International Journal of Selection and Assessment Volume 21 Number 3 September 2013

What Do We Know About

Competency Modeling?
Thomas H. Stone*, Brian D. Webster** and
Stephen Schoonover***
*Department of Management, Oklahoma State University, Stillwater, OK, USA. tom.stone@okstate.edu
**Department of Management, Oklahoma State University, Stillwater, OK, USA
***Schoonover Associates, Keswick, VA, USA

While use of competency modeling (CM) has grown considerably, evaluation of it has been
sparse. ‘Although today there is a significant amount and range of practice, there is sadly,
little research or formal description of practice . . .’ (Schippmann, J. S. (2010). In J. Scott &
D. Reynolds (Eds.), Handbook of Workplace Assessment (p. 215). San Francisco, CA: John
Wiley). CM literature is reviewed with the goal of shedding light on the prevalence, uses
and effectiveness of CM. CM adoption, use, and effectiveness were examined via surveys
and expert interviews. Results indicate extensive use of CM in corporate talent manage-
ment systems. Our survey and review highlight the need for further examination of CM, its
uses and implementation challenges.

1. Introduction methodology to develop a competency model (Campion

et al., 2011; Sanchez & Levine, 2009). Although no unify-

A t the onset of the 21st century, industrial organ-

izational (I/O) psychologists began to question the
methodological rigor of competency modeling (CM),
ing formal definition exists among scholars, most defini-
tions state or imply that ‘competencies’ are behaviors,
skills, and/or knowledge that align with organizational
specifically compared to traditional job analysis (TJA). strategies and are necessary for successful performance
‘By the mid-1990’s, practice and application had out- (Campion et al., 2011).
stripped research and reporting by such a huge margin Adding to the uncertainty regarding a formal defini-
that nobody had a clear sense of what was going on’ tion and methodology to develop a competency model
(Schippmann, 2010, p. 204). This led to a task force con- is the fact that CMs can be utilized as stand-alone
sisting of members of the Professional Practice Commit- models or can form the basis of comprehensive Talent
tee and Scientific Affairs Committee of the Society Management (TM) systems. TM refers to a comprehens-
of Industrial Organizational Psychology (SIOP). This ive method to hire talented individuals and ensure
group’s work resulted in a set of standards for CM prac- proper performance throughout their tenure with
tice and a 10-dimension level of rigor scale (Schippmann the company. Indeed, TM systems are likely the most
et al., 2000). Following interviews with 37 academic and common use of competencies. Indirect evidence re-
practitioner subject matter experts, the task force con- garding the extent to which competency models
cluded TJA is more rigorous in terms of components are used in organizations comes from two competency
such as choosing task descriptor content, determining modeling subject matter experts, Jeff Schippmann
reliability of judgments, and detail of content. The major and John Scott, and an internet search of TM com-
CM strength was its linkage to business goals and long- panies. J.S. Schippmann (personal communications,
term strategies. Additionally, CM serves as a broad tool February 10, 2012) estimated that 70–80% of Fortune
for organizational development by aligning human re- 500 companies use some form of TM system and
source (HR) systems so that employees are hired, virtually all such systems rely on competencies for de-
trained, evaluated, compensated, and promoted based fining and measuring work and performance. In a re-
on the same attributes (Campion et al., 2011). cent interview John Scott, Chief Executive Officer
Despite the reported growth of CM, controversy re- (CEO) of APTmetrics, an HR management company,
mains regarding the definition of competency and the stated:

© 2013 John Wiley & Sons Ltd,

bs_bs_banner

9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main St., Malden, MA, 02148, USA
Competency Modeling 335

‘I would venture to guess that most – if not all – of the ethics and integrity. While the Mercer study found
Fortune 100 use competency models as the foundation for competency models covered 80% of managerial and
their talent management systems (succession planning, Hi supervisory employees, and 50% of professionals, the
Po identification, selection, performance.’ ( J. C. Scott, per- Schoonover survey found less with just over one-
sonal communication, March 30, 2012).
third of professionals, middle managers and first-line su-
pervisors being covered by CM and only a quarter of
A recent review of clients of 10 prominent TM firms managers.
echoed the CEO’s comments (‘PeopleFluent,’ 2012). For
example, competencies are the core of PeopleFluent’s
TM systems, used by nearly 80% of Fortune 100 com- 2.2. Reasons for adopting competencies
panies. However, little is known about how organ-
izations use CM and its effectiveness. The two surveys discovered organizations adopt CM for
similar reasons (listed in decreasing order of frequency):
to enhance performance, to integrate HR processes, to
2. Previous user data of competency align behavior with corporate values, selection, develop-
modeling ment, and career pathing.

