This article was downloaded by:[Edwards, Jeffrey R.] On:15 July 2008 Access Details:[subscription number 794992879] Publisher:Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK The Academy of Management Annals Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t791720496 4 Person-Environment Fit in Organizations: An Assessment of Theoretical Progress Jeffrey R. Edwardsa aKenan-Flagler Business School, University of North Carolina, First Published on:01 August 2008 To cite this Article:Edwards, Jeffrey R. (2008) '4 Person-Environment Fit in Organizations: An Assessment of Theoretical Progress', The Academy of Management Annals, 2:1, 167 — 230 To link to this article: DOI:10.1080/19416520802211503 URL:http://dx.doi.org/10.1080/19416520802211503 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use:http://www.informaworld.com/terms-and-conditions-of-access.pdf Thisarticlemaybeusedforresearch,teachingandprivatestudypurposes.Anysubstantialorsystematicreproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independentlyverifiedwithprimarysources.Thepublishershallnotbeliableforanyloss,actions,claims,proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. The Academy of Management Annals Vol. 2, No. 1, 2008, 167–230 8 00 4 2 y ul J 5 1 Person–Environment Fit in Organizations: 3 3 1: At: 2 An Assessment of Theoretical Progress R.] y e effr J s, d ar w d E y: [ JEFFREY R. EDWARDS* B d Kenan-Flagler Business School, University of North Carolina e d a o nl w o D TR1A1O2T210JjAre0900eaaAcrf.400dyiyfa1gMr180wlld0eoio-0en8yaAr6r0ma r0E5a&2_dl/y 2ndb0A1sA 0dw@F09o_r r84f(aFt3au ip1rMrc2nndra6l1iccsenn5sai3.cnst2e1)iad0/s71gu8.t9es0gm42m12e-1rn61t0 5A607n3a n(oalnslicne)t Person–environment (P–E) fit is a central concept in organizational behavior research. Historically, reviews of P–E fit research have summarized empirical studies but said little about whether P–E fit research has made theoretical progress. This chapter applies criteria for evaluating theory to review and assess the theoretical status and progress of P–E fit research. The review encompasses P–E fit theories that span nearly a century and cover research on job satisfaction, job stress, vocational choice, recruitment and selection, and organizational climate and culture. This review indicates that most theories in P–E fit research fall well short of criteria for developing strong theory, and theories presented in recent years are no stronger than those developed decades earlier. Reasons for theoretical stagnation in P–E fit research are iden- tified, and ways to promote theoretical progress are discussed. *Email: [email protected] ISSN 1941-6520 print/ISSN 1941-6067 online © 2008 Academy of Management DOI: 10.1080/19416520802211503 http://www.informaworld.com 167 168 • The Academy of Management Annals Introduction 8 For decades, person–environment (P–E) fit has maintained a central position 0 0 in organizational behavior research. In general terms, P–E fit refers to the 2 y ul congruence, match, or similarity between the person and environment J 5 (Dawis, 1992; Edwards, Caplan, & Harrison, 1998; Muchinsky & Monahan, 1 3 3 1987; Schneider, Kristof, Goldstein, & Smith, 1997). Specific types of P–E fit 1: At: 2 involve the needs of the person and the rewards provided by the environment R.] (Dawis & Lofquist, 1984; Porter & Lawler, 1968), the abilities of the person y and the demands of the environment (McGrath, 1976; Sells, 1970; Shirom, e effr 1982), and the similarity between the person and the social environment, J s, which can refer to individuals, groups, organizations, or vocations (Chatman, d ar 1989; Holland, 1997; Meglino, Ravlin, & Adkins, 1989; Schneider, 1987). w d E Although the concept of P–E fit dates back to the work of Plato (Dumont & By: [ Carson, 1995), contemporary P–E fit research is often traced to Parsons (1909), ed who developed a matching model to describe the fit between attributes of the d a o person and characteristics of different vocations. Subsequent theoretical work nl w that laid the foundation for P–E fit research includes Murray’s need–press o D model (Murray, 1938, 1951) and Lewin’s field theory (Lewin, 1935, 1951), which produced the formula B = F(P,E) stating that behavior is a function of the person and environment. Following this work, P–E fit emerged as a core concept in research on job satisfaction (Katzell, 1964; Locke, 1976; Schaffer, 1953), job stress (French, Caplan, & Harrison, 1982; McGrath, 1976), vocational choice (Dawis & Lofquist, 1984; Holland, 1997), recruitment and selection (Breaugh, 1992; Wanous, 1992; Werbel & Gilliland, 1999), and organizational culture and climate (Chatman, 