Journal of Applied Psychology © 2011 American Psychological Association

2012, Vol. 97, No. 4, 719 –738 0021-9010/11/$12.00 DOI: 10.1037/a0026078

Personnel Selection as a Signaling Game

Adrian Bangerter and Nicolas Roulin Cornelius J. König

University of Neuchâtel Saarland University

Personnel selection involves exchanges of information between job market actors (applicants and
organizations). These actors do not have an incentive to exchange accurate information about their
ability and commitment to the employment relationship unless it is to their advantage. This state of
affairs explains numerous phenomena in personnel selection (e.g., faking). Signaling theory de-
scribes a mechanism by which parties with partly conflicting interests (and thus an incentive for
deception) can nevertheless exchange accurate information. We apply signaling theory to personnel
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

selection, distinguishing between adaptive relationships between applicants and organizations,

This document is copyrighted by the American Psychological Association or one of its allied publishers.

among applicants, and among organizations. In each case, repeated adaptations and counteradapta-
tions between actors can lead to situations of equilibrium or escalation (arms races). We show that
viewing personnel selection as a network of adaptive relationships among job market actors enables
an understanding of both classic and underexplored micro- and macro-level selection phenomena
and their dynamic interactions.

Keywords: personnel selection, signaling theory, adaptation, information exchange, faking

Personnel selection involves exchanges of information be- the employment relationship, but applicants may not be motivated
tween applicants and organizations. Many commentators have to provide accurate information regarding these qualities unless it
noted that this exchange is as much a competitive as a cooper- serves their candidacy. Moreover, applicants and organizations
ative endeavor; that is, applicants and organizations approach each react to their counterparts’ actions. Cycles of behavior with
employment relationships with goals that may not be perfectly potentially unanticipated consequences may then emerge. In the
aligned. Personnel selection is the moment when these goals are selection interview, for example, applicants may try to detect
confronted for the first time. For example, Porter, Hackman, selection criteria and produce the answers they think recruiters
and Lawler (1975) stated that “the search for a fit between the want to hear in order to appear qualified for the job. Repeated
goals of a particular individual and the goals of a particular exposure to such behavior may lead recruiters to be concerned
organization typically begins with the selection process. Indi- with detecting what applicants are really like behind such perfor-
viduals seek a work organization where they can fulfill their mances. Kirkwood and Ralston (1999) wrote that “interviewers’
goals, and organizations try to hire people who can help them attempts to penetrate applicants’ performances only invite more
reach their objectives” (p. 131). sophisticated applicant performances, producing an ongoing spiral
The degree of goal misalignment undoubtedly varies, and, as the of mistrust in which each party tries to outdo the other” (p. 64).
budding relation between applicant and organization unfolds, both On the basis of these observations and many others, we argue in
parties may discover ways to better align their goals. However, at this article that (a) personnel selection situations present powerful
the moment of personnel selection, this misalignment can have incentives for job market actors to adapt their behavior to those of
far-reaching consequences for information exchanges in selection other actors, (b) actual behavior of job market actors reacts to these
situations. Indeed, parties with imperfectly aligned goals have little incentives, and (c) repeated cycles of such individual-level behav-
incentive to exchange accurate information unless it is to their iors lead to signaling systems that organize information exchanges.
advantage (Frank, 2006). That is, organizations are interested in A signaling system consists of a sender, a receiver, and a signal
accurately assessing applicants’ abilities and their commitment to that correlates with an unobservable characteristic of the sender
(Spence, 1973). Signaling systems allow actors to determine what
information is reliable for making job market choices (e.g., choos-
This article was published Online First October 31, 2011. ing among applicants). Signaling phenomena and their implica-
Adrian Bangerter and Nicolas Roulin, Institute of Work and Organiza- tions have not been sufficiently recognized in research on person-
tional Psychology, University of Neuchâtel, Neuchâtel, Switzerland; Cor- nel selection, which has historically focused on only one side of
nelius J. König, Department of Psychology, Saarland University, Saar- the employment relation (either the organization’s or the appli-
brücken, Germany. cant’s perspective; e.g., Phillips, 1998; Schmidt & Hunter, 1998;
We acknowledge the helpful comments of Franciska Krings, Filip Smither, Reilly, Millsap, Pearlman, & Stoffey, 1993) and over-
Lievens, Michael McDaniel, Paul Sackett, and Neal Schmitt on earlier
whelmingly on individual-level behavior.
versions of the paper.
Correspondence concerning this article should be addressed to Adrian We use signaling theory, a general framework derived from
Bangerter, Institute of Work and Organizational Psychology, University of evolutionary biology, game theory, and economics, to describe
Neuchâtel, Rue Emile Argand 11, 2000 Neuchâtel, Switzerland. E-mail: how signaling systems in personnel selection evolve over time as
[email protected] a result of behavior of individual job market actors. This frame-

719
 720 BANGERTER, ROULIN, AND KÖNIG

work is innovative because it is explicitly based on the incentives Signaling theory therefore addresses the conditions under which
that operate in personnel selection situations. It is useful for linking exchange of accurate information is possible among rational indi-
micro-level processes (individual choices) and macro-level pro- viduals with partly divergent interests. Principles of signaling have
cesses (evolution of signals, including market trends in the emer- been successfully applied to dilemmas of cooperative behavior in
gence and decline of selection devices) relevant to personnel many disciplines, including evolutionary biology (Zahavi & Za-
selection (Morgeson & Hofmann, 1999). It therefore can lead to havi, 1999), political science (Poundstone, 1993), anthropology
theoretical progress and implications for research, including offer- (Cronk, 2005), economics (Spence, 1973), management (Con-
ing novel testable propositions and focusing attention on hitherto nelly, Certo, Ireland, & Reutzel, 2011), and organizational behav-
ignored phenomena. ior (Deutsch Salamon & Deutsch, 2006). As suggested by the
opening comments of this article, personnel selection is also a
situation where rational actors with partly incompatible interests
Signaling Theory: Basic Principles and Applications interact (Palmer, Campion, & Green, 1999), and thus it constitutes
another promising but unexplored field of application.
In this section, we review signaling theory in detail as a foun-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

dation for applying its principles to personnel selection, which we

This document is copyrighted by the American Psychological Association or one of its allied publishers.

then do in the next three sections. We first explicate the structure The Handicap Principle
of cooperative behavior and the dilemma it poses for individuals.
Individual organisms need to obtain accurate information about
We then describe the principle of honest signaling and its wide-
potential cooperation partners. Absent an altruistic incentive to
ranging explanatory power in the study of nonhuman and human
reveal such information to others, how can this be accomplished?
behavior. We then focus on signaling in economics and manage-
Accurate information (e.g., about an animal’s genetic fitness) can
ment before describing three types of adaptive relationships in
be communicated between organisms with diverging interests if
personnel selection: between applicants and organizations, among
sending a message imposes a cost on the sender that only certain
applicants, and among organizations.
individuals (e.g., a truly fit individual) can bear. This is called the
handicap principle (Zahavi, 1975), and signals that transmit such
Dilemmas of Cooperative Behavior information are variously termed honest signals, reliable signals,
costly signals, or hard-to-fake signals (Bergstrom, 2006; Cronk,
Signaling theory seeks to explain cooperative behavior between 2005).
rational organisms—rational meaning the efficient pursuit of the The handicap principle is widespread in animal behavior. One
individual organism’s interests (Frank, 1988). Two basic domains example from predator–prey interaction is so-called stotting be-
of such behavior exist. The first concerns behavior of nonhuman havior, observed in the Thomson’s gazelle and other hooved
organisms such as animals and is traditionally investigated by animals (Walther, 1969). When a gazelle spots a predator, it
evolutionary biology. Although largely instinctive, the behavioral sometimes jumps high in the air. Such behavior is apparently
repertoire of nonhuman organisms is “rational” in the sense that it maladaptive because it attracts the predator’s attention and wastes
has emerged through processes of natural and sexual selection precious energy that the gazelle would need to escape if the
(Darwin, 1871) and thereby represents successful solutions to predator did pursue it. Initially interpreted as altruistic behavior
recurrent problems posed by past environmental pressures. Be- (i.e., alerting other gazelles to the presence of a predator), stotting
cause the unit of natural and sexual selection is the individual has since been explained as a signal of fitness directed toward the
organism carrying the genes that determine a particular behavior or predator. It credibly demonstrates that the gazelle is fit because it
trait, behavior is rational if it furthers the reproductive or survival can afford to waste energy. Stotting benefits both the gazelle and
chances of the individual. The second domain concerns human the predator, because the gazelle avoids the expenditure of energy
behavior, which is of course at least partly guided by conscious related to a long and tiring chase, and the predator gains informa-
intentions (Dennett, 1987). Human behavior is also rational in the tion about which individual gazelles are fit and therefore probably
sense that it is self-interested (i.e., human beings seek to further harder to catch. Indeed, sick or weak individuals will need all their
their individual interests; Frank, 2006).1 In both domains, individ- energy in a chase and thus cannot afford to stot. Stotting allows
ual organisms may seek to cooperate with other individuals to predators to identify these individuals. Empirical studies of ga-
achieve better outcomes than they could achieve by acting in zelles in the field support these conjectures (Caro, 1986a, 1986b;
isolation (as we will see, such cooperation is even possible among FitzGibbon & Fanshawe, 1988). By stotting, then, a fit individual
individuals belonging to classes of organisms with diametrically imposes a handicap on itself. This handicap is proof of the cred-
opposed interests, like predators and prey). In situations of poten- ibility of the signal, because less fit individuals are unable to
tial cooperation, individuals are motivated to discover information produce it without incurring unbearable costs.
about the ability of the other party to cooperate usefully and about
its trustworthiness, or commitment to the relationship (Zahavi & 1
Zahavi, 1999). Indeed, humans have an evolved capacity to rapidly We do not intend to suggest that individuals are motivated purely by
self-interest or that altruistic behavior does not exist. Indeed, commentators
detect these two dimensions of warmth and competence in con-
in various fields have long noted the ubiquity of such behavior. However,
specifics (Fiske, Cuddy, & Glick, 2007). At the same time, being altruism can often be explained as being in the best interests of an
self-interested, individuals have an incentive to deceive their part- individual (i.e., as an extended form of self-interested behavior), and even
ner to exploit the relationship for their own gain. The problem that altruistically motivated individuals face the problem of assessing whether
each party must then solve is how to gain accurate information of or not potential cooperation partners have exploitative intentions (Frank,
the other’s abilities and intentions. 2006).
 PERSONNEL SELECTION AS A SIGNALING GAME 721

Another class of situations where honest communication is For example, Frank (1988) proposed that emotional displays
possible through displaying a handicap is constituted by within- constitute hard-to-fake signals of an individual’s commitment to a
species interactions, in particular by sexual signals between males particular course of action because they are difficult to consciously
and females assessing potential mating partners, as exemplified by manipulate. Thus, displays of rage are a hard-to-fake signal of an
the peacock’s tail. Naturalists since Darwin have wondered how individual’s aggressive intentions and, thus, a preemptive deterrent
the peacock’s extravagant tail, obviously a survival disadvantage to a potential attack (Boster, Yost, & Peeke, 2003). As another
because it is cumbersome and energetically expensive to maintain, example, one ethnographic study (Boster, 2003) documented the
has survived natural selection. The theory of sexual selection polite custom in the Shuar culture of the Andes of repeatedly
(Darwin, 1871; Miller, 2000) posits that ornamental characteristics spitting on the floor when visiting someone’s home. Such a display
such as antlers and tails can constitute an honest signal to females is a hard-to-fake signal of peaceful intent: A guest with aggressive
about their bearer’s genetic fitness. Indeed, that the bearer can designs would be physiologically aroused and have a dry mouth
afford to invest in and carry a wasteful ornament is proof that he and would therefore be unable to muster the saliva necessary for
has resources to squander, and thus that he is fit and is a good repeated spitting.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

mating partner. This conjecture is empirically supported (Petrie, Because humans may also be tempted to cheat by sending
This document is copyrighted by the American Psychological Association or one of its allied publishers.

1994; Petrie & Halliday, 1994; Petrie, Halliday, & Sanders, 1991). signals unrelated to their true level of ability or commitment, and
Stotting and sexual signals illustrate situations where senders because cheating can undermine cooperation, punishment of cheat-
and receivers of a signal have a broad conflict of interests, for ers is also widespread across human societies (Henrich et al.,
instance, predators and prey or male and female potential mating 2006). Just as for nonhumans, punishment can be considered as
partners. But within that conflict of interests, individual organisms indirectly raising the cost of a signal (Searcy & Nowicki, 2005).
have converging interests. The fit gazelle and the predator both
have an interest in avoiding a chase. It is to their mutual benefit if How Signaling Systems Evolve: Adaptation,
they can signal this state of affairs to each other. Likewise, the Equilibrium, and Escalation
conflict of interest between male and female arises from the fact
that, in certain species, females invest more resources into parent- Signals typically evolve from behavior originally designed for
ing than do males. But the fit peacock and the discerning peahen another function. This is the derivation principle, originally pro-
have an interest in mating together. It is also to their mutual benefit posed by Tinbergen (1952). According to this principle, a reliable
if they can signal this state of affairs to each other. but incidental correlation between an observable feature (behavior,
The evolution of handicaps as a mechanism for guaranteeing morphology) of an organism and an unobservable parameter (e.g.,
accurate communication mitigates the problem of cheating. Cheat- genetic fitness) can be detected by other organisms. Krebs and
ers are individuals that send a signal that is not related to their true Dawkins (1984) coined the metaphor of mind-reading to describe
level of ability or commitment. Examples include mimicry, as this detection process. Those other organisms might use the infor-
when certain nonpoisonous prey species mimic the coloration of mation to anticipate the future behavior of the organisms exhibit-
poisonous prey to deter potential predators (Zahavi & Zahavi, ing the observable features. In turn, these organisms might come to
1999). If a signal does not impose a cost on its sender that is related produce the observable feature in a more conspicuous way to
to fitness, cheating strategies may evolve and spread within a manipulate the mind-reading organisms (with either cooperative or
population, ultimately undermining the value of the signal as exploitative purposes). Over time, this reciprocal adaptation, or
receivers evolve to ignore it. However, potential benefits of cheat- coevolution, between mind-readers and manipulators leads to the
ing are offset by costs if the cheater is caught, for example, by the emergence of a signaling system, where a behavior or morpholog-
risk of predation or the fact that many species severely punish ical feature survives because of its informative value to other
cheaters (e.g., birds whose coloration patterns are experimentally organisms.
manipulated to mimic high-status markings get attacked by their There are many cases of the emergence of signals from nonsig-
conspecifics; Searcy & Nowicki, 2005). naling behavior. One example concerns how male toads settle
contests for possession of females. Instead of fighting rivals, they
signal information about body size (and thus about their potential
Honest Signaling in Human Behavior fighting ability) by croaking. The pitch of a croak was probably
initially an incidental but reliable signal of body size. Experiments
Honest signaling also applies to human cooperation. However, show that croaking has probably evolved to become a signal
because humans are capable of intentional action, they can make (rather than an incidental indicator) of body size because deeper
strategic decisions to invest resources in sending a signal to attain croaks are more intimidating to other toads than high-pitched
a particular outcome. At the same time, many aspects of human croaks (Davies & Halliday, 1977). The emergence of a signal often
behavior are beyond conscious control. As a result, the catchall corresponds to a process of ritualization (Tinbergen, 1952),
metaphor of an “honest” signal becomes more complex (Cronk, whereby it becomes more conspicuous, for example, by means of
2005). There are thus two kinds of signals that are honest. First, exaggerated, simplified and repetitive movements. Examples such
similarly to the domain of evolutionary biology, there are costly as gazelles’ demonstrative stotting and oversized peacock tails
signals, which are honest because they require investment of illustrate the end result of ritualization processes.
resources the cost of which only fit individuals can bear. Then, The emergence of a signal thus results from a process of
there are signals that are not costly per se but are hard to fake reciprocal interaction between organisms in an ecosystem (or
because they are beyond the conscious control of the individual human actors in a market; Krebs & Dawkins, 1984). Depending on
and thus not manipulable. the honesty of the signal and on the intentions of the actors or
 722 BANGERTER, ROULIN, AND KÖNIG

