Interests,Work Values,and Occupations

Abstract

This study examined whether a measure of person–environment (P-E) fit predicted worker ratings of work attitudes and supervisor ratings of performance. After combining extant data elements and expert ratings of interest and work value characteristics for each occupation in the O*NET system, the authors generated a “Fit Index”—involving profile correlations between individual interest and work value score profiles and corresponding occupational profiles—for each occupation. The authors then conducted an extensive field study of 503 adult workers and 219 supervisors. The construct validity of the fit scores was supported by the patterns of over 477,000 scores across same, similar, and dissimilar occupations. Results also supported the association of person–occupation fit with desirable work attitudes and outcomes, and the incremental validity of fit when including a measure of integrity. These results add support to the idea that P-E fit can be useful in personnel staffing.

Keywords

person–occupation fit career interests work values work outcomes work attitudes job performance

The concept of person–environment (P-E) fit, which is defined as the congruence, match, similarity, or correspondence between the person and the environment, has long been prevalent in the vocational behavior, organizational psychology, and human resources management literatures (e.g., Edwards, 1991; Holland, 1997; Kristof, 1996; Pervin, 1968; Schneider, 2001; Walsh, Craik, & Price, 2000). P-E fit has been examined in reference to various person and environment constructs, such as personal interests and values (Adkins, Ravlin, & Meglino, 1996; Judge & Bretz, 1992; Knafo & Sagiv, 2004), the personality of the employee and other members of the organization (Schneider, 1987), and employee abilities (Caldwell & O’Reilly, 1990; Kristof-Brown, 2000).

Efforts to apply P-E fit to workforce needs can be traced back to the 1920s, when interest inventory results first began to be reported based on the similarity between the measured interests of individuals and the measured interests of people in various occupations. In the 1960s, there were systematic and extended efforts to apply the idea of fit to workers and work values (Lofquist & Dawis, 1969). The use of fit has been accelerating since the mid-1980s, as more researchers and employers focus on the impact of fit for worker satisfaction, organizational commitment, and the social/interpersonal aspects of job performance (Erdheim, Zickar, & Yankelevich, 2007). Given the theory and research underlying P-E fit, some have questioned why the use of fit assessments for workforce needs is not more widespread (Ployhart, Schneider, & Schmitt, 2006).

Although many types of fit have been proposed in the scholarly literature, the current study focuses exclusively on the usefulness of person–occupation fit for predicting desirable work attitudes and outcomes. We used the O*NET database as a foundation to develop and validate the WorkKeys Fit Assessment (WFA), a tool designed to measure person–occupation fit based on vocational interests and work values (ACT, 2007a). We then investigated the degree to which person–occupation fit provides incremental validity above integrity testing, a widely used noncognitive selection tool for predicting job performance. To the best of our knowledge, this has not been previously examined in the literature.

Our article is organized as follows. First, we briefly review the literature on P-E fit, paying particular attention to person–occupation fit. We then discuss the role of interests and work values in assessing fit with a given occupation, and also provide a very brief review of integrity testing. Next, we detail the development of the WFA and present the results of a field study of 503 workers employed in 111 different occupations. Results focus on the validity of the WFA and the incremental validity of person–occupation fit, over a measure of integrity, for predicting job performance. We conclude by discussing implications for research and practice.

Person–Occupation Fit

The concept of P-E fit is generally characterized by the match between individuals and various levels of their work environment (Kristof, 1996; Kristof-Brown, Zimmerman, & Johnson, 2005). The levels of the work environment can range from narrow to broad. Three distinct types of fit have garnered attention: person–job (P-J) fit, person–occupation (or person–vocation, P-V) fit, and person–organization (P-O) fit. P-J fit refers to the fit between the person and the specific tasks performed in a particular work setting (e.g., Kristof, 1996; Lawrence, 2004; Smart, Elton, & McLaughlin, 1986). P-V fit refers to the fit between the person and the work tasks typically associated with jobs in an identifiable vocation or occupation (e.g., Gati, Garty, & Fassa, 1996; Mount & Muchinsky, 1987; Rounds, Dawis, & Lofquist, 1987; Spokane, Meir, & Catalano, 2000; Swaney & Prediger, 1985; Ton & Hansen, 2001). P-O fit refers to the fit between the person and aspects of the entire organization (e.g., Arthur, Bell, Villado, & Doverspike, 2006; Hoffman & Woehr, 2006; Kristof, 1996; Verquer, Beehr, & Wagner, 2003).

Research suggests that there are advantages to P-V fit relative to P-O and P-J fit. For example, seeking P-O fit may lead to organizational ineffectiveness because over time homogeneity can lead to a narrowing of perspectives. Thus, it may make sense for organizations to hire persons with a mix (high fit and low fit) of organization-relevant knowledge, skills, and personality characteristics (Kristof, 1996). Similarly, when comparing P-V to P-J fit, assessments designed to measure P-J fit can be overly complex and impractical because of the amount of variability in specific jobs. In contrast, occupations display more consistency across settings (Schneider, Kristof-Brown, Goldstein, & Smith, 1997).

Reviews of the literature have consistently found evidence that occupational fit predicts work attitude outcomes. However, validity estimates (correlations) are typically small to moderate. A review of 66 studies estimated that the correlation (uncorrected for measurement error and range restriction) between occupational fit and job satisfaction is about .25, with maximum correlations of .40 (Spokane, Meir, & Catalano, 2000). Subsequent studies have continued to find validities in this range. For example, Ton and Hansen (2001) examined the correlation between satisfaction and two types of occupational fit: one based on interests (.24) and one based on values (.53). Although few studies have looked at the relationship between occupational fit and commitment, a recent study found a small but statistically significant relationship between fit and intent to leave one’s current career (Donohue, 2006).