1989; Meglino et al., 1989; Schneider, 1987). The various streams of research that fall under the rubric of P–E fit have generated hundreds of studies. This work has been reviewed in narrative sum- maries (Edwards, 1991; Katzell, 1964; Kristof, 1996; Meglino & Ravlin, 1998; Pervin, 1968; Spokane, 1985; Spokane, Meir, & Catalano, 2000) and meta- analyses (Arthur, Bell, Villado, & Doverspike, 2006; Assouline & Meir, 1987; Chapman, Uggerslev, Carroll, Piasentin, & Jones, 2005; Hoffman & Woehr, 2006; Kristof-Brown, Zimmerman, & Johnson, 2005; Tranberg, Slane, & Ekeberg, 1993; Tsabari, Tziner, & Meir, 2005; Verquer, Beehr, & Wagner, 2003). Based on these reviews, one might conclude that P–E fit research is thriving and productive, yielding evidence that has been used to evaluate, modify, and extend theories of P–E fit. Indeed, after decades of work, it would seem inevi- table that P–E fit research has repeatedly run through the cycle linking theory and empiricism that generates scientific progress (Runkel & McGrath, 1972), such that current P–E fit theories are markedly superior to their predecessors. Despite the enormous effort invested in conducting and reviewing empiri- cal P–E fit research, little attention has been devoted to evaluating theories that guide this research. As a result, fundamental questions about P–E fit theories Person–Environment Fit in Organizations • 169 remain largely unanswered. For instance, how would P–E fit theories fare against basic criteria for evaluating theory? Are contemporary P–E fit theories 8 superior to those developed when P–E fit research was in its infancy? Have 0 0 successive statements of P–E fit theories become more refined, adding propo- 2 y ul sitions that were initially overlooked and shedding propositions that were J 5 rejected empirically? Overall, from a theoretical standpoint, has P–E fit 1 3 3 research made progress? 1: At: 2 In this chapter, I take stock of theoretical progress in P–E fit research. R.] Using basic criteria for theory development (Bacharach, 1989; Dubin, 1976, y Sutton & Staw, 1995; Weick, 1995; Whetten, 1989), I evaluate major P–E fit e effr theories, starting with their initial presentations and tracing subsequent devel- J s, opments, and examine the extent to which theories have improved over time. d ar The review starts with work by Parsons (1909), Murray (1938, 1951), and w d E Lewin (1935, 1951), and then examines primary theories of P–E fit within the By: [ domains of job satisfaction, job stress, vocational choice, recruitment and ed selection, and organizational culture and climate. This review shows that d a o many theories are stated in vague terms that obscure the meaning of P–E fit nl w and its relationship with other constructs, and theories proposed in recent o D years are often less sophisticated than those developed decades earlier. Drawing from this assessment I offer suggestions for future theory develop- ment in P–E fit research along with recommendations for empirical research that will produce stronger tests of the theories we propose. The following review is intended to be critical but not adversarial. To this end, I evaluate P–E fit theories against theoretical criteria I believe most researchers would consider relevant and uncontroversial, and I have liberally quoted from sources that present the theories to minimize interpretations of my own. Moreover, this review is not a prologue to some alternative theory I have developed that is immune from the criticisms leveled against the theories examined here. Indeed, my own theoretical work related to P–E fit either draws heavily from the work of others (Edwards et al., 1998) or is tangential to P–E fit research as it is usually conceived (Edwards, 1992). In either case, my work is subject to many of the same criticisms that apply to the theories reviewed. Furthermore, some of the scholars who developed the theories I review have become colleagues and friends over the years, and I am not in this business to hurt feelings or make enemies. Rather, we as P–E fit researchers are part of the same struggle to advance knowledge, which in turn requires strong theory. To that end, we would all benefit from taking a hard look at the theories that guide our work, noting their strengths and limitations, integrating their best features, and determining how these theories can be developed and improved. Selecting Theories for Review The theories reviewed here were selected by balancing several criteria. One criterion was whether the source material could be reasonably viewed as an 170 • The Academy of Management Annals attempt to define constructs, explain their