Schippmann (2010) recently recognized a significant

problem in the CM literature stating, ‘Although today 2.3. Effectiveness of competency modeling
there is a significant amount and range of practice, there
The Mercer and Schoonover et al. (2000) surveys ad-
is sadly, little research or formal description of practice,
dressed effectiveness of competencies for various HR
. . .’ (p. 215). One consequence of this shortcoming is
applications. In the Mercer survey, Rathbar-Daniels
statements regarding CM practice that are not sup-
et al. (2001) state, ‘The great majority reported a lack
ported by user survey data. For example, several
of formal measurement of the business impact of
sources contend competency models are used actively
competency-based HR applications’ (p. 73). Several
to align HR systems (e.g., Green, 1999; Lawler, 1994;
quotes suggest satisfaction with CM to strengthen HR
Lucia & Lepsinger, 1999), but offer no supporting data.
processes, ‘sharing a common purpose’ and facilitating
In addition to the dearth of knowledge regarding CM
the same competency language regarding performance.
users, available data are over a decade old. Indeed, two
Additionally, units with high competency ratings were
surveys, the Mercer survey (Rathbar-Daniels, Erickson,
more productive and competency-based selection led to
& Dalik, 2001) and an unpublished survey (Schoonover,
lower turnover and higher performance. Similarly,
Schoonover, Nemerov, & Ehly, 2000), are among the
Schoonover et al. (2000) found only half of those imple-
more recent surveys presenting actual user data of CM
menting CM conducted any measurement of effect-
use and effectiveness.
iveness and little consistency of practice. Approaches to
Both of these surveys attempted to better understand
measurement of effectiveness include 32% used per-
why competencies are used, their effectiveness, and the
formance ratings, 31% used employee satisfaction, 29%
barriers to effective CM implementation. Of the 130 HR
used competency ratings, 27% used management and
executives of large firms interviewed in the Mercer sur-
employee perceptions, 26% used turnover and 15%
vey, (Rathbar-Daniels et al., 2001) 75% used CM for 2
used return on investment (ROI).
years and 25% were not using CM or used it for less
The Schoonover et al. (2000) study assessed the
than 2 years. In the Schoonover et al. (2000) study con-
effectiveness of nine competency goals and nine HR
ducted with four consulting firms, approximately one-
applications using a five-point scale of 1 = very ineffect-
third of the 300 internet responders actively used CM
ive to 5 = very effective. Aligning behavior with core
and were similar in size and industries to the Mercer
values was rated highest, 3.52, followed by enhancing
sample. This survey categorized CM user’s expertise re-
performance expectations, 3.32, communicating generic
sulting in one-third ‘sophisticated’, that is, multiple CM
leadership skills, 3.26 and developing functional areas
applications over a period of years. A brief summary of
lowest, 2.88. Note that none were rated as effective.
the two surveys is described.
The most effective (percent rating ‘very effective’ or ‘ef-
fective’) CM application of an HR function was selection,
53%, consistent with the Mercer findings. Ratings of
2.1. Coverage of CM
other HR applications were performance management
Competencies were used by just over half of employees 50%, training 42%, job descriptions 38%, development
among competency users in the Mercer and Schoonover planning 36%, career pathing 31%, succession planning
samples. In the Mercer sample, 23% use functional com- 30%, strategic HR planning 29% and compensation 19%.
petencies such as finance and marketing and 21% use As with ratings of effectiveness of competency focus,
both functional and organizational competencies such as the Schoonover study reported a significant relationship

