|
COPYRIGHT 2004 Personnel Psychology, Inc.
An updated theoretical model of applicant reactions to selection procedures is proposed and tested using meta-analysis. Results from 86 independent samples (N = 48,750) indicated that applicants who hold positive perceptions about selection are more likely to view the organization favorably and report stronger intentions to accept job offers and recommend the employer to others. Applicant perceptions were positively correlated with actual and perceived performance on selection tools and with self-perceptions. The average correlation between applicant perceptions and gender, age, and ethnic background was near zero. Face validity and perceived predictive validity were strong predictors of many applicant perceptions including procedural justice, distributive justice, attitudes towards tests, and attitudes towards selection. Interviews and work samples were perceived more favorably than cognitive ability tests, which were perceived more favorably than personality inventories, honesty tests, biodata, and graphology. The discussion identifies remaining theoretical and methodological issues as well as directions for future research.
**********
Research in personnel selection traditionally has focused on understanding the process from the perspective of the organization. Studies concerning the validity and utility of selection techniques have demonstrated how organizations can benefit from using valid selection tools. Researchers have also developed an interest in examining selection from the applicant's perspective, recognizing that not only do companies select employees, but applicants also select the organizations to which they will apply and where they are willing to work (Rynes, 1993). Thus, as research continues with the goal of better estimating the predictive value of selection devices, a related concern is in understanding how applicants perceive and react to the selection process.
Studying applicant reactions is important for at least five reasons. First, applicants who find particular aspects of the selection system invasive may view the company as a less attractive option in the job search process. Maintaining a positive company image during the selection process is of significant importance as there are costs associated with losing top candidates (Murphy, 1986). Second, candidates with negative reactions to a selection experience might dissuade other potential applicants from seeking employment with the organization (Smither, Reilly, Millsap, Pearlman, & Stoffey, 1993). Third, candidates may be less likely to accept an offer from a company with selection practices that are perceived unfavorably (Macan, Avedon, Paese, & Smith, 1994). Fourth, applicant reactions may be related to the filing of legal complaints and court challenges. Applicants who perceive a particular selection technique as invasive or inappropriate may be more likely to bring suit than applicants who perceive the process as fair and face valid (Smither et al., 1993). Finally, although there is little empirical data on these issues, it is also possible that applicants may be less likely to reapply with an organization or buy the company's products if they feel mistreated during the selection process.
This study begins with a discussion of conceptual models of applicant reactions and a brief review of relevant research. Next, we offer an updated theoretical framework, suggest several hypotheses, and test the model using meta-analysis. Overall, this research provides empirical answers to questions about the theoretical and practical value of studying selection from the applicants' perspective.
Theoretical Foundation
The term applicant reactions has been used to refer to the growing body of literature that examines "attitudes, affect, or cognitions an individual might have about the hiring process" (Ryan & Ployhart, 2000, p. 566). One of the first theoretical models of applicant reactions was an effort to the existing research to organizational justice theory in order to explain how applicants' justice perceptions develop and subsequently affect various outcomes in selection settings (Gilliland, 1993). Organizational justice generally involves the perceived fairness of: (a) outcome allocations (distributive justice), (b) rules and procedures used to make those decisions (procedural justice), (c) sensitivity and respect shown to individuals (interpersonal justice), and (d) explanations and accounts given to individuals (informational justice; Greenberg, 1993). The basic premise of organizational justice theory in selection contexts is that applicants view selection procedures in terms these four facets of justice, and these perceptions influence future attitudes, intentions, self-perceptions, and behaviors.
A more recent general model of applicant reactions has emerged that builds upon this initial theoretical framework to include additional antecedent and moderator variables (Ryan & Ployhart, 2000). In addition to justice considerations, the model includes perceptions of one's affective and cognitive states during the process and general perceptions about testing and selection as possible determinants of various personal and organizational outcomes.
On the basis of these frameworks, an updated model of applicant reactions to selection procedures is proposed as the conceptual foundation for the present study. Figure 1 outlines this model, which is guided by and adapted from earlier models (Gilliland, 1993; Ryan & Ployhart, 2000). Each portion of the framework is described below and a sample of representative research is discussed before turning our attention to empirical tests of the model using meta-analysis.
Model Overview
The main premise of the model outlined in Figure 1 is that important outcomes can be best predicted by applicant perceptions of the selection process. These outcomes include performance on selection procedures, self-perceptions, and a variety of attitudes and behaviors. Applicant perceptions take into account applicant views concerning the various dimensions of organizational justice, thoughts and feelings about testing, and broader attitudes about tests and selection in general. The model also specifies four broad classes of antecedent variables that are proposed as determinants of applicant perceptions and proposes several moderators of these relationships. Each component of the model is reviewed below along with a brief discussion of studies that have tested portions of the model empirically. Because of their central importance to the model, applicant perceptions are reviewed first, followed by outcomes and antecedents. respectively.
