 

LIBRARY
Michigan State
University

‘—

This is to certify that the
thesis entitled

The Inﬂuence of Culture on Applicant Justice Perceptions in
Response to Administrative Consistency and Interpersonal
Sensitivity

presented by

Smriti Shivpuri

has been accepted towards fulﬁllment
of the requirements for the

Master of degree in I ndustrial/Organizational
Arts Psychology

 

 

////////m/
' Major/gmfessor’s Signature
4/ I [j 05

Date

MSU is an Aﬁinnative Action/Equal Opportunity Institution

—. ... _.— _--—.-— —

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/CIRC/DateDue indd

 

THE INFLUENCE OF CULTURE ON APPLICANT JUSTICE PERCEPTIONS IN
RESPONSE To ADMINISTRATIVE CONSISTENCY AND INTERPERSONAL
SENSITIVITY
By

Smriti Shivpuri

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS
Department of Industrial/Organizational Psychology

2005

ABSTRACT
THE INFLUENCE OF CULTURE ON APPLICANT JUSTICE PERCEPTIONS IN
RESPONSE TO ADMINISTRATIVE CONSISTENCY AND INT ERPERSONAL
SENSITIVITY

By
Smriti Shivpuri

A ﬁ‘amework for modeling applicant fairness perceptions during the selection
process has been established by Gilliland (1993); however, little work has examined how
individual differences affect reactions to the violation and satisfaction of procedural rules
presented in this framework. In particular, how applicants’ cultural orientations inﬂuence
their reactions to procedural rules has been largely ignored. This study examined
whether applicants’ cultural orientations, speciﬁcally, their individualism (IND) and
collectivism (COL) orientations serve as moderators of the relationship between the
violation/satisfaction of procedural rules and subsequent fairness perceptions. Three
procedural rules, namely, consistency of administration, interpersonal sensitivity of
administrator, and explanation provision were manipulated in a simulated application
context. Results indicated that IND or COL orientatidn moderated the relationship
between some-procedural rules and process fairness as well as other outcomes. Results
suggest that organizations should consider the cultural orientations of their applicant pool

when developing and structuring selection processes.

TABLE OF CONTENTS

LIST OF TABLES ..................................................................... v
LIST OF FIGURES ........................ . ............................................ vii
INTRODUCTION ..................................................................... 1
Applicant Reactions ................................................................ 5
Basic Background ............................................................... 5
F acund Findings ................................................................. 9
Momentous Model .............................................................. 11
Relevant Rule ................................................................... 14
Explanations ......................................................................... 18
Exploring Explanations ........................................................ 18
Speciﬁcally Selection .......................................................... 25
Strategic Summary ................................................................. 31
Individualism and Collectivism ................................................... 39
IND/COL and Justice .......................................................... 51
Hypotheses .......................................................................... 56
METHOD ............................................................................... 71
Sample ............................................................................... 71
Measures ............................................................................. 7 1
Procedure ............................................................................ 77
Manipulations ................................................................... 79
Debrieﬁng ........................................................................ 82
Pilot Testing ......................................................................... 83
Manipulation Checks ............................................................... 90
Hypothesis Testing ................................................................. 91
DISCUSSION ........................................................................... 1 17
Consistency Effects ................................................................. 1 l7
Sensitivity Effects ................................................................... 123
Explanation Effects ................................................................ 126
Implications .........................................................................
LIMITATIONS AND FUTURE RESEARCH ..................................... 137

iii

Context and Manipulation Issues ................................................. 140

IND and COL Measure Issues .................................................... 144

Issues Involving Individual Level IND and COL .............................. 148

Applicant Reactions Cross-Culturally ........................................... 149
CONCLUSION ......................................................................... 1 5 1
APPENDD( A: Measures ............................................................... 152
APPENDDII B: Informed Consent Form ............................................. 156
APPENDD( C: Protocol for Study ................................................... 158
APPENDD( D: Survey One ........................................................... 163
APPENDD( E: Survey Two ......................................................... 168
APPENDIX F: Debrieﬁng Form .................................................... 175
REFERENCES ......................................................................... 177

iv

LIST OF TABLES

Table 1. Scale Descriptives, Reliabilities, and Intercorrelations .................
Table 2. Correlation Between Demographics and Scale Scores .................

Table 3. Mean Differences on Manipulation Checks ..............................

Table 4. Moderated Regressions of Procedural Justice Perceptions on
Consistency, Collectivism, and Individualism .....................................

Table 5. Moderated Regressions of Distributive Justice Perceptions on
Consistency, Collectivism, and Individualism .....................................

Table 6. Moderated Regressions of Intentions towards the Organization on
Consistency, Collectivism, and Individualism .....................................

Table 7. Moderated Regressions of Procedural Justice Perceptions on
Sensitivity, Collectivism, and Individualism .......................................

Table 8. Moderated Regressions of Distributive Justice Perceptions on
Sensitivity, Collectivism, and Individualism .......................................

Table 9. Moderated Regressions of Intentions towards the Organization on
Sensitivity, Collectivism and Individualism ........................................

Table 10. Test of Mediation of Explanation Adequacy ...........................

Table 11. Moderated Regressions of Procedural Justice Perceptions on
Explanation, Collectivism, and Individualism ....................................

Table 12. Moderated Regressions of Distributive Justice Perceptions on
Explanation, Collectivism, and Individualism .....................................

Table 13. Moderated Regressions of Intentions towards the Organization on
Explanation, Collectivism, and Individualism .....................................

Table 14. Summary of Results for Hypotheses ....................................

Table 15. Moderated Regressions of Procedural Justice Perceptions Pre on
Consistency, Sensitivity, Explanation, Collectivism, and Individualism .......

86

88

91

99

103

98

101

104

105

109

111

112

113

118

135

Table 16. Moderated Regressions of Procedural Justice Perceptions Post on
Consistency, Sensitivity, Explanation, Collectivism, and Individualism ....... 136

LIST OF FIGURES

Figure 1. Model of the Interaction between IND/COL Orientation and
Consistency, Interpersonal Sensitivity, and Explanation Provision on
Procedural Fairness Perceptions ......................................................

Figure 2. Interaction Between Consistency and Individualism on Procedural
Justice Post ...........................................................................

Figure 3. Interaction Between Consistency and Collectivism on Distributive
Justice ....................................................................................