© 2013 John Wiley & Sons Ltd International Journal of Selection and Assessment
Volume 21 Number 3 September 2013
336 Thomas H. Stone, Brian D. Webster and Stephen Schoonover

between level of sophistication and performance man- 3. 2011 competency modeling survey
agement, training, and selection. For all HR applications,
sophisticated users’ effectiveness ratings were highest Our survey is web-based with some phone interviews.
while beginner and experienced users were lower and We used a convenience, snowball sampling method con-
relatively similar to one another. tacting experienced U.S. and Canadian HR and I/O
Overall, 30% of competency users in the Schoonover psychology professionals acquainted with the first au-
survey were satisfied or very satisfied with their com- thor. Of the 48 respondents, 69% use CM and half are
petency applications and their effectiveness. However, ‘sophisticated,’ meaning competencies are used for
Schoonover et al. (2000) state, ‘one of the most inter- three or more HR applications and have been used for 5
esting and potentially disturbing findings of the study’ (p. or more years. Responses were only included in our
14) was that 37% of those using CM were dissatisfied or analyses if the organization uses competency modeling.
very dissatisfied with their competency applications. As a result, our sample consisted of 33 usable re-
They suggest the level of expertise may explain this dif- sponses. For comparative purposes, the results of this
ference as effectiveness of CM was ‘very significantly survey are outlined in a similar manner to the Mercer
correlated with level of user expertise.’ And they note, and Schoonover surveys. Findings of the two previous
‘The average effectiveness gap between beginner and surveys, as well as our 2011 survey are summarized in
experienced user was .2, while the average gap between Table 1.
experienced and sophisticated users was .7’ (p. 9). Our survey examined questions similar to the Mercer
Recall that only approximately one-third of the sample and Schoonover surveys in an attempt to gain clearer in-
was very experienced or sophisticated CM users. sight as to how CM use and effectiveness has changed
Schoonover et al. (2000) conclude these findings suggest over the past decade. Those surveyed provided self-
that the ‘payoff’ for use of competencies requires per- report answers to questions related to issues such as
sistence over time. how long have competencies been used, for which HR
functions are competencies used, were consulting firms
used for CM implementation, professional HR certifica-
2.4. What are the barriers to success of CM? tions and affiliations, organization size, and effectiveness
The surveys demonstrate several common themes in re- of the CM system. Additionally, our survey contained a
gards to barriers of success of CM. Notably, lack of top qualitative component in that we analyzed open-ended
management ‘buy-in’ and CM perceived as an ‘HR thing’ responses to questions such as why competencies are
requiring too many financial and human resources. used and what are barriers to effective implementation
The two surveys are dated and are relatively small of competencies? The use of CM versus TJA was also
samples. However, our recently conducted survey, in examined to better understand their relationship.
conjunction with the Mercer and unpublished Schoon-
over surveys, represent a small step toward providing
3.1. Coverage of CM
data necessary for a ‘formal description of practice’
(Schippmann, 1999) and challenges to successful CM Although the Mercer and Schoonover surveys examined
implementation. Perhaps more importantly, the three use of CM, the 2011 survey asked a related but different
surveys, taken together, highlight the need for further question, ‘Are jobs for which CM is conducted different
examination and analysis of how CMs are currently from those conducted using TJA?’ Half of the CM users
utilized in organizations and their effectiveness. said there was no difference while others identified

Table 1. Summary of surveys

Items Mercer (1999–2000) Schoonover (2000) CM Survey (2011)

Number of 75% used CM for at least 33% used CM. 65% said likely to 69% used CM
respondents 2 years implement
utilizing CMs
How is CM applied to 56% covers all employees in the 51% covers all employees in the N/A
individuals? organization organization
Reasons for adoption Integrate HR processes, align Enhance performance, integrate HR Raise performance, align HR
organizational goals, reinforce process, align behavior with practices with organizational
behaviors, facilitate change organizational goals strategy
Measure of Satisfied with CM but no formal 50% measured effectiveness. No Mean rating (0–100) 75.2 (16.1
effectiveness? way to assess effectiveness consistent method of measurement SD)
Barriers to Top management support, Top management support, lack of Competencies not fully
implementation lack of resources resources explained and understood by
employees
CM = competency modeling; HR = human resource.

International Journal of Selection and Assessment © 2013 John Wiley & Sons Ltd
Volume 21 Number 3 September 2013
Competency Modeling 337

Table 2. Correlations among variables of interest

M SD 1. 2. 3. 4. 5. 6. 7.

