Breaugh & Starke (2000)

Journal of Management http://jom.sagepub.com

Research on Employee Recruitment: So Many Studies, So Many Remaining Questions James A. Breaugh and Mary Starke Journal of Management 2000; 26; 405 DOI: 10.1177/014920630002600303 10.1177/014920630002600303

The online version of this article can be found at: http://jom.sagepub.com/cgi/content/abstract/26/3/405

Published by: http://www.sagepublications.com

On behalf of:

Southern Management Association

Additional services and information for Journal of Management can be found at: Email Alerts: http://jom.sagepub.com/cgi/alerts Subscriptions: http://jom.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations http://jom.sagepub.com/cgi/content/refs/26/3/405

Downloaded from http://jom.sagepub.com at Universiteit van Amsterdam SAGE on November 9, 2009

Journal of Management 2000, Vol. 26, No. 3, 405–434

Research on Employee Recruitment: So Many Studies, So Many Remaining Questions James A. Breaugh and Mary Starke University of Missouri–St. Louis

Over the last thirty years, the amount of research on recruitment topics has increased dramatically. Despite this increase, recent reviews of the rec recrui ruitme tment nt lit litera eratur turee oft often en hav havee had a som somewh ewhat at pes pessim simist istic ic tone. Reviewers have concluded that we still do not know a great deal about why recruitment activities have the effects they do. In particular, recent reviews have criticized many of the studies conducted for being poorly designed, narrow in focus, and not grounded in theory. We believe that many of these criticisms are legitimate. We also believe that, in order for future studies to result in a better understanding of the recruitment process, such studies need to be designed with an appreciation of the complexity of the recruitment process (i.e., the number of variables involved and the nature of their relationships). In this regard, we offer an organizing framework of the recruitment process. In introducing this framework, we draw upon theories from a variety of research domains and give considerable attention to process variables (e.g., applic applicant ant attent attention, ion, messag messagee credib credibility ility,, applic applicant ant selfself-insigh insight) t) that mediate the relationships between recruitment activities (e.g., recruiter behavior) and recruitment outcomes (e.g., the number of applications generated). Having introduced an organizing framework, we selectively review recruitment research, giving particular attention to the topics of recruitment sources, recruiters, recruiters, and reali realistic stic job previe previews. ws. This review makes apparent a number of important issues that recruitmentt res men resear earch ch has yet to addr address ess.. © 2000 Els Elsevi evier er Sci Scienc encee Inc Inc.. All rights reser reserved. ved.

Research interest in the topic of employee recruitment has increased substantially over the last thirty years. As an example of this increasing interest, consider that in the first edition of the Handbook of Industrial and Organizational Psychology, less than one page of coverage was given to the topic of recruitment (Guion, 1976). By the time the second edition of this handbook was published,

Direct all correspondence to: James A. Breaugh, School of Business Administration, University of Missouri–St. Louis, St. Louis, MO 63121; Phone: 314-516-6287; Fax: 314-516-6420. Copyright © 2000 by Elsevier Science Inc. 0149-2063

405


406

J.A. BREAUGH AND M. STARKE

research on recruitment was seen as meriting an entire chapter (Rynes, 1991). Even though numerous recruitment studies had been published since the 1976 Handbook chapter appeared, Rynes perceived this body of research to be lacking in several ways. For example, she found most of the research that she reviewed focuse foc used d on one of onl only y thr three ee top topic icss (i. (i.e., e., re recru cruitm itment ent sou source rces, s, rec recrui ruite ters, rs, and realistic job previews) and that the research on each of these topics was “developed in isolation from the others” (Rynes, 1991: 399). In her chapter, Rynes showed how this piecemeal approach to research hindered theory development and left unanswered several questions that an employer might have. More recently, the expanding body of research on recruitment has been seen as sufficient to merit entire books (e.g., Barber, 1998). Given that dozens of studies were published between Rynes’ (Rynes, 1991) and Barber’s (Barber, 1998) reviews of recruitment research, it is not surprising that Barber felt that understanding in certain areas (e.g., recruitment source effects) had increased. However, Barber also pointed out that researchers still had failed to address adequately a number of significant issues (e.g., the site visit). She also noted that methodological weaknesses (e.g., failure to measure key variables) made it difficult to draw clear conclusions from many studies. In summary, if one peruses reviews of recruitment research (e.g., Barber, 1998; Breaugh, 1992; Rynes, 1991; Wanous, 1992), one finds a mix of optimism and pes pessim simism ism.. Rev Review iewers ers wer weree ple please ased d tha thatt res resea earch rch on rec recrui ruitme tment nt had increased, and they were confident that progress in understanding the effects of recruitment activities was being made. Yet, they also seemed frustrated that more progress in understanding the recruitment process had not occurred. Given Barber’s excellent review of recruitment research (Barber, 1998), it seemed unnecessary for us to provide a systematic review of the same research in this article. However, we also are aware that many readers of this journal will not have read Barber’s book. Thus, we felt it was important to provide a selective review of the recruitment literature. We encourage readers who are interested in more detailed coverage of recruitment research to refer to the works of Barber (1998), Breaugh (1992), Rynes (1991), and others cited herein. In addition to providing the reader with a selective review of recruitment research, a second objective in writing this paper was to stimulate the type of research that will help answer many of the research questions that remain even after so many recruitment stud st udie iess ha have ve be been en co cond nduc ucte ted. d. In at atte temp mpti ting ng to st stim imul ulat atee re rese sear arch ch,, we ha have ve highlighted unresolved issues and several overlooked topics. There are a number of possible approaches for structuring a review. For example, given that many readers may not be overly familiar with the recruitment literature, we could begin with a review of this research. However, we believe that it makes the most sense to offer an organizing framework prior to reviewing available research. This framework will make apparent the complexity of the recruitm recr uitment ent proc process, ess, comp complexi lexity ty that has frequ frequentl ently y been overl overlooke ooked d (Bar (Barber, ber, 1998). It also will show the narrowness of the recruitment literature (i.e., the number of key issues that have not been examined). On a more positive note, we hope the framework we offer will stimulate future research that will lead to a better understanding of the recruitment process (i.e., the interaction among reJOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


RESEARCH ON EMPLOYEE RECRUITMENT

407

cruitment variables and the relationship between recruitment variables and other variables). The organizing framework presented in Figure 1 is based upon models that have been offered by Barber (1998), Breaugh (1992), and Rynes (1991). We acknowledge that our framework represents a simplified view of reality. In addressing the segments of this framework, we will introduce the theoretical explanations that have been offered for the relationships between variables. We will not discuss specific research findings until the entire framework has been introduced. Before presenting our organizing framework, it seems appropriate to offer a definition of what we mean by the term recruitment. Although the term recruitment is commonly used, it is not easy to define (see Barber, 1998: 5–6). For our purpose, the definition offered by Barber will suffice: “recruitment includes those practices and activities carried on by the organization with the primary purpose of identifying and attracting potential employees” (p. 5). An Organizing Framework for Understanding the Recruitment Process

As noted by Rynes (1991), much recruitment research has focused on the effects of recruitment sources (e.g., Do individuals referred by current employees have a lower turnover rate than persons recruited via newspaper ads?), recruiters (e.g., Do recruiters who offer more information about a job make a better impression on job applicants?), and realistic job previews (e.g., Does providing accurate job information result in a higher level of job satisfaction for new employees?). As depicted in Figure 1, all of these topics can be considered recruitment activities. Although it is understandable why researchers have frequently focused on the effects of recruitment activities (e.g., most companies document how applicants heard about job openings), we believe that more attention should be focused upon the entire recruitment process in order to better understand whether an employer’s recruitment activities will accomplish its objectives. Establishing Recruitment Objectives and Strategy Development

In discussing the recruitment process, Barber (1998) delineated three phases (i.e., generating applicants, maintaining applicant status, and influencing job choice decisions). That is, (a) certain recruitment activities (e.g., advertising on a Spanish-speaking radio station) may influence the number and type of individuals who apply for a position, (b) certain activities (e.g., professional treatment during a site visit) may affect whether job applicants withdraw during the recruitment process, and (c) certain recruitment actions (e.g., the timeliness of a job offer) may influence whether a job offer is accepted. In planning a strategy for generating applicants, a fundamental question that should be addressed is, What type of individual does the organization want to recruit (e.g., What knowledge, skills, and abilities are important? Is a diverse applicant pool desired?)? Until an employer determines the type of applicants it seeks, it is difficult for it to address several other strategy-related questions: Where should the organization recruit (e.g., colleges versus state employment JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


408


s s e c o r P t n e m t i u r c e R l a n o i t a z i n a g r O e h t f o l e d o M A . 1 e r u g i F

JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



409

offices)? What recruitment sources should it use to reach the desired applicant population (e.g., the Web versus job fairs)? When should the employer begin recruiting (e.g., at the start of a college student’s senior year versus during the second semester)? What message should it convey to potential applicants (e.g., a good deal of job-related information versus information on a few key aspects of the job)? In order for an organization to intelligently answer these basic questions, it must have a clear sense of what its recruitment objectives are (Breaugh, 1992). For this reason, we have portrayed the establishment of objectives as the first phase of the recruitment process (see Figure 1). In the past, it appears that many organizations have had the simple recruitment goal of attracting a large number of job applicants (Wanous, 1992). For a variety of reasons (e.g., the cost of processing applications), several researchers have questioned the wisdom of simply trying to attract a large number of applicants. Instead, it has been suggested (Rynes, 1991) that employers would be wise to consider a wider range of possible recruitment objectives. For example, some employers might be interested in trying to influence one or more post-hire outcomes by the way they recruit. Such post-hire outcomes include: the job satisfaction of new employees, their initial job performance, whether the organization is seen as living up to the psychological contract that has been established, and the first-year retention rate of new hires. In establishing recruitment objectives, organizations might also focus their attention on post-hiring outcomes that can be measured the day employees begin work (Breaugh, 1992). Such outcomes might include: the cost of recruiting, the speed with which jobs were filled, the number of individuals hired, and/or the diversity of the new employees. Although recruitment activities have been linked to some of these post-hire outcomes, some researchers (e.g., Williams, Labig, & Stone, 1993) have argued that in recruiting many employers are not overly concerned with post-hire outcomes. Rather, they are interested in pre-hire outcomes such as the number of individuals who apply for a position, the quality of these applicants, their diversity, and the number of individuals who accept job offers that are extended. If an employer is interested in these more proximal outcomes of the recruitment process, its strategy development should be focused on how to accomplish them. In summary, we believe that the first stage of the recruitment process should be the establishment of objectives. If clear objectives have not been established, it is difficult to develop a sound recruitment strategy (Rynes & Barber, 1990). Having established a core set of objectives, an organization can more intelligently answer the strategy development questions posed earlier (e.g., What recruitment sources to use?). Having developed a recruitment strategy, an employer can then undertake the recruitment activities that are likely to lead to its desired objectives. Intervening/Process Variables

Although we believe that the recruitment process should proceed in the manner portrayed in Figure 1 (i.e., recruitment objectives 3 strategy development3 recruitment activities 3 recruitment results), it is useful to add intervening/process variables to this model and to work backwards through the model. That is, in order to make decisions about what recruitment activities to JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


410


undertake (i.e., strategy development), an employer needs to understand why certain recruitment activities may result in certain recruitment outcomes (Rynes, 1991). We use the term intervening/process variable as a label for the factors that have been hypothesized to explain the relationships between recruitment activities and outcomes. A number of potential intervening variables are portrayed in Figure 1. In terms of generating applicants, it is critical that an employer’s recruitment actions attract the attention of potential job applicants (Barber, 1998). Research suggests that the following attributes are likely to generate attention: (a) messages that are vivid in nature (e.g., include pictures) and include concrete language (Tybout & Artz, 1994), (b) messages that convey unexpected information (Kulik & Ambrose, 1993), (c) messages that provide personally relevant information (Chaiken & Stangor, 1987), and (d) messages that are conveyed in face-to-face conversations (Tybout & Artz, 1994). In addition to attracting attention, recruitment communications need to be understandable and viewed as credible by the individuals whom the organization is interested in recruiting (Breaugh & Billings, 1988). In terms of a message being understood, such mundane factors as using an appropriate level of expression and choosing the correct language (e.g., Spanish) are clearly necessary (Jablin, Putnam, Roberts, & Porter, 1987). The medium used to deliver a message can also influence comprehension. For example, Stiff (1994) suggested that, in contrast to a verbal message, a written message may increase understanding given that the message can be reread and studied. In contrast, Lengel and Daft (1988) suggested that in-person communication may lead to greater understanding, especially when the message is somewhat complex. Lengel and Daft argued that in-person communication is a richer medium in that tone of voice, non-verbal cues, and other variables exist that are lacking in a written message. With regard to message credibility, research (Stiff, 1994) has consistently shown that communicator expertise and trustworthiness lead to a message’s being believed. In terms of expertise, generally those who are closest to the work situation (e.g., job incumbents) are seen as being an informed source of jobrelated information (Fisher, Ilgen, & Hoyer, 1979). With regard to trustworthiness, research has shown that receiving information that is different than that expected from the message source results in credibility (Stiff, 1994). For example, in persuasion research, it has frequently been found that communicators who convey information that detracts from their position are rated as more trustworthy than communicators who cite only information that supports their position (Chaiken & Stangor, 1987). In contrast, communicators who are likely to derive personal benefit from misleading the communication recipient frequently lack credibility (Stiff, 1994). Research (Harkins & Petty, 1981) also has shown that receiving a consistent message from multiple sources results in an attribution of credibility. In order to be effective, a recruitment message also needs to generate initial interest from potential job applicants. Such interest is more likely to be forthcoming if a job opening (e.g., the job itself, the organization, the location) is viewed positively (Barber & Roehling, 1993). Interest is also more likely to be JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



411

forthcoming if an applicant perceives that he/she has a reasonable likelihood of receiving a job offer (Rynes, 1991). As noted by Barber (1998), the recruitment process does not end with the generation of job applicants. An organization also should be concerned about maintaining the interest of job candidates and influencing their job choice decisions. With regard to intervening variables, many of the same ones that affect the generation of job applicants also are important for these later two stages of the recruitment process. For example, individuals are likely to remain job candidates if they continue to see a job opening as attractive and if they perceive themselves as likely to receive job offers as they move through the recruitment process (Rynes, Bretz, & Gerhart, 1991). Similarly, the attractiveness of a job opening should have a major influence on a job applicant’s decision whether to accept a job offer (Turban, Eyring, & Campion, 1993). In attempting to understand the effects of recruitment activities, two key intervening variables are the accuracy of applicants’ job expectations (Wanous, 1992) and their degree of self-insight (Breaugh, 1992). If an employer is interested in influencing such post-hire outcomes as job satisfaction, initial job performance, perceptions of violations of the psychological contract, and firstyear retention rate, it should be particularly concerned with whether job applicants have an accurate perception of what a job entails and whether they have insight into their own capabilities (e.g., knowledge, skills) and desires (e.g., what they want from a position and an employer). For example, given that the labor market allows job candidates who do not perceive a good person-job fit to self-select out of job consideration, applicants who have an accurate perception of a job situation and have self-insight should, if hired, find their jobs satisfying (otherwise, they would be unlikely to accept job offers). Being satisfied, new employees should be less likely to quit during the initial employment period (Wanous, 1992). Selfselection also should favorably influence the initial performance of new hires (e.g., Downs, Farr, & Colbeck, 1978). This assertion is based upon the assumption that, if a person does not perceive he or she has the ability to perform well, the person will be less likely to accept the job offer. Regardless of whether the labor market allows applicants to self-select out of job consideration, having realistic job expectations should influence new employees’ perceptions of psychological contract violations (Rousseau, 1995). For example, if an applicant has no better options, even after receiving information that makes a job appear undesirable, he or she may accept a job offer. However, if the employer has been honest with the individual during the recruitment process, even though he or she may not like the new position, the person should not feel that the organization has failed to live up to its side of the employment contract. It also has been argued (Wanous, 1992) that first-year retention rate may be positively affected by individuals having had accurate job information during the recruitment process, even when the labor market does not allow for self-selection. This argument is based on the assumption that, having made an informed job choice decision, new employees will feel an obligation to remain in the job for at least a few months. In summary, in order to understand why certain recruitment JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


412


activities may impact certain recruitment outcomes, it is necessary to understand the role that several intervening/process variables play. Recruitment Activities that are Likely to Influence Key Intervening/ Process Variables

To date, most of the research conducted has addressed recruitment sources, recruiters, and realistic job previews (Barber, 1998). Given this fact, we will focus most of our attention on these three activities, their hypothesized relationship with intervening variables, and how such process variables influence several outcome variables. As noted earlier, in introducing our framework of the recruitment process, we are focusing upon theoretical rationales that have been offered for the relationships between recruitment activities and the intervening variables. Recruitment sources. Although employers use a variety of recruitment sources, little recent data exist on the frequency of usage. The most commonly cited study of source usage was conducted by the Bureau of National Affairs (1988). This study examined the frequency of source usage across five job types (office/clerical, production/service, professional/technical, commission sales, and managers/supervisors). Although the results of the BNA study are too complex to fully discuss here, a few general trends are worth noting. Regardless of job type, newspaper ads, employee referrals, direct applications (i.e., individuals who apply for a job without having received information that a job opening existed), and recruiting at schools were commonly used sources. A few sources were closely tied to the type of job opening. For example, state employment service offices only were commonly used for filling office and production jobs. The National Organizations Study, which involved a national probability sample of employers, reported findings similar to those of the BNA survey (Kalleberg, Knoke, Marsden, & Spaeth, 1996). Reports from a national probability sample of employees (Vecchio, 1995) concerning how they found their current positions further support the general finding that firms rely heavily upon employee referrals, newspaper ads, and direct applications. With regard to recruitment sources, two theoretical explanations (i.e., the realistic information hypothesis and the individual difference hypothesis) for why sources may be differentially associated with recruitment outcomes have attracted the most attention (Barber, 1998). The realistic information hypothesis proposes that persons recruited via certain sources are likely to have more accurate information about what a job entails (Rynes, 1991). Possessing such information is thought to enable an applicant to make a more informed decision about whether to pursue a job. For example, with regard to employee referrals, it has been suggested (Breaugh, 1992) that before referring a person for a job with their firm, current employees would provide the individual with realistic information about what a job involved. Concerning direct applications, it has been suggested (e.g., Breaugh & Mann, 1984) that they may possess more realistic information concerning a job opening as a result of having done more research on potential employers than applicants coming from other sources. Conversely, it has been argued that individuals recruited by other sources (e.g., school placement offices, newspaper ads) may lack realistic job information (Williams et al., 1993). Lacking JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