Interests and Work Values

Holland’s (1997) theory of careers posits that both people and work environments can be classified into six types: Investigative, Artistic, Social, Enterprising, Conventional, and Realistic. According to this theory, environments are dominated by a given type, and people tend to gravitate to, and remain in, work environments they resemble: Conventional types gravitate toward environments consisting primarily of Conventional types; Artistic types gravitate toward environments consisting primarily of Artistic types, and so on. P-E congruence provides opportunities for people to engage in activities that match their interests and reward their skills. Further, the theory holds that P-E congruence contributes to satisfaction and stability. Holland’s theory has received extensive empirical scrutiny and much support (Nauta, 2010), and the six types serve as the organizing framework for many of today’s career assessment applications. Schneider (2001) and others have emphasized that Holland’s model is an explicit P-E theory with important implications for understanding the consequences of P-E fit in work settings.

Research shows that work values are another important aspect of an individual’s fit with the work environment. Chatman (1989) suggested that values are an appropriate means of conceptualizing fit because individual and organizational values can be evaluated similarly. Chatman’s research shows that individuals with high P-O work value fit adjusted to the organization more quickly, were more satisfied, and intended to remain with the organization longer than those with low work value fit (Chatman, 1991). Meglino, Ravlin, and Adkins (1989) reported that congruence between employees’ work values and those of their supervisors was associated with job satisfaction, organizational commitment, and reporting to work on time. Further, Judge and Bretz (1992) reported that individuals were more likely to choose jobs with value content similar to their own value orientation.

While measures of value-based P-O fit often focus on the match between the work values of workers and supervisors, measures of value-based person–occupation fit typically involve the match between the availability of occupational attributes (e.g., earnings, prestige) and the values that people place on those attributes (e.g., Prediger & Staples, 1996; Zytowski, 2004a). Theory and research support the importance of fit between values and work attributes in achieving successful work adjustment (Dawis & Lofquist, 1984), and the effectiveness of value-based fit measures is supported by their well-established use in career counseling (e.g., Bobek & Gore, 2004; Zytowski, 2004b).

Measures of Conscientiousness or Integrity

Integrity tests (see Berry, Sackett, & Wiemann, 2007, for a review) have been shown to be among the most valid of available selection devices. For example, in a review of the validity and utility of personnel selection methods, Schmidt and Hunter (1998) pointed to integrity tests as the selection tool providing the greatest incremental validity over general mental ability, the single best predictor of job performance. A large-scale meta-analysis conducted by Ones, Viswesvaran, and Schmidt (1993) found that integrity tests are valid not only for predicting theft and nontheft counterproductive work behaviors, but also job performance. In essence, these are compound tests primarily comprised of conscientiousness, emotional stability, and agreeableness items. Integrity tests are clearly strong predictors of a variety of work outcomes, and thus can serve as a natural comparison point when examining whether fit provides incremental validity for predicting job performance and other outcomes.

Hypotheses

First, we expect that workers will display higher levels of fit to their current occupation—or occupations similar to their current occupation—than to occupations that are dissimilar to their current occupation. That is, to the degree that people gravitate toward work environments that are in accord with their personal characteristics, a valid measure of fit should demonstrate greater fit between (for example) incumbent salespersons and sales occupations than between incumbent salespersons and construction occupations. Second, we expect that fit will predict worker ratings of work satisfaction and commitment, and will also predict supervisor ratings of performance. Finally, we expect fit—a measure of noncognitive constructs related to performance, yet different from integrity—to yield incremental validity over integrity, a commonly used measure of personality when predicting task and overall performance ratings.

Method

Participants

Participants were 558 employed adults who voluntarily completed the WFA as part of a field study. The participants were from 10 organizations spanning a range of industries, including manufacturing, transportation, healthcare, education, information services, as well as educational testing and publishing. The size of participating organizations ranged from small businesses to branches of multinational companies. Of the 558 records, 55 (10%) were not used because of missing assessment item responses, inconsistent responding (e.g., random responding), or lack of variability in responses (e.g., answering extremely important to every item). The remaining records (N = 503) were used for the analyses presented below. The sample ranged in age from 17 to 68 and included more males (57%) than females. Approximately 95% of the sample had completed high school (or the equivalent), and of those, about 45% had some level of education beyond high school. The sample represented 111 of the 949 O*NET occupations, and these occupations represented 21 of the 23 O*NET major occupational groups. A more detailed breakdown of the demographic characteristics of this sample is shown in Table 1, and occupational representation is shown in Table 2.

Table 1.

Demographic Characteristics of Participants

	Fit Only^a		Fit + Integrity^b
Characteristic	N	%	N	%
Gender
Male	286	57.1	168	70.0
Female	215	42.9	72	30.0
Ethnicity
African-American/Black (non-Hispanic)	93	18.6	76	31.8
American Indian, Alaskan Native	3	0.6	1	0.4
Caucasian-American/White (non-Hispanic)	288	57.5	107	44.8
Hispanic (Mexican American, Puerto Rican, Cuban, and so on)	62	12.4	21	8.8
Asian American, Pacific Islander	25	5.0	21	8.8
Multiracial	6	1.2	4	1.7
Other	9	1.8	5	2.1
I prefer not to respond	15	3.0	4	1.7
Highest level of education
No formal education	4	0.8	5	2.1
Elementary/middle school	24	4.8	4	1.7
High school diploma	232	46.4	118	49.4
GED	29	5.8	21	8.8
Trade/proprietary school certification	60	12.0	35	14.6
Associate’s degree	43	8.6	23	9.6
Bachelor’s degree	57	11.4	17	7.1
Master’s degree	42	8.4	16	6.7
Doctorate degree	9	1.8	0	0
Tenure
Less than 6 months	101	20.1	29	12.0
6 months to less than 1 year	58	11.6	27	11.2
1 year to less than 2 years	65	13.0	26	10.8
2 years to less than 5 years	98	19.5	54	22.4
5 years to less than 10 years	94	18.7	56	23.2
10 years or more	86	17.1	49	20.3

Note. Numbers vary due to missing responses.