Figure 4. Interaction Between Sensitivity and Collectivism on Procedural
Justice Post ..............................................................................

Figure 5. Interaction Between Sensitivity and Collectivism on Intentions
Post .......................................................................................

Figure 6. Interaction Between Explanation and Individualism on Intentions
Pre ........................................................................................

Figure 7. Interaction Between Explanation and Individualism on Intentions
Post .......................................................................................

vii

57

94

99

102

106

114

114

INTRODUCTION
“No man has the right to dictate what other men should perceive, create

or produce, but all should be encouraged to reveal themselves, their
perceptions and emotions... ”

- Ansel Adams (1902-1984)

In the past decade, applicant reactions during and after the process of selection
have received greater and more focused attention. There are many factors that contribute
to this increased interest in reactions. Until recently, low unemployment rates and an
increasing demand for labor had resulted in greater competition between organizations
for qualiﬁed applicants (Greenstein, Jaffe, & Kayatin, 1999). This increased degree of
competition led organizations to reevaluate their systems of selection in order to
determine what aspects could most positively inﬂuence applicants’ attraction to the
organization (Ryan & Ployhart, 2000). In addition, many organizations have recently
adopted a total quality focus of management, in which applicants are viewed as
“customers” of a selection process because of their potential to be users and beneﬁciaries
of this process (Gilliland & Cherry, 2000). As a result, these organizations are
attempting to determine applicant preferences in terms of selection tools and processes, in
order to attract and satisfy more applicants.

Because many outcomes relevant to organizations have been shown to be related
to applicant reactions to selection systems, organizations’ concerns are not without merit.
Applicant attitudes toward the selection system, such as perceived fairness, have been
shown in numerous studies as well as a recent meta-analysis to be related to applicant
attitudes towards the organization, such as organizational attractiveness, intentions to

recommend the organization to others, intentions to use the organization’s products, and

job acceptance intentions (Fiaschetti, 1998; Gilliland, 1994; Hausknecht, Day, &
Thomas, 2004; Macan, Avedon, Paese, & Smith, 1994; Truxillo & Bauer, 1999).

Because of how important many of these outcomes are to an organization, it is no
surprise that researchers have begun looking into what factors affect applicant reactions
in a selection context. Gilliland (1993), gaining inspiration ﬁom the organizational
justice literature, has created a model of applicant reactions in response to fairness
perceptions in a selection context. This includes both reactions to the fairness of the
selection process by 'which the hiring decision was made (procedural fairness), and the
fairness of the actual hiring decision (distributive fairness). Gilliland proposes that there
are certain procedural rules that are present during a selection process, and that the
violation of these rules leads to perceptions of procedural unfairness and subsequent
negative outcomes, such as a decreased likelihood of job acceptance intentions.
However, although much research has focused on the relationship between the procedural
rules and applicant perceptions of procedural fairness (Bauer, Truxillo, Sanchez & Craig,
2001; Gilliland, 1994; Macan et al., 1994; Truxillo & Bauer, 1999; Fiaschetti, 1998),
very little has looked at how individual dzﬂ’erences may affect this relationship.

An alternative approach in determining what factors affect applicant reactions
during selection has been to look at the use of explanations in the selection context.
Much research in the organizational justice literature has shown that the use of
explanations can mitigate the negative affects of an undesirable outcome on the receiver
of that outcome (Colquitt & Chertkoff, 2001; Greenberg, 1993, 1994; Shapiro, Buttner, &
Barry, 1994; Shaw, Wild, & Colquitt, 2003). It is inevitable in a selection context that

some applicants will receive undesirable outcomes, because it is very rare that every

applicant is offered a job. In fact, studies have shown that providing an explanation for
rejecting an applicant can decrease the applicant’s perceptions of unfairness (Gilliland,
1994; Gilliland, Groth, & Baker, 2001; Horvath, Ryan, & Stierwalt, 2000; Ployhart,
(Ryan, & Bennett, 1999). .However, applicants not only react to what decision was made,
but how the decision was made. Therefore, though many studies have looked at how
explanations for the hiring decision affect fairness perceptions, few studies have
examined how explanations for the procedure used in hiring affect perceptions of fairness
(for exceptions see Gilliland, 1994, and Horvath et al., 2000). In addition, even fewer
studies have explored how individual diﬁ'erences may affect this relationship.

This paper extends previous research by taking both these approaches and
applying a new lens through which one can view both these perspectives: that lens is
culture. The assumption that both these approaches are making is that everyone will
react to a stimulus, be it the violation of a procedural rule or the explanation of why a
procedure was used, in very similar ways. These approaches both do not take into
account the fact that applicants come ﬁom different backgrounds and have different
experiences, and that this can cause them to react to the same stimulus differently.

The reason why culture is so important to consider, especially in an organizational
selection context, is that we are rapidly moving toward becoming a global community.
Multi-national corporations are becoming more common, and companies are expanding
their markets and adding overseas branches daily (Skarlicki, 2001). In addition, the
minority and immigrant population in the United States is continually growing
(Schmidley & Gibson, 1999). All these factors result in long-term, increased contact

between people of different cultures in an organizational setting. Therefore, in the same

way that it is important for American companies to understand how American applicants
will react to their particular selection systems (i.e., because it affects signiﬁcant outcomes
such as recommendation intentions, acceptance intentions, and re—application intentions),
it is also important for multinational companies to understand how their applicants of
different cultures will react, and if and why they will react differently. In fact, it is
important for any company that has a diverse pool of applicants to understand this.

This paper is an attempt to begin to answer some of these questions. In this paper,
I will examine how cultural orientations, in particular the cultural dimensions of
Individual and Collectivism, affect the relationship between the violation of certain
procedural rules and fairness during the selection process and how it alters (if at all) how
explanations for a selection process affect fairness perceptions of that process. The paper
will open with a general review of the literature on applicant reactions, and then will
focus on certain procedural elements of Gilliland’s (1993.) model of applicant reactions to
selection systems. It will then move on to provide background on literature pertaining to
explanations, and their effect on fairness perceptions in the organizational justice
literature. More speciﬁc details will be provided on ﬁndings relevant to explanations
utilized in a selection context. The cultural dimensions of individualism and collectivism
(IND and COL) will then be presented, and literature linking these dimensions to fairness
perceptions and the selection context will be discussed. After the general background of
- these concepts has been relayed, speciﬁc hypotheses pertaining to the relationship
between IND and COL, procedural rules, explanations, and applicant reactions, and the
rationale for these hypotheses will be put forth. Finally, the method by which these

hypotheses will be tested will be presented.