413

such information, these individuals are thought to be less likely to be able to self-select out of consideration for jobs that may not be a good fit in terms of their skills or interests. If hired, these less informed individuals are more likely to be unhappy with their job choice and may be more likely to resign. The second explanation commonly offered for recruitment source differences is known as the individual difference hypothesis (Rynes, 1991). This explanation is based on the premise that sources differ in the types of individuals they reach, and that these differences result in different outcomes. For example, Ullman (1966) speculated that persons recruited via employee referrals may be more capable than individuals recruited from such sources as newspapers. This assertion is based on the assumption that current employees would screen individuals before making job referrals because they may see their own reputation as being affected by the quality of the referrals. Other researchers have speculated that different recruitment sources reach applicant groups that may differ on characteristics such as motivation or perceived job mobility (Barber, 1998). Unfortunately, researchers have not specifically described which sources should be linked to which individual difference variables (e.g., Would direct applicants have a higher motivational level to find a job because applying for job openings that may or may not exist takes more effort than simply responding to advertised positions?). Given the organizing framework we presented, it should be apparent that we believe that more theory development needs to take place with regard to recruitment sources and their relationship with process variables. For example, as Barber (1998) noted, “recruitment scholars have largely ignored the question of how to attract attention to recruitment materials” (p. 39). In terms of attracting the attention of potential job applicants, one may be able to build a theoretical case that the use of direct mail (a rarely examined recruitment source) or employee referrals (which involve actively seeking out individuals who may be interested in a job opening) would be attention-getting recruitment sources compared to such passive sources as newspaper advertising (a potential job candidate may not be looking for a position). Based upon psychological research (Stiff, 1994), one might hypothesize that job advertisements placed in atypical locations in a newspaper may be attention-getting (Breaugh, 1992). With regard to message comprehension and credibility, theory development also is lacking. One might hypothesize that employee referrals would be a particularly credible source if made by job incumbents as opposed to supervisors who may be perceived by applicants as under pressure to fill vacant positions. Given that most employee referrals involve two-way interactions, referrals may be a source that facilitates message comprehension (Lengel & Daft, 1988). Although we could offer further speculation about what intervening processes explain why recruitment sources have the effects on outcome variables (e.g., job survival) that have been documented (see Wanous & Colella, 1989), such theory development is beyond the scope of this paper. In summary, in terms of the relationships between recruitment sources and key intervening variables (e.g., applicant attention), it suffices to say that much theoretical work remains. In the future, researchers need to move beyond simply JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


414


considering recruitment source effects that are associated with realistic job expectations and individual differences. Once theoretical linkages have been specified between various recruitment sources and key intervening variables, it is more likely that the type of theory-driven empirical research called for by Rynes (1991), Barber (1998), and others will be conducted. Recruiters. Researchers have offered a number of explanations for why recruiters may have an effect on job candidates. The first of these has to do with recruiter informativeness. Researchers (e.g., Powell, 1991) have theorized that some recruiters provide more information and more specific information to applicants than other recruiters. Given that the information provided is of personal relevance to the job candidate and is believed, interacting with more informative recruiters should have numerous beneficial effects (e.g., enhanced ability of applicants to remove themselves from consideration for jobs that ultimately would not be satisfying). In terms of recruiter informativeness, it has been suggested (Breaugh, 1992) that an individual’s prospective supervisor or coworkers should be particularly informative in contrast to other recruiters (e.g., a person from the human resources department) who may not possess as much information. Researchers (e.g., Maurer, Howe, & Lee, 1992) also have hypothesized that recruiter credibility helps explain the differential effects of recruiters on applicants. For example, Fisher et al. (1979) hypothesized that, in comparison to job incumbents, corporate recruiters would lack credibility. Fisher et al. based their assertion on the assumption that corporate recruiters would be viewed by applicants as lacking expertise concerning what a job involves and would be perceived as having a vested interested in filling open positions. It also has been suggested (Rynes, 1991) that recruiters may have an impact because job candidates view them as signals of unknown organizational attributes. For example, researchers (e.g., Connerley & Rynes, 1997) have suggested that recruiter personableness may be important because it “signals” how the person may be treated if hired or how likely the person is to receive a job offer. In a similar vein, interacting with female or minority recruiters may signal to an applicant that an employer values diversity. Such a value system may make a job more attractive to certain job candidates (Highhouse, Stierwalt, Bachiochi, Elder, & Fisher, 1999). Although researchers have theorized about the positive effects of recruiters who are informative, credible, personable, and demographically diverse, researchers have not focused heavily on the underlying reasons why different types of recruiters may influence the attention potential applicants give to a job opening or the interest applicants have in a position. Given the significance of attracting applicant attention and generating their interest, clearly these two issues merit future attention. A topic related to the informativeness of a recruiter is the degree to which the message conveyed by a recruiter involves sharing realistic as opposed to only positive information. Because the topic of providing realistic information is generally covered in the context of realistic job previews, we will discuss this issue in the next section. In summary, a strong case can be made that the choice of a recruiter can make a difference (Rynes et al., 1991). However, at present, although researchers JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



415

have provided some theoretical foundation for why recruiters may have beneficial effects, more theory development needs to take place. Realistic job previews. According to Phillips (1998), “no recruitment issue has generated more attention than realistic job previews (RJPs), the presentation by an organization of both favorable and unfavorable job-related information to job candidates” (p. 673). Given the attention that RJPs have attracted, it seems reasonable to ask, “Is possessing inaccurate job expectations really a major issue?” Unlike recruiting sources, national sample data on realistic expectations have not been gathered. However, data from individual studies suggest that unmet expectations are a frequent occurrence. In a review of several RJP studies, Wanous (1992) suggested that unmet expectations on the part of new employees are frequently reported. In the context of psychological contracts, Rousseau (1995) provided ample evidence that new hires often see their organizations as not having lived up to agreed-upon arrangements. Compared to the topics of recruitment sources and recruiter effects, there has been considerable theory development with regard to RJPs. For example, Breaugh (1992); Wanous (1992); Fedor, Buckley, and Davis (1997); Hom, Griffeth, Palich, and Bracker (1998); and Saks and Cronshaw (1990) have all offered models of the process by which RJPs affect such outcome variables as job survival, work attitudes, and job performance (an adaptation of Breaugh’s model is portrayed in Figure 2). Given the numerous models in existence, it is not possible to discuss all of their subtle nuances. However, it is worthwhile to discuss key conceptual variables that most of the models have in common. Most RJP models hypothesize that providing realistic job information to applicants results in their having their job expectations met. It is further hypothesized that providing an RJP influences role clarity (which is thought to affect job performance) and individuals’ perceptions that the organization was honest with them. Given that candidates perceive alternatives to accepting an undesirable job, providing an RJP is hypothesized to result in applicant self-selection, which, in turn, is hypothesized to result in a higher level of job satisfaction, a lower level of voluntary turnover, and a higher level of performance. Although the various models of RJP effectiveness include a number of variables, many models have failed to address some key process variables. For example, as pointed out by Phillips (1998), “all theories about the effectiveness of realistic job previews share the underlying assumption that their message is received and processed by applicants” (p. 673). Given the significance that we have attached to job applicant attention and comprehension, it should not be surprising that we think that more theory development should address these two variables. From our review of the research literature, it appears that Colarelli (1984) is the only researcher who has given much attention to many of the process variables we highlighted. In establishing a theoretical framework for his study, Colarelli addressed process variables only in the context of providing an RJP via a brochure versus providing an RJP by means of a face-to-face conversation with a job incumbent. Colarelli hypothesized that, in comparison to receiving an RJP in written form, RJPs received via a face-to-face interaction would attract greater attention, be JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


416


Figure 2. The Formation of Job Expectations and their Influence on Important Employee Attitudes and Behaviors: A Process Model (adapted from Breaugh, 1992) JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



417

more understandable, and be more credible. Although he did not address why he thought a face-to-face conversation would attract attention, Colarelli suggested that a conversation aided comprehension due to the fact that an applicant could ask questions to clarify ambiguous information. He argued that job incumbents have considerable credibility given that they are generally seen as “more knowledgeable than other sources of information” (Colarelli, 1984: 634) and because they are generally seen as more trustworthy. Over the years, a number of researchers have suggested some theoretical refinements to the basic processes that have been hypothesized for explaining RJP effects. We will highlight three of these suggestions. In an important (but somewhat overlooked) paper, Greenhaus, Seidel, and Marinis (1983) questioned the emphasis that researchers have placed on met expectations. Greenhaus et al. argued that, if a person’s job expectations were negative and met (e.g., the person was not able to turn down an undesirable job), one would not generally predict job satisfaction. Alternatively, a person could be pleasantly surprised about certain aspects of a job upon beginning work (i.e., not all unmet expectations are unpleasant). Instead of focusing so heavily on met expectations, Greenhaus and his colleagues argued that more attention should be paid to whether a job provided for value attainment (i.e., a match between a person’s job values and job experiences). These researchers not only proposed value attainment as an intervening variable between having realistic expectations and experiencing satisfaction, they tested their assertion. In two studies, they found that “value attainment accounted for significantly more variance in facet satisfaction than did realistic expectations” (p. 394). In order to better understand the effects of RJPs, Fedor et al. (1997) argued for more attention being given to the attributions new employees make for any discrepancies that occur between what they expected and what they experienced. For example, they suggested that new employees who perceive that an organization misled them about job features would likely react more negatively than if they attributed unmet expectations to the fact that the job had changed since the position was described to them or to some other reason. Researchers also have suggested that more attention needs to be given to characteristics of the RJP recipients. For example, Meglino, DeNisi, Youngblood, and Williams (1988) hypothesized that RJP effects would be stronger for more intelligent applicants given that they could better comprehend the information conveyed. Breaugh and Billings (1988) hypothesized that RJP effects would be stronger if the RJP information addressed aspects of a job that were seen as particularly important by job candidates. In summary, although researchers have given considerable attention to the reasons why RJPs “work,” more theory development is needed. In particular, we believe that researchers need to move beyond viewing RJPs as the primary mechanism for influencing the accuracy of applicant job expectations. The realism of job expectations can be influenced by individuals having similar prior work experience (Breaugh, 1992) or having completed internships (Taylor, 1988). Other message attributes. Although researchers have given the majority of their attention to whether a recruitment communication contains realistic (as JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