^a N = 503.

^b N = 242 subgroup with Fit data merged with integrity test and supervisor ratings data.

GED = General Educational Development.

Table 2.

Occupational Groups (Based on O*NET) of Participants

	Fit only^a		Fit + integrity^b
Occupational groups^c	N	%	N	%
Management	35	7.0	16	6.6
Business and financial operations	7	1.4	0	0
Computer and mathematics	10	2.0	2	0.8
Architecture and engineering	1	0.2	0	0
Life, physical, and social science	4	0.8	1	0.4
Community and social services	7	1.4	1	0.4
Education, training, and library	32	6.4	11	4.6
Arts, design, entertainment, sports, and media	6	1.2	0	0
Healthcare practitioners and technical	4	0.8	0	0
Healthcare support	1	0.2	0	0
Protective service	1	0.2	0	0
Food preparation and serving	28	5.6	0	0
Building and grounds cleaning and maintenance	10	2.0	3	1.2
Personal care and service	2	0.4	0	0
Sales and sales related	26	5.2	4	1.7
Office and administrative support	71	14.1	15	5.8
Farming, fishing, and forestry	2	0.4	0	0
Construction and extraction	7	1.4	0	0
Installation, maintenance, and repair	22	4.4	13	5.4
Production and manufacturing	63	12.5	51	21.2
Transportation and material moving	164	32.6	125	51.9

Note. ^a N = 503 with Fit data matching with occupational groups.

^b N = 242 subgroup with Fit data merged with integrity test and supervisor ratings data.

^cBased on O*NET major occupational groups.

Some analyses focused on a subset of the sample described above (N = 242; 48% of the 503) who also completed an integrity test (the WorkKeys Performance Assessment; ACT, 2007b), and whose supervisors had provided job performance ratings. This subset of participants came from eight organizations and ranged in age from 20 to 68. They represented 50 of the 949 O*NET occupations, and these occupations represented 11 of the 23 O*NET major occupational groups. A more detailed breakdown of the demographic and occupational characteristics of this subgroup is shown in Tables 1 and 2.

Measurement of Fit

The measurement of fit in the WFA is based on the correlation between respondents’ interests and values profiles and the corresponding interests and values profiles for each O*NET occupation (ACT, 2007a). These two correlations are combined and normed as described below.

Interest inventory

The WFA interest inventory was developed based on the ACT Interest Inventory, an established assessment included in several ACT programs and currently completed by over 4 million persons each year (ACT, 2006, 2009). The six scales cover the full spectrum of basic work tasks and parallel the six interest types in John Holland’s well-known theory of careers (Holland, 1997), which also is the basis for the O*NET occupational interest profiles (Rounds, Smith, Hubert, Lewis, & Rivkin, 1998). Based on item performance and content guidelines, the best 72 items (12 items per scale) were selected from the ACT Interest Inventory item pool (90 items) for use in the WFA. The WFA interest inventory items describe common, work-related activities that are likely to be familiar to people through participation or observation (occupational titles and specific job duties are not used). Table 3 provides descriptions and sample items for each of the six scales. For each item, respondents indicated whether they liked, disliked, or were indifferent to engaging in the activity.

Table 3.

Interest Inventory Scales and Sample Items

Scale	Description	Sample item
Administration and Sales	Persuading, influencing, directing, or motivating others through activities such as sales, supervision, and aspects of business management	Conduct a meeting
Business Operations	Developing and/or maintaining accurate and orderly files, records, accounts, and so on; designing and/ or following systematic procedures for performing business activities	Figure shipping costs for catalog orders
Technical	Working with tools, instruments, and mechanical or electrical equipment. Activities include designing, building, repairing machinery, and raising crops/animals	Assemble a cabinet from written instructions
Science and Technology	Investigating and attempting to understand phenomena in the natural sciences through reading and research	Learn how the brain works
Arts	Expressing oneself through activities such as painting, designing, singing, dancing, and writing; artistic appreciation of such activities	Make creative photographs
Social Service	Helping, enlightening, or serving others through activities such as teaching, counseling, and working in service-oriented organizations	Help someone make an important decision

Work values inventory

Items for the WFA values inventory were written by two ACT researchers, each with over 25 years of experience in career assessment development. Item development was guided by three goals. First, we sought values that are commonly recognized and used in other career assessments. Thus the first step was to build a set of candidate values from a review of the literature, emphasizing values that corresponded to “work needs” and “work contexts” in the O*NET system (National Center for O*NET Development, 2006). Second, we sought values that were likely to contribute to capturing essential differences across a wide range of occupations (e.g., working indoors vs. working outdoors). To this end, a number of WFA values were based on values from the 22-item Inventory of Work-Relevant Values (IWRV; Bobek & Gore, 2004). IWRV is an assessment component of DISCOVER®, ACT’s online career planning system designed to facilitate exploration of a broad, comprehensive range of occupations. Third, we sought values that permitted development of rating rules (described below) that minimized ambiguity. These guidelines led to a set of 18 items, 16 of which corresponded to IWRV values. The six IWRV values not used in the WFA were either judged to be uncommon (e.g., certification) or overlapping conceptually (e.g., authority was used, so management was not). The set of 18 items were also judged to be similar to 10 of the 21 O*NET work needs and 14 of the 57 O*NET work contexts. Table 4 provides descriptions of each of the 18 work values in the WFA work values inventory. For each item, respondents indicated how important it was to them using a 5-point Likert-type style scale ranging from 1 (not important) to 5 (extremely important). In total, the WFA consists of 90 scored items (72 interests and 18 values) requiring approximately 10–15 min to complete.

Table 4.