1. Years 7.1 7.6 1.0

2. Consulting .36 .49 -.02 1.0
3. Org Size 10,097 20,973 -.08 .51* 1.0
4. SIOP .35 .49 .52* .06 .08 1.0
5. PHR .05 .22 .13 .40* .57* .31 1.0
6. SPHR .15 .37 -.15 .40* -.10 -.31 -.10 1.0
7. Effectiveness 75.2 16.1 .45* -.20 .11 .40 .14 -.69* 1.0
N = 33. *Correlation is significant at p<.05 level (1-tailed). Years = Years CM has been implemented; Consulting = Consulting Firm Used; Org
Size = Organizational Size; SIOP = Society for Industrial and Organizational Psychology Member; PHR = Professional in Human Resources Member;
SPHR = Senior Professional in Human Resources Member.

differences based on job level and function with manage- surveys conducted over a decade ago, the 2011 survey
rial and professional jobs using CM and lower level jobs discovered top management buy in was consistently
requiring more physical requirements using TJA. This listed as the biggest barrier to success of CM. As a
finding is consistent with the SIOP committee task force follow-up to our survey, several major CM vendors, De-
finding that TJA is generally more task descriptive while velopment Dimensions International (DDI), Personnel
CM is more closely tied to organizational goals. One Decisions Research Institute (PDRI), APTMetrics, Hogan
comment that illustrates different focuses toward CM Assessments and Lominger, were asked their views of
and TJA is ‘The jobs are not different but the purpose major CM issues and challenges. The following repres-
dictates whether a competency modeling or more tradi- ent the major issues reported.
tional job analysis is appropriate.’ The online survey
asked about the relationship of CM to TJA. Of the 33 3.4. Implementation
respondents using CM, 39% said the two had merged,
CM supplemented TJA, 25%, 9% stated CM replaced There was a high degree of agreement that some organ-
TJA, 9% said CM and TJA were separate and 18% were izations develop competency models but do not know
categorized as ‘other.’ how to implement them. Two representative responses
are, ‘A common request we get is how to roll out the
competency models once they are built’ and ‘It really
3.2. Effectiveness of competency modeling comes down to how organizations implement the
Respondents were asked how effective their organiza- models.’ Another observed, ‘Implementation is less of
tion’s competency modeling system has been by answer- an issue with selection than performance management
ing on a scale from 0 to 100. Online survey and development as they affect jobs and careers more.’
respondents’ mean rating of overall effectiveness of
their CM effort was 75.2 (0–100 scale) (16.1 SD). 3.5. Lack of integration with other HR/talent
Table 2 presents the correlations among variables of in- management systems
terest. Consistent with Schoonover et al. (2000), we
Although Campion et al. (2011) contend integration of
found a positive and statistically significant relationship
HR systems is a virtue of CM, respondents noted chal-
between CM user-rated effectiveness and years the
lenges regarding linking different applications especially
competency modeling system has been implemented
when different models are developed for different HR
(r = .45; p < .05). One somewhat surprising finding from
functions.
the 2011 survey is that less than 30% of CM users em-
ployed a consulting firm when initiating the CM process
and those who did were large firms. Interestingly, CM 3.6. Lack of appropriate level of specificity
effectiveness and use of a consulting firm was negatively On the one hand, some competencies are too broad
correlated but non-significant (–.20) while organization and cannot be measured or communicated while others
size and use of a consulting firm was positively correl- are overly detailed and overwhelming for users. A large
ated and significant at (.51; p < .05). 2011 respondent revised and simplified competencies to
gain employee acceptance.
3.3. What are the barriers to success of CM?
Although not presented in Table 2, our survey included
3.7. Leadership changes create CM instability
a qualitative element in that we asked respondents to Strategically, top management uses CM to focus and
indicate barriers to the success of CM. Consistent with direct the organization in specific directions (Campion