APPLICANT REACTIONS
WM

One of the ﬁrst records of attention to applicant reactions to selection systems in
i the psychological literature occurred during the late 1930’s. However, at this time,
applicants’ emotional reactions were not being considered for the purpose of changing
selection systems to elicit more favorable reactions, but instead, reactions were being
used as selection criteria to determine if the applicant was ﬁt for the job (National
Institute of Industrial Psychology [NIIP], 193 8). For example, if an applicant responded
in a cool manner to a highly stressful selection situation, then he/she might have been
considered as fulﬁlling one of the criteria (i.e. having a calm temperament) in order to
become an airline pilot. Towards the late 1940’s and early 1950’s researchers became
more concerned about the “face validity” of tests, and mention is made about whether
tests were well-received, and whether tests made examinees tense (Fields, 1950; Mosier,
1947; Peterson, 1949). In the 1960’s, a book, The Brainwatchers. increased interest in
applicant reactions by addressing the personnel selection industry’s use of personality
tests and describing how applicants reacted to these types of tests by responding in
socially desirable ways and faking answers (Gross, 1962). Shortly after, the subject of
applicants’ reactions to employment tests (in particular, personality tests) in terms of
invasion of privacy came to the foreground in a congressional hearing. In 1965, the
House Special Subcommittee on Constitutional Rights and the House Special
Subcommittee on invasion of privacy, held a hearing to determine whether pre-

employment psychological tests were an unjustiﬁed invasion of applicants’ privacy; this

1.12

was as a result of negative reactions ﬁom applicants and the media concerning some of
the questions on the tests (U .S. Congress, 1990).

Although, as can be seen, the topic of applicant reactions to selection systems was
a concern at certain times in the past, it has only been in the past 20 years that interest and
research has really burgeoned in this area. Starting in the late 1980’s and early 1990’s
researchers began to realize that most of the research on selection had been viewing it
from the organization’s perspective, and completely neglecting the fact that applicants’
attitudes and perceptions were involved in the process as well (Martin & Nagao, 1989;
Rynes, 1993; Singer, 1990). Consequently, a call went out for elaboration and
advancement of research and thought in this area, and the result came in the form of
theoretical exploration and empirical inquiry into applicant reactions.

Though a few empirical studies had been done on the topic of applicant reactions
to speciﬁc selection tools before (e. g. Cunningham, 1989; Lounsbury, Bobrow, & Jensen,
1989; Martin & Nagao, 1989; Ryan & Sackett, 1987), the study that was integral in
sparking interest in applicant reactions was conducted by Arvey, Strickland, Drauden,
and Martin in 1990. This study did not look at fairness reactions, but it is included in this
review because it is a seminal study in the applicant reactions literature, and it was the
starting point for a great deal of subsequent research on the topic (Ryan & Ployhart,
2000). In the ﬁrst phase of this study, a scale, the Test Attitude Survey (TAS) was
developed to measure the attitudes and opinions of applicants towards tests they recently
took. There were nine factors developed that assessed test-takers’ motivation, lack of
concentration, belief in tests, comparative anxiety, test case, external attribution, general

need for achievement, future effects, and preparation. The survey was validated using

several studies. The second phase was designed to test various propositions and
hypotheses posited in the ﬁrst phase. One study in the second phase found that applicants
reported more positive attitudes, higher effort and motivation on tests than incumbents.
Another study had a signiﬁcant ﬁnding that test-taking attitudes are related to test
performance, and that racial differences in test scores diminished signiﬁcantly when
attitudes were controlled. Finally, there was limited evidence in a third study that found
that test motivation moderates test validity (Arvey et al., 1990).

Although Arvey et a1. (1990) were the ﬁrst to put the limelight on applicant
reactions, Singer (1990) was the ﬁrst researcher to suggest the application of a speciﬁc
theoretical perspective to applicant reactions in a selection context. Singer suggested that
applicant perceptions of fairness in selection situations be examined ﬁom an
organizational justice perspective. Organizational justice theory states that when
organizations make decisions about the allocation of outcomes, two forms of fairness, or
justice, are considered: distributive justice and procedural justice (Greenberg, 1990).
Distributive justice refers to the fairness of organizational outcome distributions, while
procedural justice refers to the fairness of procedures used to determine this distribution
of outcomes (Greenberg, 1990). Singer reviewed a large body of literature on this
perspective, and identiﬁed the Six criteria used in evaluating procedural justice that were
most thoroughly empirically validated. These six criteria, which were taken ﬁom
Leventhal’s (1980) justice judgment model, were consistency in applying rules, bias
avoidance, accuracy of information, correctability of the decision, opportunities to select
the decision-making agent, and maintenance of ethical and moral standards. Singer,

using a sample of entry-level job applicants and professionals from New Zealand,

...

identiﬁed ﬁve major determinants of perceived fairness in selection through the use of an
open-ended questionnaire and factor analysis. Singer found that the determinants in
common between Leventhal’s model and her own analysis were consistency, bias
avoidance and choice of selectors. The two criteria that both samples in Singer’s study
reported to be key determinants, but were not represented in Leventhal’s model, were
information soliciting and ethicality.

Though Singer’s (1990) study was a good starting point for investigating fairness
in selection, it was too simplistic and did not fully address how applicant perceptions of
fairness could affect future outcomes. Schmitt and Gilliland (1992) and Gilliland (1993)
also explored perceptions of fairness in selection from an organizational justice
perspective, along the same vein as Singer, but the result was a broader perspective and
more comprehensive theory that looked at not only determinants of procedural fairness
but also determinants of distributive fairness, and their relation to various outcomes.
Schmitt and Gilliland (1992) reviewed research related to applicants’ reactions to
selection while focusing on six different selection procedures: interviews, computerized
testing, integrity testing, drug testing, assessment centers and work samples.

