THE PERCEIVED IMPORTANCE OF PERSON-JOB FIT AND PERSON-ORGANIZATION FIT BETWEEN AND WITHIN INTERVIEW STAGES

Transcription

1 SOCIAL BEHAVIOR AND PERSONALITY, 2005, 33(3), Society for Personality Research (Inc.) THE PERCEIVED IMPORTANCE OF PERSON-JOB FIT AND PERSON-ORGANIZATION FIT BETWEEN AND WITHIN INTERVIEW STAGES AICHIA CHUANG National Taiwan University, Taipei, Taiwan PAUL R. SACKETT University of Minnesota, MN, USA This study explored the perceived importance of person-job fit and person-organization fit between and within three interview stages (i.e., initial, final, and single stage). The initial and final interviews refer to the sequential interviews conducted in a multiple-stage selection process and the single interview refers to the only interview conducted before a final decision is made. Using 446 campus recruiters, it was shown that: 1) P-J fit was perceived as more important than P-O fit in the initial interview; 2) the importance of P-J fit becomes lower from the initial interview to the single interview and from the initial interview to the final interview; and 3) the importance of P-O fit becomes higher from the initial interview to the single interview and from the initial interview to the final interview. Empirical implications and future directions are discussed. The desirability of fit between individuals and their work environments has long been a research interest in organizational behavior and industrial/organizational psychology (e.g., Murray, 1938; Pervin, 1968; Schneider, 1987). The often broadly defined term, person-environment fit (P-E Aichia Chuang, Department of Business Administration, National Taiwan University, Taipei, Taiwan; Paul R. Sackett, Department of Psychology, University of Minnesota, MN, USA. This study is part of the first author's doctoral dissertation under the supervision of the second author. Thanks go to the dissertation committee members: Richard D. Arvey, John P. Campbell, Deniz S. Ones, and Connie R. Wanberg. An earlier version of this paper was presented at the 1st annual conference of the New Paradigms of Management in the 21st Century, Taipei, Taiwan. The authors also thank the editor and two anonymous reviewers for their helpful comments. Keywords: person-job fit, person-organization fit, interview stage. Please address correspondence and reprint requests to: Aichia Chuang, 811, Second Building of College of Management, National Taiwan University, No. 85, Section 4, Roosevelt Road, Taipei 106, Taiwan. Phone: ; Fax: ; <achuang@management.ntu.edu.tw> 209

2 210 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES fit), is in fact composed of many particular fit indices such as person-job fit, person-organization fit, person-vocation fit, and person-group fit (Kristof, 1996). These dimensions of P-E fit have contributed to areas of literature such as motivation (Hackman & Oldham, 1980), vocational choice and preparation (Bretz, Rynes, & Gerhart, 1993; Holland, 1973), work adjustment (Dawis & Lofquist, 1984), job satisfaction (Bizot & Goldman, 1993; Breeden, 1993), turnover (Saks & Ashforth, 1997), and managerial selection decision (Adkins, Russell, & Werbel, 1994; Hitt & Barr, 1989). Among the various types of fit, person-job and person-organization fits are the most studied. The P-J and P-O fit researchers have broadly investigated the validity of fit with regard to organizational variables such as selection criteria, hiring recommendations, job satisfaction, stress, organizational commitment, organizational identification, performance, participation, intention to quit, and occupational change (e.g., Adkins et al., 1994; Bretz et al., 1993; Cable & Judge, 1997; Caldwell & O Reilly, 1990; French, Caplan, & Harrison, 1982; Kolenko & Aldag, 1989; Kristof-Brown, 2000; Meglino, Ravlin, & Adkins, 1989; Saks & Ashforth, 1997; Wilk & Sackett, 1996). While fit studies have frequently used selection interviews as the device through which data were collected, very little attention has been paid to investigating the importance of P-J and P-O fit in various stages of the interview (e.g., initial interview, final interview, and single interview). Adkins et al. (1994) had suspected that the importance of the type of fit studied in the stage of the interview used may contribute to the level of significance of the fit variables. This study represents the first attempt to investigate the perceived importance of P-J and P-O fit between and within interview stages with a wide range of sample. The present study makes a twofold contribution to extant research on fit and interview literature. First, the study tests how the importance of P-J fit and P-O fit is respectively perceived across three types of interview stages. Second, the study empirically delineates the relative importance of P-J and P-O fit in each of the three interview stages. These will assist fit researchers to test propositions regarding fit and selection outcomes at the appropriate stage to eliminate possibilities of misconstruing constructs. For instance, if P-O fit is perceived as more important in later stages of an interview process, researchers are recommended to test how P-O fit and not P-J fit leads to selection recommendations in later stages. HYPOTHESES Previous research seemed to suggest that in a sequential selection process where more than one interview is conducted before the final offer is extended, job-related attributes and/or objective qualifications were the criteria recruiters