organisms implicated (either cooperative or exploitative), the about the qualities of a given applicant. Nor are applicants partic-
emergent system can be more or less stable. It will be stable if ularly motivated to provide the employer with accurate informa-
senders and receivers’ behaviors are mutually reinforcing; such a tion unless it is to their advantage. In such a situation, the employer
system is then in a state of equilibrium. If not, outcomes other than must decide to offer the applicant high or low wages.
equilibrium may emerge, namely, escalation (Vermeij, 1994). An Spence (1973) outlined a system whereby high-quality appli-
organism that develops a superior adaptation is at an advantage cants can signal their worth to employers. Any signal whose
relative to other competing organisms in the same ecology. For production costs are negatively correlated with the quality of the
example, a peacock that develops a larger and more extravagant applicant sending it can be an honest signal. As an example,
tail is at an advantage relative to other peacocks in the competition Spence assumed that education is marginally easier to acquire for
for peahens. Or a predator species that evolves a more lethal high-quality applicants than for low-quality ones. If employers
weapon (a stronger jaw, faster running speed) is at an advantage believe that education is a credible signal of higher productivity,
relative to its prey. But this relative advantage may be short-lived, they will offer higher wages for educational credentials. This will
because it may trigger a counteradaptation that subsequently in- cause higher quality applicants to invest in acquiring an education.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

creases the selection pressure on the original organism. The re- Lower quality applicants will not do so, because the costs are too
This document is copyrighted by the American Psychological Association or one of its allied publishers.

sulting cycle of adaptations and counteradaptations, alternatively high for them (e.g., they may not possess the ability to fulfill the
described as the Red Queen phenomenon (Van Valen, 1973), an degree requirements). Employers’ beliefs about the relation be-
arms race (Dawkins & Krebs, 1979), or simply escalation (Ver- tween education and applicant quality will be confirmed, causing
meij, 1994), is one of the most important motors of the evolution employers to again offer higher wages in the next round of hiring
of species. In the domain of human behavior, many applications and applicants to differentially invest in acquiring an education
can be found. For example, human cognitive abilities may have depending on their quality. In this way, education (or any signal
evolved as the result of an intraspecies arms race (Flinn, Geary, & the production costs of which are negatively correlated with qual-
Ward, 2005). And of course, the term arms race is itself a meta- ity) emerges as an honest signal between employers and appli-
phor derived from the escalation in the production of nuclear cants.
weapons by the United States and the USSR during the Cold Although Spence pointed out in a footnote that the same rea-
War—a case famously amenable to analysis by game theory
soning applies to the decisions of applicants, he did not explore
(Poundstone, 1993).
what might constitute potential signals in their case. It is important
One possible, beneficial outcome of escalation is an overall
to note the negative relation between productivity and cost: If
increase in individual fitness. Another, less beneficial outcome is
signaling costs were the same for all individuals, then all would
that individuals may continue to invest resources into staying
invest in education, and education would cease to be a credible
ahead of competitors while their average relative benefit does not
signal (Spence, 2002). The same would happen if the costs of
increase (Frank, 2006). Arms races can lead to the domination and
education decreased (e.g., if grade inflation reduced the effort
exploitation of one party over the other, or they may lead to a state
necessary to acquire a high-quality degree). In such a situation, its
of mutually beneficial reciprocal exploitation. Dawkins and Krebs
signaling power would also decrease and employers and applicants
(1979) proposed the life– dinner principle to explain the outcome
would need to converge on other signals. Spence’s work is foun-
of an arms race. The life– dinner principle is illustrated by a race
between a fox (predator) and a rabbit (prey). The rabbit forfeits its dational in many areas of economics that describe cooperative
life if caught by the fox, whereas the fox forfeits its dinner if the dilemmas between agents with conflicting interests and asymmet-
rabbit escapes. Thus, although the fox will eventually starve if it rical information, for example, agency theory and contract theory
does not catch any rabbits, the selection pressure for running speed (Eisenhardt, 1989).
operating on it is less severe than the pressure operating on the Signaling has widely been applied in management. Early schol-
rabbit. Organisms subjected to stronger selection pressures will ars have noted the fundamental problem of aligning employees’
tend to evolve better adaptations and a relative competitive advan- goals with those of the firm (Porter et al., 1975; Simon, 1947). A
tage. recent integrative review (Connelly et al., 2011) found that interest
in signaling has increased, especially in the fields of strategy and
entrepreneurship, but also in organizational behavior (Nicholson &
Signaling in Economics and Management White, 2006) and human resources management. Deutsch Salamon
Human economic behavior has long been described along the and Deutsch (2006) described how organizational citizenship be-
lines of the handicap principle. Veblen (1899) suggested that havior (OCB) can serve as an honest signal of otherwise unob-
conspicuous consumption (the wasteful display of excess re- servable capabilities to other organizational members. OCB has
sources by the rich) and conspicuous leisure (the ostentatiously been traditionally investigated as altruistic behavior (e.g., employ-
wasteful pursuit of economically unnecessary activities) serves as ees who engage in OCB are “good soldiers”). But the handicap
a signal of social status. In a seminal paper, Spence (1973) inde- principle suggests that it may be a way of credibly signaling
pendently developed a theory of signaling similar to that of Zahavi attributes that would not be visible from in-role behavior. For
(1975) to explain the effects of information asymmetries in mar- example, a cashier who volunteers to organize an organization-
kets. Although he was an economist with no particular interest in wide social event gains the opportunity to display abilities (e.g.,
personnel selection, Spence used the domain of hiring as an organizing skills, leadership) that would not be visible in her
information asymmetry example. He conceptualized hiring from day-to-day job. In doing so, she imposes a cost on herself that less
the perspective of employers as an investment decision made capable employees may not be able to bear (Deutsch Salamon &
under uncertainty, as employers have only imperfect information Deutsch, 2006).
 PERSONNEL SELECTION AS A SIGNALING GAME 723

Signaling in Personnel Selection stand out to potential employers. Depending on how high the
competitive stakes are, such behavior can take various forms.
Ironically, although he shared the 2001 Nobel Prize in econom- Applicants may try to outdo each other by engaging in noteworthy
ics for his work on signaling using hiring as an example, Spence extracurricular activities (P. Brown & Hesketh, 2004), padding
has been little cited in the personnel psychology literature. Re- their resumés (Amare & Manning, 2009), or even directly sabo-
search in personnel selection has used signaling theory to study taging other applicants’ progress (Coombs & Virshup, 1998). On
how recruiters infer unobservable information about applicants the other side of the fence, organizations may also compete with
(e.g., value congruence) from observable attributes (e.g., cognitive each other to attract and retain the best applicants. This competi-
ability; Aguinis, Michaelis, & Jones, 2005; Cable & Judge, 1997). tion is known as the War for Talent, a term that describes the
Research in recruitment, on the other hand, has studied how widespread belief that talented employees are rare yet crucial for
applicants infer unobservable characteristics of organizations from the prosperity and survival of organizations (Michaels, Handfield-
known characteristics (e.g., inferring information about the orga- Jones, & Axelrod, 2001). Waging the War for Talent requires
nization from the characteristics of recruiters encountered during organizations to signal desirable attributes to attract applicants,
the selection process; Ehrhart & Ziegert, 2005; Ryan, Sacco,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

like high wages and bonuses, fast-track promotion systems, or

This document is copyrighted by the American Psychological Association or one of its allied publishers.

McFarland, & Kriska, 2000; Rynes, 1991; Rynes, Bretz, & Ger-
commitments to employee well-being or corporate social respon-
hart, 1991). In both cases, however, researchers have focused on
sibility.
only a part of signaling theory (i.e., how actors infer unobservable
In the next sections, we therefore examine three types of adap-
characteristics of their partners from observable characteristics;
tive relationships among job market actors. The first is between
Highhouse, Thornbury, & Little, 2007). The notion of honest
applicants and organizations and corresponds to the typical per-
signals has gone unacknowledged in the twin selection and recruit-
sonnel selection situation. The second concerns applicants in com-
ment literatures, as have the notions of reciprocal adaptation and
petition with other applicants for jobs. The third concerns organi-
escalation and their implications for the long-term stability of
zations in competition with other organizations to attract
signaling systems.
To date, then, there has been no systematic application of the applicants. In each case, cycles of individual-level adaptations and
most important aspects of signaling theory to the field of personnel counteradaptations can lead to the market-level emergence of
selection. The remainder of this article will show that it holds signaling systems, equilibrium situations, or arms races (Dawkins
important potential for understanding and integrating research & Krebs, 1979; Vermeij, 1994). We discuss these three relation-
findings in this field. We start by summarizing its main principles. ships in more detail and develop general propositions applying
First, signaling theory applies to all interactions (human or other- signaling theory to personnel selection. Some of these propositions
wise) where individuals with imperfectly aligned motives seek are reconceptualizations of phenomena already studied under cur-
cooperation. Personnel selection is such a situation, because orga- rent theoretical paradigms, and others are novel statements that can
nizations’ goals of obtaining accurate information about applicants lead to new research questions. Taken together, these propositions
are imperfectly aligned with applicants’ goals of appearing attrac- constitute a novel, high-level theoretical framework within which
tive. At the same time, however, it is in interests of applicants and many personnel selection phenomena can be studied. We also
organizations to exchange accurate information with each other in highlight relevant examples of classic, emerging and declining
order to increase the quality of the selection decision for both the signaling systems from research and practice. Relationships be-
applicant and the organization. Both parties do cooperate in ex- tween applicants and organizations are the prototypical case of
changing information to this end, but they need to solve the personnel selection, having been most studied by academics, and
dilemma of cooperation outlined above. offer many practical examples. They therefore are developed in
Second, a basic requirement for accurate communication in such more detail. But adaptive relationships among applicants and
situations is that signals either must be hard to fake or must impose among organizations are also relevant for personnel selection.
a cost on the sender such that only fit individuals can bear the cost Moreover, outcomes from one adaptive relationship may influence
(otherwise, the incentive to cheat will lead some senders to do so, another.
and receivers will learn to rely less and less on the signal over
time). Third, signals often evolve from activities originally de-
signed for purposes other than signaling, through reciprocal adap- Adaptive Relationships Between Applicants and
tation between senders and receivers. Fourth, signaling systems Organizations
consist of (a) a population of senders who produce the signal to
influence receivers’ behavior, (b) a signal that is correlated with an Adaptations between applicants and organizations constitute the
unobservable but relevant characteristic of senders, and (c) a classical situation in personnel selection where organizations se-
population of receivers who interpret the signal as an indicator of lect among applicants for a job. On the one hand, organizations try
that characteristic. Fifth, signaling systems can vary in stability to identify honest signals of two unobservable qualities of appli-
along a continuum ranging from equilibrium to escalation (an arms cants: ability and commitment to the employment relationship.
race). Identifying honest signals of applicant ability corresponds to as-
We argue that these principles constitute a framework with sessment of person–job fit (whether the abilities of the applicant
wide-reaching potential for understanding phenomena related to correspond to the abilities required by the organization). On the
personnel selection. However, not only organizations and appli- other hand, organizations’ interest in identifying honest signals of
cants have misaligned interests. Applicants may also compete with applicant commitment corresponds to assessment of person–
each other to distinguish themselves from other applicants and thus organization fit (whether the values of the applicant correspond to
 724 BANGERTER, ROULIN, AND KÖNIG

the culture of the organization).2 This distinction between two test this conjecture. A recent meta-analysis of work sample validity
kinds of fit has a long history in organizational psychology (Roth, Bobko, & McFarland, 2005) also found a decrease over
(Kristof-Brown, 2000; March & Simon, 1958; Wanous, 1978). It is time, which was not explained.
analogous to the two fundamental dimensions of competence and We now discuss costly signals, hard-to-fake signals, and cheat-
warmth evaluated in any human social relation (Fiske et al., 2007). ing costs in detail. Costly signals of ability include educational
On the other hand, applicants try to mind-read organizations credentials (Spence, 1973), job experience, professional reputa-
(Kleinmann et al., 2011; Krebs & Dawkins, 1984), that is, to detect tion, letters of recommendation, and references. Educational cre-
the criteria they are interested in and send the right signals. In turn, dentials require investment of time, money, and effort to acquire.
organizations may adapt their selection criteria. This may lead to Professional reputations must be built through mindful interactions
cycles of reciprocal adaptations between the two parties. Over with colleagues, clients, and the like. And letters of recommenda-
time, these adaptive dynamics lead to the emergence and evolution tion are costly to produce by proxy, i.e., letter writers are typically
of signaling systems. Figure 1 graphically depicts this process, high-status individuals whose time is a precious commodity. Their
distinguishing between individual-level behavior and market-level willingness to “waste” time on a letter is credible proof of their
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

outcomes and featuring Propositions 1– 6. esteem for the applicant. Providing references in one’s resumé also
This document is copyrighted by the American Psychological Association or one of its allied publishers.

