Evaluating Personality-Based Job Requirements

Joyce Hogan and Brent Holland

Hogan Assessment Systems

Paper presented at Seventeenth Annual Meeting of the

Society for Industrial-Organizational Psychology, Inc., Toronto, April, 2002.

1
 Evaluating Personality-Based Job Requirements

Although “worker characteristics” is a major focus of job analysis methods, there are few
systematic procedures for studying personality-based job requirements. This is not to say that job
analysts have ignored personality influences; there are personality footprints visible in many
worker-oriented job analysis methods. However, most available methods have not focused on
trying to solve the problem of what personality characteristics are important for effective job
performance. To answer this question requires a taxonomy of personality characteristics that
influence work, a procedure for identifying these characteristics reliably, and methods to evaluate
the validity of the job analysis results.
Our objective for developing a new job analysis method was to investigate the possibility
that an integrated system of job analysis, predictor specification, and criterion measurement could
be built on a common taxonomy of underlying Big-Five personality constructs. We wanted to be
able to link personality constructs as job requirements to predictors and outcome measures that
assess individual differences. This personality-based job analysis is intended to serve empirical
test validation.
The first question for any assessment is what to measure. The defensible answer will
specify which constructs and why they are important. We sought to identify and evaluate personal-
ity characteristics that influence work—characteristics whose manifestations are sufficiently impor-
tant that they can be observed and evaluated by others. Sixty years of personality research sug-
gests five factors can account for observers’ descriptions of individual differences in social behav-
ior. These “Big-Five” factors emerge from lexical analyses of trait descriptions evaluating others’
performance. The five dimensions as they apply to worker performance are as follows:
Emotional Stability = nervous and moody—calm and assured
Extraversion/Surgency = quiet and unassertive—active and outgoing
Conscientiousness = impulsive and careless—responsible and conforming
Agreeableness = hard-nosed and tough—tactful and sensitive
Intellect/Openness = narrow and unimaginative—curious and imaginative
The observable part of personality is a person’s reputation. At work, reputation builds from
interactions with others, it is evaluative in nature, and observers can rate targets with high reliability.
This suggests that a personality-based job analysis minimally ought to assess five dimensions
because they are the core of an adequate description of a worker’s reputation.

Construction of Performance Improvement Characteristics

The Performance Improvement Characteristics (PIC) job analysis method is designed to
evaluate personality-based job requirements that can subsequently inform test validation research
(Hogan & Rybicki, 1998). Ideally, PIC results reflect the personality dimensions that are most

2
important for job performance and these results are then used to develop hypotheses about per-
sonality measures that, potentially, are valid predictors of job performance. PIC results also have
implications for criterion specification in that personality requirements can point to the importance
of certain job performance criteria (e.g., if conscientiousness is important worker requirement, then
performance criteria such as integrity, attention to detail, and dependability ratings could be rel-
evant).
The structural model for the PIC derives from the structure of the Hogan Personality
Inventory (HPI; Hogan & Hogan, 1995), which is a measure of normal personality based on the
Big-Five model. The goal of the HPI is to predict occupational performance and its has a 20 year
history of doing so. We wished to capitalize on the measurement structure that we knew would
successfully integrate a job analysis with valid predictors of job performance. The PIC consists of
48 items that form seven scales. Each scale consists of five to nine items, with no item overlap,
and the scales, along with their corresponding Big-Five dimensions and definitions, appear in
Figure 1. The response format is a 4-point scale indicating the extent to which having the personal
characteristic improves job performance. It is interesting to note that Primoff’s Job Element
Procedure(see Primoff & Eyde, 1988) uses a very similar evaluation format. The Flesch-Kincaid
analysis indicates that the items are written at a 7th grade reading level. A sample of the PIC
rating form apprears in Figure 2.

Figure 1

PIC Scales, Corresponding Big-Five Factors, and PIC Scale Definitions

PIC Scale Big-Five Factor PIC Scale Definition

Adjustment Emotional Stability Resilient, upbeat, & remaining calm

under pressure
Ambition Surgency Competitive, self-confident, & taking
initiative
Sociability Surgency Approachable, outgoing, & social

Likeability Agreeableness Considerate, perceptive, tactful, &

good natured
Prudence Conscientiousness Planful, controlled, & attentive to
details
Intellectance Intellect/Openness Imaginative, creative, open-minded,
& analytical
School Success Intellect/Openness Staying up-to-date on business &
technical matters

3
 Figure 2
Performance Improvement Characteristic
INSTRUCTIONS
Personal characteristics affect job performance as a long distance truck driver. Below is a list of characteristics used frequently to
describe behavior. Please provide a rating of the extent to which each characteristic improves performance as a truck driver. Use
the scale below to make your ratings.