Applicant perceptions. A variety of perceptions have been studied to date, including procedural justice (e.g., Bauer, Truxillo. Sanchez, Craig, Ferrara, & Campion, 2001; Ployhart & Ryan, 1998), distributive justice (e.g., Smither et al., 1993), interpersonal justice (e.g., Ryan & Chart, 1999). informational justice (e.g., Bauer, Maertz, Dolen, & Campion, 1998). test motivation (e.g., Sanchez, Truxillo. & Bauer, 2000), test anxiety (e.g., Ryan, Ployhart, Greguras, & Schmit, 1998), attitudes towards tests in general (e.g., Chan, Schmitt, Jennings, Clause, & Delbridge, 1998), and attitudes towards selection in general (e.g., Macan et al., 1994). The justice perspective stems directly from Gilliland (1993), who proposed that applicants' perceptions of fairness directly influence subsequent attitudes and behaviors both during and after hiring. For example, the model predicts that applicants who feet that they were treated unfairly during an interview would be less likely to accept a job offer or recommend the employer to others.
Other constructs included in this portion of the model are derived from research on the internal cognitions held by applicants as they complete selection tools (Arvey, Strickland, Drauden, & Martin, 1990). The basic premise of this line of research is that applicants who are more motivated and less anxious will perform better on selection procedures. In addition, applicants who hold more positive perceptions about testing and selection in general are more likely to view favorably those organizations using such tools.
Outcomes. The types of outcomes that have been studied in the context of applicant reactions have grown steadily, although there are still relatively few studies that directly examine behavioral outcomes. More commonly, researchers have found small to moderate positive associations between applicant perceptions and actual and perceived selection procedure performance (e.g., Chan, 1997; Chan & Schmitt, 1997; Macan et al., 1994; Schmit & Ryan, 1997; Smither et al., 1993), self-efficacy (e.g., Bauer et al., 1998), and self-esteem (e.g., Bauer et al., 2001). Actual procedure performance refers to the test scores or ratings earned by applicants on a given selection device, whereas perceived procedure performance refers to self-assessed perceptions about performance on such screening tools. When considering organizational attractiveness (perceptions about the appeal or image that a company or organization maintains), past research generally shows moderate positive relationships with applicant perceptions (e.g., Bauer et al., 1998; Kluger & Rothstein, 1993; Macan et al., 1994: Rynes & Connerley, 1993). Researchers have also found positive relationships with a variety of behavioral intentions such as offer acceptance intentions (e.g., Truxillo, Bauer, Campion, & Paronto, 2002), application intentions (e.g., Rafaeli, 1999), retesting intentions (e.g., Madigan, 2000), product purchase intentions (e.g., Macan et al., 1994), litigation intentions (e.g., Bauer et al.. 2001), and recommendation and reapplication intentions (e.g., Ployhart & Ryan, 1998). Finally, although relatively rare, some researchers have studied behavioral outcomes including work performance (Gilliland, 1994; Hunthausen, 2000) and applicant withdrawal (e.g., Ryan, Sacco, McFarland, & Kriska, 2000; Schmit & Ryan, 1997). Results are mixed when considering the behavioral outcomes examined in these few studies.
Note that applicant perceptions have been linked conceptually with additional outcomes, including job satisfaction, organizational commitment, turnover, and organizational climate (Gilliland, 1993). Further, because intentions and behaviors tend to be moderately related (Armitage & Conner, 2001), applicant perceptions should be related to actual behaviors such as recommending the employer to others, reapplying, retesting, and bringing litigation claims. However, little research exists to date that tests these propositions.
Antecedents. Four classes of antecedent variables have been identified in the applicant reactions literature. These include person characteristics, perceived procedure characteristics, job characteristics, and factors associated with the organizational context. There has been considerable attention given to identifying the person characteristics and perceived procedure characteristics that give rise to the applicant perceptions described above. In particular, person characteristics that have commonly been examined include demographic variables such as gender, age, and race. For example, race has been studied in this context as a possible explanation for the Black--White test score gap (Ryan, 2001). In addition. Chan found that Blacks held less favorable perceptions of cognitive ability tests than Whites (1997). Other researchers have examined personality characteristics including several of the Big Five dimensions such as Conscientiousness and Neuroticism (Ostberg, Truxillo, & Bauer, 2001) as possible determinants of applicant perceptions. Finally, some researchers have explored the possibility that prior work experience or familiarity with testing situations could help explain applicant perceptions (Truxillo, Bauer, & Sanchez, 2001).
Perceived procedure characteristics include many of the justice rules identified by Gilliland (1993) such as job relatedness, opportunity to perform. reconsideration opportunity, two-way communication, and propriety of questions (e.g.. Bauer et al., 1998; Ryan & Chart, 1999; Truxillo et al., 2001). In particular, job relatedness has been studied extensively in previous research based on the premise that applicants will perceive selection more favorably to the extent that techniques are perceived as face valid and predictive of job performance. Thus, job relatedness is often conceptualized as a two-factor construct comprised of face validity and perceived predictive validity.
Face validity has been defined as "the extent to which applicants perceive the content of the selection procedure to be related to the content of the job" (Smither et al., 1993, p. 54). Some researchers have used the term content validity to refer to the perceived relevance of the content of the selection procedure. However, most treatments of validity consider content validity as an aspect of test development that is best assessed and influenced by the test developer or other trained experts (AERA, APA, & NCME, 1999). Face validity judgments require no such expertise and simply involve surface-level judgments about the apparent relevance of test content. Thus, content validity is primarily assessed by the test developer, whereas face validity is typically assessed by the test taker. In addition. face validity is not a psychometric property; rather, it is an individual's judgment about the job relatedness of a selection procedure.