In what follows, we discuss what constitutes an honest signal constitutes a costly signal of applicants’ job experience, because
from the organization’s point of view, what strategies applicants only experienced applicants can provide references. Organizations
use to detect and adapt to organizational selection criteria, how may also search for costly signals of applicants’ commitment to
counteradaptations emerge, and what consequences these entail for accepting the position if offered it, especially in job markets where
the evolution of applicant– organization signaling systems. unemployed applicants are required to apply regularly for posi-
tions to continue to receive unemployment benefits. Such signals
Organizational Selection Strategies: The Search for may include the applicant’s longevity in previous organizations,
Honest Signals of Applicant Ability and Commitment credible demonstrations of willingness to incur personal costs in
order to occupy the position (e.g., willingness to accept a lower
Organizations and their representatives or allies try to identify salary to work for an organization or to move to another city), or
honest signals of desirable applicant qualities. Two kinds of honest effort visibly expended to inform oneself about the organization.
signals exist in personnel selection, costly signals and hard-to-fake Many of these signals are inferred from biodata in the resumé
signals. Costly signals correspond to the handicap principle: they (B. K. Brown & Campion, 1994; Thoms, McMasters, Roberts, &
require applicants to invest resources to acquire and display them. Dombkowsky, 1999).
These we call investment costs. Signals that are hard to fake are The second kind of honest signal typically involves signals that
typically beyond the conscious control of applicants. Both kinds of are hard to fake because they are beyond conscious control. Cog-
honest signals also must be associated with cheating costs, in order nitive ability tests and work samples are examples. They have
to deter potential cheaters. A central preoccupation of personnel predictive validity (Roth et al., 2005; Schmidt & Hunter, 1998).
selection is predictive validity, or the correlation between an Ability tests are hard to fake, because the cognitive processes
observable predictor (a selection device) and an unobservable
underlying intelligent performance (e.g., working memory span or
quantity of interest to employers (typically job performance;
processing speed; Jensen, 1998) are not under conscious control.
Schmidt & Hunter, 1998). Predictive validity thus describes the
And scoring high on a work sample is impossible without the
empirical strength of a signal at a given point in time. As such,
requisite experience or knowledge. Another kind of hard-to-fake
predictive validity is a necessary (but not sufficient) condition for
signal may be constituted by structured interview questions about
a signal to be honest. In knowing predictive validity, one never-
past behavior (Janz, 1982). Structured interviews have predictive
theless remains agnostic about whether a signal is costly or hard to
validity (Huffcutt & Arthur, 1994). And because such questions
fake. For example, predictive validity may simply reflect an inci-
can require applicants to describe their past on-the-job behavior in
dental link between the signal and an unobservable characteristic,
detail, it can be difficult for them to provide high-quality answers
as in the case of the typical empirical approach to identifying
without having corresponding job experience.
biodata items (Gunter, Furnham, & Drakeley, 1993). And some
Some applicants may be tempted to cheat and mimic an honest
signals that have predictive validity may not be costly or hard to
signal. Returning to the examples above, applicants can buy a fake
fake, for example, personality tests. Nevertheless, signaling theory
degree in what has been estimated as a billion-dollar industry
suggests that the predictive validity of a selection device may
(Bear & Ezell, 2005). They can lie about their experience on their
change over time if the investment costs of the signal change or if
resumés (Aamodt, 2006). Writers of letters of recommendation can
cheating costs change. In particular, if such costs decline, validity
may also decline. There is some circumstantial evidence that “cheat” by reusing templates of previous letters that may not
predictive validity may change over time. Van Iddekinge, Roth, reflect the true qualities of the person they are recommending. And
Raymark, and Odle-Dusseau (2011) found that the predictive references can also be faked. Thus, for applicants to have a
validity of integrity tests has decreased over time. They did not genuine incentive to actually pay investment costs in the long run,
offer an explanation for this finding. A recent meta-analysis of
assessment center validity for predicting supervisor performance 2
This situation makes the adaptive relationship between applicants and
ratings (Hermelin, Lievens, & Robertson, 2007) also found a organizations analogous to mating games between male and female con-
decrease over time. Hermelin et al. suggested that this might be specifics, where ability (genetic fitness) and commitment to a relationship
due to range restriction caused by stronger preselection of appli- are also qualities females evaluate in males and where males are selected
cants in organizational settings, but they were unable to directly on their ability to send the right signals (Zahavi & Zahavi, 1999).
 PERSONNEL SELECTION AS A SIGNALING GAME 725

Propositions 5a, 5b and 6

Long-term evolution of selection systems

Escalation
market-level
outcomes
Equilibrium
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Applicants
Organizations
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Propositions 1a-1d detect

Propositions selection
Applicants
2a-2c aggregated
adapt
Organizations
detect criteria and
selection Organizations
criteria individual-level
adaptand
behavior adapt
Organizations criteria
Applicants counteradapt
adapt behavior
criteria behaviors
seek honest signals detect selection to applicants’
over time
of applicant ability criteria and behavior
and commitment adapt behavior

Propositions 3-4
Adaptive dynamics between applicants and organizations

Figure 1. Reciprocal adaptations between applicants and organizations in personnel selection and their
consequences.

there must be some risk of punishment linked to cheating. Other- organization fit or other perspectives (Sanders & Frenkel, 2011).
wise, many applicants would cheat and organizations would learn And actors may also differ in their approaches to identifying
to discount the signal over time (we discuss the dynamic evolution honest signals. Some actors may rely on experiential learning or
of signaling systems below). Organizations thus need to increase past experience (Herriott, Levinthal, & March, 1985; Huber,
cheating costs. At least part of the extensive efforts they invest in 1991). However, some organizational allies, such as academic
verifying applicant information can be considered an attempt to researchers, may engage in systematic research on behalf of orga-
increase such costs and to ensure applicants refrain from mimick- nizations (Brief, 2000) to discover and develop honest signals.
ing costly signals. For example, reference checks are a way of Thus, validation research (Schmidt & Hunter, 1998) can be con-
increasing cheating costs. Applicants generally must authorize sidered a systematic, organized attempt at signal detection.
recruiters to check references (Levashina & Campion, 2009). This Some organizational actors may intrinsically value the hard-to-
constitutes either a credible signal that the information has at least fake or costly nature of a signal to the extent that they neglect
not been blatantly faked by the applicant or a highly risky bluff. predictive validity. This can explain why recruiters are sometimes
Also, organizations can invest in running background checks to
interested in invalid methods like graphology or nonverbal behav-
investigate applicants’ past (Isaacson, Griffith, Kung, Lawrence, &
ior. For instance, if recruiters believe that verbal behavior (i.e.,
Wilson, 2008).
applicants’ interview answers) can be easily faked, they may focus
It is also theoretically possible to cheat on hard-to-fake signals.
more on nonverbal behavior. This is more difficult to manipulate
However, because such signals are beyond conscious control,
(Ekman & Friesen, 1969) and commonly believed to “leak” infor-
cheating in this case typically involves, for example, obtaining test
questions and answers before completing a mental ability test mation about unobservable states or traits of applicants (Bavelas,
(Burke, 2009; Lievens & Burke, 2011). Organizations also invest 1992). Similarly, the persistence of graphology in some settings
substantial resources in guarding the security of such test items and may be due to recruiters’ beliefs that it is both hard to fake and
identifying potential cheaters (Burke, 2009). Such actions in turn costly. Some recruiters believe graphology is hard to fake because
increase the costs would-be cheaters must incur and therefore act it is more difficult for applicants to manipulate their writing than
as a deterrent. content (Balicco, 2002). Other recruiters use handwritten applica-
Given our focus on imperfectly aligned motives of applicants tion letters because they require more effort to produce. The
and organizations, it is worth noting that organizations are them- additional effort is believed to deter potential applicants who are
selves composed of multiple actors, all of whom may differ in their not really interested in the job, and thus handwritten letters pur-
motives for hiring (e.g., their relative focus on ability vs. commit- portedly constitute costly signals of the applicants’ motivation for
ment). For example, human resources professionals may differ the job (Bangerter, König, Blatti, & Salvisberg, 2009; Driver,
from line managers in their focus on person–job fit versus person– Buckley, & Frink, 1996).
 726 BANGERTER, ROULIN, AND KÖNIG

We summarize the above discussion on organizational efforts can use advice to prepare themselves to look like more qualified
around signaling by the following propositions: ones. Yet by transmitting potentially deceptive signals, they must
also accept cheating costs: the risk of getting caught and elimi-
Proposition 1a: Organizations try to discover and exploit nated from the selection process.
honest signals of applicant quality and commitment. Applicants not only mind-read organizations but may also adapt
their behavior during selection encounters. Applicants are moti-
Proposition 1b: Organizations invest resources to keep cheat- vated to adapt their responses in personality tests (Birkeland,
ing costs high. Manson, Kisamore, Brannick, & Smith, 2006; Marcus, 2006) or
the way they present themselves in interviews to better fit the job
Proposition 1c: The predictive validity of a selection device
profile and alter interviewers’ evaluations (Stevens & Kristof,
is a function of (a) its costly or hard-to-fake nature and (b) its
1995). During interviews, they can do this using impression man-
cheating costs.
agement or faking tactics to reduce or eliminate discrepancies
Proposition 1d: Organizational actors differ in how they try between what they think they can offer and the ideal profile the
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

to identify and exploit honest signals. organization is looking for (Levashina & Campion, 2006) or
This document is copyrighted by the American Psychological Association or one of its allied publishers.

simply to be liked by the interviewer (Gilmore, Stevens, Harrell-

Applicant Adaptation Strategies: Mind-Reading Cook, & Ferris, 1999; Kristof-Brown, Barrick, & Franke, 2002).
The importance of this phenomenon explains the growing attention
Organizations and Sending the Right Signals
given to impression management and faking in selection interview
Organizations’ selection criteria are typically not made known research during the past decades (Ellis, West, Ryan, & DeShon,
to applicants in advance. However, applicants are both motivated 2002; Gilmore & Ferris, 1989; Levashina & Campion, 2007;
and able to detect what organizations are interested in (i.e., to Sackett & Lievens, 2008; Tsai, Huang, Wu, & Lo, 2010).
mind-read their criteria) and to use this information to adapt their Different kinds of honest signals may be differently vulnerable
strategies accordingly. There are also individual differences in to applicant adaptations. We argue that costly signals are easier to
applicants’ motivation and ability to engage in these activities adapt to than hard-to-fake signals. This is because shifts in cost-
(Levashina & Campion, 2006), as well as differences in the extent liness may change the accessibility of signals, whereas a hard-to-
to which applicants can adapt to different kinds of honest signals. fake signal is intrinsically difficult to manipulate. First, societal
There are individual differences in applicant motivation to and technological changes may radically decrease the cost of a
mind-read organizations. Brown and Hesketh (2004) distinguished signal and allow applicants to acquire and send it more easily. For
two applicant types that represent two prototypical categories at example, the increased accessibility of higher education has led to
opposite ends of a continuum: players and purists. Players con- an increase of university graduates in many labor markets, thereby
sider the job market as a positional game. They spend time decreasing the signaling value of educational credentials (P.
preparing themselves and mind-reading employers to maximize Brown & Hesketh, 2004). And information technologies have led
the chances of getting hired. On the other hand, purists view hiring to an increase in information about selection devices available to
as a process based on merit. They believe their individual achieve- applicants in recent years, via the mass media discussed above.
ment, capabilities, efforts, and ambition will be sufficient to get Moreover, organizations may not always invest sufficiently in
them the job they want and expect to be judged on their merit. keeping cheating costs high, thus creating opportunities for some
Applicants also differ with respect to mind-reading ability, as applicants to cheat (Levashina & Campion, 2006). On the other
shown by research on the ability to identify selection criteria hand, it seems that it is more difficult to adapt to hard-to-fake
(Kleinmann, 1993; Kleinmann et al., 2011; König, Melchers, signals. This is particularly striking for the case of ability tests.
Kleinmann, Richter, & Klehe, 2006, 2007). Applicants high in this Despite the existence of a flourishing test coaching industry, it
ability can more easily detect and adapt to criteria and thus remains controversial whether such tests can be prepared for
perform better in the selection process. effectively (Kulik, Bangert-Drowns, & Kulik, 1984; Powers, 1993;
The environment is replete with cues that facilitate mind- Ryan, Ployhart, Greguras, & Schmit, 1998).
reading. For instance, applicants can prepare for the selection This discussion on applicants’ strategies can be summarized by
process using the abundant advice literature available. Scores of the following propositions:
books, websites, or online training videos and programs tell ap-
plicants how to play the recruitment game: how to tune their Proposition 2a: Applicants try to detect organizational selec-
resumés, how to write a remarkable letter, what interview ques- tion criteria and adapt their behavior to fulfill these criteria.
tions they will be asked and what answer they should provide, how
to behave or dress during interviews, how to prepare for tests or Proposition 2b: There are individual differences in appli-
even how to cheat on them (P. Brown & Hesketh, 2004; Palmer et cants’ motivation and ability to detect organizational criteria
al., 1999). Even experienced applicants get coached by headhunt- and adapt their behavior.
ers on how to behave during job interviews to conform to hiring
Proposition 2c: Costly signals are easier to adapt to than
organizations’ expectations (Finlay & Coverdill, 2002). However,
hard-to-fake signals.
it is unclear whether advice actually improves applicants’ future
job performance or simply makes applicants better during the
Counteradaptation
selection process (Palmer et al., 1999). This situation is also
interpretable along the lines of signaling theory. Similar to mim- We showed that organizations try to identify honest signals of
icry in zoology (Zahavi & Zahavi, 1999), less qualified applicants desirable applicant qualities using selection instruments and that
 PERSONNEL SELECTION AS A SIGNALING GAME 727

applicants respond by attempting to detect organizations’ criteria The Evolution of Signaling Systems
and adapting to them. In this section we describe counteradapta-
tion (the next step in the emergence of signaling systems) and its Over time, adaptive dynamics between applicants and organi-
consequences. Organizations may counteradapt by trying to keep zations can affect the evolution of signaling systems, leading to
their selection criteria from being identified by applicants. Or they various market-level outcomes such as the decline of existing
may modify them to keep a step ahead of applicants. For instance, signaling systems or the emergence of new ones. We distinguish
if recruiters realize applicants can detect selection criteria in in- between two paths of evolution, equilibrium and escalation. As
terviews, they may change their questions or their evaluation discussed above, a signaling system is in a state of equilibrium if
process or turn to alternative selection procedures that are per- senders’ and receivers’ behaviors are mutually reinforcing. A
ceived as costlier or to harder-to-fake signals of applicant qualities. paradigm example of equilibrium is Spence’s (1973) example of
Similarly, recruiters who become aware that applicants use im- education as a signal of applicant quality. If employers believe that
pression management tactics during interviews may learn to dis- education discriminates between high-quality and low-quality ap-
count such tactics (Rosenfeld, 1997). plicants, if they structure wage differentials accordingly, and if
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