Work Values and Definitions

Value	Definition
Authority	Telling people what to do; controlling the behavior of others
Autonomy	Making my own plans and decisions as I do my work
Compensation	Earning a higher salary or wage than people working in other jobs
Creativity	Creating something new or finding new ways of doing things; original thinking
Flexible hours	Being able to choose the hours I want to work, as long as the work gets done on time
Helping people	Improving the lives of others by activities such as teaching, physically assisting, or mentoring
Influencing others	Convincing or advising people to do things, even though I have no authority over them
Intellectual stimulation	Thinking about difficult concepts and working to solve complex problems
Order	Putting things in order for others; using a system or rules to arrange things
Physical activity	Moving around in my work by walking, bending, lifting, and so on
Precision	Being exact or very accurate in the work I do
Public contact	Interacting with customers (as in sales) or the public (as in police work)
Social status	Being looked up to by others in my company or my community because of my job
Taking risks	Working in settings that involve risk, such as on high places or near vehicular traffic
Using my hands	Using my hands to skillfully control objects, tools, or machines
Variety	Using many different skills to do my work
Working independently	Working separately from coworkers and supervisors for much of the day
Working outdoors	Working outdoors much of the time, being exposed to the weather

Occupation ratings

For each examinee, the WFA provides fit results for the 949 occupations in O*NET 10.0 (ACT, 2007a). Occupational ratings corresponding to the WFA interest scores come from two sources. O*NET ratings for 748 of the 949 occupations were obtained from Rounds et al. (1998). The remaining 201 occupations were rated by an ACT occupational information expert with over 25 years of experience. Ratings for each of the 201 occupations involved three steps. First, the rater selected from two to six rated O*NET occupations that were similar to the unrated occupation on the basis of work tasks and worker skills. Second, the unrated occupation was assigned ratings based on expert judgment, guided by descriptions of O*NET occupations, descriptions of Holland types, and the ratings of the set of similar occupations. The rater was instructed to assign ratings within the range of ratings for the set of similar occupations unless there was a clear and compelling reason to exceed that range (e.g., because a business occupation was known to be more characteristic of the Investigative type than the business occupations it was being compared to). This process assured that the patterns of new ratings were generally in accord with the patterns of established O*NET ratings. Third, an initial set of ratings for 56 occupations were reviewed by an ACT career expert for appropriateness and consistency prior to completing the ratings for the remaining 145 occupations.

Occupational ratings corresponding to WFA work values scores were developed by two ACT career experts, each with over 25 years of experience in rating occupations. Although a review of O*NET work needs and work contexts identified 796 occupations with ratings for some or all of the corresponding WFA work values, ambiguities in these ratings made it difficult to formalize rating rules for the remaining 153 occupations. Therefore, a new set of rating rules was developed to assign new ratings for every value for every occupation. These rules permitted the development of operational definitions that retained the intent of O*NET work need and work context definitions, while also enhancing clarity for the raters. Raters were also provided with examples of anchor occupations, based on these new rules, for the upper, middle, and lower ratings on the scale. Rating assignment consistency was evaluated by comparing the ratings of similar sets of occupations, revising discrepant ratings as needed. The ratings were subsequently reviewed by a third ACT career expert for appropriateness and consistency (ACT, 2007a).

Norms

The norming sample was constructed from 2003 to 2006 DISCOVER user records. A total of 19,614 persons were identified who completed both the ACT Interest Inventory and the IWRV once, and who identified themselves as “Job Seeking/Working Adults.” This sample was disproportionally female; to balance the effects of gender we selected all males and a random set of females of equal number. Although 2 of the 18 WFA work values (autonomy and variety) do not correspond to IWRV values, both of these values were piloted during IWRV development, so means and standard deviations (SD) for these two values were estimated using data collected at tryout. In sum, the norming sample consisted of 12,946 adults (equal numbers of males and females) who completed both the ACT Interest Inventory and the IWRV online. Responses to WFA items (a subset of ACT Interest Inventory items and IWRV items) were used to create norms involved in generating Fit Index scores for all 949 O*NET occupations, as described below.

Fit index. Operational calculation of fit in the WFA is based on the correlation between respondent profiles and corresponding occupational profiles for each O*NET occupation. One of the advantages of using correlation to measure profile similarity is that it focuses on profile shape and does not penalize for assessment score level differences due to individual response tendencies. Technical details on the Fit Index are available from ACT (2007a). The procedure for obtaining fit in the WFA can be summarized as follows:

A correlation was calculated to measure the similarity between the person’s six interest inventory scores and the corresponding six occupational ratings.

A correlation was calculated to measure the similarity between the person’s 18 responses to the work values inventory and the corresponding 18 occupational ratings.

The two scores were summed to form a raw score. Using the norms, each raw score was converted to a standard score (with a mean of 0 and an SD of 1) and a percentile score ranging from 1 to 99. This percentile score is referred to as the Fit Index, and was used for validation and the analyses presented here.

Other Measures

Participant ratings (work attitudes)

Rating scales were created to serve as work attitude criterion measures. The three outcomes, each related to participants’ current occupations, were (a) job satisfaction, (b) perceptions of job match, and (c) job commitment. The satisfaction criterion consisted of two general satisfaction questions (e.g., Overall, how satisfied are you with your job?). The internal consistency (coefficient α) for this scale was .68. The job match criterion, consisting of three questions examining the extent to which participants perceived their current job as matching their interests and values (e.g., To what extent does your current job enable you to do the kind of work you want to do?) attained an internal consistency coefficient of .78. The job commitment criterion, consisting of two questions asking participants to estimate their commitment to their occupation (e.g., How committed are you to staying at your current place of employment?) attained an internal consistency reliability of .78. The satisfaction questions were completed by all field study participants, whereas the job match and job commitment sets of questions were added while the study was in progress and were only completed by a subset of 219 participants.