Does Not Improve Minimally Improves Moderately Improves Substantially Improves

Performance Performance Performance Performance
0 1 2 3

Example Item HPI Dimension

‹ Is not easily irritated by others Adjustment
‹ Takes initiative and gets projects moving Ambition
‹ Seems to need and enjoy social interaction Sociability
‹ Understands others’ moods Likeability
‹ Rarely deviates from standard procedures Prudence
‹ Is imaginative and open-minded Intellectance
‹ Likes to learn new things, and enjoys education and training School Success

Internal Psychometrics
Table 1 presents the PIC scales, descriptive statistics, internal consistency, and test-retest
reliabilities. As seen, the internal consistency reliabilities range from .76 (Adjustment) to .87
(Likeability) with an average alpha of .81. Test-retest reliabilities of ratings gather on the secretary
job over a three-month interval ranged for .60 (School Success) to .84 (Intellectance) with an
average of .72. The PIC scales are all intercorrelated with r’s ranging from .2 to .6 and although
this positive manifold is not unusual, lower scale to scale correlations would have been desirable.

Table 1
Descriptive Statistics and Reliabilities for the Performance Improvement Characteristics
N = 822
Scale Name Number Mean SD Alpha Inter Rtt* SE
of Items -Item r
Adjustment 9 19.13 4.39 .76 .26 .64 .16
Ambition 7 15.92 4.04 .83 .40 .80 .14
Sociability 6 7.67 4.15 .83 .44 .73 .15
Likeability 6 12.91 3.86 .87 .53 .69 .14
Prudence 8 18.50 3.89 .82 .36 .69 .14
Intellectance 7 10.68 5.08 .85 .44 .84 .18
School Success 5 11.44 2.86 .84 .52 .60 .12
* n = 79

4
 Nevertheless, the factor structure of the item correlations supports a seven factor solution. We
evaluated the PIC’s factor structure using principal axis procedures with varimax rotation. As shown in
Table 2, the PIC’s internal structure is consistent with its conceptual model (i.e., the Hogan Personality
Inventory). In nearly every case (94%), the items written to map a specific construct loaded most
heavily on the correct factor. Only two items failed to have their primary loading on the conceptual
factor and only one item (item 1) failed to load (either through primary or secondary loading) on the
proper scale. Overall, the factor analytic results support the seven-factor model that the PIC was
designed to measure. The data also provide additional evidence, at least in terms of internal structure,
supporting the congruence between the PIC and HPI. No other job analytic and predictor tool(s) that
we are aware of possess such similar features.

Table 2
PIC Factor Structure Loadings
Adjustment Ambition Sociability Likeability Prudence Intellectance School
Success
.58 (4) .67 (14) .75 (18) .74 (25) .72 (33) .70 (39) .62 (44)
.58 (3) .64 (10) .69 (22) .71 (23) .72 (35) .70 (41) .58 (48)
.53 (8) .64 (12) .68 (21) .68 (28) .61 (36) .57 (42) .57 (45)
.51 (7) .57 (13) .66 (17) .65 (24) .58 (34) .55 (38) .55 (46)
.34 (6) .51 (11) .62 (19) .63 (27) .44 (31) .42 (37) .47 (47)
.26 (5) .50 (15) .60 (16) .60 (26) .44 (32) .33 (43)
.25 (2) .30 (16) .36 (20) .41 (30) .30 (40)
.23 (9)* .25 (29)*
(1)**
Note. Item numbers are shown in parentheses. *Secondary loading on primary factor; ** Failed to
load on primary factor.

Applications
To date, we have gathered over 4,000 PICs across more that 210 jobs. Our samples of
jobs roughly represent the base rate of occupational types in the U.S. economy as reported by
Gottfredson and Holland (1996): Realistic (66.7%), Conventional (13.4%), Enterprising (11.1%),
Social (4.6%) Investigative (3.0%), and Artistic (1.2%). In all studies, subject matter experts
(SMEs) complete the PICs using paper-and-pencil forms or electronic files on the internet.
As a new methodology, we focused on basic applications to evaluation job differentiation,
reliability of ratings across raters and jobs, and validity of results. Three research questions and
example applications are presented next.

5
 Differentiating among Jobs. Any useful job analysis must be sensitive enough to identify
valid differences in job requirements. We hypothesized that, for the majority of jobs, Adjustment
and Prudence would improve performance (cf. Hogan & Holland, in press; Ones, Viswesvaran, &
Schmidt, 1993) whereas the remaining personality constructs would be differentially important by
job. To evaluate the discriminating power of the PIC, we compared jobs from the major occupa-
tional types in Holland’s (1997) vocational theory. For PIC Adjustment and Prudence personal
requirements, there was approximately a standard deviation range across SME job ratings; the
remaining PIC scale ratings varied as much as one and a half standard deviations. Figure 3 shows
the comparison of PIC profiles for selected jobs. Contrasts between sales and mail processing
jobs are particularly distinctive and informative. The dispersed SME ratings for the mail process-
ing job suggest that rater tendencies halo, leniency, and central tendency are not operating.