applicants invest differentially in education depending on their

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Proposition 3: Over time, cycles of repeated adaptations and quality, then employers’ beliefs will be confirmed by applicants’
counteradaptations (hereafter, adaptive dynamics) to a selec- behavior, and employers will continue to pay more for better
tion system will occur between organizations and applicants. educated applicants.
But reciprocal adaptations can also undermine the stability of
An important question at this point is whether one party is
signaling systems, leading to escalation (arms races). One promi-
systematically at an advantage over the other. Signaling theory
nent arena for an arms race is the selection interview. We showed
offers an answer to this question, by invoking the different selec-
above that applicants can use advice books to prepare for inter-
tion pressures put on applicants and organizations depending on
views. But recruiters can adapt their questions, for instance, by
the job market situation (the life– dinner principle as described
asking unexpected questions, posing trick questions, or using
above; Dawkins & Krebs, 1979) and the cost of failure (Vermeij,
puzzles (Poundstone, 2003). Recruiters may also counteradapt by
1994). This account predicts that applicants have more influence
using new interview techniques such as the patterned behavior
on signaling games (e.g., by preparing themselves, trying to iden-
interview, a technique designed to measure applicants’ past be-
tify selection criteria, or cheating) and are at an advantage relative
havior in job-related situations (Janz, 1982). But, as new interview
to organizations (Kador, 2006; Ralston & Kirkwood, 1999). In-
techniques are adopted, so does new advice become available to
deed, the selection pressure on organizations seems weaker than on
applicants. For example, advice books now propose ready-made
applicants. If there are few jobs available and unemployment is
techniques to help applicants adapt to behavioral interview ques-
high, failing to recruit a good applicant may not endanger the
tions (Ralston & Kirkwood, 1999). And applicants have been
organization’s survival, because there will be other qualified ap-
reported to routinely devise answers to such questions when pre-
plicants on the market. On the other hand, applicants often need to
paring for an interview (Martin & Pope, 2008). In sum, both
find a job in a relatively short period of time out of pure financial
recruiters and applicants try to find ways to take the control of the
necessity (P. Brown & Hesketh, 2004). They will thus be more
interview (Palmer et al., 1999), constantly adapting and counter-
motivated to adapt quickly, influencing the evolution of signaling
adapting, and the interview becomes a game in which both appli-
systems and developing a potential advantage over recruiters (Ral-
ston & Kirkwood, 1999). The prediction that applicants have a cants and interviewers are trying to trick and outguess the other
systematic advantage over organizations seems to be supported for (Kirkwood & Ralston, 1999).
the case of faking in personality tests, where attempts to identify Another prominent example of an arms race involves personal-
fakers and correct their scores often fail (Griffith & Peterson, ity testing. Personality tests are self-report measures, so they are
2008; Morgeson et al., 2007; but see also Ones, Dilchert, Viswes- vulnerable to faking (Cook, 2009). Thus, this arms race is driven
varan, & Judge, 2007). Another case is maintaining the security of by applicants’ well-documented propensity to fake on personality
item pools in ability testing using Internet technology, which tests (Ones & Viswesvaran, 1998). Many popular personality tests
experts have claimed will be “ultimately a losing battle” (Davey & have been leaked, and their structure and desirable responses are
Nering, 2002, p. 187). now widely available (e.g., Hoffman, 2001). Counteradaptations
Of course, the situation may be reversed when there are more by organizations consist in the development of techniques for
job openings than qualified applicants, or when fluctuations in detecting fakers, dissuading would-be fakers, and camouflaging
applicant pool quality (Connerley, Carlson, & Mecham, 2003) the selection criteria. Examples of attempts of detecting fakers
limit the number of qualified applicants on the market. Organiza- include the use of social desirability scales or trick questions to test
tions may then be subjected to more pressure to counteradapt than honesty. Examples of dissuasion include telling applicants that
are applicants. This may lead to increased competition among faking can be detected and will be punished (Dwight & Donovan,
organizations (see Adaptive Relationships Among Organizations 2003). Camouflaging selection criteria involves the use of more
below). subtly formulated items. A recent development in this respect is
the conditional reasoning test of aggression, where individuals
Proposition 4: The pressure to adapt and counteradapt is solve dilemmas camouflaged as inductive reasoning problems.
moderated by market forces: It will be stronger on applicants These allow inferences about applicants’ potential for dysfunc-
when there are few jobs available but stronger on organiza- tional behavior (Berry, Sackett, & Tobares, 2010; James, 1998;
tions when there are more jobs than qualified applicants. LeBreton, Barksdale, Robin, & James, 2007).
 728 BANGERTER, ROULIN, AND KÖNIG

Proposition 5a: Selection systems that are relatively difficult this correlation (mind-reading; Krebs & Dawkins, 1984). In the
to adapt to (i.e., that are based on costly or hard-to-fake near future, applicants might learn to manipulate this information
signals) will remain in stable use over time. to influence potential employers (e.g., self-censoring the content
they post when they go on the job market or even strategically
Proposition 5b: Selection systems that are relatively easy to posting content designed to impress recruiters).
adapt to will lead to processes of escalation. These examples suggest that, in the long run, escalation may
affect the stability of signaling systems. Less sophisticated systems
Signaling theory predicts that escalation will lead to more so- may decline, eventually becoming extinct, if receivers learn to
phisticated adaptations over time, and a cursory look at the case of discount information channeled by the signal. Another interesting
personnel selection seems to support this. Taking faking in per- moment in the evolution of a signaling system is its emergence, as
sonality testing as an example again, we can observe an evolution when job market actors try to establish the honesty of a signal.
in this direction. One of the earliest attempts to control faking was Users’ mistrust and discounting seems to be an initial sign of the
the invention of a lie scale (Ruch, 1942). Later on, researchers tried decline of a signaling system. An example of this comes from
to use response latencies to detect fakers (e.g., Holden & Hibbs,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

impression management research. Repeated exposure to impres-

1995), whereas latest developments consist of the conditional
This document is copyrighted by the American Psychological Association or one of its allied publishers.

sion management may lead recruiters to mistrust applicants in the

reasoning tests described above (James et al., 2005) or even long run or even to discount their responses. This may lead to an
eye-tracking technology (van Hooft & Born, 2011). adversarial relationship between recruiters and applicants or to a
Resumé screening seems to follow the same pattern of increas- cynical view of the interview as an empty ritual where parties
ing sophistication. Organizations have developed scanning soft- “simply go through the motions” (Ralston & Kirkwood, 1999, p.
ware to automatically select resumés based on the number of 199) and no information of use is exchanged.
appropriate keywords (Amare & Manning, 2009). The advice The history of personnel selection contains several examples of
literature then advised applicants to “write for the robot” (Amare extinct or declining signaling systems. Graphology as a selection
& Manning, 2009, p. 35) by directly copying keywords from job device is, arguably, near extinction (Bangerter et al., 2009). The
ads into their resumés to better match criteria. Some applicants letter of recommendation is a case of a system in decline. Letters
even use more subtle techniques, such as typing keywords in of recommendation are widely used (Ryan, McFarland, Baron, &
microscopic fonts or in white colors that are invisible to the human Page, 1999). However, there are many signs that their stability as
eye but detectable by a scanner. Organizations then developed an honest signal is compromised. They exhibit rather low reliabil-
more sophisticated scanning software to thwart applicants. ity and validity (Colarelli, Hechanova-Alampay, & Canali, 2002;
The rise of computerized aptitude testing constitutes yet another Hunter & Hunter, 1984; Moser & Rhyssen, 2001; Reilly & Chao,
case of an arms race. Computerized tests of aptitude show great 1982) because they are uniformly positive. As a result, many
promise because they can reduce administration costs (e.g., via recruiters question their utility in selection decisions (Nicklin &
unproctored testing) and speed up the selection process (Lievens & Roch, 2009). Letters of recommendation are amenable to a signal-
Burke, 2011). However, several issues have emerged, including ing game analysis (e.g., Farrell & Gibbons, 1989) based on diver-
applicant cheating and the threat of systematic item piracy by gent interests of the three parties involved: the applicant (or target),
unscrupulous test coaching vendors. Pirates can attempt to breach the letter writer, and the organization. Writers’ interests are often
tests by sending a large set of applicants to take the test and more aligned with the applicant and less with the organization
memorize items they encounter (Schnipke & Scrams, 1999). These (Colarelli et al., 2002). Thus, writers often face a kind of prisoner’s
items can then be deposited on so-called braindumps on the dilemma: If they are honest (e.g., by mentioning negative as well
Internet and sold to applicants. Test vendors have reacted by as positive information) while others are not, the applicants they
creating so-called web patrols, which are search devices that troll recommend will be at a disadvantage. This problem is exacerbated
the Internet to detect piracy, or by implementing forensic analyses by the fact that applicants have implicit ways of pressuring the
of test takers’ responses to detect items that have been compro- writer (e.g., threat of litigation, access to the letter; Farrell &
mised (Burke, 2009). All of these efforts may be construed as Gibbons, 1989; Paetzold & Wilborn, 1992). Interestingly, mistrust
increasing cheating costs (Proposition 1b). of the content of letters of recommendation has also led to coun-
Social networking websites (e.g., Facebook) may constitute teradaptations. For example, some efforts have focused on extract-
another emerging battleground for arms races between applicants ing reliable information from the text of the letter (Peres & Garcia,
and organizations. Currently, many applicants openly post per- 1962) or mind-reading the true intent of the writer by deciphering
sonal information on such sites, even to the point of exhibiting purportedly “coded” language (Thornton, 2003). Other possibili-
problematic content (e.g., related to sexual activity or drug or ties involve focusing on peripheral aspects of the letter, like its
alcohol abuse). Posting such information also correlates with cer- length. Writers write longer letters for applicants they favor (Meh-
tain personality traits (Karl, Peluchette, & Schlaegel, 2010). How- rabian, 1965), and readers are sensitive to this feature (Kleinke,
ever, organizations are increasingly using this information to check 1978). It seems like a good candidate for an honest signal, because
on applicants’ backgrounds, sometimes even infiltrating student it is hard to fake: Because writers are often high-status individuals,
groups or getting access to private information (Brandenburg, their time is a precious commodity. Their willingness to “waste” it
2008). This situation can be interpreted along the lines of the on a long letter is thus credible proof of their esteem for the
derivation principle (Tinbergen, 1952): Observable features (a applicant. In sum, the evolution of adaptations and counteradap-
Facebook profile) are incidentally but reliably correlated with tations has led to mistrust of letters of recommendation as an
unobservable characteristics (personality traits) of an organism (an honest signal, to a decline in their use and to the evolution of
applicant), and other organisms (recruiters) are learning to detect alternative ways of extracting credible information from them.
 PERSONNEL SELECTION AS A SIGNALING GAME 729

There are also examples of emerging but not yet established Take the example of internships. They are traditionally a source
alternative signaling systems. Aguinis et al. (2005) proposed that of practical experience for university graduates. However, a study
certification could be considered as signals of human resources with German university students has shown that they consider the
professionals’ potential productivity. They showed that the number internship as a way of distinguishing their resumé from those of
of individuals possessing such certification in the United States other job seekers, based, for instance, on the prestige of the
increased by 50% between 2000 and 2003. However, less than five companies they worked for. In general, students are also aware of
percent of human resources job announcements either required or what their colleagues are doing and of what kind of activities are
preferred such certification. Aguinis et al. (2005) concluded that useful to include in a resumé (Bloch, 2007).
“apparently, employers do not consider HR certification as a signal Extracurricular activities are a second way of signaling employ-
of employee value-added and future productivity” (p. 168). A ability. These activities can be considered a costly signal. Al-
more recent study found that certification increased job prospects though participation in some activities allows applicants to acquire
(Lester, Mencl, Maranto, Bourne, & Keaveny, 2010). These con- competencies related to future work (e.g., managing skills), this
flicting results suggest that human resources constituencies are does not apply to all activities. For instance, it is unclear how
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

trying to promote a new signal of applicant quality but that running a marathon makes an applicant a better manager. Yet
This document is copyrighted by the American Psychological Association or one of its allied publishers.

organizations have not yet uniformly accepted it. More generally, spending time and energy on these activities means having less
institutions of certification can constitute ways to guarantee the time to invest in studying. Therefore only high-quality applicants
quality of the certified individual or organization and thus consti- can bear the cost of such activities without hurting their academic
tute sophisticated signaling systems, provided that the certification results. However, for these activities to emerge as a signaling
is costly to acquire (Lizzeri, 1999). system, both senders (i.e., applicants) and receivers (i.e., recruit-
ers) have to consider these activities as a costly and thus honest
Proposition 6: Over time, escalation will lead less sophisti- signal and understand that the other party does. Several anecdotes
cated signaling systems to decline in use and new, more show that applicants do indeed perceive this, as the following
sophisticated systems to emerge. graduate argues: “I’ve been to America for a year, I’ve been doing
this, I’ve been doing that— employers go like ‘Wow! How has she
Adaptive Relationships Among Applicants been doing all that and got a degree?’” (P. Brown & Hesketh,
Adaptive relationships among applicants correspond to cases 2004, pp. 130 –131). The graduate’s argument (how has she been
where applicants compete with each other for job vacancies. In doing all that and got a degree) is an illustration of the handicap
principle, as soon as there are more applicants than vacancies, principle. On the other hand, there is evidence that recruiters use
applicants are in competition with each other. Thus, getting a job extracurricular achievements as signals of applicants’ value. Grad-
depends not only on the abilities of applicants to fulfill the re- uates with higher levels of participation in extracurricular activities
quirements of the job but also on their relative ability compared to and more leadership positions within these activities are perceived
other job seekers (P. Brown & Hesketh, 2004). Positioning oneself as being of higher quality and are invited to more job interviews
as an applicant is related to employability, which means adapt- (Chia, 2005; Nemanick & Clark, 2002). A recent international
ability and personal career-related assets like attitudes, knowledge, study (Hustinx et al., 2010) found that the motivation to engage in
skills, and abilities (Fugate, Kinicki, & Ashforth, 2004; Van Der volunteering activities was stronger in job markets where such
Heijde & Van Der Heijden, 2006). The discourse on employability activities are used by potential employers to evaluate productivity.
is increasingly present in the media and has become a preoccupa- Therefore, extracurricular activities do seem to constitute a costly
tion of individuals, organizations, and governments (Moreau & signaling system both for applicants and for recruiters.
Leathwood, 2006). The traditional way to signal employability That such activities are valued by employers is not new. What
was through education (Spence, 1973), which can be an honest is new is that applicants get involved in these activities not only
signal of applicant qualities if it reliably distinguishes high-quality out of intrinsic motivation but also with the strategic intention to
from low-quality applicants. However, the development of mass improve their resumés (Tomlinson, 2007). Organizations are sen-
higher education has led to an increasing graduation rate in many sitive to this and advertise extracurricular activities they offer
countries. Brown and Hesketh (2004) argued that as advanced using the employability argument. For example, a website writes
degrees become more common, the signaling power of education that “Getting involved in a university related activity is a great way to
decreases. We agree in part but suggest that, consistent with make new friends—and boost your CV” (http://www.manchester
Proposition 6 above, more sophisticated signals can emerge, such .ac.uk/undergraduate/studentlife/extra-curricularactivities/). The na-
as the reputation of the degree-granting institution in some coun- ture of extracurricular activities has also changed. Older studies of
tries, leading to educational arms races (Winston, 2004). Graduates recruiters’ preferences (Harcourt & Krizan, 1989; Hutchinson,
are acutely aware of this, as well as of the importance of distin- 1984) focused on traditional activities such as membership in
guishing themselves relative to their peers (Tomlinson, 2007, sports clubs or associations. It seems that the activities that were
2008). positively viewed by recruiters in the past are now considered
In response to the above development, new ways for applicants commonplace among applicants, who try to distinguish themselves
to honestly signal their abilities have emerged. Like many signal- with increasingly inventive activities (P. Brown & Hesketh, 2004).
ing systems we discussed previously, these activities were not For instance, MBA students signal their ability by running mara-
originally undertaken for signaling purposes, but under current thons, sailing regattas, making films, or climbing Mount Everest
market forces (i.e., media discourse on employability) they have while still getting top grades (Morris, 2007).
evolved to become signals interpreted as such by both recruiters Competition may be particularly intense for new job market
and applicants. entrants such as graduates, who are pushed to find means of
 730 BANGERTER, ROULIN, AND KÖNIG