3 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 211 sought in the initial interviews (e.g., Bretz et al., 1993; Rynes & Gerhart, 1990). In fact, studies have implied that different selection criteria (e.g., P-J and P-O fit) may be sought in different stages of the interviews. In Bretz et al. (1993) and Kristof-Brown (2000), the authors found that although recruiters did consider both P-J and P-O fits, P-O fit was used less frequently. Bretz et al. interviewed on-campus recruiters about their opinions on the perceived fit of the applicants they had just interviewed. The majority of the considerations on the applicants fit was job-related. These authors suspected that the reason why P-O fit was not shown significantly as an indicator of fit was probably because the research setting was on-campus recruiting where P-J fit is anecdotally assumed to be the focus. They also suggested that P-O fit may be considered more in a later stage in the selection process. Likewise, in Study 2 of Kristof-Brown P-J fit as perceived by recruiters explained a greater amount of variance in hiring recommendations than did perceived P-O fit, suggesting that in the first interviews, as the research setting was, P-J fit weighted more in recommendability. The author, however, commented that in the later interviews P-O fit should be applied to distinguish among candidates. Another study by Adkins et al. (1994) supported the notion that P-O fit may be a criterion used in a later stage. These authors used corporate recruiters who conducted on-campus interviews and found a marginally significant relationship between P-O fit and invitation to a second interview. This led them to conjecture that their research setting (i.e., initial interviews) restricted the confirmation of their hypothesis. The studies described above implied that recruiters may seek congruence cues with differential levels of weight at different stages of the selection process. Specifically, what is implied from the studies is that P-J fit was often the most important criterion in hiring recommendations when the research setting was the initial interview. What are left unknown include: 1) the importance of P-J and P- O fits between various stages of interviews, and 2) the relative importance of P-J and P-O fits within each stage. Therefore, the current study was designed to investigate these two conundrums. With regard to the first question, we wished to test the relative emphasis recruiters allocate to P-J fit across initial and final interview stages and the relative emphasis recruiters allocate to P-O fit across initial and final interview stages. In the circumstances when only one selection interview is run before job offers are made, it is reasonable to postulate that in relations to initial and final interviews, a single interview will focus on P-J fit more than a final interview and less than an initial interview. Likewise, a single interview will emphasize P-O fit more than an initial interview and less than a final interview. Assuming fixed interview length, by definition, choosing to spend more time on job-related measures indicates less time on organization-related measures. Therefore, the following hypothesis about analyses between interview stages was formed:

4 212 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES Hypothesis 1: The perceived importance of P-J and P-O fit will each vary among interview stages. The perceived importance of P-J fit will become lower as the interview stage goes from the initial interview, to the single interview, and then to the final interview. Consequently, the perceived importance of P-O fit will become higher as the interview stage goes from the initial interview, to the single interview, and then to the final interview. In relation to the second question, we investigated the relative importance of P-J and P-O fits within interview stage. Based on the suggestion from the aforementioned literature (Adkins et al., 1994; Bretz et al., 1993; Kristof-Brown, 2000), it seemed reasonable to hypothesize the following regarding analyses within interview stages: Hypothesis 2a: The perceived importance of P-J and P-O fit will vary within interview stages. P-J fit emerges as a more important criterion than P-O fit in the initial interview. Hypothesis 2b: The perceived importance of P-J and P-O fit will vary within interview stages. P-O fit emerges as a more important criterion than P-J fit in the final interview. With the notion that P-J fit is traditionally perceived as the necessary qualification of the applicants (Chatman, 1989; Rynes & Gerhart, 1990), it seemed logical to state the following hypothesis: Hypothesis 2c: The perceived importance of P-J and P-O fit will vary within interview stages. P-J fit emerges as a more important criterion than P-O fit in the single interview. METHOD PARTICIPANTS Participants were 446 campus recruiters associated with the career services in a large midwestern university. The reason that campus recruiters were chosen to be the subjects was because they would be able to identify an extensive range of job positions. There were seven campus career services and one large news service organization involved in providing access to recruiters and/or in distributing the surveys. Recruiters were approached on one of the three occasions or by one of three methods: on the day of a campus interview, at the job fair, or by mail. Each career service participated in the study in a combination of these three different methods. A total of 1993 surveys were distributed and the final response rate was.22. The response rates by career services ranged from.14 to.34. The final sample was on average years old (SD = 9.89) and a majority of the sample was White (92.3%); male respondents constituted 55.3% of the sample. Thirty-five percent of the sample was human resources staff and

