Bowen & Ostroff 2004

Understanding HRM-Firm Performance Linkages: The Role of the "Strength" of the HRM
System
Author(s): David E. Bowen and Cheri Ostroff
Source: The Academy of Management Review, Vol. 29, No. 2 (Apr., 2004), pp. 203-221
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/20159029
Accessed: 20-01-2017 08:44 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/20159029?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The
Academy of Management Review
This content downloaded from 193.147.185.18 on Fri, 20 Jan 2017 08:44:34 UTC
All use subject to http://about.jstor.org/terms
? Academy o? Management Review
2004, Vol. 29, No. 2, 203-221.
UNDERSTANDING HRM-FIRM PERFORMANCE

LINKAGES: THE ROLE OF THE "STRENGTH"
OF THE HRM SYSTEM
DAVID E. BOWEN
Thunderbird, The American Graduate School of International
Management
CHERI OSTROFF
Teachers College, Columbia University
Theory building has lagged on the intermediate linkages responsible for the relation
ship between HRM and firm performance. We introduce the construct "strength of the
HRM system" and describe the metafeatures of an HRM system that result in a strong
organizational climate, analogous to Mischel's "strong situation," in which individ
uals share a common interpretation of what behaviors are expected and rewarded.
The strength of the HRM system can help explain how individual employee attributes
accumulate to affect organizational effectiveness.
In recent years scholars have devoted a great pioyee performance to a more macro focus on
deal of attention to examining the linkage be the overall set of HRM practices and firm perfor
tween HR practices and firm performance. mance (e.g., Arthur, 1992; Huselid, 1995; Huselid
Based on research evidence to datef it is becom & Becker, 1996; Huselid, lackson, & Sch?ler,
ing increasingly clear that the HR system is one 1997). That is, the dominant trend in research on
important component that can help an organi the HRM-firm performance linkage has been to
zation become more effective and achieve a take a systems view of HRM by considering
competitive advantage (Becker & Huselid, 1998). the overall configuration or aggregation of
However, a larger question remains unan HRM practices (Ferris, Arthur, Berkson, Kaplan,
swered: How does HRM contribute to firm per Harrell-Cook, & Frink, 1998), rather than by ex
formance? amining the effects of individual HRM practices
More specifically, if there is indeed an impact of on firm performance (e.g., Delaney & Huselid,
HRM systems on firm performance, how do these 1996; Delery & Doty, 1996) or on individual per
effects occur? What are the mechanisms through formance.
which these effects manifest themselves?. . .
A second approach has been the strategic per
These questions call for theory refinement and
spective on HRM, which has taken on different
the development of more comprehensive models
of the HRM-firm performance relationship that meanings in the literature (Ferris et al., 1999). In
include intermediate linkages and boundary one strategic-based approach, researchers have
conditions.... this type of research should be examined the particular "fit" between various
given a high priority by HRM scholars (Ferris, HRM practices and the organization's competi
Hochwarter, Buckley, Harrell-Cook, & Frink, 1999:
394). tive strategy (e.g., Miles & Snow, 1994; Wright &
Snell, 1991). Embedded in this view is the notion
In research on the HRM-firm performance re that organizations must also horizontally align
lationship, scholars have often assumed two their various HRM practices toward their strate
perspectives. One has been based on a systems gic goal and that practices must complement
approach. Research in this area has moved from one another to achieve the firm's business strat
a focus on separate HRM practices and em egy (Sch?ler & lackson, 1987a,b; Wright & Snell,
1991; Wright, McMahan, & McWilliams, 1994).
The guiding logic is that a firm's HRM practices
We thank Blake Ashforth, College of Business, Arizona must develop employees' skills, knowledge, and
State University, and Art Brief, then associate editor for AMR, motivation such that employees behave in ways
for their comments and time. that are instrumental to the implementation of a
203
204 Academy of Management Review April
particular strategy. Similarly, researchers have on climate as an important mediating variable