interrelationships, and organize them in a coherent framework, thereby constituting a bona fide effort to develop 8 theory (Dubin, 1976; Sutton & Staw, 1995; Whetten, 1989). This criterion ruled 0 0 out frameworks used in literature reviews to describe relationships examined 2 y ul in empirical work (e.g., Edwards, 1991; Kristof, 1996) and models used in stud- J 5 ies to assemble hypotheses derived from existing theory and research (e.g., 1 3 3 Cable & Edwards, 2004; Day & Bedeian, 1995; Dineen, Ash, & Noe, 2002; Graves 1: At: 2 & Powell, 1995; Kristof-Brown, Barrick, & Franke, 2002; Wayne & Liden, 1995). R.] At the stage of theory selection, this criterion was intended not to evaluate theo- y ries, but rather to identify attempts to develop theory. Material that satisfied e effr this criterion was eligible for theoretical evaluation as part of the review itself J s, A second criterion was whether the theory explicitly incorporated the con- d ar cept of P–E fit. As noted earlier, P–E fit refers to the congruence, match, or w d E similarity between the person and environment. This general definition of P– By: [ E fit has been distinguished in terms of supplementary fit and complementary ed fit (Muchinsky & Monahan, 1987). Supplementary fit exists when the person d a o “supplements, embellishes, or possesses characteristics which are similar to nl w other individuals” in the environment (Muchinsky & Monahan, 1987, p. 269). o D Complementary fit means that a “weakness or need of the environment is off- set by the strength of the individual, and vice versa” (Muchinsky & Monahan, 1987, p. 271). Complementary fit has been further distinguished according to whether needs are held by the person or environment (Dawis & Lofquist, 1984; Edwards, 1991; French et al., 1982; Kristof, 1996). The degree to which the needs of the person are fulfilled by intrinsic and extrinsic rewards in the environment is termed needs–supplies fit (French et al., 1982; Kristof, 1996). The degree to which needs of the environment are fulfilled by capabilities of the person is called demands–abilities fit, such that environmental needs are reframed as demands imposed on the person (French et al., 1982; Kristof, 1996; McGrath, 1976). Theories that addressed one or more of these three forms of P–E fit were considered for review. A third criterion was the degree to which the theory provides a general foun- dation for P–E fit research or characterizes a major stream of research within the P–E fit domain. As noted earlier, the five streams that surfaced from this review involve research on job satisfaction, job stress, vocational choice, recruitment and selection, and organizational culture and climate. This criterion captured theories that anchored subsequent theory development (e.g., Lewin, 1935, 1951; Murray, 1938, 1951; Schaffer, 1953), serve as primary theoretical statements within a research stream (e.g., Chatman, 1989; French, Rodgers, & Cobb, 1974; Locke, 1969, 1976; Holland, 1959, 1997; Schneider, 1987), or are representative of other theories in the stream (e.g., Breaugh, 1992; McGrath, 1970; Wanous, 1980, 1992). This criterion also excluded some relevant theories that are largely redundant with the theories reviewed. As a result, the review presented here is selective but adequately maps the theoretical terrain of P–E fit research. Person–Environment Fit in Organizations • 171 Criteria for Evaluating Theories Principles for developing and evaluating theory have been discussed exten- 8 sively in the organizational and social sciences (Bacharach, 1989; Blalock, 0 0 2 1969; Dubin, 1978; Kaplan, 1964; Sutton & Staw, 1995; Whetten, 1989; Weick, y Jul 1995). Of these principles, the most basic stipulate that a theory should select 5 3 1 and define constructs of interest, describe how the constructs relate to one 3 1: another, explain why the focal constructs were chosen and why they relate as 2 At: predicted by the theory, and specify boundaries that denote the conditions R.] under which the predictions of the theory should hold. Theories satisfy these y e principles to varying degrees (Weick, 1995), and as the principles are better Jeffr satisfied, a theory can be regarded as stronger and more fully developed. s, d The general principles summarized above can be translated into specific ar w criteria for evaluating theories in a particular domain of inquiry. Applying d E y: [ these principles to P–E fit research suggests that, as a starting point, P–E fit B d theories should clearly define the person and environment constructs that e ad constitute P–E fit. At a minimum, these definitions should indicate whether o nl P–E fit refers