In relation to interviews, Schmitt and Gilliland (1992) concluded that the evidence
showing a relation between interviewer behavior and interviewee intentions was mixed,
but instead of focusing on intentions, they suggested that two-way communication during
the actual intervieWing process as well as the degree of structure of the interview should
be examined, as these aspects could affect the interviewee’s fairness perceptions.
Regarding computer testing, the authors presented mixed evidence about the nature of

reactions to this type of selection tool, though they indicated that getting direct feedback

L'.‘

was a factor that contributed to positive feelings towards this process. According to the
authors, both integrity testing and drug testing required more research, because if used for
selection, fairness of such tests is a great concern. As relates to both work-sarnple tests
and assessment centers, the authors stated that the high face validity of both procedures
of selection make them more likely to elicit positive reactions from applicants.

Based on their review of different selection tools and justice literature, Schmitt
and Gilliland (1992) proposed a list of procedural issues that could affect applicant
reactions to organizations’ selection practices. The list consisted of feedback regarding
performance, knowledge of how tests are used to make decisions, standardization of test
administration, face validity of procedures, information regarding validation of
procedures, possibility of retest, knowledge of how tests are scored, personalization of
process, and privacy and conﬁdentiality of safeguards. They then composed a table of
each selection procedure and predicted whether on each issue the procedure would result
in a positive or negative reaction ﬁ'om the applicant. They discussed the types of
outcomes resulting from employee reactions to selection procedures. The authors stated
that organizational attractiveness and reputation, organizational HR practices, and
applicant’s motivation to do well on the test could be affected by perceived fairness of
the selection procedure.

F acund Findings

Around the time that theories of the relevance of applicant reactions to various
outcomes were being postulated, research had already begun to test whether or not many
of these relationships actually did exist. Smither, Reilly, Millsap, Pearlman and Stoffey

(1993) conducted a study in which they asked applicants for a wide range of civil service

positions to react to the selection tests they were given both prior to and aﬁer they were
notiﬁed of their test scores. They gave applicants questionnaires that were meant to
assess many of the outcomes proposed by Schuler (1993) including their perceptions
concerning the face validity and predictive validity of the exam, knowledge of results,
likelihood of improving their test scores, affective reaction to taking the test, and
organizational attractiveness prior to notiﬁcation of their scores, and procedural and
distributive justice and recommendation intentions after they were notiﬁed. The
researchers found that applicants ’ reactions to the selection test were positively related to
procedural and distributive justice perceptions, organizational attractiveness, and
recommendation intentions, while actual test scores were weakly related to procedural
justice, and unrelated to organizational attractiveness or willingness to recommend the
employer to others.

In addition to Smither et al. (1993), other studies also found a relation between
applicant reactions to selection and their attitudes toward the organization. Macan,
Avedon, Paese, and Smith (1994) found that applicants who perceived the selection
process favorably were also more satisﬁed with the job and the organization, and had
greater intentions of accepting the job. Gilliland (1994) found a relation between
applicants’ perceptions of fairness and their intentions to recommend the company to
others, as did Bauer, Truxillo, Sanchez, Craig, et al. (2001). Truxillo and Bauer (1999)
found positive relations between perceptions of process and outcome fairness and
attraction to the organization as well as perceived positive employee relations. In a study
done by Fiaschetti (1998), an individual’s general reaction to the selection process was

found to be a strong predictor of that individual’s intentions to accept the job offer, use or

10

buy the company’s products, and tell others about the company. These relationships
between reactions and organizational attractiveness, intention to recommend the
organization to others, and intentions to buy an organization’s products have been found
cross-culturally as well (Stinglhamber, Vandenberghe, & Brancart, 1999).

The relation of outcomes other than just organizationally-focused outcomes to
applicant reactions have been assessed. Several studies have found that reactions or
attitudes towards certain selection procedures, such as tests used for selection, can affect
an applicant’s performance on that procedure (Arvey, Strickland, Drauden, & Martin,
1990; Chan, Schmitt, DeShon, Clause, & Delbridge, 1997; Ryan, Ployhart, Greguras, &
Schmitt, 1998). The validity of a selection procedure has also been found to be related to
applicants’ reactions to that procedure (Schmitt & Ryan, 1992; Hunthausen, 2000). An
applicant’s expectations for being hired, perceived performance, test-taking self-efﬁcacy,
and general self-efﬁcacy have all been found to be related to his/her attitudes towards the
selection process (Bauer, Maertz, Dolen, & Carnpion, 1998; Chan, Schmitt, Jennings,
Clause, & Delbridge, 1998; Gilliland, 1994; Hausknecht et al., 2004; Ployhart & Ryan,
1997; Ryan, Greguras, & Ployhart, 1996).

Momentous Mary

Inasmuch as a great deal of this research established that applicant reactions did
affect many outcomes that were important to both organizations (i.e. recommendation
intentions, job acceptance intentions) and the applicants themselves (i.e. performance in
selection, self-efﬁcacy), it was no surprise that a model was developed to explicate how
characteristics of a selection process and outcome can affect applicant reactions and what

consequences can result from these reactions. Using ideas from Schuler (1993), Arvey

ll

and Sackett (1993), and Schmitt and Gilliland (1992), Gilliland ( 1993) created a model
of how situational and personal conditions inﬂuence the degree to which procedural and
distributive rules are perceived as satisﬁed or violated within selection contexts.
Perceptions of the degree to which each of the rules are satisﬁed or violated combine to
form overall perceptions of fairness of the selection system. The situational and personal
conditions that Gilliland referred to are aspects of the selection process such as test type,
HR policy, and behavior of HR personnel, and aspects of the distribution of outcomes
such as hiring decision, performance expectations, salience of discrimination and locus of
special needs. The procedural justice rules corresponding to these conditions are grouped
into three categories: formal characteristics of the selection system, explanations given
during the selection process, and interpersonal treatment.

Rules under formal characteristics are job relatedness, opportunity to perform,
reconsideration opportunity, and consistency. Rules included in explanation are feedback
(timeliness and informativeness of), selection information (justiﬁcation for decision), and
honesty. Interpersonal treatment rules are interpersonal effectiveness, two-way
communication, and propriety of questions. Gilliland stated that the weighting and
salience of these rules will depend on type of selection procedure used, whether the rule
is violated (more salient and weighted) or satisﬁed (less salient and weighted), and time
in selection process when perceived fairness is assessed.