taken a contingency perspective, with the as in the HRM-firm performance relationship. The
sumption that the effectiveness of the HR system HRM system itself is discussed not so much in
depends on contextual features such as indus terms of content (e.g., the specific set of HRM
try, firm size, or manufacturing policies (e.g., practices necessary for achieving an organiza
MacDuffie, 1995; Youndt, Snell, Dean, & Lepak, tional goal) but rather process (the features of an
1996). HRM system that send signals to employees that
A related approach within the strategic per allow them to understand the desired and ap
spective on HRM pertains to how the overall set propriate responses and form a collective sense
of HRM practices is generally associated with of what is expected). We describe how a "strong
firm performance and competitive advantage climate" (Schneider, Salvaggio, & Subirats, 2002)
(Ferris et al., 1999). Central here is the resource can be viewed as a "strong situation" (Mischel,
based perspective (Barney, 1991) such that, col 1973, 1977), in which employees share a common
lectively, a firm's human resources are believed interpretation of what is important and what
to have implications for firm performance and behaviors are expected and rewarded. We then
provide a unique source of competitive advan introduce the concept of "strength of the HRM
tage that is difficult to replicate (Wright et al., system" and specify the metafeatures of the
1994). The guiding proposition is that HRM prac overall HRM system that would lead to strong
tices are socially complex and intricately linked climates, after which we examine the conse
in ways that make them difficult for competitors quences of strong versus weak HRM systems,
to copy (Boxall, 1996). More fully, the complexi arguing that the emergence of the intended or
ties of the human resource value creation ganizational climate from psychological cli
process make HRM a source of competitive ad mates is moderated by the strength of the HRM
vantage that is rare, inimitable, and nonsubsti system. We close with directions for future re
tutable (Barney, 1991; Ferris et al., 1999). The search on this new strength of the HRM system
resource-based view has prompted recent work construct and its antecedents and conse
on how HRM practices contribute to firm perfor quences. Our discussion is framed within the
mance by leveraging human capital, discretion mesoparadigm that concerns the simultaneous
ary effort, and desired attitudes and behaviors study of organizational, group, and individual
(e.g., Becker & Gerhart, 1996; Lado & Wilson, processes and specifies how levels are interre
1994; Wright et al., 1994). lated in the form of linking mechanisms (House,
Taken together, these two perspectives on the Rousseau, & Thomas-Hunt, 1995).
HRM-firm performance relationship?the sys
tems and strategic perspectives?help stage CLIMATE AS A MEDIATOR OF THE HRM-FIRM
how HRM practices and their influence on em PERFORMANCE RELATIONSHIP
ployee attributes can lead to desired outcomes
at the firm level, such as productivity, financial We begin our framework with the notion that
performance, and competitive advantage. Yet different business strategies are linked to differ
still left unanswered is the process through ent sets of HRM practices, based on the contin
which this occurs. Although both perspectives gency perspective of strategic human resource
take a macro approach, they assume implicit, management (e.g., Sch?ler & lackson, 1987b,
multilevel relationships among HRM practices, 1995). For example, a strategy of innovation
individual employee attributes, and organiza should foster adoption of HRM practices that
tional performance (Huselid, 1995; Wright et al., share a focus on innovation; a strategy of cus
1994). The features of HRM that are necessary tomer service should be linked to a set of prac
to facilitate these linkages have not been well tices that center around service. We then build
addressed. on the view that HRM systems influence em
In what follows we develop a framework for ployee attitudes and behavior, as well as organ
understanding how HRM practices, as a system, izational outcomes, through employee interpre
can contribute to firm performance by motivat tations of the work climate (Ferris et al., 1998;
ing employees to adopt desired attitudes and Kopelman, Brief, & Guzzo, 1990).
behaviors that, in the collective, help achieve Before developing climate as a mediator, it is
the organization's strategic goals. We first focus important to note that other perspectives delin
2004 Bowen and Ostroff 205
eate different variables that can operate as a HRM, climate is an appropriate construct for de
mediator in the HRM-firm performance relation veloping our framework, based on the recent
ship. For example, the technical subsystem per emphasis on climates around strategic objec
spective focuses on task requirements and task tives that are purported to enhance effective
accomplishment (Katz & Kahn, 1978) and has ness (e.g., Schneider, 2000).
historically dominated HRM research (Sch?ler & Psychological climate is an experiential
lackson, 1995). The underlying assumption is based perception of what people "see" and re
that HRM practices lead to employee knowl port happening to them as they make sense of
edge, skills, and abilities (KSAs) that, in turn, their environment (Schneider, 1990, 2000). This
influence firm performance at the collective sensemaking is relative to the goals the organi
level (Sch?ler & lackson, 1995).
zation pursues; how employees are to perform
Additionally, there are perspectives that their daily activities; the management practices
focus on "higher-order" socially interactive
under which employees work; and the percep
constructs?what Ferris and his colleagues
tions of the kinds of behaviors that management
(1998) term social context theory views of the
expects, supports, and rewards (Schneider,
relationship between HRM and performance. By
Brief, & Guzzo, 1996). Organizational climate is a
higher order, we mean social structures that
cannot be reduced to an aggregation of the per shared perception of what the organization is
ceptions of the individuals currently composing like in terms of practices, policies, procedures,
the organization. routines, and rewards?what is important and
Although we focus on climate, two examples what behaviors are expected and rewarded
of higher-order social structures are organiza (e.g., lames & Jones, 1974; Jones & James, 1979;
tional culture and the organization role struc Schneider, 2000)?and is based on shared per
ture. Culture, conceptualized as organization ceptions among employees within formal organ
ally embedded assumptions and values, can izational units.
function both as an antecedent to the HRM sys Climate researchers have acquired a strate
tem and as a mediator of its linkage to firm gic focus over the years, with the move from
performance (Denison, 1996). Organizational as viewing climate perceptions as shared percep
sumptions and values shape HRM practices, tions about global, generic issues to linking cli
which, in turn, reinforce cultural norms and rou mate perceptions to a shared, specific, strategic
tines that can shape individual and firm perfor content criterion of interest, such as a climate for
mance. Role theorists conceptualize the organi innovation (Delbecq & Mills, 1985; Klein & Sorra,
zation as a system of formal roles, existing apart 1996) or service (Schneider, 1990). Individual
from any one current occupant, which serve to level psychological climates may emerge as a
convey standardized information to employees shared organizational climate, which, in turn,
about expected patterns of activity (Ashforth, ultimately relates to organizational performance.
2001; Katz & Kahn, 1978). In this view the HRM
Climate is a critical mediating construct in
system can be seen as part of the "maintenance
exploring multilevel relationships between
subsystem" (Katz & Kahn, 1978) that defines
roles, which, in turn, influence individual and HRM and organizational performance. Because
climate is widely defined as the perception of
firm performance.
these formal and informal organizational poli
Our focus on climate complements the techni
cal and higher-order social structure perspec cies, practices, and procedures (Reichers &
Schneider, 1990), it follows that the HRM prac
tives on the HRM-firm performance relationship.
We focus on climate because of our interest in tices and HRM system will play a critical role in
multilevel relationships, since both psychologi determining climate perceptions. In turn, empir
cal climates?as individual-level perceptions? ical demonstrations have indicated that organ
and organizational climate?as a shared per izational climate is related to higher-level be
ception at the firm level?have been positioned haviors and organizational performance
as mediators of the relationship between HRM indicators, including customer satisfaction, cus
practices and performance (e.g., Kopelman et tomer service quality, financial performance, or
al., 1990; Ostroff & Bowen, 2000). Additionally, ganizational effectiveness, and total quality
given our interest in strategic perspectives on management outcomes (e.g., Borucki & Burke,
1999; Johnson, 1996; Ostroff & Schmitt, 1993; how the HRM system can be designed and ad
Schneider & Bowen, 1985). ministered effectively by defining metafeatures
Although the above variables are well estab of an overall HRM system that can create strong
lished in the literature, the mechanisms by situations in the form of shared meaning about
which they interrelate are poorly understood. the content that might ultimately lead to organ
For example, as Boxall (1996) has observed, izational performance.
knowledge of HRM practices is widespread, but Given a desired content of the HRM system,
knowledge of how to refine and implement them the HRM system may still not elicit appropriate
within a particular context (e.g., a particular collective behaviors and attitudes needed for
strategic focus) may not be. With respect to cli effectiveness, because individuals may inter
mate, Schneider (2000) has observed that there is pret the HRM practices idiosyncratically, lead
little research or understanding of how organi ing to variability in psychological climate per
zational climate actually develops. Intuitive ceptions. HRM practices can be viewed as a
acceptance of an HRM-climate linkage far ex symbolic or signaling function by sending mes
ceeds theory development of the mechanisms sages that employees use to make sense of and
responsible. to define the psychological meaning of their
work situation (e.g., Rousseau, 1995). All HRM
INTEGRATING HRM CONTENT AND PROCESS practices communicate messages constantly
and in unintended ways, and messages can be
Two interrelated features of an HRM system understood idiosyncratically, whereby two em