© 2013 John Wiley & Sons Ltd International Journal of Selection and Assessment
Volume 21 Number 3 September 2013
338 Thomas H. Stone, Brian D. Webster and Stephen Schoonover

et al., 2011). ‘CM changes as leaders change’. However, The third reason for study of CM users is the lack of
experts observed this may lead to instability in basic evidence of CM effectiveness. The Schoonover et al.
applications such as selection. (2000) finding of dissatisfaction among over one-third of
CM users and the 2011 survey’s 75.2% effectiveness rat-
ing suggests CM may have some implementation issues.
3.8. Competencies are not validated Future research should use more objective measures of
effectiveness rather than the mainly subjective ratings
Some organizations adopt competencies, but fail to ex-
used in existing research. Additionally, responses of
amine how they predict job behaviors, particularly when
major CM vendors suggest several sources of challenge
CM libraries are used.
to effectiveness of CM. Of the challenges mentioned,
the lack of validation and appropriate level of specificity
4. Conclusion issues are examples of potential gains from use of TJA
methods suggested by Sanchez and Levine (2009) while
The extant CM literature is primarily prescriptive others are management issues. In the rush to speed ap-
advocating adoption of CM. A number of I/O psycho- plication, CM has not received the critical review its
logy articles question the methodological rigor of the widespread adoption merits.
CM process while also arguing traditional job analysis
methods would benefit CM. This review suggests the
need for more current information regarding CM users, References
effectiveness and challenges of CM adoptions.
Adoption and use of CM merits further research Campion, M., Fink, A., Ruggeberg, B., Carr, L., Phillips, G., &
Odman, R. (2011). Doing competencies well: Best practices
for at least three reasons. First, although published
in competency modeling. Personnel Psychology, 64, 225–262.
empirical data regarding CM adoption are lacking, in- Green, P. (1999). Building robust competencies: Linking human re-
formal evidence suggests growing adoptions by private source systems to organizational strategies. San Francisco, CA:
and public sector organizations of all sizes. Favorable Jossey-Bass.
and prescriptive CM information proffered in books, Lawler, E. (1994). From job-based to competency-based
articles and influential organizations such as the organizations. Journal of Organizational Behavior, 15, 3–15.
Society for Human Resource Management (SHRM), Lucia, A., & Lepsinger, R. (1999). The art and science of compet-
WorldatWork and the American Society of Training and ency models. San Francisco, CA: Jossey-Bass.
Development (ASTD) encourage adoption. PeopleFluent. (2012). About people fluent. Available at
Second, as a major tool of work analysis that may be peoplefluent.com./about (accessed 4 April 2012).
replacing TJA in some contexts, CM merits critical ana- Rathbar-Daniels, D., Erickson, M., & Dalik, A. (2001). Here to
stay: Taking competencies to the next level. Worldatwork
lysis. Extensive adoption of TM systems, our survey and
Journal, First qtr, 70–77.
CM vendor interviews suggest this may be occurring. Sanchez, J., & Levine, E. (2009). What is (or should be) the
Specifically, the consultant panel stated the vast majority difference between competency modeling and traditional
of their business is CM not TJA. However, little is job analysis? Human Resource Management Review, 19, 53–63.
known about the effect of replacing TJA with CM. Is the Schippmann, J., Ash, R., Battista, M., Carr, L., Eyde, L., Hesketh,
CM process less reliable and more open to bias than B., Kehoe, J., Pearlman, K., Prien, E., & Sanchez, I. (2000).
TJA? The Schippmann et al. (2000) panel felt CM was The practice of competency modeling. Personnel Psychology,
better for training and development and that TJA should 53, 703–740.
be used for selection, performance appraisal and related Schippmann, J. S. (1999). Strategic job modeling: Working at the
HR applications because TJA would better withstand core of integrated human resources. Mahwah, NJ: Erlbaum.
legal challenges. Campion et al. (2011) argue that legal Schippmann, J. S. (2010). Competencies, job analysis, and the
next generation of modeling. In J. Scott & D. Reynolds
defensibility of HR systems is a key reason many large
(Eds.), Handbook of Workplace Assessment (pp. 197–231). San
organizations conduct TJA. Sanchez and Levine (2009) Francisco, CA: John Wiley & Sons.
provide excellent suggestions for useful cross- Schoonover, S., Schoonover, H., Nemerov, D., & Ehly, C.
fertilization of CM and TJA. Future research should ex- (2000). Competency-based HR applications: Results of a com-
amine the extent to which such cross-fertilization is prehensive survey, personal correspondence from Stephen
occurring and is successful. Schoonover, Aug. 23, 2011.

International Journal of Selection and Assessment © 2013 John Wiley & Sons Ltd
Volume 21 Number 3 September 2013
Copyright of International Journal of Selection & Assessment is the property of Wiley-
Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.