418


opposed to only positive) information (Phillips, 1998), three other message attributes (i.e., the breadth of topics addressed, the specificity of the information provided, the timing of communications) merit attention. It has been well documented that during the job search process individuals frequently lack information about job and organizational attributes. For example, Maurer et al. (1992) found that engineering students who were job hunting reported they lacked information about such issues as starting salary, how raises are determined, benefits, and the success of new hires. The students who reported they lacked information about positions also reported they were less likely to accept job offers. Similarly, Barber and Roehling (1993) found that individuals who reported having more information about a job and/or an organization were more attracted to the organization. In attempting to explain why a lack of information may result in a negative reaction from recruits, Barber and Roehling suggested that job applicants may see the failure of an organization to provide sufficient information as an “indicator of sloppy, disinterested recruiting practices” (Barber & Roehling, 1993: 853). In other words, an employer’s failure to supply information may be viewed as a signal of its lack of professionalism or of its lack of interest in the candidate. Although some researchers (Breaugh & Billings, 1988) have theorized that an employer’s failure to provide information about a range of important job attributes (e.g., supervisory style, coworker relations) can hurt its ability to recruit, Highhouse and Hause (1995) have qualified this general prediction. These authors agreed that decision makers seek to avoid uncertainty. Thus, in general, job applicants should tend to devalue a job opening for which job attribute information is lacking. However, Highhouse and Hause suggested that there is an exception to this general tendency to avoid uncertainty. They hypothesized that decision makers might actually prefer a job opening with missing information on a job attribute over an alternative job opening for which information was available on the attribute but the attribute was viewed as being below average. Concerning information specificity, Breaugh and Billings (1988) argued that, even when an employer tried to present realistic job information, frequently the information presented was so general (“salaries are competitive”) that it did not allow for informed recruit decision-making. More recently, Barber and Roehling (1993) addressed the issue of information specificity from a different perspective. These researchers hypothesized (and confirmed) that job applicants pay more attention to specific than to general information. Given that attracting applicant attention is a key variable in the recruitment process, it appears that by providing more specific information an employer may be able to generate more applicant interest. With regard to the timing of recruitment communications, relatively little theory development has taken place (Barber, 1998). Based upon research in decision making, Breaugh (1992) hypothesized that a desire for uncertainty avoidance on the part of job applicants might make it wise for an organization to make job offers early in the recruitment cycle. Thurow (1975) suggested that the advantage of early recruitment may only hold for more desirable employers. Rynes et al. (1991) focused upon the timing of recruitment communications in terms of signaling. These authors hypothesized that delays in responding to JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



419

applicant inquiries or in making job offers may be viewed as a sign that the organization does not have a strong interest in the candidate. A Review of Recruitment Research

Having introduced a framework for understanding the recruitment process and having summarized theoretical rationales that have been offered to explain the relationships portrayed in our model, we now will provide a selective review of empirical recruitment literature. Research on Recruitment Sources

One of the first issues that an organization needs to address in trying to fill job openings is how to reach individuals who may be qualified for and interested in the openings. Over the years, several researchers have investigated the use of different recruitment sources and whether certain sources are linked to beneficial outcomes. A study by Ullman (1966) was one of the first to examine recruitment sources. He found that new employees who were recruited by means of informal sources (i.e., employee referrals, direct applications) had a lower turnover rate than individuals recruited via formal sources (i.e., newspaper advertisements, employment agencies). The publication of Ullman’s study (Ullman, 1966) stimulated several additional recruitment source studies (e.g., Blau, 1990; Gannon, 1971; Saks, 1994). Some of these studies attempted to test whether the two explanations (the realistic information hypothesis and the individual difference hypothesis) that have been offered to explain source effects were supported by empirical data. For example, Taylor and Schmidt (1983) examined source differences in absenteeism, turnover, and performance for employees who were recruited by a variety of sources (e.g., referrals, rehired former employees). They also measured several individual difference variables that might explain any source differences they found. Taylor and Schmidt reported that, compared to the other sources, rehires had longer tenure and lower absenteeism. They also found some support for the individual difference explanation for their findings. One would also presume that those in the rehire group had a fairly accurate view of what the job involved. Although the results of recruitment source studies with regard to outcome variables are not entirely consistent, it frequently has been found that employee referrals and direct applications had lower levels of turnover and higher levels of satisfaction than persons recruited by more formal sources (Breaugh, 1992). However, empirical data have not consistently supported either the information realism or the individual difference explanations for the recruitment source differences reported (Barber, 1998). The failure of research to support these explanations for source differences clearly raises concern about their validity. However, methodological weaknesses of studies is an alternative explanation for the failure to find support for the information realism and individual difference hypotheses. We will discuss two of these methodological limitations (i.e., the samples utilized, the measurement of variables). In order to adequately test whether different recruitment sources result in job applicants who differ in terms of the realism of job expectations or with regard to JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


420


individual differences, a researcher needs to gather data from job applicants. Yet, as noted by Barber (1998), in the great majority of recruitment source studies, data have been gathered from new employees. Consider the study by Breaugh and Mann (1984). In this study, it was reported that recruitment sources did not affect the ability level of new employees. Although this may be the case, one can question how relevant the data reported by Breaugh and Mann (1984) are to testing the individual difference hypothesis. Although new employees in this study may not have differed in terms of their ability, applicants recruited from different sources may have. However, such an applicant ability difference may have been eliminated as a result of selection procedures used by the hiring organization. In summary, given that Breaugh and Mann tested for ability differences using new employees rather than job applicants, it is not clear what one can conclude from their results. In contrast to the results reported by Breaugh and Mann, the two studies that have examined source differences on applicants suggest that informal recruitment sources were linked to more qualified job applicants. For example, Kirnan, Farley, and Geisinger (1989) found that persons recruited from informal sources (e.g., employee referrals) were of higher quality (and were more likely to be offered jobs). Williams, Labig, and Stone (1993) also found that informal sources “reached differently qualified applicants in terms of nursing experience and education which, in turn, were valid predictors of subsequent nurse performance” (p. 163). In summary, in order to advance our understanding of why recruitment sources may influence work outcomes, more attention needs to be focused on applicant differences on key intervening variables. A second methodological weakness of many recruitment source studies concerns the measurement of variables. In order to provide a fair test of the realism and the individual explanation for source effects, it is critical that these variables be measured precisely and measured at the appropriate time. With regard to measurement precision, researchers have suggested that recruitment source effects may be due to individual differences in such factors as motivation, ability, or perceived ease of movement. However, instead of measuring variables such as ability, “investigators all too often chose demographic proxies bearing little theoretical relevancy to criteria of source effectiveness” (Griffeth, Hom, Fink, & Cohen, 1997: 20). Even when researchers have attempted to measure individual difference variables more directly, problems emerge. For example, although Griffeth et al. assessed quality in a rigorous fashion (i.e., during interviews, ratings were made of work experience, training, etc.), these authors reported data on new hires rather than applicants. Measurement precision also has been a problem in testing the realism explanation for source differences. For example, although Williams et al. (1993) gathered data from job applicants, they tested the realism explanation for source differences indirectly (i.e., they measured the amount of pre-hire information the applicant possessed, rather than the accuracy of this information). Conceivably, the reason that Williams et al. focused upon information quantity is due to the time frame for measuring variables in their study. These authors measured several key variables during the job application process. Given JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