can be distinguished: content and process. By ployees interpret the same practices differently
content, we mean the individual practices and (Guzzo & Noonan, 1994). Although much has
policies intended to achieve a particular objec been written about the substantive content of
tive (e.g., practices to promote innovation or au HRM?that is, the specific practices that can
tonomy). The content of the HRM system refers to build task-relevant skills and motivations such
the set of practices adopted and, ideally, should as those for a climate for innovation (Delbecq &
be largely driven by the strategic goals and Mills, 1985), service (Schneider, 1990), change
values of the organization. That is, given some (Schneider et al., 1996), or safety (Zohar, 2000)?
strategic goal such as service, efficiency, or qual little attention has been given to the social con
ity, a set of HRM practices should be devised to structions that employees make of their inter
help direct human resources in meeting this goal. actions with HRM across practices and time
To be effective in terms of content, the foci of the (Rousseau & Greller, 1984).
HRM practices must be designed around a partic In what follows we focus on how HRM can
ular strategic focus, such as service or innovation. send unambiguous messages to employees that
While a number of different models detailing result in a shared construction of the meaning of
the appropriate HRM practices for different the situation. Thus, we concentrate on under
strategies have been offered (e.g., Dyer & standing what features of HRM process can lead
Holder, 1988; Miles & Snow, 1994; Sch?ler & Jack employees to appropriately interpret and re
son, 1987b), rhetoric about this contingency spond to the information conveyed in HRM prac
perspective outpaces data supporting it (cf. tices. We develop the notion that characteristics
Huselid, 1995; MacDuffie, 1995; Sch?ler & Jack of a strong HRM system must be present in order
son, 1987a; Youndt et al., 1996). It is likely that for a shared, strong organizational climate to
there is not a single most appropriate set of emerge (at the aggregate level) from psychologi
practices for a particular strategic objective. cal climates (at the individual level) and propose
Rather, different sets of practices may be that the strength of the HRM system is a linking
equally effective (Delery & Doty, 1996), so long as mechanism that builds shared, collective percep
they allow a particular type of climate around tions, attitudes, and behaviors among employees.
some strategic objective (e.g., climate for inno
vation or service) to develop (Klein & Sorra, 1996). CLIMATE AS THE SITUATION: THE CONCEPT
We propose that HRM content and process OF SITU?TIONAL STRENGTH
must be integrated effectively in order for pre
scriptive models of strategic HRM actually to Kurt Lewin's early work on climate is the foun
link to firm performance. By process, we refer to dation of discussions of situationism in social
psychology (Ross & Nisbett, 1991). Lewin and his meaningful construct (James, 1982). Recently, the
associates (Lewin, Lippit, & White, 1939) demon notion of strong or weak climates has begun to
strated that different leadership styles created emerge in the literature, with a focus on the
different climates, which, in turn, led to different extent to which employees interpret the situa
behavioral reactions and attitudes of members tion similarly, thereby producing low variance
in the groups studied. As Ross and Nisbett sum in perceptions about the situation (Jackofsky &
marize, "The main point of Lewin's situationism Slocum, 1988; Payne, 2000; Schneider et al., 2002).
was that social context creates potent forces As such, an organizational climate can acf as a
producing or constraining behavior" (1991: 9). strong situation when employees develop a
shared interpretation of the organization's poli
Strength of Situation cies, practices, procedures, and goals and de
velop shared perceptions about what behaviors
The situation, as developed in situationism, are expected and rewarded in the organization.
entails the psychological meaning of situations Additionally, the work on strategic climate
for the individual and the behavior potential of content?for example, for safety and innovation
situations for the individual (Endler & Magnus (e.g., Schneider, 1990)?assumes that the more
son, 1976). The interest is not in the physical or HRM practices send strong signals about what
actual situation per se but, rather, the situation strategic goals are most important and what
individuals "see" based on their perceptions, employee behaviors are expected, supported,
cognitive maps, schemata, enactments, and and rewarded relative to those goals, the more
even behavior in the situation (Drazin, Glynn, & likely it is those goals will be achieved.
Kazanjian, 1999).
In an attempt to explain when the character
STRENGTH OF THE HRM SYSTEM
istics of a situation would most likely lead to
consistency in behaviors, Mischel developed the What are the features of an HRM system that
concept of the relative power of situations to allow for the creation of a strong situation? Al
control individual behavior: though suggestions for appropriate process for
Psychological "situations" and "treatments" are separate practices have been offered (e.g., em
powerful to the degree that they lead all persons ployee participation in the design and adminis
to construe the particular events the same way, tration of performance appraisal), metafeatures
induce uniform expectancies regarding the most
appropriate response pattern, provide adequate
of an HRM system overall have not been identi
incentives for the performance of that response fied. Using social cognitive psychology and so
pattern, and instill the skills necessary for its cial influence theories, we propose a set of char
satisfactory construction and execution. Con acteristics that allow HRM systems to create
versely, situations and treatments are weak to strong situations in which unambiguous mes
the degree that they are not uniformly encoded,
do not generate uniform expectancies concerning sages are communicated to employees about
the desired behavior, do not offer sufficient incen what is appropriate behavior. These character
tives for its performance, or fail to provide the istics refer to the process by which a consistent
learning conditions for successful construction of message about HRM content can be sent to
the behavior (Mischel, 1973: 276).
employees.
In sum, situational strength deals with the HRM practices can be viewed as communica
extent to which a situation induces conform tions from the employer to employee (Guzzo &
ity?a strong situation?or is interpreted as Noonan, 1994; Rousseau, 1995; Tsui, Pearce, Por
ambiguous?a weak situation (Mischel & Peake, ter, & Tripoli, 1997). The literature on message
1982). The interest is in specifying situational based persuasion (Chaiken, Wood, & Eagley,
contingencies that identify when individual dif 1996) has its roots in McGuire's (1972) two-step
ferences will or will not control individual be process of "reception"?encoding of the mes
havior Mischel (1997). sage (exposure to the message, attention to its
content, comprehension of the content)?and
"yielding"?acceptance of the message (agree
Strong Climates
ing with the message and storing it in memory).
Only when perceptions are shared across For a message to have its desired effect, both
people does organizational climate become a reception and yielding are necessary. Yet mak
ing sense of the environment often entails nu idiosyncratic psychological climate perceptions.
merous cycles of attending to information, inter The strength of the HRM system can be concep
preting information, acting on it, and receiving tualized in terms of its effectiveness in convey
feedback to clarify one's sense of the situation, ing the types of information needed to create a
particularly when events are highly ambiguous strong situation.
or subject to change (Weick, 1995; Wicker, 1992).
Attribution theory has been useful in helping
explain message-based persuasion and in help Distinctiveness
ing identify key features that will allow for mes
sages to be received and interpreted uniformly Distinctiveness of the situation generally re
among employees (Fiske & Taylor, 1991). In the fers to features that allow it to stand out in the
HRM context, employees are required to infer environment, thereby capturing attention and
cause-effect attributions from these communica arousing interest. We elucidate four character
tions to determine what behaviors are impor istics of HRM that can foster distinctiveness: vis
tant, expected, and rewarded. Causal inference ibility, understandability, legitimacy of author
can be understood not solely as the inner work ity, and relevance.
ings of the mind but also as a process by which Visibility. Visibility of the HRM practices re
people gather and elicit causal explanations fers to the degree to which these practices
from others and communicate their explana are salient and readily observable. This is a
tions to others (Fiske & Taylor, 1991). basic prerequisite for interpretation involving
In order to function effectively in a social con whether an HRM practice and its component
text and make accurate attributions about a sit parts are disclosed to employees, affording
uation, an employee must have adequate and them the opportunity for sensemaking. Visibility
unambiguous information. Although attribu or salience has long been identified as an im
tional frameworks have been used to explain portant characteristic in determining not only
whether an individual attributes the cause of whether people attend to information but how
another person's behavior to internal or external they cognitively organize it (e.g., Tajfel, 1968)
factors, Kelley's (1967) attribution theory details and make cause-effect attributions (Taylor &
the process for making attributions not only to Fiske, 1978). For example, if performance criteria
other people but to situational factors as well. are not transparent or if pay administration out
According to Kelley's (1967) covariation model, comes are withheld, such as with pay secrecy,
an individual can make confident attributions this certainly will not create Mischel's (1973)
about cause-effect relationships in situations strong situation, in which everyone has shared
depending on the degree of distinctiveness (the constructions of the situation and uniform ex
event-effect is highly observable), consistency pectancies regarding the most appropriate
(the event-effect presents itself the same across response pattern and what incentives are
modalities and time), and consensus (there is available.
agreement among individuals' views of the The creation of a strong organizational situa
event-effect relationship). Indeed, Mischel's tion requires that situational characteristics be