The procedural justice rules Gilliland (1993) identiﬁed mirror the rules of
distribution in the organizational justice literature. These rules are equity, equality, and
special needs. An assessment of equity is made by comparing the outcome of the hiring

decision to the performance expectations of the applicant. Gilliland hypothesized that

12

applicants will perceive a selection decision as unfair or inequitable when they expect to
get the job and do not receive it or when they do not expect the job and do receive it.
Equality in a selection decision is met when all individuals have an equal chance of
getting the position. Gilliland stressed that this rule may be more signiﬁcant in terms of
its violation (i.e. adverse impact) than its satisfaction. He stated that when this rule is
violated based on job-relevant differentiating characteristics, it will not affect fairness
perceptions, but when it is violated on the base of differentiating characteristics not
relevant to the job, it will lead to perceptions of unfairness. Special needs is assessed by
the degree to which hiring is distributed based on the individual needs of others. This
would refer to preferential selection of a disadvantaged subgroup (i.e. handicapped, racial
minorities). Gilliland hypothesized that the degree to which the reason for the special
needs consideration is made salient to job applicants will affect the dominance of the rule
in fairness evaluations, which will affect fairness perceptions. The more salient the
reason is, the less unfair the applicants will evaluate it. Gilliland acknowledged that
satisfaction of one rule may lead to the violation of another, but reasoned that satisfaction
of the most salient rule is most important in increasing fairness perceptions.

Other aspects of the model include moderators of the rule-fairness relationship
and outcomes of perceptions of fairness of the process and outcome. Gilliland (1993)
proposed that applicants’ prior experiences with selection and the stage of the selection
process the applicant is in will moderate the relationship between satisfaction of justice
rules and fairness perceptions. The outcomes of fairness perceptions are grouped into

three categories: reactions during hiring, reactions after hiring, and self-perceptions.

13

Reactions during hiring include job application and acceptance decisions,
application recommendations, test motivation, and legal battles. Performance,
organizational citizenship behavior, job satisfaction, and organizational climate belong to
the category of reactions after hiring. Finally self-perceptions include self-esteem, self-
efﬁcacy, and future job-search intentions. Gilliland made several propositions in regards
to the relationship between these outcomes and fairness perceptions, but it is beyond the
scope of this paper to discuss them all, and therefore the reader is referred to Gilliland
(1993) for further details. However, many of the propositions Gilliland has made have
remained empirically untested.

Relevant Rule

One of the procedural rules in Gilliland’s (1993) model that this paper will focus
on is the consistency rule, regarded in the model as a formal characteristic of the selection
process. Consistency refers to “ensuring that decision procedures are consistent across
people and over time” (pg. 705; Gilliland, 1993). The reason why this particular rule is
of interest is because it is both relevant to issues in research as well as practice. In terms
of research, from the very beginning, consistency has been considered one of the main
rules used in determining fairness of a selection process (Leventhal, 1980; Schmitt &
Gilliland, 1992). In fact, in her survey of entry-level job applicants and professionals
from New Zealand, Singer (1990) found that consistency was rated the most important of
the six factors that were determined to affect fairness perceptions in the selection context
(others included bias avoidance, honest communication, and information soliciting).
Leventhal (1980) stated that it is important to have safeguards to ensure objectivity in the

decision process in order to ensure procedural fairness, and having a selection process

14

that is consistent is one way to do this in a selection context. As Schmitt and Gilliland
(1992) have stated, consistency in the selection process is important so “that personal
biases or self interests [do] not inﬂuence the allocation process” (pg. 36).

In addition to rational arguments for the importance of consistency during
selection, there is empirical evidence that supports this notion. There have been
countless studies showing a strong positive relation between consistency and process
fairness perceptions (Bauer, Maertz, Dolan & Carnpion, 1998; Oliver, 1998; Ployhart &
Ryan, 1998; Ryan & Chan, 1999; Ryan, Greguras, & Ployhart, 1996), as well as 3 meta-
analysis (Hausknecht et al., 2004). Perceptions of consistency have been shown to vary
depending upon what selection procedure is used. For instance, Truxillo, Bauer, and
Sanchez (2001), using a sample of 379 applicants for an entry-level law enforcement
position, found that a written test was considered as more fair in terms of consistency
than a video-based test was. Structured interviews, which ensure a greater degree of
consistency than unstructured interviews, have yielded higher ratings of fairness than
unstructured interviews (J anz & Mooney, 1993; Seijts & Jackson, 2001).

Consistency is an important process rule to examine in terms of practical interests
as well. Several studies have demonstrated that consistency is viewed as an issue of
importance to real applicants in ﬁeld settings, not just to those simulating applicants in a
lab setting. Sheppard and Lewicki (1987) asked a sample of managers to give examples
of unfair organizational actions and to state what justice rule they felt was being violated
in these examples. Results indicated that consistency in applying rules across people and
settings was rated as a key predictor for perceptions of process fairness. As mentioned

previously, Singer (1990) collected statements from students who had all in the past

15

LJ'

 

applied for at least one entry-level job. They were asked to list the one most important
factor they believed would make a selection process fair. The authors then factor
analyzed and reﬁned the responses. One factor that emerged in this group as well as a
validation group of managers was consistency of administration. In particular, statements
about “equal opportunity regardless of age, sex, or race” and using the “same selection
procedure for every applicant” were mentioned. Along a different vein, Truxillo and
Bauer (1999) distributed a survey to a large group of candidates for an entry-level police
position in an organization where banding was being used to increase diversity. Results
showed that applicants’ race interacted with their afﬁrmative action perceptions to affect
their fairness reactions to banding. The authors proposed that this may be due to the
phenomenon that those less likely to beneﬁt from banding (usually Whites) may view it
as causing inconsistent treatment of applicants, whereas those more likely to beneﬁt from
it (in this case Aﬁican-Americans) may focus on how the company may be compensating
for past discrimination, not on inconsistent treatment.

Although a strong link has been found between consistency and fairness
perceptions, research indicates that many organizations are still using inconsistent
selection practices (Neuman, 1996; Novey, Kopel, & Swank, 1996; Novick & Ellis,
1977). Bauer et al. (1998) conducted a study where they surveyed applicants for an ofﬁce
position in an accounting department. They found that procedural consistency ratings
predicted procedural fairness perceptions both after the test, but before feedback, and
after the test, after feedback (when some applicants were selected for the job while others
were not and procedural fairness ratings were expected to vary). In a recent meta-

analysis of the applicant reactions literature, Hausknecht, Day and Thomas (2004) found

16

that although the mean observed correlation between consistency perceptions and
procedural justice perceptions were higher in “hypothetical” or lab contexts (r=.44), it
was still moderate in “authentic” or ﬁeld settings (r=. 17). Consequently, it is evident that
consistency is an issue that determines fairness perceptions in the real world, not only in
contrived settings.