Figure 3

PIC Profile Comparisons

100

90

80

70
Cashier

60 Office
Percentiles

Manager
Crrct Officer
50
Processor

40 Sales

Delivery
30

20

10

0
Adjustment Ambition Sociability Likeability Prudence Intellectance School Success
PIC Scales

Reliability of Job Ratings across Locations. To evaluate the consistency of SME PIC ratings for
the same job in different locations, data from a large utility company in the southeastern U.S. were
gathered from first line supervisors. SMEs (N > 10) from five locations evaluated the personality-
based requirements of the supervisor job. Figure 4 shows these profiles. As seen in Figure 4 for
the various locations, the profile shapes are quite consistent with high scores for Adjustment,
Ambition, and Intellectance and low average scores for the other scales. A one-way ANOVA
indicated one significant difference within the scales for the Adjustment mean ratings (F (4, 90) =
6
3.46, p. < .01); post hoc analysis revealed that the only PIC mean rating differences were between
locations with the highest and lowest mean Adjustment ratings. With substantial congruence of
mean PIC ratings within scales, the results suggest that the personality-based requirements of the
supervisor job in this company are the same regardless of geographic or operating location.

Figure 4

PIC Profiles for Supervisors in Different Locations

100

90

80

70

60
Percentiles

50

40

30

20

10

0
Adjustment Ambition Sociability Likeability Prudence Intellectance School Success
PIC Scales

Validity of the PIC for Differentiating Job Performers

If PIC results are valid, then SMEs ratings should provide the personality profiles of
effective job performers. These profiles should be distinctive from ineffective job performers. For
this evaluation, it is necessary to have PIC job analysis data, personality predictor data from
incumbents using the same personality constructs, and criterion data reflecting adequacy of overall
job performance. Although a number of these validation studies exist (cf. Tett, Holland, & Hogan,
2002), one example with well developed criterion data illustrates the general findings. Figure 5
shows three personality profiles for convenience store cashiers—one summarizes SMEs mean PIC
ratings for the personality-based job requirements, and the other two are HPI mean scale scores
for top and bottom performing store cashiers. The PIC and the HPI profiles of the top performers
are similar and the HPI profile of the bottom performers is quite different. The PIC profile indi-
cates that the requirements for improved performance include Adjustment, Ambition, Sociability,

7
Likeability, and Prudence. This translates into being calm under pressure, being action-oriented,
meeting the public well, and being dependable and reliable. The HPI profile of the highest rated
cashiers closely parallels the PIC requirements and is significantly different from the lowest rated
cashiers on HPI Adjustment, Ambition, and Prudence. In the concurrent validation research, HPI
Adjustment (r = .50, p < .01), Ambition (r = .36, p < .05), and Prudence (r = .43, p < .05)
significantly predicted cashier job performance. This provides some preliminary evidence that the
PIC yields profiles that closely match the characteristics of workers who are seen as effective.

Figure 5

Cashier Profiles for the PIC and HPI Profiles for Top and Bottom Performers

100

90

80

70

60
Percentiles

PIC Profile
50 Top Performers
Bottom Performers

40

30
20

10

0
Adjustment Ambition Sociability Likeability Prudence Intellectance School Success
PIC and HPI Scales

Summary
To summarize, the PIC is a worker-oriented job analysis method designed to evaluate
personality-related job requirements. It is designed primarily for use in conjunction with test
validation research for personnel selection. PICs are scored by aggregating SME ratings to form a
seven dimension profile. These results can then be used to develop hypotheses about personality
measures that, potentially, are valid predictors of job performance. Results to date suggest the
instrument is reliable and can yield valid results. Continued research is needed to extend the
database with additional jobs and more applications.

8
 References

Gottfredson, G. D., & Holland, J. L. (1996). Dictionary of Holland occupational codes (3rd
ed.). Odessa, FL: Psychological Assessment Resources.
Hogan, J. & Holland, B. (in press). Using theory to evaluate personality and job performance
relations: A socioanalytic perspective. Journal of Applied Psychology.
Hogan, J., & Rybicki, S. (1998). Performance Improvement Characteristics job analysis
manual. Tulsa, OK: Hogan Assessment Systems.
Hogan, R., & Hogan, J. (1995). Hogan Personality Inventory manual. Tulsa, OK: Hogan
Assessment Systems.
Holland, J. L. (1997). Making vocational choices. Odessa, FL: Psychological Assessment
Resources.
Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (1993). Comprehensive meta-analysis of integrity
test validation: Findings and implications for personnel selection and theories of job perfor
mance. Journal of Applied Psychology, 78, 679-703.
Primoff, E. S., & Eyde, L. D. (1988). Job Element Method. In S. Gael (Ed.). The job analysis
handbook for business, industry, and government (Vol. II, pp. 807-824). New York:
John Wiley.
Tett, R. P., Holland, B. D., Hogan, J., & Burnett, D. (April, 2002). Validity of trait-based job
analysis using moderator correlations. Paper presented at the Seventeenth Annual
Meeting of the Society for Industrial-Organizational Psychology, Inc., Toronto.