(1973, 1977) explication of a strong situation im salient and visible throughout much of employ
plies that it is one in which there is distinctive ees' daily work routines and activities. When the
ness, consistency, and consensus. HRM system includes a wide spectrum of HRM
We propose that when the HRM system is per practices?for example, selection, training, di
ceived as high in distinctiveness, consistency, versity programs, employee assistance pro
and consensus, it will create a strong situation. grams, and so forth?that affect a large number
Using literature on message-based persuasion of employees, visibility is likely to be higher.
and social influence, we elucidate nine Expanding the number and range of practices
metafeatures of HRM systems that build distinc should enhance salience and visibility, because
tiveness, consistency, and consensus, thereby it increases complexity and allows for the set of
creating a strong influence situation in which practices to be more figurai relative to other
employees share constructions of the situation. stimuli?both of which are principles of sa
As such, the features help foster the emergence lience (Fiske & Taylor, 1991). Additionally,
of a strong organizational climate, as opposed to shared meanings cannot be developed unless
most or all employees are subjected to and can management is that HRM is "legitimate" or
perceive the same practices. "credible."
Understandability. Understandability of HRM This notion is related to message source in
content refers to a lack of ambiguity and ease of social cognition, since the characteristics of the
comprehension of HRM practice content. An or message source are linked to attributions made
ganizational communication that cannot be un and the outcomes of persuasion (Fiske & Taylor,
derstood can have no authority (Barnard, 1938). 1991). Communicator credibility (Chaiken et al.,
Features of the stimulus or situation evoke cog 1996) is a critical component in attribution, per
nitive categories (e.g., sch?mas, scripts, cogni suasion, and influence attempts. However, the
tive maps), drawing attention to some features elaboration likelihood model of persuasion
and away from others. Sometimes profound dif (Petty & Cacioppo, 1986) indicates that persua
ferences exist in category systems across people sion and influence are not simply functions of
(Kelley, 1955). To the extent that the situational features of the communicator and credibility
stimulus is ambiguous or unclear, multiple cat but, rather, joint functions of the communicator's
egorizations are likely (Feldman, 1981). That is, credibility and the recipients' involvement in
different people are likely to use different cog the outcomes (Hass, 1981). Relatedly, obedience
nitive categories to attend to different aspects of to legitimate authority is a function of more than
the information, resulting in different attribu the individual's subordination to a position of
tions. For example, employees must be able to "higher office"; it also involves an individual's
understand how the practice works. HRM prac interpretation of the relevance of influence at
tempts to them (Kelman & Hamilton, 1989).
tices such as benefit plans, gain-sharing plans,
Relevance. Relevance of the HRM system re
and succession plans are easily misunderstood fers to whether the situation is defined in such a
or at least open to multiple interpretations.
way that individuals sees the situation as rele
Legitimacy of authority. Legitimate authority
vant to an important goal (Kelman & Hamilton,
of the HRM system and its agents leads individ
1989). Relevance, coupled with legitimate au
uals to consider submitting to performance ex
thority, means that influence is based on both a
pectations as formally sanctioned behaviors. In
perception of superordinate authority and what
fluence by legitimate authority is essentially a Kelman and Hamilton (1989) term motivational
perceptual process?that is, one sees the behav
significance. For the latter, individuals must
ioral requirements of one's own role as subordi
perceive the situation as relevant to their impor
nate to another that stands out as the legitimate
tant goals, that the desired behaviors are clear
authority (Kelman & Hamilton, 1989). It is the and optimally suited for goal attainment, and
concept of authority whereby individuals are that influencing agents have the personal
willing to submit to the necessities of coopera power to affect the achievement of these goals
tive systems (Barnard, 1938). (Kelman & Hamilton, 1989).
The HRM system is most likely to be perceived Here, consideration of both individual goals
as an authority situation when the HRM function and organizational goals?in our case, the stra
is perceived as a high-status, high-credibility tegic goal desired in the form of HRM con
function and activity. This is most likely when tent?is important in that individual goals
HRM has significant and visible top manage should be fostered to align with those of the
ment support in the firm and can be achieved organization. Alignment or congruence between
through investments in HR practices or the HRM individuals' and managers' goals has been
function, or perhaps by placing the director of shown to have important consequences for both
HRM in a high-level managerial position. This individual attitudes and behaviors, as well as
fits the observation about the requirements for for effective organizational functioning (Vancou
the success of HRM systems generally; namely, ver & Schmitt, 1991). Thus, the situation must be
success depends largely on top management defined in such a way that individuals are will
support, including top managers' beliefs about ing to work toward goals that not only allow
the importance of people, investment in human them to meet their own needs but, in doing so,
resources, and involvement of HRM profession also allow the organization to achieve its goals.
als in the strategic planning process (Ostroff, For example, if the organization has a strategic
1995). In such a way, the signal sent from top goal of customer service and an employee val
ues financial gain, then service-based bonuses nicator, thereby increasing the probability that
will heighten relevance and allow both the in the HRM message will be encoded and inter
dividual and organization to achieve their preted uniformly among employees. However,
goals. Relatedly, the relevant desired behaviors distinctiveness alone is not likely sufficient
must be specified and obstacles to their perfor enough for people to view the situation uni
mance removed. formly and to respond to the message sent by
Additionally, relevance is a function of the the set of HRM practices. For employees to make
perceived power of the influencing agent(s) to accurate attributions about what behaviors are
help individuals achieve relevant goals expected and rewarded, attributional principles
(Kelman & Hamilton, 1989). Influence is based on of causation must be present. Fundamental
the extent to which an agent (e.g., HRM staff principles for causal attribution include priority,
member or line manager enacting HRM prac whereby causes precede effects, and contiguity
tices) is perceived as possessing personal capa with the effect, whereby causes occur close in
bilities and is willing to use them to aid goal time to an effect (Kassin & Pry or, 1985).
achievement?separate from his or her influ Similarly, as alluded to above, the literature
ence based on position power and legitimate on authority and influence indicates that indi
authority. Perceived power of the influencing viduals who are to be influenced must perceive
agent(s) depends on two factors. One is whether instrumentalities in the situation whereby be
the agent can affect some of the conditions nec haviors lead to rewards. That is, the distinctive
essary for the achievement of relevant goals ness characteristics ensure that the HRM system
through, for example, the application of unique is viewed, overall, as significant in defining the
expertise or the allocation of necessary re social context for employee behavior; a consis
sources. Characteristics of the agent that bear tent pattern of instrumentalities across HRM
on this issue include his or her prestige, special practices, time, and employees that link specific
knowledge or expertise, representativeness, events and effects further enhances the likeli
control of resources, and ability to apply sanc hood that desired specific behaviors will be dis
tions. A second is the perceived likelihood that played.
the agent will actually use his or her relevant These notions are related to Kelley's (1967)
capabilities in ways that will affect the likeli concept of consistency. Consistency generally
hood of goal achievement. refers to establishing an effect over time and
Taylor and Fiske (1991) explain the relation modalities whereby the effect occurs each time
ship between relevance and the credibility or the entity is present, regardless of the form of
legitimacy of the message source. If outcomes the interactions. Thus, we focus on features that
(rewards, punishments, goal attainment) de establish consistent relationships over time,
pend on someone else's actions as well as the people, and contexts: instrumentality, validity,
individual's actions, then this creates a condi and consistent HRM messages.
tion of outcome dependency, which, in turn, af Instrumentality. Instrumentality refers to es
fects perceptions and attributions. When people tablishing an unambiguous perceived cause
are more outcome dependent, particularly when effect relationship in reference to the HRM sys
the outcomes are relevant, they direct more ac tem's desired content-focused behaviors and
tive attention to the person or source of commu associated employee consequences. It ensures
nication. At the same time, when outcomes are that there are adequate incentives associated
particularly relevant, credibility of the message with performance of the desired behavioral pat
source has less of an influence. Thus, it appears tern. Strong instrumentalities, combined with
relevance alone can enhance distinctiveness; the earlier "relevance" of social influence, lever
when relevance is not strongly established, le age influence within an expectancy theory of
gitimacy plays a greater role. motivation perspective (e.g., Vroom, 1964).
Perception plays a central role in instrumen
tality because it emphasizes how employees an
Consistency ticipate likely consequences of behavior. Instru
The above features of visibility, understand mentalities are shaped largely by reinforcement
ability, legitimacy of authority, and relevance consistency and are established by consistency
help draw attention to the message and commu and repetition over time, particularly through
application of reinforcement principles. Employ dictory. Consequences of inconsistency can be