421

this time frame, it would have been difficult for individuals to assess the accuracy of their job expectations. In contrast to Williams et al., who measured variables during the application process, several researchers can be criticized for waiting too long to measure variables. For example, in attempting to measure recruitment source effects, many researchers (e.g, Griffeth et al., 1997) have measured job satisfaction, performance, and turnover at the end of one year of employment. It is quite likely that recruitment effects would be stronger if measured closer to the time of hiring (Wanous, 1992). Similarly, some researchers (Breaugh & Mann, 1984) measured the accuracy of job expectations after a year on the job. In contrast to relying so heavily on a person’s memory, Saks (1994) measured the realism of job expectations after one month of employment. Having reviewed the research on recruitment sources, Barber (1998) concluded that past research has not made a strong case for the importance of source differences. We agree. However, based upon the findings of the two studies (Kirnan et al., 1989; Williams et al., 1993) that presented data on job applicants, we believe that it is premature to conclude that recruitment sources do not merit attention. Instead, we believe that research in this area needs to be reoriented. Future research should focus more heavily upon job applicants, rather than new hires, and should give more attention to the measurement of key variables. With regard to future research, there are two remaining themes we would like to emphasize (i.e., expanding the range of intervening variables that are considered, giving more attention to the employer’s perspective). If one reflects on the reasons why recruitment sources have been hypothesized to be associated with certain work outcomes, several factors that may explain inconsistent results become apparent. Consider the use of employee referrals. Researchers have suggested that one reason that employee referrals may be associated with beneficial work outcomes is due to the fact that employees making referrals may pre-screen potential applicants. As was noted by Breaugh (1992), pre-screening is most likely to occur when the person making the referral is concerned about his or her reputation in the organization. Given that some recruitment source studies have involved work situations (e.g., summer jobs) in which the employees making referrals may not always have been concerned about their reputations, one may question how much pre-screening actually took place in certain studies. In order to better understand the recruitment process, in future research, it is important that data be gathered from those who made referrals to assess whether pre-screening actually occurred. In a similar vein, it has been suggested that individuals recruited via different sources may differ in terms of their motivation to find a job and/or their perceptions of job mobility. Clearly, it is time for researchers to start measuring such explanatory variables directly rather than measuring demographic variables that may be related to them (Griffeth et al., 1997). In the first part of this paper, we argued that more attention should be given to several process variables. With regard to recruitment sources, we think this is a critical need. For example, consider the issues of attracting applicant attention (Does an attendee at a job fair stop at a particular employer’s booth?) and generating applicant interest (Does an attendee submit an application?). To our JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


422


knowledge, no recruitment source study has addressed the issues of attracting attention and generating interest. Yet, in a labor market in which jobs are plentiful, these are critical issues for employers. In addition to focusing upon process variables such as applicant attention, researchers also need to focus more heavily upon important pre-hire variables on which recruitment sources are likely to have a more immediate impact than they do on post-hire variables. For example, corporate recruiters tend to be interested in such factors as whether the quality of job applicants is higher for certain recruitment sources, whether certain sources are more likely to yield a higher percentage of new hires, and whether certain sources are likely to generate minority applicants (Barber, 1998). Given that only two studies (Kirnan et al., 1989; Williams et al., 1993) have presented data on applicants as well as new hires, it is difficult for current research to shed much light on these key concerns (Kirnan et al.’s findings suggested that individuals recruited via referrals tended to be of higher quality than persons recruited by means of more formal sources and that minorities were more likely to use formal recruitment sources). As noted by Barber (1998), it is curious that little research has investigated why employers recruit as they do. Two studies by Rynes and her colleagues are exceptions. Rather than looking at how employers determined what recruitment sources to use, Rynes and Boudreau (1986) examined how 145 companies conducted college recruiting programs. Among the issues they investigated were the factors that affected the companies’ selection of colleges. Rynes and Boudreau reported that the factors given the most weight were a college’s reputation in critical skill areas, the general reputation of the school, and the performance of previous hires from the school. As pointed out by Rynes, Orlitsky, and Bretz (1997), much of recruitment research has focused on recruiting college students. In an attempt to understand how employers make decisions about whether to recruit new college graduates as opposed to targeting more experienced individuals, Rynes et al. gathered data from staffing professionals at 251 companies. They found that experienced hires were evaluated more highly in terms of their technical skills, their interpersonal skills, the realism of their job expectations, and their likelihood of success on the job. New college graduates received higher ratings for being willing to learn new things. Rynes et al. also asked the staffing professionals to rate the overall effectiveness of several recruitment sources. Among the sources receiving above-average ratings were informal referrals, newspaper ads, private search firms, and direct applications. College placement services, professional associations, and on-line employment exchanges received below-average ratings. Given the problems that have plagued recruitment source research (e.g., the focus on new employees rather than applicants), it is difficult to draw many firm conclusions. On a more positive note, this is an area that is ripe for real progress to be made. In this paper, we have emphasized the importance of further theory development. However, we also think that more researchers should follow the lead of Rynes and her colleagues in investigating why employers make the decisions that they do (e.g., why certain recruitment sources are selected). JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



423

Research on Recruiters

Given the central role that recruiters play in the recruitment process, it is not surprising that considerable research has addressed the issue of recruiters and their impact on job candidates. Alderfer and McCord (1970) conducted one of the first studies of recruiter characteristics. They found that applicant perceptions of recruiters’ behaviors (e.g., their willingness to provide information) and attitudes (e.g., their interest in the candidate) were associated with applicants’ expectations of receiving a job offer and their reported probability of accepting the offer. Following the publication of Alderfer and McCord’s study, several other researchers (e.g., Harris & Fink, 1987; Macan & Dipboye, 1990; Taylor & Bergmann, 1987) examined the impact of recruiter informativeness and personableness. Although these studies differed in terms of their research designs (e.g., Taylor and Bergmann gathered longitudinal data while most researchers gathered cross-sectional data) and their data analysis strategies (e.g., unlike most researchers, in order to better evaluate recruiter effects, Harris and Fink controlled for pre-interview impressions of job attributes), they generally replicated the results of Alderfer and McCord. That is, in several studies, job applicants who perceived recruiters as being informative and personable also rated the prospective job and the prospective employer as being more attractive. These positive impressions of the job and the employer were, in turn, related to applicants’ reporting a higher expectancy of receiving a job offer and a higher likelihood of accepting a job offer. At the same time that researchers investigated the relationship between applicant perceptions of recruiter behavior and their reactions to the recruitment process, they frequently also examined whether recruiter demographic characteristics or other recruiter attributes (e.g., training) were related to applicant reactions. In terms of recruiter demographic characteristics, a few studies have linked them to applicant reactions to the recruitment process. However, in general, studies have either not found a relationship or the relationships reported were small in magnitude (Barber, 1998). As an example, consider recruiter gender. Barber and Roehling (1993) and Taylor and Bergmann (1987) found that interacting with male recruiters had a small positive association with applicants’ reactions to the recruitment process. In contrast, Connerley (1997), Connerley and Rynes (1997), Harris and Fink (1987), and Turban and Dougherty (1992) found no evidence that recruiter gender was related to applicant reactions. Instead of looking directly at recruiter demographic variables, a few studies have examined whether being similar to the recruiter on certain dimensions might be important to job applicants. For example, Turban and Dougherty (1992) studied whether applicants and recruiters’ being of the same gender or from the same university might have a beneficial effect on applicants. They found that both types of similarity were related to job attractiveness. In contrast, Cable and Judge (1996) did not find demographic similarity (e.g., age, race, gender) between applicants and recruiters to be associated with such variables as job choice intention. JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


424


In trying to understand why certain recruiters may have a more positive effect on applicants than other recruiters, researchers have looked at several other possible explanatory variables. With regard to recruiter experience and training, results have been mixed. For example, Connerley and Rynes (1997) reported that more experienced recruiters were viewed more positively by job applicants. In contrast, Connerley (1997) found no effect for experience. In general, recruiter training has not been linked to applicant reactions (e.g., Connerley, 1997; Stevens, 1998). Researchers also have investigated whether a recruiter’s functional area (e.g., personnel department vs. a line function) made a difference. Connerley (1997) and Taylor and Bergmann (1987) reported that the use of line managers as recruiters had a positive influence on applicants. In contrast, Harris and Fink (1987) and Connerley and Rynes (1997) did not find a recruiter’s functional area to make a difference. In trying to make sense out of research on recruiters, it is useful to consider the reasons that have been offered by researchers for why recruiters might have an effect on job candidates. Recruiters have been hypothesized to be important because they serve a key role in directly communicating information about a position (e.g., types of duties) and an organization (e.g., its values) to applicants. In order to carry out this communication role effectively, a recruiter needs to have both the ability to communicate effectively (e.g., sufficient time) and personal credibility. In addition to intentionally conveying information to a job candidate, a recruiter may also have an impact on applicants for unintentional reasons. It has been argued (Rynes, 1991) that applicants often view a recruiter as a signal of unknown aspects of the organization. Concerning recruiters being used as a signal, Rynes, Bretz, and Gerhart (1991) reported evidence of signaling. In particular, they reported that applicants were likely to view the recruiter as being “symbolic of broader organizational characteristics” (p. 487) when applicants lacked detailed knowledge of the hiring organization and when the recruiter was from their functional area. This signaling effect occurred both during the initial campus interview and during site visits. Although there is considerable evidence that recruiters who are perceived as being more personable and more informative are associated with beneficial outcomes for the organization (e.g., applicants reporting a higher likelihood of accepting a job offer), the results of some studies have not found such associations. Furthermore, in many studies these associations have been modest in magnitude. There are a number of possible explanations for inconsistent findings and for the modest magnitude of some of the results reported (e.g., recruiters frequently do not have sufficient time to present all of the information they would like to). Although space limitations do not allow for a discussion of potential limiting factors, the topic of recruiter credibility does merit some attention. Research on recruiter credibility has shown that job incumbents are perceived as a more credible source of information than are full-time corporate recruiters (Coleman & Irving, 1997; Fisher et al., 1979). If one considers the two factors (expertise and trustworthiness) that are seen as resulting in perceptions of credibility, this finding is not surprising. During the recruitment process, applicants want specific information about several job-related issues. Recruitment JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