5 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 213 on average the sample had 6.88 initial interviews (SD = 9.96), 2.58 final interviews (SD = 4.13) and 1.82 single interviews (SD = 4.55) per month. SURVEY DEVELOPMENT Three versions of the survey were developed with each being correspondent to one of the three interview scenarios tested: initial interview scenario, final interview scenario, and single interview scenario. Each survey described its particular scenario and asked respondents to rate a total of 42 job- or organization-related criteria. The following paragraphs describe the 42 criteria and other variables used. Job-Related Criteria (P-J Fit) There were 28 applicant s characteristics designed to capture P-J fit which included six dimensions: the individual s possession of knowledge, skills, and abilities (KSAs), interests desired by the job, characteristics of the job, and personality to fit the job. The knowledge dimension was measured by 3 items: the applicant s prior course work, the applicant s job knowledge, and applicant s previous job-relevant work experience (Bretz et al., 1993; Hitt & Barr, 1989). The skill and ability dimensions were assessed by 9 items: problem-solving skills, social skills, skills in using relevant tools and equipment, writing skills, speaking skills, personnel management skills, time management skills, quantitative skills, and memorization abilities. They were items selected from the Occupational Information Network (O*Net). Essentially, most of the skill and ability items were interchangeable, so the assignment of items into the skill or ability category was based on the understandability of the term by the general cohort of recruiters. Sample items of the KSA criteria included: Whether the applicant s prior course work in educational programs matches what the job requires. and Whether the applicant s speaking skills (talking to others to effectively convey information) matches what the job requires. Applicants interest attributes employed Holland s (1973) hexagon traits (realistic, investigative, artistic, social, enterprising, and conventional) as they represent interests required by the job. A sample item included: Whether the applicant has high or low interests in jobs that are analytical, complex, and emphasize sophisticated thinking. Characteristics of the job came from the five attributes in Hackman and Oldham s (1976) job characteristics theory (skill variety, task identity, task significance, autonomy, and job feedback). A sample item included: Whether the applicant s desire for autonomy will be met by the autonomy level the job offers. Personality criteria adopted Big Five personality traits. A sample item included: Whether the applicant has a high or low level of personality of conscientiousness, reliability, and dependability as needed to be a good fit to the job. The reasons why these three models were chosen (i.e.,

6 214 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES Holland s hexagon, the job characteristics model, and the Big Five personality traits) were that they represent a comprehensive taxonomy in their own domain and when they were studied in a fit setting, they possessed solid criterionrelated validities with work outcomes. All items were selected so that they were broad characteristics that recruiters in any occupation would potentially pick to be selection criteria. For instance, a criterion regarding whether the applicant has the knowledge of manufacturing and production will be very unlikely for recruiters interviewing for a financial analyst s job to consider asking in an interview. Besides, all P-J fit characteristics were described in such a way that (1) they were requirements or supplies of the job, (2) they were desires of or characteristics possessed by the applicant, (3) they were framed as fit between the applicant and the job, and (4) they were easily understood by the recruiters. Organization-Related Criteria (P-O Fit) A total of 14 characteristics were used which included measures of values (11 items) and goals (3 items) as operationalizations of P-O fit. Seven of the value measures came from the factor analysis on the original 54 Organizational Culture Profile (OCP) items in O Reilly, Chatman, and Caldwell (1991). They included: respect for people, innovation, team orientation, employment stability, outcome orientation, attention to detail, and aggressiveness. Four of the value items came from the 4 work values in Ravlin and Meglino (1987) which included: fairness, honesty, helping others, and achievement. Sample items included: Whether the applicant places a high or low value on achievement as needed to fit the organization s work culture and Whether the applicant places a high or low value on innovation as needed to fit the organization s work culture. The reasons why these two sets of values were used were that they represented organization values generated by a broad range of people at different levels in a variety of organizations, that they were frequently employed when it comes to organization fit and that they had criterion-related validities with desirable outcome variables. The goal measures included 3 variables from Van Vianen (2000): reward, effort, and competition. A sample item read: Whether the applicant will feel comfortable with the reward system (i.e., base pay, performance standards, promotion, bonus, benefits) in my organization. As mentioned previously, the P-O fit characteristics were chosen because they are organizational values and goals common to many organizations. P-O fit criteria were described in such a way that (1) they were work cultures or goals valued by the organization, (2) they were values or goals possessed by the applicant, (3) they were framed as fit between the applicant and the organization, and (4) they were easily understood by the recruiters. In an effort to ensure that respondents were unaware of the categorization of the items to avoid any response set, the 42 items were randomly shuffled so that