ees are more likely to perceive the instrumen severe (Lidz, 1973).
tality when behavior and outcomes are closely Three types of consistency are required, each
linked in time (evoking the contiguity causation of which entails the need to avoid sending
attribution principle) and when they are admin double-bind communications to employees and
istered consistently over some time schedule to allow for HRM content to be perceived consis
tently. One is between what senior managers
(evoking the priority causation attribution prin
ciple). To the extent that HRM staff and line say are the organization's goals and values and
managers have the resources and power to link what employees actually conclude those goals
outcomes to behavior or performance on a and values are based on their perceptions of
timely and consistent schedule, they will be HRM practices. Inconsistency here is a differ
able to influence cause-effect attributions. ence between what has been termed espoused
Validity. Validity of HRM practices is impor values and inferred values (Martin & Siehl,
tant because message recipients attempt to de 1983). For example, managers may espouse a
termine the validity of a message in making value of risk taking, but employees may infer
attributions (Fiske & Taylor, 1991). Thus, HRM that performance appraisal and reward system
practices must display consistency between practices reinforce playing it safe.
what they purport to do and what they actually A second requirement for avoiding double
do in order for them to help create a strong bind communication is internal consistency
situation. Selection tests, for example, must among the HRM practices themselves. In recent
validly screen on desired employee abilities, years, much has been written on the importance
thereby making a substantive contribution to of designing an HRM system with practices that
human capital development. Recall that one as complement one another and fit together as a
pect of a strong situation is that employees have whole in achieving the organization's goals
the skills necessary to execute the behaviors (e.g., Becker & Gerhart, 1996; Delery & Doty, 1996;
expected of them. Barnard (1938) long ago ob Sch?ler & Jackson, 1995; Wright & McMahan,
served that employees would view a communi 1992; Wright & Snell, 1991). Internal alignment
cation as authoritative only if they were able among practices should result in performance
mentally and physically to comply with it. advantages for firms, because the different sets
Validity also makes a symbolic contribution of HRM practices will elicit, reward, and control
by signaling to employees what KSAs are val the appropriate employee behaviors for achiev
ued in a setting and by adding more employees ing strategic objectives (Arthur, 1992; Ulrich &
with specified skills to the workforce. Further, Lake, 1991; Wright et al., 1994). For example, if
when a practice is implemented and advertised the ability to work in teams is a screening focus
to have certain effects, and then does not do in selection, then internal consistency will be
what it was intended to do, the message sent to ensured if group, rather than individual, perfor
employees is contradictory, and employees are mance is the basis for rewards. Furthermore, if
left to develop their own idiosyncratic interpre each employee encounter with an HRM practice
tations. (e.g., hiring decision, performance appraisal in
Consistent HRM messages. These convey terview) is conceptualized as a separate situa
compatibility and stability in the signals sent by tion, then, following Mischel (1968), the func
the HRM practices. Considerable evidence indi tional similarity of these situational stimuli will
cates that individuals desire consistency in or influence the generalizability of team-oriented
ganizational life (e.g., Kelley, 1973; Lidz, 1973; behavior across on-the-job situations.
Siehl, 1985). The lack of consistency in "double A third dimension of consistency is stability
bind" communication can lead to particularly over time. HRM practices are situational stimuli,
intense cognitive dissonance (Siehl, 1985). the meaning of which is acquired across time.
Double-bind communication occurs when a per Certainly, how one responds to a situation de
son is faced with significant communication in pends on one's prior history with the stimulus
volving two separate messages (Bateson, Jack (e.g., Mischel, 1968). Behaviors and behavioral
son, Haley, & Weakland, 1956). The messages consequences remain stable when the evoking
are related to each other and deal with the same conditions remain stable. In organizations
content area, but they are incongruent or contra where practices have been in place a long time,
there is stronger agreement among employees bility). As more employees "see" the practice
as to what is expected of them and what they and perceive that top decision makers agree on
expect of the organization in return (Rousseau & it, consensus can be facilitated. Further, integra
Wade-Benzoni, 1994). tion and close interactions among HRM profes
sionals, managers, and top managers foster the
Consensus exchange of tacit knowledge for the formulation
and implementation of an organizational strat
Consensus results when there is agreement egy and HRM system that reflect the firm's stra
among employees?the intended targets of in tegic direction (Lado & Wilson, 1994). These in
fluence by the HRM system?in their view of the tegrations among decision makers can help
event-effect relationship. More accurate attribu promote relevance by clearly identifying impor
tions about what behaviors and responses lead tant goals and means to goal attainment, as
to what consequences are more likely to be well as enhance legitimacy of authority of the
made when there is consensus (Kelley, 1972). HR managers and line managers enacting the
Several factors can help foster consensus HRM policies.
among employees and can influence whether Second, to the extent that members of the top
individuals perceive the same effect with re management team disagree among themselves
spect to the entity or situation in question. about the goals of HRM and/or disagree with
Among these are agreement among message HRM professionals or managers, and to the ex
senders, which can foster consensus (Fiske & tent that HRM managers and staff members dis
Taylor, 1991), and the fairness of the HRM sys agree among themselves, it becomes difficult to
tem, which can also influence consensus inas
send unambiguous and internally consistent
much as fairness involves whether employees messages to employees. Low consistency of
understand the distribution rules by which they
HRM practices and lack of consensus are related
do, or do not, receive what they feel they deserve
in that disagreement among decision makers is
for their contributions.
likely to produce poor consistency in delivering
It is also important to point out that consis
practices; thus, different employees will experi
tency and consensus are distinct but interre
ence different event-consequence relationships.
lated concepts. For example, when individuals
Overall, then, agreement among top decision
throughout the organization experience consis
makers can help foster greater consensus
tency in HRM practices, consensus is more likely
to be fostered. At the same time, when message among employees, since it allows for more vis
ible, relevant, and consistent messages to be
senders cannot agree among themselves on the
intended message, consistency is likely to be conveyed to employees.
Fairness. Fairness of the HRM system is a
hampered.
Agreement among principal HRM decision composite of employees' perceptions of whether
makers. Agreement among these message send HRM practices adhere to the principles of deliv
ers helps promote consensus among employees. ering three dimensions of justice: distributive,
Within a strategic HRM perspective, the princi procedural, and interactional (e.g., Bowen, Gilli
pal decision makers in the organization (e.g., top land, & Folger, 1999; Folger & Cropanzano, 1998).
managers, HR executives) set the strategic goals Research indicates that the perceived fairness
and design the set of HRM practices for achiev of HRM affects how positively HRM activity is
ing those goals. When individuals view mes viewed and the capability of the HRM system to
sage senders as strongly agreeing among them influence employee attitudes and behaviors. Re
selves on the message, they are more likely to searchers have argued that there is a positive
form a consensus (Fiske & Taylor, 1991). This relationship between perceptions of HRM fair
perception of agreement can be facilitated in ness and what has been termed the acceptabil
several ways and is related to distinctiveness ity criterion of HRM practices (Bretz, Milkovich, &
and consistency. Read, 1992; Waldman & Bowen, 1998), which re
First, when multiple decision makers agree on fers to the extent to which employees contribute
the message, distinctiveness can be enhanced to and utilize HRM (e.g., complete 360 degree
because a larger number of individuals can appraisals and use feedback from it to shape
send similar communications (increasing visi their behavior).
Agreement among employees' perceptions of viewpoints such that the situation leads every
event-effect relationships will be influenced by one to "see" the situation similarly, induces uni
whether employees have similar perceptions of form expectancies about responses, provides
what distribution rules?principles of distribu clear expectations about rewards and incen
tive justice?apply in what situations. Outcomes tives for the desired responses and behaviors,
such as rewards can be distributed based on an and induces compliance and conformity through
"equality" rule, in which all receive the same social influence. Therefore, we propose that a
outcome; an "equity" rule, in which subsets of strong HRM system process can enhance organ
employees receive different amounts based on izational performance owing to shared mean
relevant differences, such as in a merit pay sys ings in promotion of collective responses that
tem; or an "individual need" rule, such as flexi are consistent with organizational strategic
ble working hours for a single mother in unique goals (assuming the appropriateness of those
circumstances (Bowen et al., 1999). goals). More specifically, an HRM system high in
Management practices that lead to employee distinctiveness, consistency, and consensus
perceptions of procedural and interactional jus should enhance clarity of interpretation in the
tice increase the transparency of these distribu setting, thereby allowing for similar "cognitive
tion rules (Bowen et al., 1999) and, by so doing, maps" or "causal maps" to develop among peo
increase the likelihood that the HRM system will ple, as well as to create an "influence situation"
be characterized by consensus about event whereby individuals yield to the message and
effect relationships. Procedural justice can be understand the appropriate ways of behaving.
enhanced by giving employees a voice in deter Further, while interactions and communica
mining the methods by which outcome deci tion among employees are likely to result in
sions are made?for example, involving employ collective sensemaking (Jackofsky & Slocum,
ees in designing behavior or outcome-based 1988), regardless of the strength of the HRM sys
performance appraisals. Interactional justice in tem, we argue that in cases where the strength
volves managers' openly and respectfully ex of the HRM system is strong, the sensemaking
plaining to employees the reasons behind deci process will be most likely to result in the in
sions and the distribution of outcomes. It can tended organizational climate. If the HRM sys
include clarifying what distribution formula tem is weak, HRM practices will send messages
was used in making individual pay increase that are ambiguous and subject to individual
decisions in situations where not all employees interpretation. Given ambiguity, one of two
received the same pay increase. things may happen: variability or unintended
sensemaking.
First, with a weak system, variability of indi
CONSEQUENCES OF THE STRENGTH OF THE
HRM SYSTEM vidual responses may be large (Mischel, 1973).
Considerable variance across individuals' per
HRM practices influence employee percep ceptions of psychological climates will exist,
tions of climate at the individual level. Further, and shared perceptions in the form of organiza
the characteristics of strong HRM systems are tional climate will not emerge. Individuals can
more likely to promote shared perceptions and construct their own version of reality (House et
give rise to the emergence of a strong organiza al., 1995) or their own version of what messages
tional climate about the HRM content. That is, are being communicated by HRM practices and
we propose that the strength of the HRM system use this to guide their own behavior. Thus, in
will foster the emergence of organizational cli weak situations (low distinctiveness, consis