to needs–supplies fit, demands–abilities fit, or supplementary fit. w o D The definitions should also specify whether the person and environment are conceptualized as objective, subjective, or both, as these distinctions are rele- vant to the meaning and effects of P–E fit (French et al., 1982; Kristof, 1996; Locke, 1976). Definitions should also describe the units on which the person and environment are conceived. For example, needs can refer to the desired amount or importance of an attribute, a distinction that has significant impli- cations for theories of needs–supplies fit (Katzell, 1964; Locke, 1969, 1976; Schaffer, 1953). P–E fit theories can be considered stronger to the extent they address these definitional issues.1 A second criterion involves the manner in which relationships between P– E fit and other constructs are described. Weaker theories express P–E fit relationships in general terms, merely saying that P–E fit is positively or neg- atively related to another construct. Stronger theories go further by describing the form of this relationship, such as whether the function relating P–E fit to an outcome is symmetric about the point of P–E fit (French et al., 1974; Locke, 1976) and whether the effects of P–E fit depend on the absolute levels of the person and environment (Edwards & Shipp, 2007; Harrison, 1978). P–E fit theories are also stronger when they indicate whether and how relationships between P–E fit and other constructs differ across types of P–E fit (i.e., needs– supplies fit, demands–abilities fit, supplementary fit) and the content of person and environment dimensions (e.g., needs and supplies for pay versus travel). Third, P–E fit theories should explain the conceptual logic for choosing the person and environment constructs included in the theory and the reasoning behind the P–E relationships described by the theory. Theories that select 172 • The Academy of Management Annals person and environment constructs with little justification are weaker than theories that explain why the selected constructs were chosen. Similarly, theo- 8 ries that posit P–E fit relationships without explanation are weaker than theo- 0 0 ries that articulate the conceptual logic behind the relationships. Typically, 2 y ul theories that carefully explain P–E fit relationships also describe the relation- J 5 ships in detail, because when the reasoning behind P–E fit relationships is 1 3 3 critically examined, the result often points to complexities that require 1: At: 2 detailed description, as when the effects of P–E misfit differ depending on R.] whether the environment exceeds or falls short of the person (Harrison, 1978; y Locke, 1976). e effr Finally, stronger P–E fit theories identify boundaries that establish condi- J s, tions under which P–E fit relationships should occur. These conditions can be d ar expressed as moderators that influence the form or strength of P–E fit rela- w d E tionships, as when the effects of demands–abilities fit are stronger when fail- By: [ ure to meet demands has important consequences (McGrath, 1970, 1976). ed Boundary conditions can also refer to limits beyond which the theory does not d a o apply, as when the theory explains the effects of subjective rather than objec- nl w tive P–E fit (French et al., 1982) or the outcomes of P–E fit are restricted to the o D organizational level rather than the individual level (Schneider, 1987). P–E fit theories with explicit boundaries are stronger than theories without stated boundaries, and theories that accommodate a wide range of conditions are stronger that theories with conditions that are restrictive and narrow. The foregoing criteria were applied to the P–E fit theories selected for review, focusing on the conceptualization of the person and environment and the relationship between P–E fit and other constructs, most of which are cast as outcomes of P–E fit. Theories were considered stronger to the extent they clearly defined the person and environment, justified the selection of person and environment constructs, described the form of the relationship between P–E fit and other constructs, explained the logic behind P–E fit relationships, and specified conditions that influenced the form and strength of the relation- ship between P–E fit and other constructs. By focusing on these issues, the scope of the review is kept manageable, and it concentrates directly on the concept of P–E fit that lies at the core of each theory. Foundations of Person–Environment Fit We begin with theories that are generally considered the roots of contempo- rary P–E fit research, starting with Parson’s (1909) matching model and