7 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 215 they were not in any particular order or categorization. The respondents were asked to rate each of the 42 items according to the stem: In an initial interview scenario, for the job I interview for the most on campus, how likely is this question to be among my top 10 questions? (the stem for initial interview scenario). The scale anchors ranged from 1 (not at all likely) to 5 (absolutely). Other Variables In an effort to examine whether a direct question captured the same information regarding recruiters fit emphasis as the 42 multiple items, a question: When you conduct an initial interview, what is the relative emphasis on the fit of the applicant to the job vs. the fit of the applicant to the organization? (the question for the initial interview scenario) was asked. The rating scale of this particular question ranged from 1 to 5 which corresponded with the 5 responses: I emphasize only the fit to the job, I emphasize fit to the job more than fit to the organization, I put equal emphasis on fit to the job and fit to the organization, I emphasize fit to the organization more than fit to the job, and I emphasize only the fit to the organization. The means of this direct question for the initial, final, and single stage were 2.89, 3.05, and 3.00 respectively, and the standard deviations were.86,.67, and.78 separately. Other variables were for control purposes and consisted of: gender (1 = male, 0 = female), age, ethnicity, position (1 = human resources staff, 0 = staff outside the human resources department), interview experiences (i.e., the number of initial, final, and single interviews conducted per month), career services, industry of the organization, scope of the organization (i.e., local, regional, national, multinational, and government), the occupation of the job used to help rate the job- and organization-related items, and intern job (1 = intern job, 0 = nonintern job). PROCEDURES The respondents were approached on one of the three following occasions: campus interview, job fair, or by mail. The first method involved approaching recruiters on the day that they were on campus for interviews with job candidates. Since there were three versions of the survey (i.e., initial, final, and single interviews), career services were told to randomly assign the surveys to recruiters who came on the same day and to those who were from the same organization. The randomization was applied to all three methods of approach. The total number of surveys distributed via this method was 528 of which 164 (.31) were returned. With regard to approaching recruiters at the job fair, the primary investigator was permitted by the career services to invite the recruiters to participate in the study as they were awaiting students at the table. They were provided with a business reply envelope with the survey. The total number of surveys distributed through the fairs was 265 of which 59 (.22) were returned. With regard to surveying recruiters by mail, mailing lists of the previous year s

8 216 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES campus recruiters, job fair recruiters, and/or organizational contacts were obtained from six career services. A cover letter was included in the mail out to explain the reason that they were invited to participate in the study and to ask them to direct the survey to their colleague who had had campus interview experiences. The total number of surveys distributed via mailing was 1184 of which 213 (.18) were returned. In an effort to improve the response rates, two rounds of reminders were sent to participants with available addresses, and one round of post card reminders to those without. Although a distinction was not made on surveys returned before and after the reminders, it was felt that the reminders helped bring back a fair number of surveys. Each participant was asked to fill out a survey which assigned him/her to an initial interview, final interview, or a single interview scenario. Participants were guided by the instruction, which familiarized them with the scenario and their tasks. For instance, for an initial interview survey, respondents were presented with the definitions of different types of interviews and asked to pretend that they were a recruiter conducting an initial interview in their organization. They had to assess 20 applicants, 5 of whom they would choose to invite for a final interview. They were told that a typical resumé of each applicant was provided to them prior to the interview. This was to help replicate the selection procedure in real organizations and help explain why some apparent criteria, such as Grade Point Average (GPA) and education, were not included in the 42 P-J and P-O list. In addition, subjects were told that they had only 30 minutes to interview each applicant and could choose only 10 questions out of the 42 to ask due to the time constraint. Before proceeding to the stem, respondents were asked to recall the job title for which they interview the most on campus and keep that in mind while completing the survey. The stem then read: In an initial interview scenario, for the job I interview the most on campus, how likely is this question to be among my top 10 questions? They were asked to read all 42 items before starting to fill out the survey. Out of the total 446 subjects, 157 of them took the initial interview survey, 133 the final interview survey, and 156 the single interview survey. RESULTS It was noticed that the mean of each recruiter s 28 P-J fit items and the mean of each recruiter s 14 P-O fit items had a significant and fairly high correlation (r =.63, p <.01). The fact that the one-way ANOVA showed that rater did have an effect in the ratings (sum of squares = , p <.01) confirmed that the high correlation of these two variables was due to the rating style of the individual raters which inflated or deflated the ratings. This rater effect clouded the real relationship between P-J and P-O fits. To partial out this effect, the 42 ratings were standardized within rater which then yielded a significant