mate (collective perceptions) from psychological tency, and consensus), constructs at the individ
climates (individual-level perceptions). ual but not the organizational level are likely to
In a strong situation, variability among em show strong relationships; psychological climate
ployees' perceptions of the meaning of the situ perceptions will have a significant association
ation will be small and will reflect a common with individual attitudes and behaviors.
desired content. In turn, organizational climate While a weak situation is produced by low
will display a significant association with em distinctiveness, consistency, and consensus, we
ployee attitudes and behaviors. This occurs be also argue that the most ambiguous or weakest
cause a strong HRM system can foster similar situation is produced when distinctiveness is
high, coupled with low consistency and consen interactions among employees are also relevant
sus. Distinctiveness drives up attention. That is, (Jackofsky & Slocum, 1988). Morgeson and Hof
HRM practices are salient or visible, and em mann (1999) provide rationales for the impor
ployees are aware of them. However, if the mes tance of these interactions in forming collective
sages that employees are now attending to are constructs. Within any collective, individuals
inconsistent or conflicting, as different individ are likely to meet one another and interact. Each
uals are subjected to different experiences with interaction results in a discrete event, and sub
the HRM practices, confusion, disillusionment, sequent interactions are termed event cycles.
or other negative reactions will likely result. In The structure of any collective group can be
such a case, not only will shared perceptions viewed as a series of ongoing events, activities,
about the practices and climate particularly be and event cycles among the individuals. These
unlikely to emerge, but many employees may interdependencies and interactions among indi
have negative attitudes. viduals over time can result in jointly produced
Alternatively, the ambiguity inherent in weak responses, and it is this structure that forms the
situations may cause employees to engage in basis for the eventual emergence of collective
collective sensemaking (House et al., 1995). constructs?one that can transcend individuals,
When faced with an equivocal situation or attri individual behaviors, and individual percep
butional uncertainty, individuals may attempt tions.
to reduce this uncertainty by engaging in a so This process is similar to the emergence of
cial process of interacting and consulting with overlapping "causal maps" through cognitive
one another to develop their own shared inter processing (e.g., Weick, 1995; Wicker, 1992). Indi
pretations (Drazin et al., 1999; Fiske & Taylor, viduals develop causal maps, which are cogni
1991; Weick, 1995). The danger here is that the tive representations of the entities in the situa
collective interpretation that employees draw tion, certain qualities of those entities, and
from the ambiguous situation is not the one in perceived linkages among them. Overlapping
tended by the organization. That is, the "strong" causal maps can be facilitated through social
climate that emerges does not match the in exchange and transactions among employees.
tended climate content; hence, it may conflict In such a way, employees can coIJecfively agree
with organizational goals and strategies and on the appropriate aspects of the environment to
may ultimately lead to conflicts, poor productiv attend to, as well as how to interpret these as
ity, or low effectiveness. This is particularly pects and how to respond to them appropriately.
likely to occur when "distinctiveness" is low (al Thus, we propose that a strong HRM system fa
though low consensus and consistency will also cilitates interactions, interdependencies, and
play a role). When practices are not made sa event cycles such that fewer event cycles are
lient, visible, and understandable, ambiguity is needed to develop shared interpretations.
high, and employees are more likely to refer to
one another in an attempt to define the situation
in their own way. Thus, we propose that Jow CONTEXT AND HRM SYSTEM STRENGTH
distinctiveness of the HRM system contributes to
a collective sensemaking process that may re In the preceding discussion we implicitly as
sult in unintended organizational climates. Fur sumed an organizational climate. Yet research
ther, a weak HRM system process is unlikely to ers and theorists recognize the multidimen
promote organizational effectiveness because it sional nature of climate such that multiple types
creates a weak situation in which either individ of organizational climates can exist within a
ual processes dominate or collective sensemak firm and at different levels of analysis in the
ing results in shared interpretations that may be organization (Schneider, 1990). That is, different
inconsistent with organizational strategic goals. functional areas, departments, or groups may
It is important to note that this process of develop different subclimates (e.g., Payne, 2000).
emergence of similar perceptions of climate Likewise, cluster analysis has been used to
does not occur in a vacuum. While the HRM demonstrate different collective climates within
system and the strength of this system form the an organization?climates that represent clus
fundamental basis of whether similar percep ters of employees who perceive the organization
tions will be derived, scholars have argued that similarly and span formal organizational units
(e.g., Jackofsky & Slocum, 1988; Joyce & Slocum, the specific practices that make up different sys
1984). tems). That is, research is needed to delineate
We acknowledge that the content of the cli how these processes influence the attributes of
mate can vary across groups within the organi the work situation as perceived by employees.
zation. Further, different HRM practices around Little is known about the important parame
a different content might be applied to different ters underlying organizational situations (e.g.,
groups of employees. We propose that if the Bern & Funder, 1978; Chatman, 1989; Fredrickson,
process of the HRM system is strong, a shared 1972). We have proposed a set of features, based
perception of the climate will emerge in organ on social influence and social cognition theo
izational subunits, albeit with some differences ries, that should help create a strong situation
in content or strategic focus across groups. In and shared meaning. It is critical that the via
deed, for many firms this may be strategically bility of these metafeatures of the organization
desirable?for example, in diversified firms, be tested as important elements that create
firms with multiple locations, international strong situations. Frederiksen (1972) proposes a
firms, or firms pursuing multiple strategic objec number of different means for attempting to
tives in different parts of the organization. It is classify and develop taxonomies of situations.
also likely that, for some groups in the organi In this case, it may be useful to attempt to group
zation, a shared climate will emerge, whereas or cluster situations on the basis of their ten
for others it will not, owing to differences in the dency to elicit similar behaviors. This would
HRM process across different groups. require a three-dimensional data matrix, with
Another concern is the possibility that a the dimensions representing person, behavior,
strong climate might be inflexible and resistant and situational attributes (Frederiksen, 1972).
to change, thereby compromising organization With such a procedure, one could derive clusters
al effectiveness. The literature on strong cul of responses or behaviors that differentially cor
tures offers a resolution of this issue. A culture respond with the nine HRM process features.
whose content comprises values and beliefs In addition, research is needed to determine
that support flexibility can be strong, without the most appropriate means for "combining" the
limiting the organization's ability to adapt to its metafeatures of the HRM system. As suggested
environment (e.g., Sathe & Davidson, 2000). Sim earlier, it is likely that some features are more
ilarly, we propose that a strong climate that has critical than others in creating a strong situa
elements of what has been termed a climate for tion. For example, without consistent HRM mes
innovation (e.g., Klein & Sorra, 1996), for exam sages, distinctiveness and consensus may lose
ple, can be simultaneously strong and adapt impact. Alternatively, although we believe this
able. In other words, the process of the HRM is less likely, a compensatory model may be
system can create a strong climate adaptable to appropriate in that a high level of one feature
change, if the content of the climate includes will make up for a low level of another feature.
elements that focus on flexibility and innova Thus, one could compare and test the viability of
tion. Although individual employees' behaviors an additive model (i.e., the sum across all fea
may differ so as to be innovative or flexible, all tures), a configurai model (i.e., different profiles
employees should still share the idea that this of features), and a multiplicative or contingency
type of adaptability is what is expected of them. model (i.e., interactions among the features).
Thus, perceptions of the climate will be the Further, it is important to determine the rela
same with a strong system that encourages in tive impact of and interrelationships between
novation or flexibility, but there may be vari HRM system strength and other determinants of
ance and changes in actual behavior over time. strong situations or climates. Factors such as
leadership, social relationships, and structural
FUTURE RESEARCH AND THEORY design features can also affect the strength of
DEVELOPMENT the situation and can foster the development of
a shared climate (Ashforth, 1985; Ostroff, Kinicki,
Research is needed on the properties of the & Tamkins, 2003). HRM features are likely to
HRM process, as distinct from research on the interact with these other factors to further foster
properties of practices (e.g., reliability) and a shared sense of the situation. For example,
the content of HRM practices and systems (e.g., supervisors can serve as interpretive filters of
HRM practices, and when they are visible in a climate for service, given that the intangibility
implementing practices or promote high-quality of service makes it difficult to specify service
exchanges with employees, they can introduce quality goals and the employee behaviors that
a common interpretation among unit members will lead to them (Bowen & Schneider, 1988). This
(Kozlowski & Doherty, 1989; Naumann & Bennett, may either complicate the ability to create a
2000). strong HRM system or moderate the relationship
Thus, a strong HRM system coupled with a between that strength and the uniformity of
visible supervisor may foster stronger relation employees' perceptions in the form of organiza
ships among HRM, climate, and performance tional climates.
than each would individually. Similarly, while
our primary intent was to elucidate the charac
teristics of an HRM process that would allow for
Methodological and Measurement Issues
shared perceptions of climate to emerge, addi
tional research is needed to determine the ex Two interrelated methodological issues are
tent to which these HRM system characteristics raised by our proposals. The first of these con
can also impact other social structures such as cerns appropriate measurement for the strength
culture, roles, communication patterns and net of the HRM system. The second concerns levels
works, and social capital, all of which may en of analysis and aggregation issues in moving
hance the relationship between HRM and per from individual-level perceptions of climate to
formance. collective constructs. A full discussion of these
issues is beyond the scope of this article.
New measures will need to be developed to
Relationships Between Content and Process
assess the strength of the HRM system. It is
Research is needed to test interrelationships important to note that this construct is a situa
between HRM process strength and content. The tional context variable, and, as we have defined
configurai approach examines how a pattern of it, it represents a higher-level construct. In past
numerous HRM practices is related to firm per research on HRM practices and systems, schol
formance so that the total effect of HRM is ars have typically relied on reports from a
greater than the sum of the individual practices higher-level manager or HR executive. In our
themselves (Becker & Gerhart, 1996; Delery & case, HR directors and top managers could be
Doty, 1996; Ichniowski, Shaw, & Prennushi, 1997). asked to evaluate the dimensions of strength of
The focus of this approach is on the sets of the system. This procedure has the obvious ad
mutually reinforcing practices that may be re vantage of obtaining a single, global measure