425

research (e.g., Maurer et al., 1992) regularly has shown that applicants report not having received as much job-related information as they would like. If one considers the issue of expertise, the reason for this lack of information is apparent. Frequently, recruiters are not very familiar with the job and the job environment. One can imagine the difficulty that a full-time recruiter would have trying to answer questions about a particular supervisor’s style or a work group’s climate. A few researchers (e.g., Harris & Fink, 1987) have investigated whether interacting with a recruiter from the candidate’s functional area as opposed to one from the human resource department makes a difference. Although such a distinction may be worth examining, a better line of inquiry may be whether the person with whom a job candidate interacts is the candidate’s potential supervisor or coworker (e.g., a person in one’s functional area may know little about important aspects of a particular job). One would expect that such a supervisor or coworker would be seen as having much more expertise than other types of recruiters. It is not sufficient for a person to have expertise to be seen as a credible source; the person also needs to be seen as trustworthy. Research (e.g., Coleman & Irving, 1997) has shown that recruiters from a human resource department are seen as less trustworthy than job incumbents. Given that full-time recruiters may be evaluated in terms of the number of individuals they bring into the selection process (Breaugh, 1992), it is not surprising that applicants may sometimes see them as presenting an overly positive view of a job opening. In contrast, potential coworkers may be seen as being more trustworthy because they will have to interact with the job candidate if the person is hired. From this brief discussion of recruiter credibility, the need for future research that examines the influence of whether the recruiter is one’s potential supervisor or coworker should be apparent. It is likely that recruiter effects would be stronger if distinctions were made between potential supervisors and coworkers versus full-time corporate recruiters. Over time, recruiter-oriented research has become more complex and more methodologically sound. For example, early studies of recruiters (e.g., Alderfer & McCord, 1970) often involved simply gathering survey data from college students immediately following a campus interview. More recently, research on recruiters has involved: (a) gathering data both before and after interviews (e.g., Harris & Fink, 1987), (b) gathering longitudinal data over various stages of the recruitment process (e.g., Taylor & Bergmann, 1987), (c) gathering data on recruiter behavior from both recruiters and job applicants (e.g., Connerley & Rynes, 1997), and (d) more intensively studying recruitment interactions by means of in-depth interviews (e.g., Rynes et al., 1991) and the audiotaping of applicant-recruiter interactions (e.g., Stevens, 1998). This increasing sophistication of research has resulted in a better understanding of recruiters and their potential effects. For example, in the 1980s, some researchers questioned whether recruiters had an important effect on job applicants (Rynes, 1991). Today, this view is less commonly held (Barber, 1998). For instance, in their intensive study of the recruitment process, Rynes et al. (1991) found that recruiters had a powerful impact. In a number of cases, poor recruiters resulted in individuals’ losing interest in jobs. JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


426


In terms of future research directions, clearly the trend toward more complex research should continue. As others have suggested (e.g., Barber, 1998), we too believe that relatively less attention should be given to recruitment on college campuses and to the initial stage of the recruitment process. In the future, researchers need to explore the influence of recruiters at job fairs or during site visits (Turban, Campion, & Eyring, 1995 have presented evidence that the host of a site visit can influence job applicant perceptions of job attractiveness and their likelihood of accepting a job offer). Research on Realistic Job Previews

Over the last thirty years, realistic job previews (RJPs) have attracted substantial attention. As described earlier, several different RJP models have been offered. These models suggest that providing realistic information to job candidates should have beneficial effects for a number of reasons (e.g., RJPs allow candidates to self-select out of consideration for jobs that would not meet their expectations, RJPs cause individuals to be more committed to their job acceptance decision). To provide a sense of the research on RJPs, we will review a few typical studies. In one of the early RJP studies, Wanous (1973) examined the effects of applicants for the job of telephone operator viewing either a traditional recruitment film or an experimental film that provided both positive and negative information about this position. He found that viewing the RJP film had no effect on job offer acceptance rate (78 of the 80 individuals in his study accepted offers), but did result in lower job expectations. Wanous reported there was no RJP effect for turnover, but that receiving an RJP did positively affect job satisfaction. RJPs generally have been provided by means of a booklet or a video (Phillips, 1998). Colarelli (1984) felt that an interview with a job incumbent might be a more powerful RJP medium given that such a two-way conversation may (a) increase an applicant’s attention and comprehension, (b) enhance the amount of personally relevant information conveyed, and (c) increase the importance attached to the information conveyed as a result of job incumbents’ being a highly credible source of information. Colarelli’s study involved applicants for the position of bank teller. His research design involved comparing a control group against individuals who received an RJP via a brochure or through an interview with a teller or a teller trainer. Given that everyone offered a job accepted it, Colarelli found no evidence of self-selection. He did find that the incumbent group had a lower turnover rate than the other two groups. With regard to the realism of job expectations, Colarelli reported that both of the RJP groups felt they had more realistic expectations than those in the control group. In terms of the personal relevance of the information received, the incumbent group had higher scores than the brochure group, which had higher scores than the control group. Colarelli did not find an RJP effect for job satisfaction. Unfortunately, although Colarelli hypothesized that receiving an RJP from a job incumbent would increase applicant attention and comprehension, he did not measure these variables. In 1985, two RJP meta-analyses were published. McEvoy and Cascio (1985) compared the effects on turnover of providing an RJP versus using a job enrichJOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



427

ment strategy. They reported the average phi coefficient between turnover and the use of an RJP was .09. They also reported that RJPs had a stronger effect on turnover for more complex jobs. Premack and Wanous (1985) reported a similar phi coefficent (.06) between the use of an RJP and turnover to that reported by McEvoy and Cascio. Premack and Wanous also examined RJP effects on several other variables (e.g., job satisfaction, performance, initial expectations). For the most part, the RJP effects they reported were in the expected direction but were small in magnitude. Despite the somewhat pessimistic conclusions of some authors (e.g., McEvoy & Cascio, 1985) concerning RJP effects, RJPs have continued to attract attention. During the 1990s, researchers examined such issues as (a) whether the RJP medium (written vs. oral) makes a difference (Saks & Cronshaw, 1990), (b) whether having prior exposure to a job moderates the impact of receiving an RJP (Meglino, Ravlin, & DeNisi, 1997), and (c) whether RJPs make higher quality applicants less likely to pursue job openings (Bretz & Judge, 1998). Although it is not possible for us to provide a review of all of the recent RJP studies, a recent paper by Phillips (1998), which includes a meta-analysis of empirical research, provides an excellent summary of the current state of RJP research. Overall, Phillips found that RJPs were related to higher job performance and to lower levels of initial job expectation, attrition from the recruitment process, and voluntary turnover. Phillips also reported that RJP effects were stronger when the RJP was provided verbally rather than via a videotape or a booklet. Phillips emphasized that many of the RJP effects were quite modest in magnitude. A review of Phillips’ paper (Phillips, 1998) may help explain some of the confusion in the RJP literature. For example, in almost one-half of the studies included in her meta-analysis, the RJP was provided after an individual had been hired. Given that RJPs have been hypothesized to “work” because they increase self-selection, provide a sense the employer is trustworthy, and increase commitment to the job choice decision, providing an RJP after hiring should lessen its effect (e.g., having already accepted a job offer, a person may not see the organization as being particularly trustworthy and may not feel committed to the “uninformed” decision he or she made). RJPs also are thought to be most effective when an applicant has unrealistic expectations and feels capable of turning down a job opportunity that is not seen as desirable. Given the visibility of several of the jobs (e.g., bank teller, grocery store checker, gas station cashier) that have been addressed in RJP studies (see Phillips, 1998: Table 1), one has to consider the degree to which the job applicants in a number of these studies had unrealistic job expectations. In a similar vein, it appears that economic conditions may have made self-selection unlikely in some RJP studies. For example, 78 of the 80 individuals offered jobs in Wanous’ study (Wanous, 1973) accepted them. Wanous noted that a difficult labor market “probably reduced an individual’s freedom to reject a job offer on the basis of a preview” (p. 331). Another possible reason for some of the modest RJP effects that have been reported is the nature of the job previews that have been provided. Research (e.g., Barber & Roehling, 1993) suggests that applicants want specific information about job and organizational attributes. Dean and Wanous (1984) pointed out that JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