Perceived predictive validity has been defined as perceptions about "how well the procedure predicts future job performance, regardless of how it looks" (Smither et al., 1993, p. 54). These assessments are also made from the perspective of the test taker and involve beliefs about whether people who score better on the test also perform better on the job. In the present context of the applicant reactions literature, perceived predictive validity also is not a psychometric property, but an individual's judgment about the predictive ability of a selection procedure.
As noted by Ryan and Ployhart (2000), other procedure characteristics such as the length of the selection process and actual outcome favorability (e.g., pass/fail information) can influence perceptions such that applicants perceive selection more favorably when procedures are not excessively long and when applicants receive positive outcomes. Providing applicants with an adequate explanation for the use of selection tools and decisions may also foster positive perceptions among applicants. In addition, researchers have proposed that applicant perceptions may be positively related to perceived test ease (e.g., Wiechmann & Ryan, 2003) and the transparency of selection procedures (e.g., Madigan, 2000). Perceived procedure characteristics and applicant perceptions have different conceptual meanings in that each of the perceived procedure characteristics is predicted to influence overall judgments of fairness, global test-taking perceptions, and general attitudes towards tests and selection. In other words, applicant perceptions are more general judgments about the process, and perceived procedure characteristics refer to specific factors associated with the selection process or procedures.
Finally, two additional classes of antecedent variables have been proposed. These include job characteristics (e.g., industry norms for selection, job attractiveness, KSA requirements) and the organizational context (e.g., selection ratio, organizational history). With the exception of a handful of studies (e.g., Macan et al., 1994; Thorsteinson & Ryan, 1997), there have been few systematic attempts to study these two classes of potential antecedent influences.
Main Hypotheses
According to the theoretical model described above, person characteristics and perceived procedure characteristics should be related to applicants' perceptions of selection procedures, which should in turn be related to a variety of attitudinal and behavioral outcomes.
Hypothesis 1: There will he nonzero relationships between person characteristics (i.e., age, gender, ethnic background, Conscientiousness, and Neuroticism) and applicant perceptions (i.e., procedural justice, distributive justice, and test motivation).
Hypothesis 2: There will be positive relationships between perceived procedure characteristics (i.e., consistency, job relatedness, face validity, perceived predictive validity, opportunity to perform, explanations/accounts, outcome favorability, and transparency) and applicant perceptions (i.e., procedural justice, distributive justice, test motivation, attitude towards tests, and attitudes towards selection).
Hypothesis 3: Applicant perceptions will be positively related to the outcomes of actual and perceived procedure performance, organizational attractiveness, recommendation intentions, offer acceptance intentions. self-efficacy, and self-esteem. There will be negative associations for those relationships involving test anxiety.
Moderators
Applicants' prior experience, hiring expectations, perceived alternatives, and the stage in the selection process are among the variables that have been proposed as moderators of the antecedent-perception link (Gilliland, 1993; Ryan & Ployhart, 2000). Overall, the number of occasions in which these variables have been measured is small and there have been even fewer attempts at evaluating them as moderators as opposed to estimating simple bivariate relationships. In this study, selection context and the stage of the selection process are systematically evaluated to help explain potentially heterogeneous relationships contained in the model. In addition, the favorability ratings of selection tools are examined to determine whether perceptions differ based on test type, as suggested in previous research (Gilliland, 1993).
Three levels of selection context are proposed including authentic, hypothetical, and descriptive contexts. Studies conducted in authentic selection contexts involve actual job applicants seeking positions with real organizations (e.g.. Bauer et al., 1998; Macan et al., 1994; Smither et al., 1993). Typically, applicants are given surveys to complete at one or more points during the selection process to assess their perceptions and attitudes towards the selection tools and the company. The second approach to examining applicant reactions involves hypothetical selection scenarios, wherein participants (often college undergraduates) assume the role of an applicant for a particular job or company and complete selection tools and reactions surveys (e.g., Chan, 1997; Ployhart & Ryan, 1998).
Examining the inherent qualitative differences associated with authentic and hypothetical selection research suggests that relationships between applicant reactions variables and various work correlates may differ depending on the study context. There are several plausible patterns of results that might emerge when examining selection context. Effect sizes may be stronger for studies conducted in authentic selection contexts as opposed to those carried out in hypothetical hiring situations because applicants in authentic selection contexts have more at stake and may be more sensitive to the types of selection tools used during the hiring process. On the other hand, relationships drawn from authentic contexts might be attenuated due to range restriction from self-selection into the hiring process. Therefore, the hypothesis regarding selection context was framed to be exploratory in nature.
Hypothesis 4: Selection context will moderate the relationship between applicant reactions and organizational...
Read the full article for free courtesy of your local library.
|
Bookmark this article
Print this article
Link to this article
Email this article
Digg It!
Add to del.icio.us
RSS