distinguishing themselves because they lack job experience and research on internships (Bloch, 2007), but much more work could
because of the steady decrease in the signaling power of their be done.
primary credentials (education). More experienced job seekers are Although individuals may differ in their propensity to distin-
likely to rely on other costly or hard-to-fake signals of quality, guish themselves from other applicants (Proposition 7b), all
such as job experience or reputation. applicants are subject to market pressure resulting indirectly
Job experience is valued by employers because of its link with from the choices of other applicants. Recall the abovemen-
performance (Schmidt, Hunter, & Outerbridge, 1986). However, tioned distinction between player and purist applicants (P.
experience is difficult to display in detail. Typically, experience is Brown & Hesketh, 2004). Players view their employability
showcased by applicants in their resumés, in an attempt to induce relative to others, whereas purists do not. However, even purists
recruiters to invite them for an interview. Recruiters are indeed may be pressured to switch strategies in order to avoid being
sensitive to various aspects of experience, such as statements of crowded out of the job market. Frank (2006) discussed the case
accomplishments (Thoms et al., 1999), in deciding which appli- of legislation prohibiting recruiters from asking female appli-
cants to interview (Behrenz, 2001). The interview itself is often cants about plans to marry or have children. This legislation can
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

focused on evaluating experience (Salgado & Moscoso, 2002). be ineffective to the degree that women who do not have such
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Given these incentives, applicants may be motivated to seek dis- plans may realize they have an advantage relative to rivals if
tinctiveness through displays of experience, perhaps to the point of they spontaneously disclose such information, which may in-
exaggerating their past accomplishments or responsibilities. The duce some of them to do just that (these women can be called
difficulty of converting experience to a visible market signal players). If enough players do this, other women (who are
becomes clear when considering the many degree mills that offer purists) may be pressured to do so as well in order to not invite
bogus degrees based on “life experience” (U.S. Department of unfavorable inferences about their future family-related plans
Education, 2011). and thus jeopardize their hiring prospects.
A primary means of signaling reputation is by enlisting third A similar logic applies to faking. Earlier, we discussed how
parties to vouch for oneself, as in letters of recommendation or faking during the selection process (Levashina & Campion, 2007;
reference checks. A special case of this is when reputations are Ones & Viswesvaran, 1998) can be considered part of an arms race
guaranteed by institutional membership (e.g., a physician who is a between applicants and organizations. But faking is also part of the
member of a professional society). However, like all costly signals, competition among applicants. In this context, faking can be seen
third-party enlistment can be faked. There are even companies that as a prisoner’s dilemma. Applicants’ behavior in selection situa-
help applicants fake job references, some going so far as to provide tions will depend on what they believe rivals will do. Because
bogus employers, complete with bogus contacts who will answer faking can modify selection decisions depending on the proportion
recruiters’ phone calls in order to bypass reference checks. Coun-
of applicants who fake, the extent of faking, and the selection ratio
teradaptations to these tactics include cross-checking companies
(Levashina & Campion, 2007; Marcus, 2006; Stewart, Darnold,
and phone numbers to make sure they are real (Leonard, 2009). As
Zimmerman, Parks, & Dustin, 2010), applicants who do not fake
suggested by Proposition 2c, technological innovations can signif-
when many of their competitors do can sometimes get eliminated
icantly decrease the cheating costs associated with managing rep-
by their honesty (Morgeson et al., 2007). Thus, assuming that
utation, thereby leading to arms races (Tennie, Frith, & Frith,
others may fake, applicants may reason that they improve their
2010).
own chances by doing so as well.
As described in Proposition 2a, applicants’ adaptive behaviors
Using the example of extracurricular activities again, given
may also depend on individual characteristics, such as the moti-
enough market pressure and an abundance of applicants with
vation or ability to engage in faking (Levashina & Campion, 2006;
similar formal qualifications, such activities are a signal that
McFarland & Ryan, 2000) or the extent to which applicants engage
can potentially lead to an arms race among applicants. In other
in player or purist strategies (P. Brown & Hesketh, 2004). We
words, applicants might allocate resources to engaging in in-
therefore suggest:
creasingly impressive extracurricular activities over time. Such
Proposition 7a: Applicants try to send signals that distinguish a process of escalation may also motivate cheaters to try and
them from other applicants to appear more attractive to em- cheaply mimic these signals by falsely claiming to engage in
ployers. impressive activities. For example, one student blithely admit-
ted to adding expertise in martial arts to her resumé depending
Proposition 7b: There are individual differences in the degree on the position she applies for (P. Brown & Hesketh, 2004). In
to which applicants try to send signals that distinguish them general, the above considerations suggest that adaptive behav-
from other applicants. ior among applicants can lead to escalation, if the pressure from
the job market is severe enough.
Proposition 7a goes beyond current conceptualizations of sig-
naling in selection research (e.g., Cable & Judge, 1997) because it Proposition 8: The higher the market pressure, the more
emphasizes that applicants do not just try to appeal to organiza- applicants will attempt to distinguish themselves from other
tions (Proposition 2a) but also position themselves relative to each applicants, leading to escalation.
other. It has several implications for research, for example, that
applicants will try to be aware of what other applicants are doing One way to test Proposition 8 is by analyzing archives of
and that they will integrate this awareness into their own job applicants’ resumés over time, quantifying the efforts invested in
market choices. Initial evidence for this conjecture comes from extracurricular activities, or their originality, and tracking their
 PERSONNEL SELECTION AS A SIGNALING GAME 731

evolution as a function of past job market pressure.3 Another attractiveness) that constitutes a potential signal they are truly
possibility that follows from Proposition 8 is that job market committed to a long-term relationship.
pressure may affect rates of applicant faking. Robie, Emmons, Of course, not all organizations can easily counteradapt to
Tuzinski, and Kantrowitz (2011) found that mean levels of appli- competitors’ conditions, nor may they want to. Organizations may
cant personality scores increased across three time periods with compete in a different institutional environment (Klehe, 2004) and
increasing unemployment rates. They suggested that unemploy- may have different dynamic adaptation capabilities (Teece, Pisano,
ment may increase market pressure and lead to higher applicant & Shuen, 1997), or may be less inclined to engage in the War for
motivation or levels of faking, in line with Proposition 8. Talent, depending on their organizational culture or values (Pfef-
fer, 2001).

Adaptive Relationships Among Organizations Proposition 9a: Organizations try to send signals that distin-
guish them from each other to appear more attractive to
The third situation we examine is analogous to the previous one: applicants.
Adaptations among organizations correspond to cases where or-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ganizations compete with each other to attract applicants. This is a

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Proposition 9b: Organizations differ in the degree to which

special case of more general processes whereby organizations seek they try to send signals that distinguish them from each other.
ways to improve their performance relative to their rivals (Barnett
& Hansen, 1996). For example, if company A differentiates itself Proposition 9a goes beyond current conceptualizations of sig-
from others by developing a competitive advantage (e.g., better naling in organizational attraction research (e.g., Ehrhart &
brand image through a new marketing program), its competitor B Ziegert, 2005) because it emphasizes that organizations do not just
will face performance shortfalls. It will develop improvements try to attract applicants but position themselves relative to each
(e.g., its own marketing campaign) to reduce the difference with A. other. However, this process may only confer short-lived advan-
This move will put greater competitive pressure on A to respond, tages because, as described above (Barnett & Hansen, 1996),
leading to escalation (Van Valen, 1973). becoming more attractive increases the selective pressure on com-
In recruitment, such an arms race exists: the War for Talent (P. petitors who can counteradapt in turn, thus eliminating the orga-
Brown & Hesketh, 2004; Larkan, 2007; Michaels et al., 2001; nization’s competitive advantage. For instance, some companies
Resto, Ybarra, & Sethi, 2007). It is part of the more general began offering signing bonuses not only to MBAs but also to
problem of labor market shortage that is a prime concern of both undergraduate students after learning that their competitors were
practitioners and academics (Lievens, van Dam, & Anderson, making such offers (Gardner, 2002). Another development in the
2002). The War for Talent suggests that talented employees are a arms race among organizations are “golden handcuffs,” loyalty
scarce resource that organizations must compete for in order to bonuses offered by organizations to retain key employees. But, in
survive.4 Successful organizations are those that adapt successfully a counteradaptation, competitors offer bonuses, called “golden
to this situation by mind-reading applicants’ requirements. High hellos,” to explicitly compensate for the loss of the loyalty bonus
wages and bonuses, fast-track promotions systems based on em- (Cappelli, 2000). At the executive level, CEO compensation is
ployees’ potential, responsibilities given to talented junior manag- strongly influenced by what competitors are paying (Fulmer,
ers, and selective hiring have all been used to attain such objectives 2009).
(Michaels et al., 2001). In addition to employing such economic
tactics, organizations can try to appeal to talented applicants by Proposition 10: The higher the market pressure, the more
signaling social reputation. For instance, they can portray them- organizations will attempt to distinguish themselves from
selves as being environmentally responsible (Behrend, Baker, & each other, leading to escalation.
Thompson, 2009), supportive of diversity (Ng & Burke, 2005), or
committed to stakeholders (Turban & Greening, 1997), even to the One way researchers could track this process is by analyzing the
point of triggering arms races (Starr, 2008). content of companies’ recruitment websites (e.g., mentions of
Another signal organizations can send is a realistic job preview environmentally responsible or diversity supportive claims, de-
(Wanous, 1973). Realistic job previews feature candid information scription of fast-track promotion opportunities) over time or by
about both positive and negative aspects of a position. Inclusion of comparing different job markets.
negative information decreases applicants’ initial expectations As noted above, if signaling systems are to develop, both parties
about a job and favors self-selection on the part of applicants. This need to converge on interpreting the signal. Research suggests that
in turn has the benefit of increasing commitment and reducing applicants interpret recruitment initiatives as signals of unobserv-
turnover on the part of those applicants who remain in the selection
process (Premack & Wanous, 1985). Realistic job previews can be 3
interpreted as honest signals of an organization’s commitment to a This research strategy is similar to that employed by paleontologists
who examine the fossil record to investigate evolutionary pressures and
long-term relationship based on transparency of information. They
adaptations of organisms over time (Vermeij, 1994).
constitute handicaps because they are costly to design and their 4
The War for Talent rhetoric has been criticized as being based on
utility in a narrow economic sense has been disputed (Buckley, incorrect facts, flawed assumptions, and hype (Pfeffer, 2001). We take no
Fedor, Carraher, Frink, & Marvin, 1997). Moreover, they disclose stance on these issues here but note that even skeptical organizations may
negative information about the job and the organization. As such, find themselves pressured into engaging in the War for Talent if most of
organizations that use realistic job previews impose a cost on their competitors do so as well (as purists may be pressured into more
themselves (i.e., going out of their way to decrease their own player tactics in adaptive dynamics among applicants).
 732 BANGERTER, ROULIN, AND KÖNIG

able organizational characteristics. Individuals with high academic selection in a novel way by specifying analogous adaptive pro-
achievement (grades and cognitive ability) prefer organizations cesses in each domain. Adaptive relationships among organiza-
offering selective hiring practices, merit-based pay, praise and tions to attract applicants (an aspect of recruitment) are analogous
recognition, or fast-track promotions (Trank, Rynes, & Bretz, to adaptive relationships among applicants trying to maximize
2002). Even less ambitious applicants may interpret signals about their relative attractiveness to potential employers (an aspect of
the organization as a good or socially responsible employer. In- selection). In other words, recruitment and selection may serve
deed, the literature on organizational attraction and applicant re- similar functions while being accomplished by different structures
actions (Ryan & Ployhart, 2000) suggests that the selection process (Morgeson & Hofmann, 1999). Our approach is similar to related
is used by applicants to infer characteristics of organizations. Thus, approaches in personnel economics that consider recruitment and
applicants’ preferences both reflect and affect the arms race among selection as two facets of the same problem, namely, matching
organizations. firms and workers (Lazear & Oyer, in press).

Advantages of Signaling Theory for Personnel

Implications for Individual-Level Theoretical
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Selection
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Approaches Within Selection

Signaling theory (Spence, 1973; Zahavi, 1975; Zahavi & Za-
havi, 1999) is a broad framework (Cronk, 2005) that describes in A signaling framework complements existing theoretical ap-
a principled and parsimonious manner the incentives involved in proaches in personnel selection, such as the psychometric ap-
cooperative interactions between rational individuals with partly proach, the applicant reactions approach, and the social process
divergent goals. It explains how exchange of accurate information approach. The psychometric approach focuses on systematically
is possible under such conditions (i.e., by making the signals costly documenting the properties (reliability and validity) of selection
or hard to fake) and how repeated cycles of micro-level phenom- devices. It has been immensely successful (Schmidt & Hunter,
ena (reciprocal adaptations of individual job market actors) can 1998). However, it is based on several assumptions that have been
affect macro-level phenomena (the evolution of personnel selec- questioned in recent years, for example, that selection is a unilat-
tion signaling systems). It also describes how macro-level phe- eral process on the part of the organization. Research has shown
nomena (market selection pressures) affect micro-level, individual that applicants’ choices of organizations (Murphy, 1986) affect the
adaptive behavior (e.g., faking). This articulation of phenomena at utility of selection procedures, thus suggesting that selection is a
different levels of analysis (a classic example of the reciprocal bilateral process. And applicant reactions research (Ryan & Ploy-
relationship between structure and interaction; Morgeson & Hof- hart, 2000) as well as related theoretical work focused on under-
mann, 1999) makes a signaling framework theoretically innova- standing the effects of applicants’ perceptions of selection proce-
tive, because most theoretical work in personnel selection research dures has shown that selection procedures are not neutral
focuses on individual-level processes, and the few studies on predictors but also act as transmitters of information about the
macro-level processes (e.g., Klehe, 2004) are not articulated with organization to applicants (Anderson, 2001). Critics have argued
individual-level theories. We now examine the implications of that it has not yet convincingly been shown that applicant reactions
signaling theory for three levels of theoretical development: the really matter (Sackett & Lievens, 2008). Signaling suggests a
level of neighboring domains of study (recruitment and selection), theoretical rationale for why they should matter, because applicant
the level of individual-level theoretical approaches within selec- adaptations are a prime motor of the long-term evolution of sig-
tion (the psychometric approach, the applicant reactions approach, naling systems. But signaling would also suggest studying the
and the social process approach), and the level of macro-level effects of applicants’ repeated interactions on their perceptions of
theoretical approaches within selection (institutional theory). We and attitudes toward selection systems in general and about the job
also examine an example of how signaling theory can open up new market, rather than only toward a particular selection device or a
areas of investigation as well as implications for the practice of particular organization.
selection. Seen through the lens of signaling theory, the psychometric
approach and the applicant reactions approach can be seen as
Implications for Neighboring Domains of Study: embodying complementary perspectives on the selection relation-
ship. However, they are both silent about the interactive and
Recruitment and Selection
adaptive nature of that relationship. Some recent theoretical per-
The neighboring domains of recruitment and selection have spectives on personnel selection, such as social process models
often been treated separately (Barber, 1998). Few theoretical ac- (Derous & De Witte, 2001; Herriot, 1993), do emphasize the
counts systematically and comprehensively examine interactions adaptive, interpersonal, and motivational nature of selection. For
or similarities between these fields (but see Wanous, 1980). Par- example, Herriot (1993, p. 372) asserted about the selection rela-
ticularly striking is the fact that research in selection has focused tionship that “clearly, information is being processed by both
on how organizations interpret signals from applicants (Cable & parties, and how each processes the information provided by the
Judge, 1997) and research in recruitment has focused on how other’s behavior affects how each behaves and is consequently
applicants interpret signals sent by organizations (Ehrhart & perceived.” But though social process models focus on individual-
Ziegert, 2005), without any recognition of this similarity. And yet, level processes and outcomes, the signaling framework links in-
recruitment and selection often occur simultaneously and are in- dividual behavior of job market actors and emergent collective
terdependent in their outcomes (Barber, 1998). The signaling phenomena and also extends the scope of relevant phenomena to
perspective we have developed links aspects of recruitment and repeated interactions and their long-term, macro-level effects.
 PERSONNEL SELECTION AS A SIGNALING GAME 733