9 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 217 TABLE 1 MEANS, STANDARD DEVIATIONS, AND INTERCORRELATIONS Measure M SD Emphasis on Fit a 1. Importance of P-J Fit KSAs Interests Job Characteristics Personality Importance of P-O Fit Goals Values Direct Fit Emphasis Interview Stages 10. Initial Final Single Recruiter Variables 13. Age Gender Ethnicity Position Initial Interview. Experience Final Interview. Experience Single Interview. Experience Job Variable 20. Intern Job Note: N = 446. Correlations greater than.093 are significant at p <.05. a Emphasis on Fit variables is standardized within rater except that the direct fit emphasis variable is unstandardized.

10 218 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES correlation of 1 (p <.01). Table 1 presents means, standard deviations, and intercorrelations for the measures. BETWEEN INTERVIEW STAGE ANALYSES The aim of hypothesis 1 was to test whether the importance of P-J fit becomes lower from the initial interview, to the single interview, and then to the final interview, and consequently whether the importance of P-O fit becomes higher from the initial interview, to the single interview, and then to the final interview. Due to the fact that the job and organization ratings were standardized, results of the P-O fit regression were identical to those of the P-J fit regression except that the signs of the coefficients and t-statistics were reversed. Therefore, only the results for the importance of P-J fit are presented (see Table 2). To test this hypothesis, the perceived importance of P-J fit (the mean of each recruiter s 28 standardized job-related items) was used as the dependent variable. The interview stages (i.e., initial, final, and single interviews) were coded as three dummy variables to insert as regressors. The direct fit emphasis question was also run with the same set of independent variables. It was used to determine whether or not the direct question yielded the same results as the multipledimension job and organization items. TABLE 2 REGRESSION ESTIMATES FOR THE PERCEIVED IMPORTANCE OF P-J FIT AND THE DIRECT FIT EMPHASIS MODEL Variables Importance of P-J Fit β Direct Fit Emphasis β Stage a Single (vs. Initial) *.157 Final (vs. Single) Final (vs. Initial) **.228 Demographics Age *.001 Gender Ethnicity Position *.267 ** Interview Experiences Initial.002 * Final *.014 Single Intern Job Occupation Administrative and Managerial Professional.075 ** Technicians and Support.111 ** -.468

11 Table 2 continued P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 219 Variables Importance of P-J Fit β Direct Fit Emphasis β Administrative Support Other Career Services Technology Management Human Ecology Agricultural, Food, & Environmental Science Public Affairs.058 * Journalism b Biological Science Organization Scope Local Regional National Industry Agricultural, Forestry, & Fisheries Construction Manufacturing.040 *.200 Transportation & Communications.067 * Wholesale Trade Retail Trade Business and Repair Services.043 *.212 Other Services.045 * Government.102 ** Interactions Initial Experience X Single Stage (vs. Initial Stage) c.004 * Initial Experience X Single Stage (vs. Final Stage) d Final Experience X Single Stage (vs. Final Stage) e Total R **.121 Adjusted R n Note: The dependent variable was standardized within rater; p <.10, * p <.05, ** p <.01 (twotailed tests). a Weights presented are the difference between the two stages (stage in the parenthesis is the omitted variable). b Represents campus recruiters surveyed via the news service organization who were approached at the job fair held by the organization for university students. c, d, e Only significant interactions between interview experience and interview stage are reported (stage in the parenthesis is the omitted variable).