Integrity

A subset of 242 field study participants completed the WorkKeys Performance Assessment (WPA; ACT, 2007b), a self-report assessment of integrity designed to measure personality and behavior characteristics related to counterproductive work behaviors such as absenteeism, violation of work rules, and hostility in the workplace. The WPA is used in employee screening and selection to identify individuals who may be prone to such counterproductive work behaviors. It contains 60 self-report questions written at the fifth-grade reading level, yielding an overall score—the Performance Index—based on two subscale scores: General Work Attitudes and Risk Reduction. The general work attitudes subscale assesses an individual’s overall approach to their work and their work environment. Items focus on the ability to communicate and relate to others, and level of productivity. An example item is “I am easily irritated by coworkers.” The risk reduction subscale contains items that focus on compliance with safety rules and procedures, as well as unnecessary risk-taking in a work environment. The risk reduction subscale focuses on an individual’s attitudes about safety procedures or improper operation of machinery. An example item is “Some safety regulations are overprotective and should not be followed.” Internal consistency of the scales was examined for 208 workers representing nine organizations spanning a range of industries, such as manufacturing, health care, and education (ACT, 2007b). Coefficient αs were .79 (General Work Attitudes), .82 (Risk Reduction), and .89 (Performance Index).

Supervisor ratings (job performance)

Supervisor rating scales were created to serve as criterion measures of job performance for the 242 individuals who completed the integrity test. These scales enabled supervisors to assess employee performance along a standard set of workplace performance dimensions. The primary reason for creating and using these research-only job performance scales was previous findings that typical organizational/administrative ratings tend to suffer more from halo effects than ratings collected for research purposes (Jawahar & Williams, 1997). Further, Rotundo and Sackett (2002) found that supervisors’ global ratings of performance were affected by their differential weighting of employee task performance, prosocial/organizational citizenship behaviors, and counterproductive work behaviors. Accordingly, scales were created to capture each of these three dimensions. A fourth scale was created to measure employee safety behaviors (e.g., adherence to safety protocols, frequency of workplace accidents) because a large number of employees included in the sample worked in occupations where safety is important (e.g., transportation and materials moving). Each supervisor scale (ranging in length from 5 to 8 items) demonstrated moderate to excellent internal consistency reliability (α range = .72 to .96). Finally, an overall measure of job performance was created by aggregating scores from the four supervisor scales.

Results

Reliability

Internal consistency reliability estimates (Cronbach’s α) across the six scales of the interest inventory ranged from .77 to .85, with a mean of .81. Internal consistency reliability estimates are not appropriate for the work values inventory, because the items are not intended to relate to one another.

Validity

Differentiation

To the degree that people gravitate toward work environments that are in accord with their personal characteristics, we would expect a worker to display higher levels of fit to their current occupation (or occupations that are similar to their current occupation) than to dissimilar occupations. Thus, a valid measure of fit should differentiate between occupations that are similar to individuals’ current occupations and those that are not. One way to examine this is presented in Table 5. This table shows, for three levels of occupational similarity, the percentage of Fit Index scores falling in each of three score levels: low (1–25), medium (26–75), and high (76–99). The first row, called “Same Occupation,” refers to Fit Index scores for the current occupation of study participants. The second row, called “Similar Occupations,” refers to Fit Index scores for all occupations in the same O*NET major occupational group as the current occupation of participants (excluding the current occupation itself). Thus, it shows Fit Index scores for similar occupations as defined by O*NET. The third row, called “Dissimilar Occupations,” refers to Fit Index scores for all occupations not in the O*NET major occupational group of the current occupation of participants—that is, for dissimilar occupations as defined by O*NET.

Table 5.

Percentage of Fit Index Scores by Score Level and Occupational Similarity

		Fit score level
Occupational Similarity	N of Scores^a	Low	Medium	High	Median fit score
Same occupation	503	8	42	50	77
Similar occupations	29,154	10	47	43	72
Dissimilar occupations	447,690	22	52	26	50

Note. N = 503.

^aNumber of fit score calculations.

As seen in Table 5, Fit Index scores vary considerably by level of occupational similarity. Fifty percent of scores based on current occupation fell in the high level, dropping to only 26% for dissimilar occupations. This pattern reverses for low levels of fit. This provides evidence that the Fit Index score differentiates occupations in ways we would expect given the assumptions underlying the concept of occupational fit. We used Friedman’s Nonparametric Test to test the null hypothesis that the average ranks of the fit scores were the same for the three levels of occupational similarity. This hypothesis was rejected (χ ² = 30.6, p < .0001), confirming that the Fit Index scores vary by level of occupational similarity.

Relations with job satisfaction, job match, and commitment

Table 6 displays observed (uncorrected) and operational (corrected) validities for the three work attitude outcomes related to participants’ current occupations: job satisfaction, perceptions of job match, and job commitment. Because observed validity estimates tend to be attenuated by a variety of biasing effects, such as measurement error in the criterion and range restriction, one cannot rely on observed validity as a final estimate of the operational validity of a test. To obtain the “true” (i.e., operational) validity of a test, psychometric techniques can be used to correct for biasing effects. We corrected the observed validities of the Fit Index shown in Table 6. First, we corrected for criterion unreliability using coefficient α. Second, we corrected the validity estimates for predictor range restriction in the Fit Index scores. This restriction occurred because the field study involved incumbent employees whose occupational fit is greater, and less variable, than the fit that would be observed in a general applicant pool. We assumed a range restriction ratio (u _x) of .88 given the field sample SD was 0.88 and the norm sample SD was 1.00. As shown in Table 6, the Fit Index was a statistically significant predictor of all three work attitude outcomes. The corrections resulted in modest gains in predictive power, for example, the observed validity of the Fit Index on job commitment (.17) increased to .24.

Table 6.

Observed and Corrected Correlations of Fit Index With Work Attitudes

		Operational validity
Work Attitude Criterion	Obs r	cME	cRR
General Satisfaction^a	.09	.11	.14
Job Match^b	.21	.24	.29
Job Commitment^b	.17	.19	.24

Note. Criterion items are discussed in the text. cME = corrected only for measurement error in criterion; cRR = cME further corrected for indirect range restriction in the predictor; Obs r = observed correlation.