Implications for Macro-Level Theoretical Approaches from researchers because it is not a side effect of personnel
Within Selection selection; rather, it plays a systematic role in affecting individual-
level adaptive behaviors of applicants.
Signaling theory complements the phenomena accounted for by
institutional theory, which has recently become more prominent in
personnel selection. For example, Klehe (2004) applied institu- Implications for Selection Practice
tional theory to the question of how organizations choose selection
devices. She developed a model of the various environmental A signaling approach to personnel selection also has implica-
pressures that influence the adoption of selection devices as well as tions for selection practice. Many of the issues discussed in this
the types of responses organizations may adopt. An example article are well known to practitioners but have been largely
response is imitation, whereby organizations adopt selection pro- ignored by academics. Palmer, Campion, and Green (1999) sug-
cedures only after other organizations have done so, to reduce gested that the notion of the selection interview as a competitive
uncertainty. arena is much more of an issue in the practitioner literature on
Institutional theory predicts limits to signaling dynamics. For interview preparation than it has been for academics: “A sense of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

inherent conflict between interviewers and applicants is evident in

This document is copyrighted by the American Psychological Association or one of its allied publishers.

example, the selection interview is arguably an institutionalized

aspect of personnel selection (i.e., it is taken for granted by all the practice literature, although it is generally downplayed or
actors) that will not foreseeably become extinct; it is therefore ignored in the extant research” (p. 346). This claim is consistent
likely to remain a battleground for arms races between applicants with recurrent examples of competitive rhetoric from advice books
and recruiters. On the other hand, signaling also constrains insti- for recruiters. For example, Kador (2006) writes, “If you repeat
tutional pressures. For example, imitation may not be a rational questions or use generalized interview questions, you will most
strategy for certain actors when the costs of adopting a signal are assuredly be fooled by a group of well-rehearsed applicants” (p.
too high (e.g., not all applicants can afford to imitate their col- xi). And Kanter (1995) writes, “Few people view the job interview
leagues’ more exotic extracurricular activities). In general, then, as a joint effort to find a good match or to determine that a match
signaling theory and institutional theory describe opposing pres- does not exist. Instead, both sides view it as a game” (p. xvii). On
sures and make complementary predictions (Terlaak & King, the other hand, qualitative research on applicants’ experiences on
2010). We speculate that the specific nature of institutional theory the labor market often reveals a sense of mistrust, frustration, and
makes it inherently more applicable to adaptive relationships cynicism (e.g., Billsberry, 2007; P. Brown & Hesketh, 2004).
among organizations, whereas signaling is more applicable to These examples suggest that dilemmas of cooperation are a reality
adaptive relationships between applicants and organizations. The for many job market actors. At the same time, the aggregated
links between institutional theory and signaling theory in person- individual decisions of these actors affect adaptive dynamics de-
nel selection should be explored in more detail. scribed in this article.
Currently, it seems premature to offer specific recommendations
for practice. However, if there is one broad practical issue that
Illuminating Blind Spots of Selection Research:
emerges from signaling theory as applied to the domain of per-
The Advice Industry
sonnel selection (writ large, i.e., also encompassing analogous
Signaling theory can stimulate research on previously underre- processes of applicant choice), it is unquestionably that of trust as
searched areas in selection. In particular, macro-level factors driv- a means of defusing escalation. Trust is the mechanism that keeps
ing escalation have largely been ignored by selection research so information exchanges reliable; in its absence, would-be cooper-
far, and the lack of theory capturing these phenomena has most ators cannot be sure they will not be exploited or duped.
likely contributed to these blind spots. A prominent example is the Organizations and applicants, as well as other job market actors
advice industry for potential applicants. The adaptive dynamics such as consultants, should be sensitized to the importance of
between applicants and organizations leading to applicant adapta- building trust and opportunities for generating trust and coopera-
tions and counteradaptations has driven the emergence of a huge tion in the selection process should be pursued (Pearce, 2000).
industry that claims to help applicants to cheaply mimic otherwise This may entail, among other things, renouncing the rhetoric of
costly signals of ability and commitment. This industry operates conflict (e.g., the War for Talent) when creating or diffusing
through media such as web forums where one can purportedly discourses about selection (e.g., advice books) or attempting to
learn how to beat commercially sold mental ability or personality analyze and thus better align actors’ incentives. Further, long-term
tests (e.g., www.jobtestprep.co.uk), books on how to prepare a guarantees of trustworthiness in a particular market (e.g., reputa-
resumé or answer interview questions, or headhunters who train tions) are a way of solving the dilemma of cooperation (Tennie et
assessment center participants (see Finlay & Coverdill, 2002). al., 2010). By this logic, both applicants and organizations should
There is even a thriving market for fake degrees of higher educa- invest resources to maintain their reputational standing in the job
tion (see Bear & Ezell, 2005) that has been estimated to have market.
generated more than one billion dollars of sales and to have sold Developing and maintaining a reputation as a trustworthy actor
degrees to at least one million customers. If faked degrees repre- does not imply blindly trusting one’s partners or opening oneself
sent such a large market (Bear & Ezell, 2005), it is easy to imagine up to exploitation by cheaters. In the prisoner’s dilemma, one of
how huge the advice industry in total must be and how many the most successful strategies is tit for tat, which initially trusts
applicants have been willing to spend money for it. Such topics are cooperation partners but punishes them if they cheat (Poundstone,
rarely mentioned in the academic literature on selection. However, 1993). This implies that organizations should both invest in man-
according to signaling theory, the advice market deserves attention aging reputation but also use selection devices that tap into honest
 734 BANGERTER, ROULIN, AND KÖNIG

(either costly or hard-to-fake) signals and seek to keep cheating Berry, C. M., Sackett, P. R., & Tobares, V. (2010). A meta-analysis of
costs high to deter and punish cheating by applicants. conditional reasoning tests of aggression. Personnel Psychology, 63,
361–384. doi:10.1111/j.1744-6570.2010.01173.x
Billsberry, J. (2007). Experiencing recruitment and selection. Chichester,
Conclusion England: Wiley.
Signaling theory offers a simple yet powerful set of mechanisms Birkeland, S. A., Manson, T. M., Kisamore, J. L., Brannick, M. T., &
for charting the interactive, adaptive, and thus dynamic nature of Smith, M. A. (2006). A meta-analytic investigation of job applicant
personnel selection relationships, going beyond the current theo- faking on personality measures. International Journal of Selection and
Assessment, 14, 317–335. doi:10.1111/j.1468-2389.2006.00354.x
retical approaches to personnel selection and linking personnel
Bloch, R. (2007). “Natürlich möchte man es auch gern im Lebenslauf
psychology to related fields of study in management, economics,
stehen haben . . .”—Bedeutungen des Praktikums für Studierende [“Nat-
and other disciplines. We have explored three kinds of adaptive urally people want to have that in the resumés too . . .” —The meaning
relationships— between applicants and organizations, among ap- of internships for students]. Beiträge zur Hochschulforschung, 4, 82–
plicants, and among organizations—showing how many important 106.
phenomena can be described in terms of signaling. Signaling Boster, J. S. (2003). Blood feud and table manners: A neo-Hobbesian
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

theory has important implications for theoretical development be-

This document is copyrighted by the American Psychological Association or one of its allied publishers.

approach to Jivaroan warfare. Antropológica, 99–100, 153–164.

tween neighboring fields of study and at the level of theoretical Boster, J. S., Yost, J., & Peeke, C. (2003). Rage, revenge, and religion:
approaches in personnel selection. Signaling theory leads to new Honest signaling of aggression and nonaggression in Waorani coali-
predictions about selection phenomena, focuses attention on un- tional violence. Ethos, 31, 471– 494. doi:10.1525/eth.2003.31.4.471
derresearched but important topics, and can inform selection prac- Brandenburg, C. (2008). The newest way to screen job applicants: A social
networker’s nightmare. Federal Communication Law Journal, 60, 597–
tice. The benefits of viewing personnel selection as a network of
626.
adaptive relationships among job market actors are numerous.
Brief, A. P. (2000). Still servants of power. Journal of Management
Inquiry, 9, 342–351. doi:10.1177/105649260000900403
References Brown, B. K., & Campion, M. A. (1994). Biodata phenomenology: Re-
cruiters’ perceptions and use of biographical information and resumé
Aamodt, M. G. (2006). Industrial/organizational psychology: An applied
screening. Journal of Applied Psychology, 79, 897–908. doi:10.1037/
approach. Belmont, CA: Wadsworth.
0021-9010.79.6.897
Aguinis, H., Michaelis, S. E., & Jones, N. M. (2005). Demand for certified
Brown, P., & Hesketh, A. (2004). The mismanagement of talent: Employ-
human resources professionals in Internet-based job announcements.
ability and jobs in the knowledge economy. Oxford, England: Oxford
International Journal of Selection and Assessment, 13, 160 –171. doi:
University Press.
10.1111/j.0965-075X.2005.00310.x
Buckley, M. R., Fedor, D. B., Carraher, S. M., Frink, D. D., & Marvin, D.
Amare, N., & Manning, A. (2009). Writing for the robot: How employer
(1997). The ethical imperative to provide recruits realistic job previews.
search tools have influenced resumé rhetoric and ethics. Business Com-
Journal of Managerial Issues, 9, 468 – 484.
munication Quarterly, 72, 35– 60. doi:10.1177/1080569908330383
Anderson, N. (2001). Towards a theory of socialization impact: Selection Burke, E. (2009). Preserving the integrity of online testing. Industrial and
as pre-entry socialization. International Journal of Selection and Assess- Organizational Psychology, 2, 35–38. doi:10.1111/j.1754-
ment, 9, 84 –91. doi:10.1111/1468-2389.00165 9434.2008.01104.x
Balicco, C. (2002). L’utilisation de la graphologie dans le recrutement de Cable, D. M., & Judge, T. A. (1997). Interviewers’ perceptions of person–
cadres au sein des cabinets conseils [The use of graphology for man- organization fit and organizational selection decisions. Journal of Ap-
agement selection in consulting firms]. L’ Orientation Scolaire et Pro- plied Psychology, 82, 546 –561. doi:10.1037/0021-9010.82.4.546
fessionnelle, 31, 195–222. Cappelli, P. (2000). A market-driven approach to retaining talent. Harvard
Bangerter, A., König, C. J., Blatti, S., & Salvisberg, A. (2009). How Business Review, 78, 103–111.
widespread is graphology in personnel selection practice? A case study Caro, T. M. (1986a). The functions of stotting: A review of the hypotheses.
of a job market myth. International Journal of Selection and Assessment, Animal Behaviour, 34, 649 – 662. doi:10.1016/S0003-3472(86)80051-3
17, 219 –230. doi:10.1111/j.1468-2389.2009.00464.x Caro, T. M. (1986b). The functions of stotting in Thomson’s gazelles:
Barber, A. E. (1998). Recruiting employees: Individual and organizational Some tests of the predictions. Animal Behaviour, 34, 663– 684. doi:
perspectives. Thousand Oaks, CA: Sage. 10.1016/S0003-3472(86)80052-5
Barnett, W. P., & Hansen, M. T. (1996). The Red Queen in organizational Chia, Y. M. (2005). Job offers of multi-national accounting firms: The
evolution. Strategic Management Journal, 17, 139 –157. doi:10.1002/ effects of emotional intelligence, extra-curricular activities, and aca-
smj.4250171010 demic performance. Accounting Education, 14, 75–93. doi:10.1080/
Bavelas, J. B. (1992). Research into the pragmatics of human communi- 0693928042000229707
cation. Journal of Strategic and Systemic Therapies, 11, 15–29. Colarelli, S. M., Hechanova-Alampay, R., & Canali, K. G. (2002). Letters
Bear, J., & Ezell, A. (2005). Degree mills: The billion dollar industry that of recommendation: An evolutionary perspective. Human Relations, 55,
has sold over a million fake diplomas. Amherst, NY: Prometheus. 315–344. doi:10.1177/0018726702553002
Behrend, T. S., Baker, B. A., & Thompson, L. F. (2009). Effects of Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011).
pro-environmental recruiting messages: The role of organizational rep- Signaling theory: A review and assessment. Journal of Management, 37,
utation. Journal of Business and Psychology, 24, 341–350. doi:10.1007/ 39 – 67. doi:10.1177/0149206310388419
s10869-009-9112-6 Connerley, M. L., Carlson, K. D., & Mecham, R. L. I. (2003). Evidence of
Behrenz, L. (2001). Who gets a job and why: An explorative study of differences in applicant pool quality. Personnel Review, 32, 22–39.
employers’ recruitment behavior. Journal of Applied Economics, 4, doi:10.1108/00483480310454709
255–278. Cook, M. (2009). Personnel selection: Adding value through people.
Bergstrom, C. T. (2006, October 23). An introduction to the theory of Chichester, England: Wiley.
honest signaling. Retrieved from http://octavia.zoology.washington.edu/ Coombs, R. H., & Virshup, B. (1998). Surviving medical school. Thousand
handicap Oaks, CA: Sage.
 PERSONNEL SELECTION AS A SIGNALING GAME 735