In addition, consistency may be easy to establish when using very basic selection
tools such as tests, but it is more difﬁcult to establish when using more complex selection
techniques (i.e. interviews, letters of recommendation, essays), which are more prone to
subjectivity and personal biases (Cesare, 1996; Schalrn, 2001; Williamson, Campion,
Malos, Roehling, Carnpion, 1997). Therefore, in real-life, there are many situations in
which establishing perfect consistency in selection will be difﬁcult. Thus, in light of
these circumstances, it is interesting to explore what variables could affect the relation
between consistency and fairness. One step in this direction is a study by Ployhart and
Ryan (1998), which found that inconsistency only affected fairness perceptions
negatively when the applicant was somehow put at a disadvantage by the consistency.
However, when the applicant was actually favored due to the inconsistency, the result on
fairness was the same as if no rule had been violated at all (Ployhart & Ryan, 1998)! In
addition, others have noted that there has been very little research on the effect of
individual moderators on justice perceptions (Gilland & Beckstein, 1996; Greenberg,
1990). Therefore, in the interest of furthering knowledge in the research area and possibly
providing some insight to practitioners, this paper will further explore two potential
moderators of the relationship between consistency and fairness perceptions,

individualism and collectivism.

l7

In addition to consistency, there are other rules that can have a strong effect on the
fairness perceptions of applicants during the selection process. The impact of one of
these rules, explanation, is discussed in the following section.

EXPLANATIONS
W

In a selection context it is almost inevitable that some individuals will leave with
a negative outcome (i.e. not getting the job offer), so it is natural that researchers have
turned to examining ways to somehow mitigate the negative effects a negative outcome
or rejection can produce on an applicant’s view of the organization, the selection process,
or perceptions of fairness in general (Gilliland, 1994; Gilliland, Groth & Baker, 2001;
Horvath, Ryan, & Stierwalt, 2000; Ployhart, Ryan, & Bennett, 1999). Therefore, in
recent years, a topic that has gained considerable attention is the use of explanations or
social accounts.

The use of explanations in order to mitigate the negative effects of an undesirable
outcome and change fairness perceptions in an organizational context was ﬁrst addressed
in the organizational justice literature (Bies & Moag, 1986; Bies & Shapiro, 1988; Bies,
Shapiro, & Cummings, 1988; Greenberg, 1991; Greenberg, 1993). Explanations were
initially seen as a valuable tool in order to increase perceptions of a type of justice
separate ﬁom procedural and distributive justice, called interactional justice. The term
interactional justice was ﬁrst coined by Bies and Moag (1986) to describe the quality of
interpersonal treatment one receives during the implementation of a procedure. Bies and
Moag (1986) posited, based on anecdotal evidence and some previous studies, that even

if an individual perceives the outcome he/she received as being fair, and the procedure as

18

having no unfair aspects to it, that individual can still have negative general fairness
perceptions if the interpersonal interaction required to come to the outcome distribution
was undesirable.

Based on interviews conducted earlier with MBA job candidates (Bies, 1985)
conceming critical incidents of fair and unfair recruiting practices, Bies and Moag (1986)
identiﬁed four criteria that were necessary for an encounter to be considered fair. These
four criteria were 1) truthfulness, which was the expectation that recruiters would be open
and honest, 2) respect, which was the expectation that recruiters would avoid being
purposefully rude or personally attacking the candidate, 3) propriety of questions, which
was the expectation that no question based on race, religion, sex, or age would be asked,
and 4) justiﬁcation, which was the expectation that when a rejection was given, there
would be adequate justiﬁcation for the rejection. Bies and Moag (1986) suggested that
these are criteria used to judge the fairness of interpersonal communication during the
selection process, and that these criteria refer to interactional aspects of the situation, not
procedural aspects. The researchers urged further research in the area of interactional
justice, and how it distinct it is from procedural justice, as well as research on how
justiﬁcations used for allocating an outcome or using a procedure by the allocator can
affect the fairness perceptions of the receiver.

Taking his own advice, Bies (1987) explored the area of justiﬁcations ﬁrrther and
pointed out inadequacies in the research in the area of organizational justice. He
explained that theories in organizational justice had ignored 1) that when an injustice
occurs, the harrndoer is caught in a “social” predicament because he/she has violated a

social norm, and 2) that individuals like to know the reasons behind allocation decisions.

19

The implication of taking account of the ﬁrst issue, according to Bies, is that attention
can then be focused on how the harrndocr defends him/herself when caught in the social
predicament. This is where social accounts come into play. Bies (1987) states that
employing a social account allows “the harrndoer to defend his or her social identity

’99

before he or she, too, becomes a ‘victim (pg. 294). The implication of not ignoring the
second issue is a new view of how people determine fairness. This will allow a shift
from viewing people as determining fairness through a input versus outcome model (as
equity theory relays), to viewing people as determining fairness by also analyzing the
reasons for which an injustice occurred in order to judge whether these reasons are
adequate.

Aﬂer presenting this argument, Bies (1987) then proposes a typology for
characterizing various types of social accounts. The ﬁrst type is the causal account
(known in layman’s terms as an excuse) and is an explanation that mitigates the
harrndoer’s responsibility for the harm by asserting that the hanndoer’s actions were
compelled by external circumstances. The second type of account, the ideological
acco_un_t, redeﬁnes the action so that the behavior no longer goes against moral principles.
This can be done by appealing to receivers’ moral convictions by accepting responsibility
for the action, but stressing the role of superordinate goals in the decision, thereby
placing the action in a broader framework that legitimates the action (Bies, 1987). It can
also be done by relabeling the action or outcome in more positive terms. The third type
of account is the referential account, which minimizes the impact of the harm by

providing a more favorable referent standard by which to evaluate it. For example, if a

10% pay cut is in store for members of the claims department, and a 25% pay cut is in

20

store for members of the HR department, a referential account would state “you folks in
the claims department are actually really well off, your coworkers in the HR department
are getting more than twice the pay cut you are!” This would lessen the impact or
undesirability of the outcome.