^aBased on 503 persons representing 21 O*NET major occupational groups. Correlations ≥ .09 are significant (p ≤ .05). That is, the 95% confidence intervals do not include zero for correlations ≥ .09.

^bBased on 219 persons representing 11 O*NET major occupational groups. Correlations ≥ .13 are significant (p ≤ .05). That is, the 95% confidence intervals do not include zero for correlations ≥ .13.

Relations with job performance

Table 7 presents the operational validities of fit for predicting supervisor ratings of job performance. The Fit Index was a statistically significant predictor of both task performance (corrected r = .29) and overall job performance (corrected r = .22). Table 8 shows the results of incremental validity models in which we calculated multiple Rs for an integrity test (the WPA), as well as for the integrity test plus the Fit Index, on the four specific (i.e., task, citizenship, counterproductive, and safety) and overall criterion measures of job performance described above. Hierarchical regression was used to examine the incremental operational validity of the fit measure above the integrity test. Using Hunter’s (1992) regression program, we regressed each of the job performance dimensions onto integrity test scores (Step 1) followed by fit scores (Step 2). There was a significant increase in the magnitude of multiple R for the task (74%) and overall job performance (26%) criteria. Significant increases in multiple R were not found for the other three performance dimensions (citizenship, counterproductive, and safety), suggesting that the incremental explanatory power of fit with respect to overall job performance was based primarily on task performance.

Table 7.

Observed and Corrected Correlations With Supervisor Ratings of Job Performance

		Operational validity
Work Criterion	Obs r	cME	cRR
Task performance	.19	.24	.29
Prosocial performance (OCBs)	.02	.03	.04
Counterproductive behaviors (CWB)	−.04	−.06	−.07
Safety	.09	.12	.15
Overall supervisor ratings	.13	.18	.22

Note. N = 242. cME = corrected only for measurement error in criterion; cRR = cME further corrected for indirect range restriction in predictor; Obs r = observed correlation. Correlations ≥ .12 are significant (p ≤ .05). That is, the 95% confidence intervals do not include zero for correlations ≥ .12.

Table 8.

Incremental Validity of Fit Over Integrity

		Multiple R ^a
		Task	OCB	CWB	Safety	Overall

Integrity	Observed	.14	.21	.04	.27	.20
	Corrected	.19	.30	.05	.38	.27
Standard error (corrected)		.09	.09	.09	.09	.09
95% CI lower bound		.02	.13	−.12	.21	.10
95% CI upper bound		.37	.47	.22	.55	.44
Integrity + Fit	Observed	.22	.21	.06	.28	.22
	Corrected	.33	.32	.09	.42	.34
Standard error (corrected)		.09	.09	.09	.09	.09
95% CI lower bound		.16	.14	−.09	.24	.16
95% CI upper bound		.51	.49	.27	.59	.51
Increase in multiple R		.14	.02	.04	.04	.07
% Increase in multiple R		74%	^b	^c	^b	26%

Note. N = 242.

^aCorrected for measurement error and indirect range restriction. CI = confidence interval; CWB = counterproductive work behaviors; OCB = organizational citizenship behaviors.

^bCare should be taken in interpreting the incremental validity for these performance dimensions as the 95% confidence intervals for the Fit Index included zero.

^cCare should be taken in interpreting the validity of Integrity as well as the incremental validity of the Fit Index for this dimension as the 95% confidence intervals for both Integrity and the Fit Index included zero.

Discussion

The results of the current study provide support for the idea that person–occupation fit can be a useful concept to consider in personnel staffing. Differentiation analyses offer evidence of construct validity for the person–occupation fit measure used in this study. Likewise, criterion-related validity correlations show that person–occupation fit predicts desirable work attitudes including job satisfaction, job match, and job commitment, as well as work outcomes including task and overall performance. Furthermore, fit was shown to provide incremental validity over integrity testing for task and overall performance. This suggests that organizations may be able to achieve increased validity and utility by adding an assessment of person–occupation fit to their existing selection arsenal (Van Iddekinge, Putka, & Campbell, 2011). Indeed, exploring this possibility in greater detail presents an exciting opportunity for future research.

There is a growing appreciation that personality and interests play an important role in understanding the multifaceted and interactive complexities of work behavior (Barrick & Mount, 2005; Murphy, 1996). The current study, along with other recent studies (e.g., Le et al., 2011; Van Iddekinge et al., 2011) shed light on some of these complexities. Nevertheless, our results appear to run counter to some prevailing views on the role of interests in work settings. It has been argued that personality and interests influence work behavior in different ways, specifically, that interest-environment fit primarily influences the choice of work environment type, whereas personality primarily influences behaviors on the job and plays the more central role among noncognitive variables in explaining outcomes in organizational contexts (Barrick & Mount, 2005). If these views are correct, then the results of the current study suggest the unlikely scenario that “pre-environment” choice behaviors add incremental validity over “in-environment” personality attributes—which in this case is integrity, widely considered to be the strongest noncognitive predictor of job performance (Schmidt & Hunter, 2004).

The incremental validity observed in the current study may instead indicate that the role of interests and values extends beyond the choice of work setting. Our results would be expected if person–occupation fit plays an important role in everyday, in-environment behaviors (putting effort into work tasks, setting goals, etc.) associated with job performance. In this regard, it is useful to call attention to the idea, proposed by many researchers over the years, that vocational interests are an expression of self-concept (Bordin, 1943; Hogan & Blake, 1996; Super, 1957). An individual’s personal narratives about who they are and how they wish to be regarded by others serve as standards for self-evaluation, setting the stage for goal-directed behavior. For example, when work environments change, self-evaluation of fit may be impacted, motivating people to modify their work behavior or environment to increase fit. These behaviors could relate to work commitment (such as seeking a different work environment) or job performance (such as learning new job skills). Seen in this light, self-evaluations of interest/value fit with an environment (imagined or experienced) evoke motivational, regulatory processes that impact judgments, attributions, and actions. The incremental validity observed in the current study suggests that these processes share commonalities with the role of integrity in regulating work behaviors, but do not fully overlap.