Cronk, L. (2005). The application of animal signaling theory to human interaction for scarce human resources. Human Resource Management,
phenomena: Some thoughts and clarifications. Social Science Informa- 41, 225–237. doi:10.1002/hrm.10033
tion, 44, 603– 620. doi:10.1177/0539018405058203 Gilmore, D. C., & Ferris, G. R. (1989). The effects of applicant impression
Darwin, C. (1871). The descent of man and selection in relation to sex. management tactics on interviewer judgments. Journal of Management,
London, England: Murray. 15, 557–564. doi:10.1177/014920638901500405
Davey, T., & Nering, M. (2002). Controlling item exposure and maintain- Gilmore, D. C., Stevens, C. K., Harrell-Cook, G., & Ferris, G. R. (1999).
ing item security. In C. N. Mills, M. T. Potenza, J. J. Fremer, & W. C. Impression management tactics. In R. W. Eder & M. M. Harris (Eds.),
Ward (Eds.), Computer-based testing: Building the foundation for future The employment interview handbook (pp. 321–336). Thousand Oaks,
assessments (pp. 165–191). Mahwah, NJ: Erlbaum. CA: Sage.
Davies, N. B., & Halliday, T. (1977, September 1). Optimal mate selection Griffith, R. L., & Peterson, M. H. (2008). The failure of social desirability
in the toad Bufo bufo. Nature, 269, 56 –58. doi:10.1038/269056a0 measures to capture applicant faking behavior. Industrial and Organi-
Dawkins, R., & Krebs, J. R. (1979). Arms races between and within zational Psychology, 1, 308 –311. doi:10.1111/j.1754-9434.2008
species. Proceedings of the Royal Society of London: Series B: Biolog- .00053.x
ical Sciences, 205, 489 –511. doi:10.1098/rspb.1979.0081 Gunter, B., Furnham, A., & Drakeley, R. (1993). Biodata: Biographical
Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press. indicators of business performance. London, England: Psychology
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Derous, E., & De Witte, K. (2001). Looking at selection from a social Press.
process perspective: Towards a social process model on personnel Harcourt, J., & Krizan, A. C. (1989). A comparison of resumé content
selection. European Journal of Work and Organizational Psychology, preferences of Fortune 500 personnel administrators and business com-
10, 319 –342. doi:10.1080/13594320143000708 munication instructors. Journal of Business Communication, 26, 177–
Deutsch Salamon, S., & Deutsch, Y. (2006). OCB as a handicap: An 190. doi:10.1177/002194368902600206
evolutionary psychological perspective. Journal of Organizational Be- Henrich, J., McElreath, R., Barr, A., Ensminger, J., Barrett, C., Bolyanatz,
havior, 27, 185–199. doi:10.1002/job.348 A., . . . Ziker, J. (2006, June 23). Costly punishment across human
Driver, R. W., Buckley, M. R., & Frink, D. D. (1996). Should we write off societies. Science, 312, 1767–1770. doi:10.1126/science.1127333
graphology? International Journal of Selection and Assessment, 4, 78 – Hermelin, E., Lievens, F., & Robertson, I. T. (2007). The validity of
86. doi:10.1111/j.1468-2389.1996.tb00062.x assessment centres for the prediction of supervisory performance rat-
Dwight, S. A., & Donovan, J. J. (2003). Do warnings not to fake ings: A meta-analysis. International Journal of Selection and Assess-
reduce faking? Human Performance, 16, 1–23. doi:10.1207/ ment, 15, 405– 411. doi:10.1111/j.1468-2389.2007.00399.x
S15327043HUP1601_1 Herriot, P. (1993). A paradigm bursting at the seams. Journal of Organi-
Ehrhart, K. H., & Ziegert, J. C. (2005). Why are individuals attracted to zational Behavior, 14, 371–375. doi:10.1002/job.4030140407
organizations? Journal of Management, 31, 901–919. doi:10.1177/ Herriott, S. R., Levinthal, D., & March, J. G. (1985). Learning from
0149206305279759 experience in organizations. American Economic Review, 75, 298 –302.
Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Highhouse, S., Thornbury, E. E., & Little, I. S. (2007). Social-identity
Academy of Management Review, 14, 57–74. functions of attraction to organizations. Organizational Behavior and
Ekman, P., & Friesen, W. V. (1969). The repertoire of nonverbal behavior: Human Decision Processes, 103, 134 –146. doi:10.1016/j.obhdp
Categories, origins, usage and coding. Semiotica, 1, 49 –98. .2006.01.001
Ellis, A. P. J., West, B. J., Ryan, A. M., & DeShon, R. P. (2002). The use Hoffman, E. (2001). Ace the corporate personality test. New York, NY:
of impression management tactics in structured interviews: A function of McGraw-Hill Professional.
question type? Journal of Applied Psychology, 87, 1200 –1208. doi: Holden, R. R., & Hibbs, N. (1995). Incremental validity of response
10.1037/0021-9010.87.6.1200 latencies for detecting fakers on a personality test. Journal of Research
Farrell, J., & Gibbons, R. (1989). Cheap talk with two audiences. American in Personality, 29, 362–372. doi:10.1006/jrpe.1995.1021
Economic Review, 79, 1214 –1223. Huber, G. P. (1991). Organizational learning: The contributing processes
Finlay, W., & Coverdill, J. E. (2002). Headhunters: Matchmaking in the and the literatures. Organization Science, 2, 88 –115. doi:10.1287/
labor market. Ithaca, NY: Cornell University Press. orsc.2.1.88
Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). Universal dimensions of Huffcutt, A. I., & Arthur, W. J. (1994). Hunter and Hunter (1984) revisited:
social cognition: Warmth and competence. Trends in Cognitive Sci- Interview validity for entry-level jobs. Journal of Applied Psychology,
ences, 11, 77– 83. doi:10.1016/j.tics.2006.11.005 79, 184 –190. doi:10.1037/0021-9010.79.2.184
FitzGibbon, C. D., & Fanshawe, J. H. (1988). Stotting in Thomson’s Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of alternative
gazelles: An honest signal of condition. Behavioral Ecology and Socio- predictors of job performance. Psychological Bulletin, 96, 72–98. doi:
biology, 23, 69 –74. doi:10.1007/BF00299889 10.1037/0033-2909.96.1.72
Flinn, M. V., Geary, D. C., & Ward, C. V. (2005). Ecological dominance, Hustinx, L., Handy, F., Cnaan, R. A., Brudney, J. L., Pessi, A. B., &
social competition, and coalitionary arms races: Why humans evolved Yamauchi, N. (2010). Social and cultural origins of motivations to
extraordinary intelligence. Evolution and Human Behavior, 26, 10 – 46. volunteer. International Sociology, 25, 349 –382. doi:10.1177/
doi:10.1016/j.evolhumbehav.2004.08.005 0268580909360297
Frank, R. H. (1988). Passions within reason: The strategic role of the Hutchinson, K. L. (1984). Personnel administrators’ preferences for re-
emotions. New York, NY: Norton. sumé content: A survey and review of empirically based conclusions.
Frank, R. H. (2006). Microeconomics and behavior. New York, NY: Journal of Business Communication, 21, 5–14. doi:10.1177/
McGraw-Hill. 002194368402100401
Fugate, M., Kinicki, A. J., & Ashforth, B. E. (2004). Employability: A Isaacson, J. A., Griffith, R. L., Kung, M.-C., Lawrence, A., & Wilson, K.-A.
psycho-social construct, its dimensions, and applications. Journal of (2008, April). Liar, liar: Examining background checks and applicants who
Vocational Behavior, 65, 14 –38. doi:10.1016/j.jvb.2003.10.005 fail them. Paper presented at the twenty-third conference for the Society for
Fulmer, I. S. (2009). The elephant in the room: Labor market influence on Industrial and Organizational Psychology, San Francisco, CA.
CEO compensation. Personnel Psychology, 62, 659 – 695. doi:10.1111/ James, L. R. (1998). Measurement of personality via conditional reasoning.
j.1744-6570.2009.01154.x Organizational Research Methods, 1, 131–163. doi:10.1177/
Gardner, T. M. (2002). In the trenches at the talent wars: Competitive 109442819812001
 736 BANGERTER, ROULIN, AND KÖNIG

James, L. R., McIntyre, M. D., Glisson, C. A., Green, P. D., Patton, T. W., Measurement issues associated with conditional reasoning tests: Indirect
LeBreton, J. M., . . . Williams, L. J. (2005). A conditional reasoning measurement and test faking. Journal of Applied Psychology, 92, 1–16.
measure for aggression. Organizational Research Methods, 8, 69 –99. doi:10.1037/0021-9010.92.1.1
doi:10.1177/1094428104272182 Leonard, B. (2009, November 20). Fake job reference services add new
Janz, T. (1982). Initial comparisons of patterned behavior description wrinkle to screening. HR News. Retrieved from http://www.shrm.org/
interviews versus unstructured interviews. Journal of Applied Psychol- Publications/HRNews/Pages/FakeJobReferences.aspx
ogy, 67, 577–580. doi:10.1037/0021-9010.67.5.577 Lester, S. W., Mencl, J., Maranto, C., Bourne, K. A., & Keaveny, T.
Jensen, A. R. (1998). The g factor: The science of mental ability. Westport, (2010). The impact of passing the professional in human resources exam
CT: Praeger. on early career success for undergraduates entering the human resource
Kador, J. (2006). The manager’s book of questions: 1001 great interview field. International Journal of Selection and Assessment, 18, 282–290.
questions for hiring the right person. New York, NY: McGraw-Hill. doi:10.1111/j.1468-2389.2010.00511.x
Kanter, A. B. (1995). The essential book of interviewing: Everything you Levashina, J., & Campion, M. A. (2006). A model of faking likelihood in
need to know from both sides of the table. New York, NY: Times Books. the employment interview. International Journal of Selection and As-
Karl, K., Peluchette, J., & Schlaegel, C. (2010). Who’s posting Facebook sessment, 14, 299 –316. doi:10.1111/j.1468-2389.2006.00353.x
faux pas? A cross-cultural examination of personality differences. In- Levashina, J., & Campion, M. A. (2007). Measuring faking in the employ-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ternational Journal of Selection and Assessment, 18, 174 –186. doi: ment interview: Development and validation of an interview faking
10.1111/j.1468-2389.2010.00499.x behavior scale. Journal of Applied Psychology, 92, 1638 –1656. doi:
Kirkwood, W. G., & Ralston, S. M. (1999). Inviting meaningful applicant 10.1037/0021-9010.92.6.1638
performances in employment interviews. Journal of Business Commu- Levashina, J., & Campion, M. (2009). Expected practices in background
nication, 36, 55–76. doi:10.1177/002194369903600103 checking: Review of the human resource management literature. Em-
Klehe, U.-C. (2004). Choosing how to choose: Institutional pressures ployee Responsibilities and Rights Journal, 21, 231–249. doi:10.1007/
affecting the adoption of personnel selection procedures. International s10672-009-9111-9
Journal of Selection and Assessment, 12, 327–342. doi:10.1111/j.0965- Lievens, F., & Burke, E. (2011). Dealing with the threats inherent in
075X.2004.00288.x unproctored Internet testing of cognitive ability: Results from a large-
Kleinke, C. L. (1978). Perceived approbation in short, medium, and long scale operational test program. Journal of Occupational and Organiza-
letters of recommendation. Perceptual and Motor Skills, 46, 119 –122. tional Psychology. Advance online publication. doi:10.1348/
doi:10.2466/pms.1978.46.1.119 096317910X522672
Kleinmann, M. (1993). Are rating dimensions in assessment centers trans- Lievens, F., van Dam, K., & Anderson, N. (2002). Recent trends and
parent for participants? Consequences for criterion and construct valid- challenges in personnel selection. Personnel Review, 31, 580 – 601.
ity. Journal of Applied Psychology, 78, 988 –993. doi:10.1037/0021- doi:10.1108/00483480210438771
9010.78.6.988 Lizzeri, A. (1999). Information revelation and certification intermediaries.
Kleinmann, M., Ingold, P. V., Lievens, F., Jansen, A., Melchers, K. G., & RAND Journal of Economics, 30, 214 –231. doi:10.2307/2556078
König, C. J. (2011). A different look at why selection procedures work: March, J. G., & Simon, H. A. (1958). Organizations. New York, NY:
The role of candidates’ ability to identify criteria. Organizational Psy- Wiley.
chology Review, 1, 128 –146. doi:10.1177/2041386610387000 Marcus, B. (2006). Relationships between faking, validity, and decision
König, C. J., Melchers, K. G., Kleinmann, M., Richter, G. M., & Klehe, criteria in personnel selection. Psychology Science, 48, 226 –246.
U.-C. (2006). The relationship between the ability to identify evaluation Martin, P. D., & Pope, J. (2008). Competency-based interviewing: Has it
criteria and integrity test scores. Psychology Science, 48, 369 –377. gone too far? Industrial and Commercial Training, 40, 81– 86. doi:
König, C. J., Melchers, K. G., Kleinmann, M., Richter, G. M., & Klehe, 10.1108/00197850810858910
U.-C. (2007). Candidates’ ability to identify criteria in nontransparent McFarland, L. A., & Ryan, A. M. (2000). Variance in faking across
selection procedures: Evidence from an assessment center and a struc- noncognitive measures. Journal of Applied Psychology, 85, 812– 821.
tured interview. International Journal of Selection and Assessment, 15, doi:10.1037/0021-9010.85.5.812
283–292. doi:10.1111/j.1468-2389.2007.00388.x Mehrabian, A. (1965). Communication length as an index of communicator
Krebs, J. R., & Dawkins, R. (1984). Animal signals: Mind-reading and attitude. Psychological Reports, 17, 519 –522. doi:10.2466/pr0.1965
manipulation. In J. R. Krebs & N. B. Davies (Eds.), Behavioural .17.2.519
ecology: An evolutionary approach (pp. 380 – 402). Sunderland, MA: Michaels, E., Handfield-Jones, H., & Axelrod, B. (2001). The war for
Sinauer. talent. Boston, MA: Harvard Business School Press.
Kristof-Brown, A. L. (2000). Perceived applicant fit: Distinguishing be- Miller, G. (2000). The mating mind: How sexual choice shaped the
tween recruiters’ perceptions of person–job and person– organization fit. evolution of human nature. New York, NY: Anchor.
Personnel Psychology, 53, 643– 671. doi:10.1111/j.1744-6570.2000 Moreau, M.-P., & Leathwood, C. (2006). Graduates’ employment and the
.tb00217.x discourse of employability: A critical analysis. Journal of Education and
Kristof-Brown, A. L., Barrick, M. R., & Franke, M. (2002). Applicant Work, 19, 305–324. doi:10.1080/13639080600867083
impression management: Dispositional influences and consequences for Morgeson, F. P., Campion, M. A., Dipboye, R. L., Hollenbeck, J. R.,
recruiter perceptions of fit and similarity. Journal of Management, 28, Murphy, K. R., & Schmitt, N. (2007). Reconsidering the use of person-
27– 46. ality tests in personnel selection contexts. Personnel Psychology, 60,
Kulik, J. A., Bangert-Drowns, R. L., & Kulik, C. C. (1984). Effectiveness 683–729. doi:10.1111/j.1744-6570.2007.00089.x
of coaching for aptitude tests. Psychological Bulletin, 95, 179 –188. Morgeson, F. P., & Hofmann, D. A. (1999). The structure and function of
doi:10.1037/0033-2909.95.2.179 collective constructs: Implications for multilevel research and theory
Larkan, K. (2007). The talent war. Singapore: Cavendish Business. development. Academy of Management Review, 24, 249 –265.
Lazear, E. P., & Oyer, P. (in press). Personnel economics. In R. Gibbons Morris, S. (2007, October 18). A stand-out CV: So much to do, so little
& D. J. Roberts (Eds.), Handbook of organizational economics. Prince- time. The Independent.
ton, NJ: Princeton University Press. Retrieved from http://faculty- Moser, K., & Rhyssen, D. (2001). Referenzen als eignungsdiagnostische
gsb.stanford.edu/oyer/wp/handbook.pdf Methode [References as a fit diagnostic method]. Zeitschrift für Arbeits-
LeBreton, J. M., Barksdale, C. D., Robin, J., & James, L. R. (2007). und Organisationspsychologie, 45, 40 – 46.
 PERSONNEL SELECTION AS A SIGNALING GAME 737