lated to firm performance. The strength of the for each dimension of strength of the system.
HRM system may be a factor influencing However, this measurement technique focuses
whether the configurai approach to HRM-firm only on measures of the attributes from a single
performance relationships is supported in em source that is at a higher level in the organiza
pirical studies. The likelihood that individual tion, while our primary theoretical focus lies in
HRM practices would function as a set, in a the impact these practices have on perceptions
mutually reinforcing manner, may be a function of employees. Because the concept of strength
of the internal consistency of those practices requires judgments and perceptions of employ
and the effectiveness with which they are im ees, we suggest that a better alternative is to
plemented together. assess these characteristics of the HRM system
A similar case can be developed for assessing from employees themselves. That is, the ap
interactions between strength and content propriate unit of measurement of assessing
across climates for different strategic foci. For strength is the individual, since employee
example, on the one hand, it may not be difficult attributions and perceptions reside in the indi
to incorporate features of a strong HRM system vidual.
for a climate focused on cost leadership or Future work should be directed at developing
safety, given that the desired outcomes and a valid measure of HRM strength. For example,
behaviors associated with those criteria can be to assess visibility, employees could be given a
specified clearly. On the other hand, it may be list of a variety of HRM practices and asked to
more difficult to create a strong HRM system for indicate the extent to which each is utilized in
the firm. A comparison between those practicesclimate and individual responses, the level of
that agents of the HRM function assert are in the individual's responses on the variables is
place and those that employees indicate are most useful. However, when moving to higher
used would provide some assessment of how levels of analysis, additional measurement is
visible the practices are to employees. sues emerge. Strong and well-designed HRM
Similarly, to assess consistency, employees systems produce greater homogeneity of per
could be asked to what extent they have actu ceptions and responses within the organization,
ally participated in or experienced each of these resulting in organizational climate. The strength
practices (e.g., received a semiannual perfor of the climate is indicated by the degree of vari
mance review). The percent of people indicating ability in responses, regardless of the level of
they experienced the practice would provide the aggregate rating on the content of climate.
some indication of how consistently the practice An indication of whether the HRM system cre
is administered across employees in the organi ates a strong situation is the extent of agree
zation. As an alternative, employees could be ment on climate ratings (Payne, 2000).
asked to indicate the extent to which they be
lieve the practice applies to all employees.
FINAL THOUGHTS
Agreement might be assessed by asking top
decision makers to delineate the strategic goals In listing challenges that the HRM community
related to HRM and the intended message of the faces in the future, Ulrich cites the need for HR
HRM practices (e.g., promote innovation and risk practice to be guided by HR theory. He reminds
taking, promote loyalty and longevity, promote HRM professionals that theory helps explain the
safety). High agreement among decision makers manner in which outcomes emerge:
should be related to higher consensus among
To make HR practices more than isolated acts,
employees as to what practices are salient, vis managers and HR professionals must master the
ible, administered consistently, and so forth. theory behind HR work; they need to be able to
Such measures would be useful from multiple explain conceptually how and why HR practices
perspectives. First, the mean score on the di lead to their outcomes ... Regardless of the pre
mension would provide an indication as to the ferred theory, managers and HR professionals
should abstract from it a higher level of reason
level at which these characteristics are present. ing for their day-to-day work and thus better ex
That is, a higher mean score on measures tap plain why their work accomplishes its goals
ping distinctiveness, consistency, and consen (1997: 238; emphasis added).
sus would be one indicator of strong HRM pro
cess. Second, researchers could assess the Recently, in the literature scholars have de
veloped "why" HR practices lead to sustainable
extent to which employees perceive character
competitive advantage. Hopefully, this present
istics in the same way?that is, they could as
effort at theory building on the strength of the
sess the extent of agreement or variability in
HRM system can begin to help explain "how"
responses among employees. Higher agreement
would support consensus and a strong system, HRM practices lead to outcomes the organiza
tion desires.
whereas high variance in responses would in
dicate a weak system.
As to assessments of climate, agreement REFERENCES
among employees about their perceptions must
Arthur, J. B. 1992. The link between business strategy and
be demonstrated before aggregated measures industrial relations systems in American steel mini
of psychological climate perceptions can be mills. Industrial and Labor Relations Review, 45: 488
used to represent a unit-level or organizational 506.
level climate construct (James, 1982). Further, it Ashforth, B. E. 1985. Climate formation: Issues and exten
is important to examine both the level (e.g., the sions. Academy of Management Review, 10: 837-847.
level of rating on a dimension of climate) and Ashforth, B. E. 2001. Hole transitions in organizational life.
the variability in responses. Level is an indica Mahwah, NI: Lawrence Erlbaum Associates.
tor of "content," whereas variability is an indi Barnard, C. I. 1938. The functions of the executive. Cam
cator of situational "strength." At the individual bridge, MA: Harvard University Press.
level of analysis, if one is interested in examin Barney, J. 1991. Firm resources and competitive advantage.
ing the relationship between perceptions of the Journal of Management, 17: 99-120.
Bateson, G., Jackson, D. D., Haley, J., & Weakland, J. H. 1956. Dyer, L., & Holder, G. W. 1988. A strategic perspective of
Toward a theory of schizophrenia. Behavioral Science, 1: human resource management. In L. Dyer (Ed.), Human
251-264. resource management: Evolving roles and responsibili
Becker, B., & Gerhart, B. 1996. The impact of human resource ties: 1-45. Washington, DC: Bureau of National Affairs.
management on organizational performance: Progress Endler, N. S., & Magnusson, D. 1976. Personality and person
and prospects. Academy of Management Journal, 39: by situation interactions. In N. S. Endler & D. Magnusson
779-801. (Eds.), Interactional psychology and personality: 1-25.
Becker, B. E., & Huselid, M. A. 1998. High performance work New York: Hemisphere.
systems and firm performance: A synthesis of research Feldman, J. M. 1981. Perception, cognition, and the organiza
and managerial implications. Research in Personnel tion. Journal of Applied Psychology, 66: 128-138.
and Human Resources Management, 16: 53-101.
Ferris, G. R., Arthur, M. M., Berkson, H. M., Kaplan, D. M.,
Bern, D. J., & Funder, D. C. 1978. Predicting more of the people Harrell-Cook, G., & Frink, D. D. 1998. Toward a social
more of the time?assessing personality of situations. context theory of the human resource management
Psychological Review, 85: 485-501. organization effectiveness relationship. Human Re
Borucki, C. C, & Burke, M. J. 1999. An examination of service source Management Review, 8: 235-264.
related antecedents to retail store performance. Journal Ferris, G. R., Hochwarter, W. A., Buckley, M. R., Harrell-Cook,
of Organizational Behavior, 20: 943-962. G., & Frink, D. D. 1999. Human resource management:
Bowen, D. E., Gilliland, S. W., & Folger, R. 1999. HRM and Some new directions. Journal of Management, 25: 385
415.
service fairness: How being fair with employees spills
over to customers. Organizational Dynamics, 27(3): 7-23. Fiske, S. T., & Taylor, S. E. 1991. Sociai cognifion. New York:
McGraw-Hill.
Bowen, D. E., & Schneider, B. 1988. Services marketing and
management: Implications for organizational behavior. Folger, R., & Cropanzano, R. 1998. Organizational justice and
Research in Organizational Behavior, 10: 43-80. human resource management. Newbury Park, CA: Sage.
Boxall, P. 1996. The strategic HRM debate and the resource Frederiksen, N. 1972. Toward a taxonomy of situations.
based view of the firm. Human Resource Management American Psychologist, 26: 114-123.
Journal, 6: 59-75.
Guzzo, R. A., & Noonan, K. A. 1994. Human resource practices
Bretz, R. D. Jr., Milkovich, G. T., & Read, W. 1992. The current as communications and the psychological contract. Hu
state of performance appraisal research and practice: man Resource Management, 33: 447-462.
Concerns, directions, and implications. Journal of Man
Hass, R. G. 1981. Effects of source characteristics on cogni
agement 18: 321-352.
tive responses and persuasion. In R. E. Petty, T. M. Os
Chaiken, S., Wood, W., & Eagly, A. H. 1996. Principles of trom, & T. C. Brock (Eds.), Cognitive responses in persua
persuasion. In E. T. Higgins & A. W. Kruglanski (Eds.), sion: 141-172. Hillsdale, NJ: Lawrence Erlbaum
Social psychology: Handbook of basic principles: 702 Associates.
744. New York: Guilford Press.
House, R., Rousseau, D. M., & Thomas-Hunt, M. 1995. The
Chatman, J. A. 1989. Improving interactional organizational meso-paradigm: A framework for the integration of mi
research: A model of person-organization fit. Academy cro and macro organizational behavior. Research in Or
of Management Review, 14: 333-349. ganizational Behavior, 17: 41-114.
Delaney, J. T., & Huselid, M. A. 1996. The impact of human Huselid, M. A. 1995. The impact of human resource manage
resource management practices on perceptions of organ ment practices on turnover, productivity, and corporate
izational performance. Academy of Management Jour financial performance. Academy of Management Jour
nal, 39: 949-969. nal, 38: 635-672.
Delbecq, A., & Mills, P. K. 1985. Managerial practices that Huselid, M. A., & Becker, B. E. 1996. Methodological issues in
enhance innovation. Organizational Dynamics, 14(1): cross-sectional and panel estimates of the human re
24-34. source management-firm performance link. Industrial
Relations, 35: 400-422.
Delery, J. E., & Doty, D. H. 1996. Modes of theorizing in stra
tegic human resource management: Tests of universal Huselid, M. A., Jackson, S. E., & Sch?ler, R. S. 1997. Technical
is?e, contingency, and configurai performance predic and strategic human resource management effective
tions. Academy of Management Journal, 39: 802-835. ness as determinants of firm performance. Academy of
Denison, D. R. 1996. What is the difference between culture Management Journal, 40: 171-188.
and organizational climate? A native's point of view on Ichniowski, C, Shaw, K., & Prennushi, G. 1997. The effect of
a decade of paradigm wars. Academy of Management human resource management practices on productivity:
Review, 21: 619-654. A study of steel finishing lines. American Economic Re
view, 87: 291-313.
Drazin, R., Glynn, M. A., & Kazanjian, R. K. 1999. Multilevel
theorizing about creativity in organizations: A sense Jackofsky, E. F., & Slocum, J. W., Jr. 1988. A longitudinal study
making perspective. Academy of Management Review, of climates. Journal of Organizational Behavior, 9: 319
24: 286-307. 334.
James, L. R. 1982. Aggregation bias in estimates of percep MacDuffie, J. P. 1995. Human resource bundles and manufac
tual agreement. Journal of Applied Psychology* 67: 219? turing performance: Organizational logic and flexible
229. production systems in the world auto industry. Indus
trial and Labor Relations Review, 48: 199-221.
James, L. R., & Jones, A. P. 1974. Organizational climate: A
review of theory and research. Psychological Bulletin, Martin, J., & Siehl, C. J. 1983. Organizational customer and
81: 1096-1112. counterculture: An uneasy symbiosis. Organizational
Dynamics, 12(2): 52-64.
Johnson, J. W. 1996. Linking employee perceptions of service
climate to customer satisfaction. Personnel Psychology* McGuire, W. J. 1972. Attitude change: The information pro
49: 831-852. cessing paradigm. In C. G. McClintock (Ed.), Experimen
tal social psychology: 108-141. New York: Holt, Rinehart
Jones, A. P., & James, L. R. 1979. Psychological climate: Di & Winston.
mensions and relationships of individual and aggre
Miles, R. E., & Snow, C. C. 1994. Fit failure and the hall of
gated work environment perceptions. Organizational
Behavior and Human Decision Processes, 23: 201-250. fame. New York: Free Press.
Joyce, W., & Slocum, J. 1984. Collective climate: Agreement as