moving to Murray’s (1938, 1951) needs–press model and Lewin’s (1935, 1951) field theory. Parsons’ Matching Model of Career Decision-making Parsons (1909) has been credited as the founder of vocational psychology (Hartung & Blustein, 2002), and his matching model of career decision-making Person–Environment Fit in Organizations • 173 is considered one of the first theories of P–E fit (Kristof-Brown et al., 2005). Parson’s model is summarized by the following quote, which often appears in 8 accounts of his work: 0 0 2 y In the wise choice of a vocation there are three broad factors: (1) a clear ul J understanding of yourself, your aptitudes, abilities, interests, ambitions, 5 1 3 resources, limitations, and their causes; (2) a knowledge of the require- 3 1: ments and conditions of success, advantages and disadvantages, 2 At: compensation, opportunities, and prospects in different lines of work; R.] (3) true reasoning on the relations of these two groups of facts. (Parsons, y e 1909, p. 5) effr J s, In this passage, Parsons foreshadows demands–abilities fit by referring to the d ar requirements of work and the aptitudes, abilities, resources and limitations of w d E the person. He also presages needs–supplies fit by describing the interests and By: [ ambitions of the person and the advantages, disadvantages, compensation, ed opportunities, and prospects of work. The consequences of these types of fit d a o are mentioned in a separate passage: nl w Do An occupation out of harmony with the worker’s aptitudes and capaci- ties means inefficiency, unenthusiastic and perhaps distasteful labor, and low pay; while an occupation in harmony with the nature of the man [sic] means enthusiasm, love of work, and high economic values – superior product, efficient service, and good pay. (Parsons, 1909, p. 3) The remainder of the Parsons (1909) volume is essentially a manual for voca- tional counseling, outlining questions counselors might ask job applicants to guide their occupational choices. Parsons also listed personal qualities suited to various lines of work, although he did not define the person and environ- ment in formal terms, describe the form of the relationship between P–E fit and outcomes, or address other theoretical issues. Hence, Parsons expressed ideas that have parallels in contemporary theories of P–E fit, but he used these ideas to inform vocational counseling, not to develop a theory of P–E fit. Murray’s Needs–Press Model Murray (1938, 1951) is regarded as a founder of P–E fit research for his devel- opment of the needs–press model. A major emphasis of this model was a typology that described different needs and organized them into broader categories, such as whether needs are conscious or unconscious, viscerogenic or psychogenic (i.e., physiological or psychological), and latent or manifest (i.e., hidden or openly expressed). Press refers to stimuli that can benefit or harm the person, based on whether the stimuli promote or inhibit the fulfill- ment of needs. Murray noted that press can be described in terms that paral- lel those used to describe needs, as when the need for achievement is thwarted by the press of failure (Murray, 1938, p. 123). Murray defined a thema as the 174 • The Academy of Management Annals combination of a particular press with its corresponding need, which he argued was key to understanding affect, behavior, and well-being. Murray 8 also distinguished between alpha press and beta press, where the former refers 0 0 to the environment as it actually exists and the latter describes the environ- 2 y ul ment as perceived by the person. J 5 From a theoretical standpoint, Murray’s (1938, 1951) primary contribu- 1 3 3 tions to P–E fit research are his typology for describing needs, the notion that 1: At: 2 needs and press can refer to the same content dimension, the distinction R.] between actual and perceived press, and the idea that needs can be fulfilled by y press, a notion that parallels contemporary thinking on needs–supplies fit. e effr However, Murray’s work said little about how and why the match between J s, needs and press influence outcomes, nor did it consider the form of this rela- d ar tionship, how it varies across needs, or boundary conditions surrounding the w d E relationship. Thus, Murray’s needs–press model is a useful starting point for By: [ P–E fit theories concerned with needs–supplies fit, but his work focused pri- ed marily on describing needs rather than explaining the nature and effects of d a o needs–press match. nl w o D Lewin’s Field Theory Lewin (1935, 1951) is widely considered a pioneer in P–E fit research. The influence of Lewin’s work