12 220 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES Hypothesis 1 was partially confirmed. The standardized regression coefficient, comparing the importance of fit between the initial interview and the single interview, equaled (p <.05). This indicated that the importance of P-J fit was lower in the single interview than in the initial interview and that the importance of P-O fit was higher in the single interview than in the initial interview. The standardized regression coefficient, comparing the importance of fit between the initial interview and the final interview, equaled (p <.01). This indicated that the importance of P-J fit was lower in the final interview than in the initial interview and that the importance of P-O fit was higher in the final interview than in the initial interview. The initial interview experience was also found to moderate how importance of P-J/P-O fit was perceived from the initial interview to the single interview. With respect to the importance of P-J fit, based on the positive coefficient on the variable initial experience X single stage (vs. initial stage) and the previous conclusion that the importance of P-J fit dropped from the initial interview to the single interview, one can conclude that although generally speaking recruiters emphasized P-J fit less from the initial to the single interview, those with more initial experience did not change their emphasis on P-J fit as much as did those with less initial experience. By the same token, with respect to the importance of P-O fit, although generally speaking recruiters put a greater focus on P-O fit from the initial interview to the single interview, those with more initial experience put less. The part of hypothesis 1 that was not confirmed was the comparison between final and single stages on the importance of P-J/P-O fit. This relationship was found to be moderated by the recruiter s initial and final interview experiences [see variables initial experience X single stage (vs. final stage) and final experience X single stage (vs. final stage) in Table 2]. Based on the positive coefficient of the variable initial experience X single stage (vs. final stage), it can be concluded that the more initial interview experience a recruiter had, the more he or she added to the emphasis on P-J fit from a final interview to a single interview. For P-O fit, it will be: the more initial interview experience a recruiter had, the less he or she retained the emphasis of P-O fit from a final interview to a single interview. Also the negative coefficient of the variable final experience X single stage (vs. final stage) can be interpreted as the more final interview experience a recruiter had, the less he or she added to the emphasis on P-J fit from a final interview to a single interview. For P-O fit, it can be interpreted as: the more final interview experience a recruiter had, the more he or she emphasized P-O fit from a final interview to a single interview. These observations indicated that the more experience a recruiter had with a certain type of interview, the more effect that experience would have on the level of emphasis he or she placed on P-J fit.

13 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 221 With regard to the direct fit emphasis question, the issue is whether it captures the same constructs as the multiple ratings. In Table 1, the direct question, with higher scores indicating emphasis on P-O fit and lower scores indicating P-J fit, was found to be positively correlated with the importance of P-O fit (r =.19, p <.01) and negatively with P-J fit (r = -.19, p <.01). However, the magnitudes were not particularly large. Furthermore, a multiple regression was conducted with the direct question being the dependent variable and the results are shown in Table 2. They indicate that the direct question and the multiple ratings did not tap the same constructs. The interview stage variables hardly predicted the emphasis on P-J and P-O fit released by the direct question. Also, only a couple of other control variables were significant. This finding was startling and suggested that a different conclusion may result when one chooses to use a direct measure over multiple measures. A pattern with almost nothing predicting the fit emphasis seems to pose a warning against using an overall question, which is determined by an idiosyncrasy of the recruiters. It suggests that multiple measures, which are more elaborate, should be employed whenever it is possible. Therefore, the current authors chose to interpret results with the model using the multiple P-J and P-O fit measures. WITHIN INTERVIEW STAGE ANALYSES Hypotheses 2a, 2b, and 2c aimed to test the relative importance of P-J and P- O fit within the initial interview, final interview, and the single interview, respectively. A one-sample t-test using the standardized ratings was performed for each hypothesis. Because the job-related items and organization-related items were perfectly correlated (r = -1), a one-sample t-test on the mean of job-related items was judged to be sufficient for testing each hypothesis. The value used to test against was 0 due to the fact that the ratings were standardized within rater. Hypothesis 2a predicted that in the initial interview, P-J fit would be a more important criterion than P-O fit. The one-sample t-test on the mean of job-related items showed that there was a mean difference between job- and organizationrelated items in the initial interview stage (M =.015, p <.05, one-tailed). Therefore, hypothesis 2a was supported. Hypotheses 2b and 2c were not confirmed by the one-sample t-tests (M = -.004, p >.10, one-tailed; M =.001, p >.10, one-tailed) which indicated no difference between the importance of P-J fit and the importance of P-O fit in the final and the single interviews, respectively. DISCUSSION The past decade has witnessed a steady progress in assessing personenvironment fit both in terms of its definitions and its impact on outcome