Soon after Bies (1987) characterization of the different types of explanations,
Bies, Shapiro, and Cummings (1988) came out with another treatise stating that just using
an explanation (whether it be causal, referential, or ideological) is not enough to mitigate
the negative affects of an undesirable outcome, but two other factors contribute to
whether the explanation will have the intended effects: adequacy and interpersonal
sensitivity. The researchers surveyed evening business students and asked them to
recount resent episodes where they had made a request of their boss, and that request had
been denied. Whether the boss claimed mitigating circumstances was measured, as well
the boss’s interpersonal sensitivity in communicating the reason for rejection and the
adequacy of the account. The reactions that were measured were the degree of anger the
participant felt, how much the participant disapproved of his/her boss, the participant’s
procedural justice perceptions, and whether the participant complained to superiors. The
results showed that the perceived sincerity of boss was signiﬁcantly negatively associated
with all the reactions and that adequacy of the account was signiﬁcantly negatively
associated with all reactions except complaints. Whether or not the boss claimed
mitigating circumstances alone was not signiﬁcantly related to any of the reactions,
which was contrary to what previous research had indicated (Bies & Shapiro, 1988).

Shapiro (1991) then argued that more than just the characteristics of the

explanation and the explainer are involved in determining whether an explanation will

21

it

I c

.I‘

 

have a mitigating effect on negative reactions. She found that there are “boundary”
conditions involved when an explanation is being used, and whether or not an
explanation will mitigate negative reactions is contingent upon how many and which of
these conditions are at play. In addition to the perceived adequacy of the explanation and
the perceived honesty of the explainer, Shapiro found that the outcome severity and the
type of negative reaction experienced by the receiver (injustice, disapproval,
punitiveness, or unforgiveness) were boundary conditions. Type of explanation
(altruistic or selﬁsh) inﬂuenced negative reactions, but when adequacy of explanation I
was accounted for, this relationship disappeared. Outcome severity moderated the effect
of explanations on negative reactions, such that no matter what explanation was given,
negative reactions were greatest when participants received the harshest outcomes, and
less negative the less severe the outcome. Perceived adequacy and honesty mediated the
mitigating effects of explanation on feelings of disapproval and punitiveness, but not
unforgiveness and injustice, and therefore, type of reaction was considered a contingency
factor.

Greenberg came out with a series of studies that examined the relationship
between characteristics of explainers and explanations for outcomes, and how these
affected not only negative reactions in general, but fairness reactions. Greenberg (1991)
found that in a performance appraisal context, those explanations for decisions which
were considered to be most fair were those that enhanced the receiver’s self-image. This
supported the idea of the self-serving bias or the egocentric bias, which is the notion that
what an individual considers to be fair is whatever beneﬁts him/her (Greenberg, 1987).

In another study, Greenberg (1993) found that when participants were underpaid for

22

 
 

participating in an experiment, they stole more money ﬁom the experimenter than when
they were paid equitably. However, when the explanation for the underpayment was
highly inforrnationally valid (i.e. from an expert source, directly acquired, publicly
revealed, & double checked) and delivered in an interpersonally sensitive manner,
participants stole less than when the information was not highly valid or delivered
sensitively. Validity was positively related to procedural justice perceptions, and
sensitivity was positively related to interactional justice perceptions. In a subsequent
study (Greenberg, 1994) the researcher found similar effects of interpersonal sensitivity
and informational thoroughness on increasing acceptance of a work-site smoking ban,
and increasing procedural justice perceptions of the ban.

Because so much attention had been paid to how the perceived adequacy of an
explanation determined whether or not it mitigated negative consequences of an
undesirable outcome, Shapiro, Buttner, and Barry (1994) conducted a study in order to
determine which factors caused an explanation to be perceived as adequate. The
researchers discovered that both the speciﬁcity of the content of the explanation as well
as the sincerity of its delivery inﬂuenced perceptions of adequacy, although the former
had more of an inﬂuence than the latter. They also found that the effect of these two
factors was additive when outcome severity was low, but had more of a complex,
interactive effect when outcomes were very severe. The moderating effects of outcome
characteristics were also explored by Colquitt and Chertkoff(2001). These researchers
found that individual’s outcome expectations and the favorability of the outcome both
moderated the relationship between explanation for an outcome and procedural fairness

perceptions as well as task motivation. An explanation produced the highest process

23

 

 

uh

an

iv]

fairness ratings and task motivation when used to explain an unexpected, unfavorable
outcome.

In order to make sense of the various ﬁndings on the beneﬁcial effects of using
explanations to promote fairness and other outcomes throughout the years, Shaw, Wild,
and Colquitt (2003) recently conducted a meta-analysis. The researchers’ hypotheses
centered around the use of fairness theory (Folger & Cropanzano, 1998) in explaining
how and when explanations will have beneﬁcial effects. Faimess theory focuses on how
receivers of the explanation interpret the information they are being given in terms of
different counterfactuals. A could counterfactual comes into play when the receiver feels
that the explainer or hanndoer could have done something different in order to avoid the
harm. A should counterfactual is activated when the receiver feels that the explainer
should have done something different to avoid the harm. Finally a would counterfactual
is activated when the receiver can imagine how his/her life would have been different if
the explainer had employed other means. According to Folger and Cropanzano (1998),
all these counterfactuals have to be activated in order for an injustice to be perceived.
Explanations work to reduce one or more of the activation of these counterfactuals.

Basing their rationale on this theory, the researchers posited that both explanation
provision and explanation adequacy would have beneﬁcial effects, that is, positive effects
on procedural justice, distributive justice, and cooperation, but negative effects on
withdrawal and retaliation (Shaw, Wild, & Colquitt, 2003). The meta-analysis supported
both these hypotheses. Another hypothesis which was supported was one that stated that
explanations in the form of excuses have more beneﬁcial effects than in the form of

justiﬁcations. The last two hypotheses tested dealt with which moderators inﬂuence the

24

 

,, .
as '1

...~' My
.

.33 137

”H... ~

‘ I
c...» n.”

