for selection, performance
appraisal, compensation, and training purposes. By way of comparison, traditional job
analysis was perceived as leading to a narrower, more specific set of skills and abilities
that tend to be useful for only one or two specific purposes (e.g., selection) and just a
small group of jobs. From what I can tell, this may be a potential strength of competency
modeling over traditional job analysis. The need for more general, rather than specific,
job analyses, seems to mesh with validity generalization findings as well.
A third difference compared to traditional job analysis indicated by
several respondents, across the two camps, was that competency modeling does not require
quantitative ratings by SMEs. According to one respondent, this was an advantage for
certain job groups, such as executives, where the SMEs would be unwilling to complete
lengthy questionnaires. Likewise, not having a requirement for ratings was seen as an
advantage in terms of time and expense. I point out, however, that several of my
respondents did collect quantitative ratings for the purposes of competency modeling. It
would be unfair, then, to characterize competency modeling as being purely qualitative,
just as I believe it would be inappropriate to characterize job analysis as being
nonquantitative just because some methods do not use ratings.
My personal response to this third difference is to be somewhat
skeptical. While I can see the advantages of saving time and costs, it seems to me that
some degree of accuracy may be lost. Minimally, I view the rating process as a
double-check on the appropriateness of the KSAOs that have been obtained. While I am not
convinced that different conclusions will be drawn if ratings are used, I do think it is
an important part of this process. My final concern is a legal one; I believe that it
would be much more difficult to defend a selection process that is not based on
quantitative ratings, particularly if the organization is relying on a content validity
strategy. My conversations with the respondents indicated that no one was aware of any
discrimination lawsuits involving competency modeling.
Finally, a major distinction between both camps’ description of
competency modeling and traditional job analysis was that tasks were not usually discussed
or they were generated in the third step, when the competencies were defined. One
respondent indicated that this was due to time and cost constraints. Another respondent
indicated that, in essence, the skills and abilities were generated first and the tasks
were generated afterwards. Do you think that the failure to explicitly consider the tasks
first is a problem? I’m not sure; an empirical study would be interesting here.
So What’s the Next Step?
One of my last calls was to Jeff Schippmann, who, I discovered, is
chairing a SIOP committee on competency modeling. According to Jeff, this task force was
formed in light of the explosion of interest in competency modeling. The purpose of the
task force is to define what competency modeling is and to identify good practices
associated with it. In addition to doing a massive literature search, the committee has
conducted in-depth interviews with experts in the field of job analysis and competency
modeling in order to get an better understanding of what it is and how it is conducted. If
you are interested in getting much more information in much greater detail about
competency modeling, wait to hear from Jeff’s committee!
To wrap things up, I think all of my respondents would define
competency modeling more or less the same way, but they would go about the process
somewhat differently. However, as one respondent indicated, there is no one "right
way" for conducting competency modeling. This bothered me at first, but I realized
that the same can be said for many other I-O procedures, such as job evaluation and even
job analysis itself. Well, I’m out of time now. To conclude, as to whether competency
modeling is viagraized job analysis or an impotent imposter, that depends on how you use
it.
Reactions to this column? Have you had a different experience with
competency modeling? Topics you would like to see covered in the future? Please e-mail at c1994@umslvma.umsl.edu; phone (314) 516-6280,
fax; (314) 516-6420; or snailmail me, Michael Harris, School of Business Administration,
University of Missouri–St. Louis, St. Louis, MO 63121. I look forward to hearing from
you!
I thank the following individuals who were helpful in providing
information to me: Bob Jones, Southwest Missouri State University; Allan Patterson,
Bryan and Associates; David Pollack, US Immigration and Naturalization; Leslie
Pollack, US Office of Personnel Management; Jeff Schip-pmann, PDI; Michael
Stevens, Psychological Associates; Suzanne Tsacoumis, HumRRO; Deborah
Whetzel, US Postal Service.
TIP
Vol. 36/No. 2 October, 1998
Return to Table of Contents