14 222 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES variables (Edwards, 1991; Kristof, 1996). Research has revealed some common themes of person-environment fit and two essential components have been identified: person-job fit and person-organization fit. The studies of P-J and P-O fits have included not only delineating the dimensions but also searching for criterion-related validity (e.g., Adkins et al., 1994; Cable & Judge, 1997; Kristof- Brown, 2000; O Reilly et al., 1991; Saks & Ashforth, 1997). The purpose of this study was to contribute to this field by empirically testing the perceived importance of P-J and P-O fits in the employment selection process. IMPORTANCE OF P-J AND P-O FIT The study tested recruiters perceived importance of P-J and P-O fit between and within three interview stages. An interesting pattern of findings emerged from the results. Recruiters are inclined to view applicants P-J fit as more important than their P-O fit in the initial interview of a sequential selection process. They then tend to increase the weight they put on the P-O fit as the applicants come back for final interviews and at the same time to decrease the focus on P-J fit due to time constraints. However, this propensity does not increase their emphasis on P-O fit to such an extent that it takes over P-J fit in the final interviews. In addition, this study also found that in the circumstances when there is only one interview conducted before selection decisions are made, recruiters tend to focus more on P-O fit and less on P-J fit than in the initial interview probably to make sure that they capture both types of characteristics. However, the P-O fit still does not precede the importance of P-J fit in the single interview situation. The findings imply that in any case, an applicant s fit with the job is still the number one criterion to fulfill from the organization s point of view. A goal of this current study was to answer the question fit researchers have been pondering: where does P-O fit fit? Previous studies have demonstrated a possible role of P- J fit in initial interviews (Adkins et al., 1994; Bretz et al., 1993; Kristof-Brown, 2000), but no one has found evidence of where P-O fit plays a role. This study has an answer to that question. As previously explained, the importance of P-O fit is found to become higher from an initial interview to a final interview. Although the study does not find P-O fit to have a more important role than P-J fit in any of the stages, its significance has been recognized. This may be an accurate reflection of the observation that organizations have been looking for candidates with organization-related characteristics to adjust to the change of job specifications. This suggests that future research could investigate whether the greater emphasis on P-O fit does indeed result in positive outcomes for the employees and the organization.

15 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 223 EMPIRICAL IMPLICATIONS AND FUTURE DIRECTIONS The finding that P-J fit is looked at more than P-O fit in the initial interview implies that recruiters do use a screening test for the first round interview and that the screening criteria are job-related characteristics (i.e., P-J fit). This may explain why Adkins et al. (1994), using campus recruiters, found a marginal relationship between P-O fit and invitation to a second interview. Campus interviews are generally initial interviews in which recruiters focus more on P-J fit than on P-O fit. When the variation of P-O fit is small and not reliable, a pattern of relationship between P-O fit and other variables may not be found. Therefore, future research should adopt the more appropriate method; that is to assess only P-J fit in the initial interview and to leave P-O fit for later stages. This study also described the appropriateness of using a direct (overall) question to gauge fit. When the direct question was correlated with the index of the multiple job and organization items, they related significantly but yet to a marginal degree. When a multiple regression was run with the multiple-item index and the direct question being two separate dependent variables with the same set of regressors, the direct question model showed no pattern compared to a clean model with the multiple items. The implication is that when the dimensions of fit are not specified and recruiters are left guessing what fit means, they form their own understanding of fit which is elusive even when it has been the subject of long-term study (Rynes & Gerhart, 1990). Thus, it is very likely that the direct question captures nothing but recruiters idiosyncrasy. This finding about single-item vs. multiple-item questions may shed light on the comparison between subjective fit and objective fit. The subjective fit asks respondents a direct question about their perceptions of the fit between themselves and the job or organization. The objective fit uses a mathematical congruence fit index calculated from a computation of paired attribute profiles with the aid of instruments. The finding of the present study supports the idea of using objective fit (i.e., using multiple job and organization fit dimensions) rather than a subjective fit. Although some studies have used an overall fit question, they have also employed a different sample to secure the construct validity. It should still be recommended that, in future research, a multiple-item scale be used to measure fit. STUDY STRENGTHS AND LIMITATIONS One of the strengths of this study is that it surveyed a broad range of recruiters from different career services, organizational scopes, and industries, and therefore, employment selection information on jobs of diverse occupations was obtained. Also, it is possible to generalize the results to a wide range of population. However, since the literature in the importance of P-J and P-O fits is scant, the present study is exploratory in nature in terms of the control variables