Mischel, W. 1968. Personality and assessment New York:
a basis for defining aggregate climates in organiza Wiley.
tions. Academy of Management Journal, 27: 721-742. Mischel, W. 1973. Toward a cognitive social learning concep
tualization of personality. Psychological Review, 80:
Kassin, S. M., & Pryor, J. B. 1985. The development of attribu 252-283.
tion processes. In J. Pryor & J. Day (Eds.), The develop
ment of social cognition: 3-34. New York: Springer Mischel, W. 1977. The interaction of person and situation. In
Verlag. D. Magnusson & N. S. Endler (Eds.), Personality at the
crossroads: Current issues in interactional psychology:
Katz, D., & Kahn, R. L. 1978. The social psychology of organ 333-352. Hillsdale, NJ: Lawrence Erlbaum Associates.
izing. New York: Wiley.
Mischel, W. 1997. Personality dispositions revisited and re
Kelley, G. A. 1955. A theory of personality: The psychology of vised: A view after three decades. In R. Hogan, J. John
personal constructs. New York: Norton. son, & S. Briggs (Eds.), Handbook of personality psychol
Kelley, H. H. 1967. Attribution theory in social psychology. In ogy: 113-132. New York: Academic Press.
D. Levine (Ed.), Nebraska symposium on motivation: 192? Mischel, W., & Peake, P. K. 1982. Beyond d?j? vu in the search
240. Lincoln: University of Nebraska Press. for cross-situational consistency. Psychological Review,
Kelley, H. H. 1972. Causal schemata and the attribution pro
89: 730-755.
cess, ?n E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E. Morgeson, F. P., & Hof mann, D. A. 1999. The structure and
Nisbett, S. Valins, & B. Weiner (Eds.), Attribution: Per function of collective constructs: Implications for multi
ceiving the causes of behavior: 151-174. Morristown, NJ: level research and theory development. Academy of
General Learning Press. Management Review, 24: 249-265.
Kelley, H. H. 1973. The processes of causal attribution. Amer Naumann, S. E., & Bennett, N. 2000. A case for procedural
ican Psychologist, 28: 107-128. justice climate: Development and test of a multilevel
Kelman, H. C, & Hamilton, V. C. 1989. Crimes of obedience: model. Academy of Management Journal, 43: 881-889.
Toward a social psychology of authority and responsi Ostroff, C. 1995. SHRM/CCH survey. Human Resources Man
bility. New Haven, CT: Yale University Press. agement: Ideas and Trends in Personnel (356): 1-12.
Klein, K. J., & Sorra, J. S. 1996. The challenge of innovation Ostroff, C & Bowen, D. E. 2000. Moving HR to a higher level:
implementation. Academy of Management Review, 21: Human resource practices and organizational effective
1055-1080. ness. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel
theory, research, and methods in organizations: 211-266.
Kopelman, R. E., Brief, A. P., & Guzzo, R. A. 1990. In B. Schnei
San Francisco: Jossey-Bass.
der (Ed.), Organizational climate and culture: 282-318.
San Francisco: Jossey-Bass. Ostroff, C, Kinicki, A. J., & Tamkins, M. M. 2003. Organiza
tional culture and climate. In W. C. Borman, D. R. Ilgen,
Kozlowski, S. W. J., & Doherty, J. L. 1989. Integration of climate & R. J. Klimoski (Eds.), Comprehensive handbook of psy
and leadership: Examination of a neglected issue. Jour chology, volume 12: Industrial and organizational psy*
nal of Applied Psychology, 74: 721-742. chology: 565-594. New York: Wiley.
Lado, A. A., & Wilson, M. C. 1994. Human resource systems Ostroff, O, & Schmitt, N. 1993. Configurations of organiza
and sustained competitive advantage: A competency tional effectiveness and efficiency. Academy of Man
based perspective. Academy of Management Review, agement Journal, 36: 1345-1361.
19: 699-727.
Payne, R. L. 2000. Culture and climate: How close can they
Lewin, K., Lippit, R., & White, R. 1939. Patterns of aggressive get? In N. M. Ashkanasy, C. P. M. Wilderom, & M. F.
behavior in experimentally created social climates. Peterson (Eds.), Handbook of organizational culture and
Journal of Social Psychology, 10: 271-299. climate: 163-176. Thousand Oaks, CA: Sage.
Lidz, T. 1973. Origin and treatment of schizophrenic disor Petty, R. E., & Cacioppo, J. T. 1986. The elaboration likelihood
ders. New York: Basic Books. model of persuasion. In L. Berkowitz (Ed.), Advances in
experimental social psychology, vol. 19: 123-205. San culture. In P. Frost, L. Moore, M. Louis, C. Lundberg, &
Diego: Academic Press. J. Martin (Eds.), Organizational culture: 125-140. Beverly
Reichers, A. E., & Schneider, B. 1990. Climate and culture: An Hills, CA: Sage.
evolution of constructs. In B. Schneider (Ed.), Organiza Tajfel, H. 1968. Social and cultural factors in perception. In
tional climate and culture: 5-39. San Francisco: Jossey G. Lindzey & E. Aronson (Eds.), Handbook of social psy
Bass. chology: 315-394. Reading, MA: Addison-Wesley.
Ross, L., & Nisbett, R. E. 1991. The person and the situation: Taylor, S. E? & Fiske, S. T. 1978. Salience, attention, and
Perspectives of social psychology. Philadelphia: Temple attributions: Top of the head phenomena. In L. Berkowitz
University Press. (Ed.), Advances in experimental social psychology: 249
287. New York: Academic Press.
Rousseau, D. M. 1995. Psychological contracts in organiza
tions. Thousand Oaks, CA: Sage. Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. M. 1997.
Rousseau, D. M., & Greller, M. M. 1994. Human resource Alternative approaches to employee-organization rela
practices: Administrative contract-makers. Human Re tionship: Does investment in employees pay off? Acad
source Management, 33: 385-402. emy of Management Journal, 40: 1089-1121.
Rousseau, D. M., & Wade-Benzoni, K. A. 1994. Linking strat Ulrich, D. 1997. Human resource champions: The next agenda
egy and human resource practices: How employee and for adding value and delivering results. Boston: Harvard
customer contracts are created. Human Resource Man Business School Press.
agement 33: 463-490. Ulrich, D., & Lake, D. 1991. Organizational capability: Creat
Sathe, V., & Davidson, E. J. 2000. Toward a new conceptual ing competitive advantage. Academy of Management
ization of culture change. In N. M. Ashkanasy, C. P. M. Executive, 5(1): 77-91.
Wilderom, & M. F. Peterson (Eds.), Handbook of organi
Vancouver, J. B., & Schmitt, N. W. 1991. An exploratory exam
zational culture and climate: 279-296. Thousand Oaks,
ination of person-organization fit: Organizational goal
CA: Sage. congruence. Personnel Psychology, 44: 333-352.
Schneider, B. 1990. The climate for service: An application of
Vroom, V. 1964. Work and motivation. New York: Wiley.
the climate construct. In B. Schneider (Ed.), Organiza
tional climate and culture: 383-412. San Francisco: Jos Waldman, D. A., & Bowen, D. E. 1998. The acceptability of
sey-Bass. 360-degree appraisals: A customer-supplier relationship
perspective. Human Resource Management, 37: 117-130.
Schneider, B. 2000. The psychological life of organizations. In
N. M. Ashkanasy, C. P. M. Wilderom, & M. F. Peterson Weick, K. E. 1995. Sensemaking in organizations. Thousand
(Eds.), Handbook of organizational culture and climate: Oaks, CA: Sage.
xvii-xxii. Thousand Oaks, CA: Sage. Wicker, A. W. 1992. Making sense of environments. In W. B.
Schneider, B., & Bowen, D. E. 1985. Employee and customer Walsh, K. H. Craik, & R. H. Price (Eds.), Person
perceptions of service in banks: Replication and exten environment psychology: 157-192. Hillsdale, NJ: Law
sion. Journal of Applied Psychology, 70: 423-433. rence Erlbaum Associates.
Schneider, B., Brief, A. P., & Guzzo, R. A. 1996. Creating a Wright, P. M., & McMahan, G. C. 1992. Theoretical perspec
climate and culture for sustainable organizational tives for strategic human resource management. Journal
change. Organizational Dynamics, 24(4): 7-19. of Management, 18: 295-320.
Schneider, B., Salvaggio, A. N., & Subirais, M. 2002. Climate Wright, P. M., McMahan, G. C, & McWilliams, A. 1994. Hu
strength: A new direction for climate research. Journal of man resources and sustained competitive advantage: A
Applied Psychology, 87: 220-229. resource-based perspective. International Journal of Hu
Sch?ler, R. S., & Jackson, S. E. 1987a. Organizational strategy man Resource Management, 5: 301-326.
and organization level as determinants of human re Wright, P. M., & Snell, S. A. 1991. Toward an integrative view
source management practices. Human Resource Plan of strategic human resource management. Human He
ning, 10: 125-141. source Management Review, 1: 203-225.
Sch?ler, R. S., & Jackson, S. E. 1987b. Linking competitive Youndt, M. A., Snell, S. A., Dean, J. W., Jr., & Lepak, D. P. 1996.
strategies and human resource management practices. Human resource management, manufacturing strategy,
Academy of Management Executive, 1(3): 207-219. and firm performance. Academy of Management Jour
Sch?ler, R. S., & Jackson, S. E. 1995. Understanding human nal, 39: 836-866.
resource management in the context of organizations Zohar, D. 2000. A group-level model of safety climate: Testing
and their environment. Annual Review of Psychology, the effect of group climate on microaccidents in manu
46: 237-264.
facturing jobs. Journal of Applied Psychology* 85: 587
Siehl, C. J. 1985. After the founder: An opportunity to manage 596.
David E. Bowen is dean of faculty and programs and professor of management at

Thunderbird, The American Graduate School of International Management. His re
search interests are organizational behavior issues in service quality and the linkage
between human resource management effectiveness and competitive advantage.
Cheri Ostroff is a professor of psychology and education at Teachers College, Colum

bia University. She received her Ph.D. in industrial-organizational psychology from
Michigan State University. Her current research interests include levels of analysis
issues, human resource management systems, and person-environment congruence.

Bowen & Ostroff 2004

Uploaded by

Copyright:

Available Formats

Bowen & Ostroff 2004

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bowen & Ostroff 2004

Uploaded by

Copyright:

Available Formats

Understanding HRM-Firm Performance Linkages: The Role of the "Strength" of the HRM

UNDERSTANDING HRM-FIRM PERFORMANCE

particular strategy. Similarly, researchers have on climate as an important mediating variable

application of reinforcement principles. Employ dictory. Consequences of inconsistency can be

Joyce, W., & Slocum, J. 1984. Collective climate: Agreement as

David E. Bowen is dean of faculty and programs and professor of management at

Cheri Ostroff is a professor of psychology and education at Teachers College, Colum

You might also like