is widespread, and there is little dispute that his research on field theory, group dynamics, and related topics have significantly influenced social, organizational, and applied psychology. Lewin’s main contribution to P–E fit research was his statement that behavior is a function of the person and environment, as expressed in the classic formula B = F(P,E). By this formula, Lewin meant that “behavior (B) is determined by the person and the environment (B = F(P,E)) and not by the person or the environment alone” (Lewin, 1951, p. 339). This formula is pervasive in P–E fit research, and its conceptual and intuitive appeal are undeniable. Despite the widespread reference to Lewin’s formula in P–E fit research, Lewin (1935, 1951) did not posit that behavior results from the fit between the person and environment. Rather, his claim was more general, asserting that the person and environment jointly affect behavior without specifying the nature of the effect. Hence, B = F(P,E ) could be interpreted as saying that the person and environment combine additively, interactively, proportionally, or in other ways that do not signify P–E fit (Schneider, 2001). Moreover, Lewin did not specify the particular person and environment constructs that combine to determine behavior, such as whether P and E refer to needs and supplies, demands and abilities, or the self and others. Hence, Lewin’s formula B = F(P,E ) is shorthand for saying that the person and environment jointly influ- ence behavior, but the effect represented by this formula did not refer to P–E fit. Beyond this formula, Lewin’s work says little that pertains to P–E fit theory and research, even though the broader impact of his work is indisputable. Person–Environment Fit in Organizations • 175 Job Satisfaction Discrepancy theories of job satisfaction are based on the premise that job satis- 8 faction results from the comparison between what the job provides and what 0 0 2 the employee needs, wants, or desires from the job (Katzell, 1964; Lawler, 1973; y Jul Locke, 1969, 1976; Morse, 1953; Porter & Lawler, 1968; Schaffer, 1953). This 5 1 comparison corresponds to needs–supplies fit as discussed in the P–E fit liter- 3 1:3 ature. As such, discrepancy theories of job satisfaction fall squarely within the 2 At: domain of P–E fit. This section examines discrepancy theories of job satisfac- R.] tion proposed by Schaffer (1953), Katzell (1964), and Locke (1969, 1976). y e effr Schaffer’s Theory of Job Satisfaction J s, d Schaffer (1953) outlined a theory of job satisfaction that has been cited as an ar w early formulation of P–E fit (Dawis, 1992; Latham & Pinder, 2005). Schaffer d E y: [ (1953) described the core elements of his theory as follows: B ed For any individual in any given situation the amount of tension or dissat- d oa isfaction generated is determined by (a) the strength of his [sic] needs or nl w drives, and (b) the extent to which he [sic] can perceive and utilize o D opportunities in the situation for the satisfaction of those needs. (p. 2) As implied by this passage, tension and dissatisfaction were considered inter- changeable, as Schaffer (1953, p. 2) confirmed by defining dissatisfaction as “a conscious recognition of a state of tension”. Need strength was not explicitly defined, although Schaffer (1953, p. 4) indicated that it could be measured by asking respondents to rank or rate each need “in terms of its importance”. Need satisfaction was also not defined, and its operationalization blurred the distinction between the comparison of what the employee wants and the job provides with emotions that might result from this comparison (Hulin & Blood, 1968). This confound is evident in the need satisfaction measure devel- oped by Schaffer, which contained statements describing need fulfillment (e.g., “I have as much freedom as I want on my job”) but used a response scale ranging from “not at all satisfied” to “completely satisfied” with the instruc- tions “Let your feelings be your guide in rating these items” (Schaffer, 1953, p. 26). The needs and opportunities that constitute need satisfaction were also not defined, although the need satisfaction measure referred to these concepts as the desired and perceived amount, frequency, or duration of various job attributes. Schaffer (1953) described the combined effects of need satisfaction and need strength on job satisfaction as follows: Over-all job satisfaction will vary directly with the extent to which those needs of an individual which can be satisfied in a job are actually satis- fied; the stronger the need, the more closely will job satisfaction depend on its fulfillment. (p. 3)