The current study has a number of strengths, including the inclusion of both interests and work values, extensive use of O*NET, the collection of supervisor ratings on multiple performance dimensions, and the assessment of incremental validity beyond another noncognitive predictor. On the other hand, our research is somewhat limited by the fact that a number of O*NET occupational groups were not represented in our sample. In addition, the use of the WFA to measure fit did not permit us to examine the separate impacts of interests and work values. Nevertheless, we believe our study advances knowledge regarding person–occupation fit and its effect on work outcomes.

To date, the concept of person–occupation fit has been used primarily within the sphere of career and vocational counseling. This type of fit underlies the logic of most assessments used in career planning, where the exploration of good-fit occupations can help counselees focus on a manageable set of personally relevant options. However, we hope our work encourages both researchers and practitioners to more fully explore the potential benefits of person–occupation fit within organizational contexts. There is certainly no shortage of questions awaiting research. Does job complexity and/or the situational strength of the work environment (structured vs. unstructured) moderate the relationship between person–occupation fit and work outcomes? Are there specific aspects of fit that can work synergistically with certain personality characteristics to better predict work outcomes? While the results of the current study suggest that person–occupation fit may be a useful concept in organizational contexts, a fuller picture of the benefits of fit in organizations awaits further research.

Footnotes

Acknowledgments

Thanks to Xuan Wang, Tamera McKinniss, and Ben Postlethwaite, for contributing to the validation and implementation of the WorkKeys Fit Assessment.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

ACT, Inc. (2006). Research support for DISCOVER assessment components. Retrieved from http://www.act.org/discover/pdf/research_support.pdf

ACT, Inc. (2007a). WorkKeys fit assessment: User and technical guide. Retrieved from http://www.act.org/WorkKeys/assess/pdf/FitUserGuide.pdf

ACT, Inc. (2007b). WorkKeys performance assessment: User and technical guide. Retrieved from http://www.act.org/WorkKeys/assess/pdf/PerformanceUserGuide.pdf

ACT, Inc. (2009). The ACT Interest Inventory Technical Manual. Retrieved from http://www.act.org/research/researchers/techmanuals.html

Adkins

C. L.

Ravlin

E. C.

Meglino

B. M.

(1996). Value congruence between co-workers and its relationship to work outcomes. Group and Organizational Management, 21, 439–460.

Arthur

Bell

S. T.

Villado

A. J.

Doverspike

(2006). The use of person-organization fit in employment decision making: An assessment of its criterion-related validity. Journal of Applied Psychology, 91, 786–801.

Barrick

M. R.

Mount

M. K.

(2005). Yes, personality matters: moving on to more important matters. Human Performance, 18, 359–372.

Berry

C. M.

Sackett

P. R.

Wiemann

(2007). A review of recent developments in integrity test research. Personnel Psychology, 60, 271–301.

Bobek

B. L.

Gore

P. A.

(2004). Inventory of work-relevant values: 2001 Revision (Report No. 2004-3). Iowa City, IA: ACT.

10.

Bordin

E. S.

(1943). A theory of vocational interests as dynamic phenomena. Educational and Psychological Measurement, 3, 49–65.

11.

Caldwell

D. F.

O’Reilly

C. A.

(1990). Measuring person-job fit with a profile-comparison process. Journal of Applied Psychology, 75, 648–657.

12.

Chatman

J. A.

(1989). Improving interactional organizational research: A model of person-organization fit. Academy of management Review, 14, 333–349.

13.

Chatman

J. A.

(1991). Matching people and organizations: Selection and socialization in public accounting firms. Administrative Science Quarterly, 36, 459–484.

14.

Dawis

R. V.

Lofquist

L. H.

(1984). A psychological theory of work adjustment. Minneapolis: University of Minnesota Press.

15.

Donohue

(2006). Person-environment congruence in relation to career change and career persistence. Journal of Vocational Behavior, 68, 504–515.

16.

Edwards

J. R.

(1991). Person-job fit: A conceptual integration, literature review, and methodological critique. In Cooper

Robertson

(Eds.), International review of industrial and organizational psychology (Vol. 6; pp. 283–357). New York, NY: Wiley.

17.

Erdheim

Zickar

M. J.

Yankelevich

(2007). Remembering Donald G. Paterson: Before the separation between industrial–organizational and vocational psychology. Journal of Vocational Behavior, 70, 205–221.

18.

Gati

Garty

Fassa

(1996). Using career-related aspects to assess person-environment fit. Journal of Counseling Psychology, 43, 196–206.

19.

Hoffman

B. J.

Woehr

D. J.

(2006). A quantitative review of the relationship between person-organization fit and behavioral outcomes. Journal of Vocational Behavior, 68, 389–399.

20.

Hogan

Blake

R. J.

(1996). Vocational interests: matching self-concept with the work environment. In Murphy

K. R.

(Ed.), Individual differences and behavior in organizations (pp. 89–144). San Francisco, CA: Jossey-Bass.

21.

Holland

J. L.

(1997) Making vocational choices: A theory of vocational personalities and work environments (3rd ed.). Lutz, FL: Psychological Assessment Resources.

22.

Hunter

J. E

. (1992). REGRESS: A multiple regression program in BASICA. User’s manual. Department of Psychology, Michigan State University, East Lansing, MI.

23.

Jawahar

I. M.

Williams

C. R.

(1997). Where all the children are above average: The performance appraisal purpose effect. Personnel Psychology, 50, 905–925.

24.

Judge

T. A.

Bretz

R. D.

(1992). Effects of work values on job choice decisions. Journal of Applied Psychology, 77, 261–271.

25.