428


the RJPs provided in many studies did not involve the communication of specific RJP information. In summary, there are several potential explanations for the modest RJP effects that have been reported. One can only imagine the potential effects RJPs might have if future studies involve situations in which RJPs (a) provide specific information about a wide range of important job attributes, (b) are conveyed by a credible source such as a future coworker, and (c) are targeted at applicants (not new employees) who lack realistic job information and have other viable job opportunities. Recently, Meglino et al. (1997) and Bretz and Judge (1998) have raised the issue of whether providing realistic job preview information may hurt an employer’s chances of hiring desirable job applicants. For example, based upon their own research and a meta-analysis of other relevant research, Meglino et al. (1997) concluded that providing realistic information to applicants who had prior exposure to that type of position resulted in their overestimating the negative aspects of the job. This overestimation could result in individuals with prior job experience, individuals who are frequently viewed as desirable job candidates, rejecting job offers. Bretz and Judge (1998) also suggested that higher quality job candidates may be less willing to pursue jobs for which negative information has been provided. They provided some evidence that supports this position. Although neither set of authors recommended that RJP information should not be provided, others may come to this conclusion from their research. In a recent paper, Buckley, Fedor, Carraher, Frink, and Marvin (1997) took a very different position. These authors argued that an employer has an “ethical imperative to provide recruits with realistic job previews” (p. 468). One of the reasons that RJPs are thought to have positive effects on turnover and satisfaction is because they lower job expectations. In a recent paper, Buckley, Fedor, Veres, Wiese, and Carraher (1998) advocated the wider use of RJPs but also noted that creating an RJP can be an expensive process. Given this fact, they experimented with an alternative expectation lowering procedure (ELP). In their study, an ELP involved an individual participating in a workshop that (a) stressed the fact that individuals tend to have unrealistically high job expectations, (b) addressed the likelihood that such expectations are likely to be unfulfilled, and (c) discussed the common outcomes attached to unfulfilled expectations. To test the impact of receiving an ELP, Buckley et al. conducted an experiment with newly hired employees who had yet to start work. These individuals were divided into four groups. One group went through the traditional orientation program. A second group received a written RJP that was based upon a detailed job analysis. A third group was composed of those who participated in the ELP workshop. The fourth group was a control group. Buckley et al. found that the RJP and the ELP were equally effective in lowering job expectations. They reported a strong association between expectation level and subsequent employee turnover. In our opinion, ELPs merit future research as a potential low-cost alternative to RJPs. Given that many new employees appear to have incorrect expectations concerning what their new jobs entail (Wanous, 1992), we believe that a realistic job preview represents an important recruitment mechanism. However, we think JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



429

there may be value in focusing less on an RJP as a specific event (e.g., showing a 10-minute video) and instead focusing more on the process of providing accurate job information at several points during the recruitment process. For example, an employer could begin to provide realistic information with its job advertisement. Following this, it could increase the amount of information conveyed during the initial interview by selecting recruiters who are likely to really understand the job situation. Additional realistic job information could be communicated by a visit to the work site and/or through applicants taking part in a work sample (Wanous, 1992). By viewing realistic recruitment as a process, an employer may more thoughtfully plan out recruitment events. Trying to provide realistic job information at a series of points via different mechanisms may also lessen the likelihood of a recruit’s becoming overloaded with information (Barber, 1998). In summary, we believe research on RJPs has resulted in a better understanding of the consequences of hiring individuals who have realistic expectations. Our hope is that in future years researchers will focus more attention on the conditions in which RJPs are utilized (e.g., Did job candidates lack accurate job information?) and on the psychological variables (e.g., the credibility of the message) that are hypothesized to underlie RJP effectiveness. Other Recruitment Topics

Recruitment advertising. Although firms spend considerable money on job advertising, little research has addressed this topic. Previously, we reviewed research (e.g., Barber & Roehling, 1993) that documented the positive effects of providing a recruitment message that conveyed both more information and more specific information. Other research dealing with recruitment ads has also shown the benefits of their including specific information (Mason & Belt, 1986) and more information (Yuce & Highhouse, 1998). Recently, Highhouse, Beadle, Gallo, and Miller (1998) examined whether the wording of an advertisement for a restaurant position affected college students’ perceptions of the estimated hourly pay, the company’s image, and their job pursuit intentions. The only difference Highhouse et al. reported was a position scarcity effect on pay (e.g., when only a few jobs were advertised, pay was estimated to be over $1.70 more than if many openings were advertised). The results of Highhouse et al.’s study suggest that job applicants may infer certain information (i.e., pay rate) from the wording of an ad. It is important for future research to investigate what other inferences may be drawn from recruitment ads. The site visit. To date, little research has addressed the importance of the site visit. The findings of the studies that have are mixed. For example, Taylor and Bergmann (1987) reported that, after entering expected job attributes as predictor variables in regression analysis, the site visit (e.g., how well recruits were treated) had no effect on applicants’ ratings of company attractiveness or of their likelihood of accepting job offers. In contrast, Rynes, Bretz, and Gerhart (1991) found that for their sample the site visit played a significant role in the recruitment process. In particular, job seekers noted the importance they attached to the status of the people they met, whether they felt “specially treated,” and the organizaJOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


430


tion’s professionalism during their visit. Turban, Campion, and Eyring (1995) reported that various ratings of the site visit (e.g., host likeableness) were predictive of the recruits’ ultimate decision of whether to accept a job offer. An interesting feature of this study was the fact that the site visit host would be a coworker of the recruit if he or she were hired. As noted by Barber (1998), in comparison to the issue of generating job applicants, the topic of maintaining applicant interest has attracted relatively little research attention. Given that a site visit may have an important influence on whether job applicants remain interested in a position, we believe the site visit deserves more research attention. Timing issues. Although being timely in terms of recruitment actions is generally viewed as being important (Barber, 1998), only a few studies have addressed timing issues. With regard to the timing of recruitment actions, Rynes, Bretz, and Gerhart (1991) reported that for their sample: (a) long delays between various phases of the recruitment process were common, (b) job applicants frequently drew negative inferences about the delays (e.g., “I am not their favorite candidate”), (c) delays affected the willingness of individuals to accept job offers, and (d) the most marketable job seekers frequently saw delays as reflecting something was wrong with the organization. In contrast to the results reported by Rynes et al. (1991), Taylor and Bergmann (1987) reported no timing effect for their study (conceivably, large sample attrition in this study could have affected results). Given the limited amount of research that has been conducted, it is premature to attach too much significance to the timing of recruitment actions. However, this appears to be yet another area that is ripe for research. Concluding Remarks

Over the last three decades, the amount of research on recruitment topics has increased substantially. In spite of this increase, recent reviews of the recruitment literature (e.g., Barber, 1998; Rynes, 1991) have had a somewhat pessimistic tone. For example, reviewers frequently have concluded that we still do not know a great deal about why recruitment activities (e.g., using certain recruitment sources) have the effects they do. In particular, reviewers have criticized many recruitment studies for being poorly designed, narrow in focus, and not grounded in theory. Based upon our review of recruitment research, we think that these criticisms are valid. We also believe that, in order for future research to lead to a better understanding of recruitment issues, studies need to be designed with an appreciation of the complexity of the recruitment process. In order to stimulate the type of research that has been called for in previous reviews, we presented an organizing framework of the recruitment process. In introducing this framework, we gave particular attention to the process variables (e.g., applicant attention, self-insight) that theory suggests should mediate the relationships between recruitment activities and recruitment outcomes. In reviewing the empirical recruitment literature, we highlighted how often these process variables have been ignored. Given Barber’s recent review of recruitment research (Barber, 1998), we were able to be selective in reviewing research. In selecting studies to review, we JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



431

tried to choose studies that (a) gave the reader a sense of the typical study that has been conducted (e.g., Wanous’ paper [Wanous, 1973] on realistic job previews), (b) called into question conventional thinking (e.g., Greenhaus, Seidel, & Marinis’s paper [Greenhaus et al., 1983] on the issue of value attainment), (c) highlighted a better research design for investigating an issue (e.g., the focus by Williams et al. [1993] upon recruiting source effects on job applicants), or (d) suggested new streams of research. Although one can view the existing recruitment literature as a hodgepodge of inconsistent results, we feel there are certain themes that emerge. For example, research on recruiters makes apparent the critical role these individuals can play both in terms of being informative and in terms of treating applicants in a personable fashion. These themes of informativeness and personable treatment also permeate other areas of recruitment research, such as research on the site visit and the timing of recruitment actions. The importance of providing realistic job information (e.g., via employee referrals or realistic job previews) has also been an important research theme in the recruitment literature. Another theme that emerged in our review of the literature is the importance of “signals” that employers may unintentionally be sending to job applicants. For example, Rynes et al. (1991) showed that applicants used information with regard to recruiter friendliness and informativeness as an indicator of how a firm treated an employee. Although our review frequently criticized past research, our objective in doing this was not to chastise previous researchers. Rather, our intent was to use the deficiencies of past research to stimulate future research that is not subject to these criticisms. Having critiqued past research, we should acknowledge the difficulty of doing research on recruitment topics. To begin with, it is difficult to define the term recruitment and to distinguish what falls within the definition. For example, we began this paper by citing Barber’s definition of recruitment (“recruitment includes those practices and activities carried on by the organization with the primary purpose of identifying and attracting potential employees”; Barber, 1998: 5). This definition says nothing about recruiting individuals who are likely to be successful on the job. Some would argue that the notion of success should be part of this definition. In a similar vein, consider an employer that uses a work sample as a selection device. If going through a work sample provides a job candidate with realistic information about a position, should the work sample be considered a selection device, or recruitment device, or both? The topic of organizational image also blurs the boundary between recruitment and other topics. For example, Turban, Forret, and Hendrickson (1998) found that an organization’s reputation influenced job applicants. We chose not to review the organizational image literature given that an employer’s reputation frequently is not directly a function of recruitment activities (e.g., media accounts of business successes or failures are more likely to affect a company’s image than are job advertisements). However, an organization’s image can clearly influence organizational recruitment. In closing, we choose to end on an optimistic note. In our opinion, this is an exciting time for recruitment researchers. The difficulty in filling jobs has focused JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