Murphy, K. R. (1986). When your top choice turns you down: Effect of preview experiments. Journal of Applied Psychology, 70, 706 –719.
rejected offers on the utility of selection tests. Psychological Bulletin, doi:10.1037/0021-9010.70.4.706
99, 133–138. doi:10.1037/0033-2909.99.1.133 Ralston, S. M., & Kirkwood, W. G. (1999). The trouble with applicant
Nemanick, R. C. J., & Clark, E. M. (2002). The differential effects of impression management. Journal of Business and Technical Communi-
extracurricular activities on attributions in resumé evaluation. Interna- cation, 13, 190 –207. doi:10.1177/1050651999013002004
tional Journal of Selection and Assessment, 10, 206 –217. doi:10.1111/ Reilly, R. R., & Chao, G. T. (1982). Validity and fairness of some
1468-2389.00210 alternative employee selection procedures. Personnel Psychology, 35,
Ng, E. S. W., & Burke, R. J. (2005). Person– organization fit and the war 1– 62. doi:10.1111/j.1744-6570.1982.tb02184.x
for talent: Does diversity management make a difference? International Resto, C., Ybarra, I., & Sethi, R. (2007). Recruit or die: How any business
Journal of Human Resource Management, 16, 1195–1210. doi:10.1080/ can beat the big guys in the war for young talents. New York, NY:
09585190500144038 Penguin.
Nicholson, N., & White, R. (2006). Darwinism: A new paradigm for Robie, C., Emmons, T., Tuzinski, K. A., & Kantrowitz, T. (2011). Effects
organizational behavior? Journal of Organizational Behavior, 27, 111– of an economic recession on leader personality and general mental
119. doi:10.1002/job.345 ability scores. International Journal of Selection and Assessment, 19,
Nicklin, J. M., & Roch, S. G. (2009). Letters of recommendation: Contro- 183–189. doi:10.1111/j.1468-2389.2011.00545.x
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

versy and consensus from expert perspectives. International Journal of Rosenfeld, P. (1997). Impression management, fairness, and the employ-
Selection and Assessment, 17, 76 –91. doi:10.1111/j.1468-2389 ment interview. Journal of Business Ethics, 16, 801– 808. doi:10.1023/
.2009.00453.x A:1017972627516
Ones, D. S., Dilchert, S., Viswesvaran, C., & Judge, T. A. (2007). In Roth, P. L., Bobko, P., & McFarland, L. A. (2005). A meta-analysis of
support of personality assessment in organizational settings. Personnel work sample test validity: Updating and integrating some classic liter-
Psychology, 60, 995–1027. doi:10.1111/j.1744-6570.2007.00099.x ature. Personnel Psychology, 58, 1009 –1037. doi:10.1111/j.1744-
Ones, D. S., & Viswesvaran, C. (1998). The effects of social desirability 6570.2005.00714.x
and faking on personality and integrity assessment for personnel selec- Ruch, F. L. (1942). A technique for detecting attempts to fake performance
tion. Human Performance, 11, 245–269. on a self-inventory type of personality test. In Q. McNemar & M. A.
Paetzold, R. L., & Wilborn, S. L. (1992). Employer (ir)rationality and the Merrill (Eds.), Studies in personality (pp. 229 –234). New York, NY:
demise of employment references. American Business Law Journal, 30, McGraw-Hill.
123–142. doi:10.1111/j.1744-1714.1992.tb00657.x Ryan, A. M., McFarland, L., Baron, H., & Page, R. (1999). An interna-
Palmer, D. K., Campion, M. A., & Green, P. C. (1999). Interviewing tional look at selection practices: Nation and culture as explanations for
training for both applicant and interviewer. In R. W. Eder & M. M. variability in practice. Personnel Psychology, 52, 359 –392. doi:10.1111/
Harris (Eds.), The employment interview handbook (pp. 337–351). j.1744-6570.1999.tb00165.x
Thousand Oaks, CA: Sage. Ryan, A. M., & Ployhart, R. E. (2000). Applicants’ perceptions of selection
Pearce, J. L. (2000). Employability as trustworthiness. In C. R. Leana & procedures and decisions: A critical review and agenda for the
D. M. Rousseau (Eds.), Relational wealth: The advantages of stability in future. Journal of Management, 26, 565– 606. doi:10.1177/
a changing economy (pp. 79 –90). Oxford, England: Oxford University 014920630002600308
Press. Ryan, A. M., Ployhart, R. E., Greguras, G. J., & Schmit, M. J. (1998). Test
Peres, S. H., & Garcia, J. R. (1962). Validity and dimensions of descriptive preparation programs in selection contexts: Self-selection and program
adjectives used in reference letters for engineering applicants. Personnel effectiveness. Personnel Psychology, 51, 599 – 621. doi:10.1111/j.1744-
Psychology, 15, 279 –286. doi:10.1111/j.1744-6570.1962.tb01622.x 6570.1998.tb00253.x
Petrie, M. (1994, October 13). Improved growth and survival of offspring Ryan, A. M., Sacco, J. M., McFarland, L. A., & Kriska, S. D. (2000).
of peacocks with more elaborate trains. Nature, 371, 598 –599. doi: Applicant self-selection: Correlates of withdrawal from a multiple hur-
10.1038/371598a0 dle process. Journal of Applied Psychology, 85, 163–179. doi:10.1037/
Petrie, M., & Halliday, T. (1994). Experimental and natural changes in the 0021-9010.85.2.163
peacock’s (Pavo cristatus) train can affect mating success. Behavioral Rynes, S. L. (1991). Recruitment, job choice, and post-hire consequences:
Ecology and Sociobiology, 35, 213–217. doi:10.1007/BF00167962 A call for new research directions. In M. D. Dunnete & L. M. Hough
Petrie, M., Halliday, T., & Sanders, C. (1991). Peahens prefer peacocks (Eds.), Handbook of industrial and organization psychology (2nd ed.,
with elaborate trains. Animal Behaviour, 41, 323–331. doi:10.1016/ pp. 399 – 444). Palo Alto, CA: Consulting Psychologists Press.
S0003-3472(05)80484-1 Rynes, S. L., Bretz, R. D. J., & Gerhart, B. (1991). The importance of
Pfeffer, J. (2001). Fighting the war for talent is hazardous to your organi- recruitment in job choice: A different way of looking. Personnel Psy-
zation’s health. Organizational Dynamics, 29, 248 –259. doi:10.1016/ chology, 44, 487–521. doi:10.1111/j.1744-6570.1991.tb02402.x
S0090-2616(01)00031-6 Sackett, P. R., & Lievens, F. (2008). Personnel selection. Annual Review of
Phillips, J. M. (1998). Effects of realistic job previews on multiple orga- Psychology, 59, 419 – 450. doi:10.1146/annurev.psych.59.103006
nizational outcomes: A meta-analysis. Academy of Management Jour- .093716
nal, 41, 673– 690. doi:10.2307/256964 Salgado, J. F., & Moscoso, S. (2002). Comprehensive meta-analysis of the
Porter, L. W., Lawler, E. E., & Hackman, J. R. (1975). Behavior in construct validity of the employment interview. European Journal of
organizations. New York, NY: McGraw-Hill. Work and Organizational Psychology, 11, 299 –324. doi:10.1080/
Poundstone, W. (1993). Prisoner’s dilemma: John von Neumann, game 13594320244000184
theory, and the puzzle of the bomb. New York, NY: Anchor Books. Sanders, K., & Frenkel, S. (2011). HR–line management relations: Char-
Poundstone, W. (2003). How would you move Mount Fuji? Microsoft’s cult acteristics and effects. International Journal of Human Resource Man-
of the puzzle: How the smartest companies select the most creative agement, 22, 1611–1617. doi:10.1080/09585192.2011.565644
thinkers. New York, NY: Little Brown. Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection
Powers, D. E. (1993). Coaching for the SAT: A summary of the summaries methods in personnel psychology: Practical and theoretical implications
and an update. Educational Measurement: Issues and Practice, 12, of 85 years of research findings. Psychological Bulletin, 124, 262–274.
24 –30. doi:10.1111/j.1745-3992.1993.tb00530.x doi:10.1037/0033-2909.124.2.262
Premack, S. L., & Wanous, J. P. (1985). A meta-analysis of realistic job Schmidt, F. L., Hunter, J. E., & Outerbridge, A. N. (1986). Impact of job
 738 BANGERTER, ROULIN, AND KÖNIG

experience and ability on job knowledge, work sample performance, and Trank, C. Q., Rynes, S. L., & Bretz, R. D. J. (2002). Attracting applicants
supervisory ratings of job performance. Journal of Applied Psychology, in the War for Talent: Differences in work preferences among high
71, 432– 439. doi:10.1037/0021-9010.71.3.432 achievers. Journal of Business and Psychology, 16, 331–345. doi:
Schnipke, D., & Scrams, D. (1999). Representing response-time informa- 10.1023/A:1012887605708
tion in item banks (Computerized Testing Rep. No. 97– 09). Newtown, Tsai, W.-C., Huang, T.-C., Wu, C.-Y., & Lo, I.-H. (2010). Disentangling
PA: Law School Admissions Council. the effects of applicant defensive impression management tactics in job
Searcy, W. A., & Nowicki, S. (2005). The evolution of animal communi- interviews. International Journal of Selection and Assessment, 18, 131–
cation: Reliability and deception in signaling systems. Princeton, NJ: 140. doi:10.1111/j.1468-2389.2010.00495.x
Princeton University Press. Turban, D. B., & Greening, D. W. (1997). Corporate social performance
Simon, H. A. (1947). Administrative behavior: A study of decision-making and organizational attractiveness to prospective employees. Academy of
processes in administrative organizations. New York, NY: Free Press. Management Journal, 40, 658 – 672. doi:10.2307/257057
Smither, J. W., Reilly, R. R., Millsap, R. E., Pearlman, K., & Stoffey, U.S. Department of Education. (2011). Diploma mills and accreditation.
R. W. (1993). Applicant reactions to selection procedures. Personnel Retrieved from http://www2.ed.gov/students/prep/college/diplomamills/
Psychology, 46, 49 –76. doi:10.1111/j.1744-6570.1993.tb00867.x diploma-mills.html
Spence, M. (1973). Job market signaling. Quarterly Journal of Economics, Van Der Heijde, C. M., & Van Der Heijden, B. I. J. M. (2006). A
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

87, 355–373. doi:10.2307/1882010

This document is copyrighted by the American Psychological Association or one of its allied publishers.

competence-based and multidimensional operationalization and mea-

Spence, M. (2002). Signaling in retrospect and the informational structure surement of employability. Human Resource Management, 45, 449 –
of markets. American Economic Review, 92, 434 – 459. doi:10.1257/ 476. doi:10.1002/hrm.20119
00028280260136200 van Hooft, E. A. J., & Born, M. P. (2011). Intentional response distortion
Starr, M. A. (2008). Socially responsible investment and pro-social change. on personality tests: Using eye-tracking to understand response pro-
Journal of Economic Issues, 62, 52–73. cesses when faking. Journal of Applied Psychology. Advance online
Stevens, C. K., & Kristof, A. L. (1995). Making the right impression: A publication. doi:10.1037/a0025711
field study of applicant impression management during job interviews. Van Iddekinge, C. H., Roth, P. L., Raymark, P. H., & Odle-Dusseau, H. N.
Journal of Applied Psychology, 80, 587– 606. doi:10.1037/0021- (2011). The criterion-related validity of integrity tests: An updated
9010.80.5.587
meta-analysis. Journal of Applied Psychology. Advance online publica-
Stewart, G. L., Darnold, T. C., Zimmerman, R. D., Parks, L., & Dustin,
tion. doi:10.1037/a0021196
S. L. (2010). Exploring how response distortion of personality measures
Van Valen, L. (1973). A new evolutionary law. Evolutionary Theory, 1,
affects individuals. Personality and Individual Differences, 49, 622–
1–30.
628. doi:10.1016/j.paid.2010.05.035
Veblen, T. (1899). The theory of the leisure class. New York, NY:
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and
MacMillan.
strategic management. Strategic Management Journal, 18, 509 –533.
Vermeij, G. J. (1994). The evolutionary interaction among species: Selec-
doi:10.1002/(SICI)1097-0266(199708)18:7⬍509::AID-SMJ882⬎
tion, escalation, and coevolution. Annual Review of Ecology and Sys-
3.0.CO;2-Z
tematics, 25, 219 –236. doi:10.1146/annurev.es.25.110194.001251
Tennie, C., Frith, U., & Frith, C. D. (2010). Reputation management in the
Walther, F. R. (1969). Flight behaviour and avoidance of predators in
age of the world-wide web. Trends in Cognitive Sciences, 14, 482– 488.
Thomson’s gazelle (Gazella Thomsoni Guenther 1884). Behaviour, 34,
doi:10.1016/j.tics.2010.07.003
184 –220. doi:10.1163/156853969X00053
Terlaak, A., & King, A. (2002, April). Contrasting institutional theory and
signaling theory to explain the adoption process of an industry standard. Wanous, J. P. (1973). Effects of a realistic job preview on job acceptance,
Paper presented at the second annual Wharton Technology Miniconfer- job attitudes, and job survival. Journal of Applied Psychology, 58,
ence, University of Pennsylvania, Philadelphia. 327–332. doi:10.1037/h0036305
Terlaak, A., & King, A. A. (2010). Contrasting institutional theory and Wanous, J. P. (1978). Realistic job previews: Can a procedure to reduce
signaling theory to explain the adoption process of an industry standard. turnover also influence the relationship between abilities and perfor-
Manuscript submitted for publication. mance? Personnel Psychology, 31, 249 –258. doi:10.1111/j.1744-
Thoms, P., McMasters, R., Roberts, M. R., & Dombkowsky, D. A. (1999). 6570.1978.tb00444.x
Resumé characteristics as predictors of an invitation to interview. Jour- Wanous, J. P. (1980). Organizational entry: Recruitment, selection, orien-
nal of Business and Psychology, 13, 339 –356. doi:10.1023/A: tation and socialization of newcomers. Reading, MA: Addison-Wesley.
1022974232557 Winston, G. C. (2004). Differentiation among U.S. colleges and universi-
Thornton, R. J. (2003). The lexicon of intentionally ambiguous recommen- ties. Review of Industrial Organization, 24, 331–354. doi:10.1023/B:
dations. Naperville, IL: Sourcebooks. REIO.0000037539.78539.02
Tinbergen, N. (1952). Derived activities: Their causation, biological sig- Zahavi, A. (1975). Mate selection: A selection for a handicap. Journal of
nificance, origin and emancipation during evolution. Quarterly Review Theoretical Biology, 53, 205–214. doi:10.1016/0022-5193(75)90111-3
of Biology, 27, 1–32. doi:10.1086/398642 Zahavi, A., & Zahavi, A. (1999). The handicap principle: A missing piece
Tomlinson, M. (2007). Graduate employability and student attitudes and of Darwin’s puzzle. Oxford, England: Oxford University Press.
orientations to the labour market. Journal of Education and Work, 20,
285–304. doi:10.1080/13639080701650164
Tomlinson, M. (2008). Students’ perceptions of the role of higher educa- Received June 8, 2010
tion credentials for graduate work and employability. British Journal of Revision received July 5, 2011
Sociology of Education, 29, 49 – 61. doi:10.1080/01425690701737457 Accepted September 16, 2011 䡲