Knafo

Sagiv

(2004). Values and work environment: Mapping 32 occupations. European Journal of Psychology and Education, 19, 255–273.

26.

Kristof

A. L.

(1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49, 1–49.

27.

Kristof-Brown

A. L.

(2000). Perceived applicant fit: Distinguishing between recruiters’ perceptions of person-job and person-organization fit. Personnel Psychology, 53, 643–671.

28.

Kristof-Brown

Zimmerman

R. D.

Johnson

E. C.

(2005). Consequences of individuals’ fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58, 281–342.

29.

Lawrence

A. D.

(2004). Screening for person-job fit: Incremental validity of congruence based approach to assessment (Doctoral dissertation, University of Akron). Dissertation Abstracts International, 65, 1060.

30.

Robbins

S. B.

Ilies

Holland

Westrick

(2011). Too much of a good thing: Curvilinear relationships between personality traits and job performance. Journal of Applied Psychology, 96, 113–133.

31.

Lofquist

Dawis

(1969). Adjustment to work. New York, NY: Appleton-Century-Crofts.

32.

Meglino

B. M.

Ravlin

E. C.

Adkins

C. L.

(1989). A work values approach to corporate culture: A field test of the value congruence process and its relationship to individual outcomes. Journal of Applied Psychology, 74, 424–432.

33.

Mount

M. K.

Muchinsky

P. M.

(1978). Person-environment congruence and employee job satisfaction: A test of Holland’s theory. Journal of Vocational Behavior, 13, 84–100.

34.

Murphy

K. R.

(1996). Individual differences and behavior in organizations: Much more than g. In Murphy

(Ed.), Individual differences and behavior in organizations (pp. 3–30). San Francisco, CA: Jossey-Bass.

35.

Nauta

M. M.

(2010). The development, evolution, and status of Holland’s theory of vocational personalities: Reflections and future directions for counseling psychology. Journal of Counseling Psychology, 57, 11–22.

36.

National Center for O*NET Development (2006). O*NET Production Database. Retrieved from http://www.onetcenter.org/database.html

37.

Ones

D. S.

Viswesvaran

Schmidt

F. L.

(1993). Comprehensive meta-analysis of integrity test validities: Findings and implications for personnel selection and theories of job performance. Journal of Applied Psychology Monograph, 78, 679–703.

38.

Pervin

L. A.

(1968). Performance and satisfaction as a function of individual-environment fit. Psychological Bulletin, 69, 56–68.

39.

Ployhart

R. E.

Schneider

Schmitt

(2006). Staffing organizations: Contemporary practice and theory. Mahwah, NJ: Lawrence Erlbaum.

40.

Prediger

D. J.

Staples

J. G.

(1996). Linking occupational attribute preferences to occupations (Report No. 96-3). Iowa City, IA: ACT.

41.

Rotundo

Sackett

P. R.

(2002). The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy-capturing approach. Journal of Applied Psychology, 87, 66–80.

42.

Rounds

J. B.

Dawis

R.V.

Lofquist

L. H.

(1987). Measurement of person-environment fit and prediction of satisfaction in the Theory of Work Adjustment. Journal of Vocational Behavior, 31, 297–318.

43.

Rounds

Smith

Hubert

Lewis

Rivkin

(1998). Development of occupational interest profiles (OIPs) for the O*NET. Raleigh, NC: Southern Assessment Research and Development Center, Employment Security Commission of North Carolina.

44.

Schmidt

F. L.

Hunter

J. E.

(1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262–274.

45.

Schmidt

F. L.

Hunter

J. E.

(2004). General mental ability in the world of work: occupational attainment and job performance. Journal of Personality and Social Psychology, 86, 162–173.

46.

Schneider

(1987). The people make the place. Personnel Psychology, 40, 437–453.

47.

Schneider

(2001). Fits about fit. Applied Psychology: An International Review, 50, 141–152.

48.

Schneider

Kristof-Brown

Goldstein

H. W.

Smith

D. B.

(1997). What is this thing called fit? In Anderson

Herriot

(Eds.), International handbook of selection and assessment. New York, NY: Wiley.

49.

Smart

J. C.

Elton

C.F.

McLaughlin

G. W.

(1986). Person-environment congruence and job satisfaction. Journal of Vocational Behavior, 29, 216–225.

50.

Spokane

A. R.

Meir

E. I.

Catalano

(2000). Person-environment congruence and Holland’s Theory: A review and reconsideration. Journal of Vocational Behavior, 58, 137–187.

51.

Super

D. E.

(1957). The psychology of careers. New York, NY: Harper.

52.

Swaney

Prediger

. (1985). The relationship between interest-occupation congruence and job satisfaction. Journal of Vocational Behavior, 26, 13–24.

53.

Ton

M. N.

Hansen

J. C.

(2001). Using a person-environment fit framework to predict satisfaction and motivation in work and marital roles. Journal of Career Assessment, 9, 315–331.

54.

Van Iddekinge

C. H.

Putka

D. J.

Campbell

J. P.

(2011). Reconsidering vocational interests for personnel selection: The validity of an interest-based selection test in relation to job knowledge, job performance, and continuance intentions. Journal of Applied Psychology, 96, 13–33.

55.

Verquer

M. L.

Beehr

T. A.

Wagner

S. H.

(2003). A meta-analysis of the relations between person-organization fit and work attitudes. Journal of Vocational Behavior, 63, 473–489.

56.

Walsh

W. B.

Craik

K. H.

Price

R. H.

(Eds.). (2000). Person-environment psychology: New directions and perspectives. Hillsdale, NJ: Lawrence Erlbaum.

57.

Zytowski

D. G.

(2004a). Work roles, values. In Spielberger

(Ed.), Encyclopedia of applied psychology (Vol. 3; pp. 715–717). San Diego, CA: Academic Press.

58.

Zytowski

D. G.

(2004b). Kuder career search user manual. Adel, IA: National Career Assessment Services.