432


organizational attention on the importance of well-designed recruitment activities. Recent research has been more theory-based and more methodologically sophisticated (e.g., Stevens [1997] used audiotapes to better understand recruiter/ applicant interactions; Barber & Roehling [1993] used verbal protocol analysis). The continued use of such methods should prove fruitful in investigating the many unanswered questions that surround the recruitment process. References Alderfer, C. P., & McCord, C. G. 1970. Personal and situational factors in the recruitment interview. Journal of Applied Psychology, 54: 377–385. Barber, A. E. 1998. Recruiting employees . Thousand Oaks, CA: Sage Publications. Barber, A. E., & Roehling, M. V. 1993. Job posting and the decision to interview: A verbal protocol analysis. Journal of Applied Psychology , 78: 845–856. Blau, G. 1990. Exploring the mediating mechanisms affecting the relationship of recruitment source to employee performance. Journal of Vocational Behavior , 37: 303–320. Breaugh, J. A. 1992. Recruitment: Science and practice . Boston: PWS-Kent. Breaugh, J. A., & Billings, R. S. 1988. The realistic job preview: Five key elements and their importance for research and practice. Journal of Business and Psychology , 2: 291–305. Breaugh, J. A., & Mann, R. B. 1984. Recruiting source effects: A test of two alternative explanations. Journal of Occupational Psychology , 57: 261–267. Bretz, R. D., & Judge, T. A. 1998. Realistic job previews: A test of the adverse self-selection hypothesis. Journal of Applied Psychology, 83: 330–337. Buckley, M. R., Fedor, D. B., Carraher, S. M., Frink, D. D., & Marvin, D. 1997. The ethical imperative to provide recruits realistic job previews. Journal of Managerial Issues , 9: 468–484. Buckley, M. R., Fedor, D. B., Veres, J. G., Wiese, D. S., & Carraher, S. M. 1998. Investigating newcomer expectations and job-related outcomes. Journal of Applied Psychology , 83: 452–461. Bureau of National Affairs. 1988. Recruiting and selection procedures. Personnel Policies Forum , Survey No. 146 (May). Washington, D.C. Cable, D. M., & Judge, T. A. 1996. Person-organization fit, job choice decisions, and organizational entry. Organizational Behavior and Human Decision Processes , 67: 294–311. Chaiken, S., & Stangor, C. 1987. Attitudes and attitude change. In.M. Rosenzweig & L. W. Porter (Eds.), Annual review of psychology: 575–630. Palo Alto, CA: Annual Reviews Inc. Colarelli, S. M. 1984. Methods of communication and mediating processes in realistic job previews. Journal of Applied Psychology, 69: 633–642. Coleman, D. F., & Irving, P. G. 1997. The influence of source credibility attributions on expectancy theory predictions of job choice. Canadian Journal of Behavioral Science , 29: 122–131. Connerley, M. L. 1997. The influence of training on perceptions of recruiters’ interpersonal skills and effectiveness. Journal of Occupational and Organizational Psychology , 70: 259–272. Connerley, M. L., & Rynes, S. L. 1997. The influence of recruiter characteristics and organizational recruitment support on perceived recruiter effectiveness: Views from applicants and recruiters. Human Relations, 50: 1563–1586. Dean, R. A., & Wanous, J. P. 1984. Effects of realistic job previews on hiring bank tellers. Journal of Applied Psychology, 69: 61–68. Downs, S. R., Farr, M., & Colbeck, L. 1978. Self-appraisal: A convergence of selection and guidance. Journal of Occupational Psychology , 51: 271–278. Fedor, D. B., Buckley, M. R., & Davis, W. D. 1997. A model of the effects of realistic job previews. International Journal of Management , 14: 211–221. Fisher, C. D., Ilgen, D. R., & Hoyer, W. D. 1979. Source credibility, information favorability, and job offer acceptance. Academy of Management Journal , 22: 94–103. Gannon, M. J. 1971. Sources of referral and employee turnover. Journal of Applied Psychology , 55: 226–228. Greenhaus, J. H., Seidel, C., & Marinis, M. 1983. The impact of expectations and values on job attitudes. Organizational Behavior and Human Performance , 31: 394– 417. Griffeth, R. W., Hom, P. W., Fink, L. S., & Cohen, D. J. 1997. Comparative tests of multivariate models of recruiting source effects. Journal of Management , 23: 19–36. Guion, R. M. 1976. Recruiting, selection, and job placement. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology : 777–828. Chicago: Rand-McNally. Harkins, S. G., & Petty, R. E. 1981. The multiple source effect in persuasion. Personality and Social Psychology Bulletin, 7: 627–635. JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000



433

Harris, M. H., & Fink, L. S. 1987. A field study of applicant reactions to employment opportunities: Does the recruiter make a difference? Personnel Psychology , 40: 765–784. Highhouse, S. Beadle, D., Gallo, A., & Miller, L. 1998. Get ’em while they last! Effects of information scarcity in job advertisements. Journal of Applied Social Psychology , 28: 779–795. Highhouse, S., & Hause, E. L. 1995. Missing information in selection: An application of the Einhorn-Hogarth ambiguity model. Journal of Applied Psychology , 80: 86–93. Highhouse, S., Stierwalt, S. L., Bachiochi, P., Elder, A. E., & Fisher, G. 1999. Effects of advertised human resource management practices on attraction of African American applicants. Personnel Psychology , 52: 425–442. Hom, P. W., Griffeth, R. G., Palich, L. E., & Bracker, J. S. 1998. An exploratory investigation into theoretical mechanisms underlying realistic job previews. Personnel Psychology , 51: 421–451. Jablin, F. M., Putnam, L. L., Roberts, K. H., & Porter, L. W. 1987. Handbook of organizational communication . Newbury Park, CA: Sage. Kalleberg, A. L., Knoke, D., Marsden, P. V., & Spaeth, J. L. 1996. Organizations in America . Thousand Oaks, CA: Sage Publications. Kirnan, J. P., Farley, J. A., & Geisinger, K. F. 1989. The relationship between recruiting source, applicant quality, and hire performance: An analysis by sex, ethnicity, and age. Personnel Psychology , 42: 293–308. Kulik, C. T., & Ambrose, M. L. 1993. Category-based and feature-based processes in performance appraisal: Integrating visual and computerized sources of performance data. Journal of Applied Psychology, 78: 821–830. Lengel, R. H., & Daft, R. L. 1988. The selection of communication media as an executive skill. Academy of Management Executive, 11: 225–232. Macan, T. H., & Dipboye, R. L. 1990. The relationship of interviewers’ preinterview impressions to selection and recruitment outcomes. Personnel Psychology , 43: 745–768. Mason, N. A., & Belt, J. A. 1986. The effectiveness of specificity in recruitment advertising. Journal of Management , 12: 425–432. Maurer, S. D., Howe, V., & Lee, T. W. 1992. Organizational recruiting as marketing management: An interdisciplinary study of engineering graduates. Personnel Psychology , 45: 807–833. McEvoy, G. M. & Cascio, W. F. 1985. Strategies for reducing employee turnover: A meta-analysis. Journal of Applied Psychology, 70: 342–353. Meglino, B. M., DeNisi, A. S., Youngblood, S. A., & Williams, K. J. 1988. Effects of realistic job previews: A comparison using an enhancement and a reduction preview. Journal of Applied Psychology , 73: 259–266. Meglino, B. M., Ravlin, E. C., & DeNisi, A. S. 1997. When does it hurt to tell the truth? The effect of realistic job reviews on employee recruiting. Public Personnel Management , 26: 414–421. Phillips, J. M. 1998. Effects of realistic job previews on multiple organizational outcomes: A meta-analysis. Academy of Management Journal, 41: 673–690. Powell, G. N. 1991. Applicant reactions to the initial employment interview: Exploring theoretical and methodological issues. Personnel Psychology , 44: 67–83. Premack, S. L., & Wanous, J. P. 1985. A meta-analysis of realistic job preview experiments. Journal of Applied Psychology, 70: 706–719. Rousseau, D. R. 1995. Psychological contracts in organizations . Thousand Oaks, CA: Sage Publications. Rynes, S. L. 1991. Recruitment, job choice, and post-hire consequences. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology (2nd. ed.): 399–444. Palo Alto, CA: Consulting Psychologists Press. Rynes, S. L., & Barber, A. E. 1990. Applicant attraction strategies: An organizational perspective. Academy of Management Review, 15: 286–310. Rynes, S. L., & Boudreau, J. W. 1986. College recruiting in large organizations: Practice, evaluation, and research implications. Personnel Psychology , 39: 729–757. Rynes, S. L., Bretz, R. D., & Gerhart, B. 1991. The importance of recruitment in job choice: A different way of looking. Personnel Psychology , 44: 487–521. Rynes, S. L., Orlitsky, M. O., & Bretz, R. D. 1997. Experienced hiring versus college recruiting: Practices and emerging trends. Personnel Psychology , 50: 309–339. Saks, A. M. 1994. A psychological process investigation for the effects of recruitment source and organizational information on job survival. Journal of Organizational Behavior , 15: 225–244. Saks, A. M., & Cronshaw, S. F. 1990. A process investigation of realistic job previews: Mediating variables and channels of communication. Journal of Organizational Behavior , 11: 221–236. Stevens, C. K. 1997. Effects of preinterview beliefs on applicants’ reactions to campus interviews. Academy of Management Journal, 40: 947–966. Stevens, C. K. 1998. Antecedents of interview interactions, interviewers’ ratings, and applicants’ reactions. Personnel Psychology , 51: 55–85. Stiff, J. B. 1994. Persuasive communication . New York: Guildford Press. JOURNAL OF MANAGEMENT, VOL. 26, NO. 3, 2000


Breaugh & Starke (2000)

Recommend Documents