16 224 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES included. Otherwise, a more sophisticated model could be built to elicit the effect of interview stages on the importance of P-J and P-O fits. A caution in interpreting the results of the study is that they pertain only to entry level jobs. The subjects used in this study are campus recruiters who were asked to recall the job for which they interview the most on campus as an anchor in completing the survey. Based on the fact that a substantial number of campus interviews are for entry level jobs, the results cannot be generalized to higher level jobs. Future research is suggested to obtain job information on a greater range of levels to add to the findings of this study. REFERENCES Adkins, C. L., Russell, C. J., & Werbel, J. D. (1994). Judgments of fit in the selection process: The role of work value congruence. Personnel Psychology, 47, Bizot, E. B., & Goldman, S. H. (1993). Prediction of satisfactoriness and satisfaction: An 8-year follow up. Journal of Vocational Behavior, 43, Breeden, S. A. (1993). Job and occupational change as a function of occupational correspondence and job satisfaction. Journal of Vocational Behavior, 43, Bretz, R. D., Jr., Rynes, S. L., & Gerhart, B. (1993). Recruiter perceptions of applicant fit: Implications for individual career preparation and job search behavior. Journal of Vocational Behavior, 43, Cable, D. M., & Judge, T. A. (1997). Interviewers perceptions of person-organization fit and organizational selection decisions. Journal of Applied Psychology, 82, Caldwell, D. F., & O Reilly, C. A. (1990). Measuring person-job fit with a profile-comparison process. Journal of Applied Psychology, 75, Chatman, J. A. (1989). Improving interactional organizational research: A model of personorganization fit. Academy of Management Review. 14, Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work adjustment: An individualdifferences model and its applications. Minneapolis: University of Minnesota Press. Edwards, J. R. (1991). Person-job fit: A conceptual integration, literature review, and methodological critique. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology (pp ). New York: John Wiley & Sons. French, J. R. P., Jr., Caplan, R. D., & Harrison, R. V. (1982). The mechanisms of job stress and strain. New York: John Wiley & Sons. Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Decision Processes, 16(2), Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley. Hitt, M. A., & Barr, S. H. (1989). Managerial selection decision models: Examination of configural cue processing. Journal of Applied Psychology, 74, Holland, J. L. (1973). Making vocational choices: A theory of careers. Englewood Cliffs, NJ: Prentice Hall. Kolenko, T. A., & Aldag, R. J. (1989). Congruence perceptions and managerial career/work outcomes: An exploratory analysis. Paper presented at the meeting of the National Academy of Management. Kristof, A. L. (1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49, 1-49.

17 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES 225 Kristof-Brown, A. L. (2000). Perceived applicant fit: Distinguishing between recruiters perceptions of person-job and person-organization fit. Personnel Psychology, 53, 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, Murray, H. A. (1938). Explorations in personality. New York: Oxford University Press. O Reilly, C. A., Chatman, J. A., & Caldwell, D. F. (1991). People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Journal, 34, Pervin, L. A. (1968). Performance and satisfaction as a function of individual-environment fit. Psychological Bulletin, 69, Ravlin, E. C., & Meglino, B. M. (1987). Effect of values on perception and decision making: A study of alternative work value measures. Journal of Applied Psychology, 72, Rynes, S., & Gerhart, B. (1990). Interviewer assessments of applicant fit : An exploratory investigation. Personnel Psychology, 43, Saks, A. M., & Ashforth, B. E. (1997). A longitudinal investigation of the relationships between job information sources, applicant perceptions of fit, and work outcomes. Personnel Psychology, 50, Schneider, B. (1987). The people make the places. Personnel Psychology, 40, Van Vianen, A. E. M. (2000). Person-organization fit: The match between newcomers and recruiters preferences for organizational cultures. Personnel Psychology, 53, Wilk, S. L., & Sackett, P. R. (1996). Longitudinal analysis of ability-job complexity fit and job change. Personnel Psychology, 49,

18 226 P-J FIT & P-O FIT BETWEEN AND WITHIN INTERVIEW STAGES