"a

n A,
.-
Iei 'h-‘l

u "at.
>-

"'|‘b~

a n-
ﬁn, \
'u.
u
; s.
0‘ . [
u
.>,‘l-
l.l l"
‘1
I "
. .TI-l‘
u":
“. _
. .III
5" ‘x’
'.\l‘
I. t
‘I‘ "4..
‘.
.M‘.‘
u
n-
: a
K. ~‘:_
‘5
. I
|
\
v-_ 1
l ~-
. my
1 W.

1“

beneﬁcial effects of explanations. The authors found support for the notion that both

outcome favorability and context affect whether explanations have beneﬁcial effects.

They found that explanations had more beneﬁcial effects when outcome favorability was

low versus when it varied, and that they were more beneﬁcial in contexts where the

instrumental, relational, and moral virtue implications were high rather than low.
Speciﬁcally Selection

Based on the above ﬁndings, one could argue that the selection context is a very
useful context in which to explore the use of explanations since 1) outcome favorability
has a good chance of being low (especially when unemployment is high, as is currently
the case), and 2) it is a context that has a high instrumental impact, that is, the outcome of
the context has important economic consequences for the individual. In spite of the
extensive research supporting the beneﬁcial effects of explanation use on fairness
perceptions, as well as numerous studies indicating the importance of favorable applicant
reactions in the selection context to various outcomes, relatively few studies have been
done on the use of explanations in the selection context. There are a few noteworthy
exceptions, however, and these will be discussed presently.

The studies done on the use of explanations in selection contexts look mainly at
two types of explanations: explanations given for the selection process, and explanations
given for the selection outcome. Gilliland (1994) looked at how an explanation given for
the validity of the selection test being used affected applicants’ fairness perceptions.
Participants in the study were asked to complete either a work sample test, cognitive
ability test, or overt integrity test (depending on what condition they were in) in order to

determine whether they would be “selected” to complete a paid coding task the following

25

week. Before completing the test, they were given a written explanation that detailed
why the particular test that was being used was a valid means of selecting individuals for
the coding task. Participants were then selected for the coding task randomly, and both
“selected” and “rejected” individuals completed questionnaires regarding fairness
perceptions, recommendation intentions, and their self-efﬁcacy. The “selected”
individuals were then called back to complete two separate coding sessions. The results
showed that explanation for the procedure had no effect on fairness perceptions, but had
an interactive effect with selection decision on recommendation intentions. That is,
explanations led to higher recommendation intentions for individuals who were rejected,
but did not have an effect on this relationship for individuals who were selected.
Horvath, Ryan, and Stierwalt (2000) also looked at explanations for procedure;
Horvath et al. (2000) manipulated the type of explanation that was given for using the
selection instrument (i.e. causal, ideological, or referential) and the outcome favorability
(hired or not hired). The researchers found that participants’ self-efﬁcacy interacted with
both type of explanation and outcome favorability to determine faimcss perceptions. In
terms of type of explanation, speciﬁcally, there was no difference in fairness perceptions -
between high and low self-efﬁcacy participants when some sort of explanation was
given, however, when no explanation was given (control) participants with lower self-
efﬁcacy perceived procedures as being less fair than those with higher self-efﬁcacy.
Ployhart, Ryan, and Bennett (1999) and Gilliland, Groth, and Baker (2001) both
examined how explanation for selection decision affected various outcomes. Ployhart et
al. (1999) conducted a study in which they asked undergraduates at a large university to

react to various scenarios. The participants were told to imagine that they had just

26

applied for a desirable job and had taken a cognitive ability and job knowledge test as
part of the selection process, and were about to receive the decision telling them whether
they were hired or not. The selection decision was revealed in the scenarios, which were
structured as letters given by the employer to the prospective employee. In these
scenarios, informational and sensitivity facets of explanations were manipulated. The
informational manipulation concerned whether the information in the explanation was
procedural information, justiﬁcation information, or personal information, and the
sensitivity manipulation dealt with whether the information was given in a sensitive
manner or in a regular fashion. The results indicated that there was a main effect of both
sensitivity and information on process fairness perceptions, and that there was also an
interaction between the two. The fairest perceptions occurred in the personal
information, sensitive group, and the least fair in the control group, which received no
explanation. Gilliland et al. (2001) also looked at type of explanation, but their
explanation “types” followed ﬁom fairness theory. Therefore, they looked at the effects
of could, would, and should reducing explanations on fairness. They found that both
would and could reducing explanation signiﬁcantly positively affected fairness
perceptions, but the results for should reducing explanations were equivocal.

Although these studies provide valuable insights on the utility of explanations in
the selection context, there is still a great deal that needs to be examined. Only two
(Gilliland, 1994; Horvath et al., 2000) of the four studies on the use of explanations in
selection have looked at explanations for procedure. The ﬁrst study (i.e. Gilliland) found
that explanation had no effect on fairness perceptions but did have an effect on

recommendation intentions for rejected participants, while the second study (i.e. Horvath

27

et al.) found that type of explanation had an effect on fairness, but there was no control
condition in this study, so there could be no comparison made between fairness
perceptions of those who received no information and those who received information.
In addition, all of these studies look at explanations for why a certain type of procedure is
used, but there is no study that the author knows of that looks at explanations for why a
procedural rule was violated. The meta-analysis found that the use of explanations was
positively related to procedural justice perceptions (p=. 17), however, this ﬁnding was for
explanations in the selection context in general, it did not indicate the useﬁrlness of any
particular type of explanation (Hausknecht etal., 2004).

According to Gilliland’s (1993) model, the violation of procedural rules during
selection can lead to many negative consequences, and this has been demonstrated in the
literature (Bauer et al., 2001; Gilliland, 1994; Hausknecht et al., 2004; Ployhart et al.,
1999; Truxillo et al., 2001). In addition, these rules often do get violated during the
selection process, ergo the variability that has been found in studies looking at the impact
of these rules on fairness in the ﬁeld where no manipulation was present (Bauer et al.,
2001; Gilliland & Beckstein, 1996). Therefore, it would be valuable for another study to
look at whether the explanation for the process being used in selection affects process
fairness perceptions. This is important for both theoretical and practical purposes.
Theoretically, it will help expand knowledge on how explanations for process ﬁrnction in
improving or not improving fairness perceptions, because only two studies investigated
this topic, and those studies show conﬂicting results. It would also be an expansion of
Gilliland’s (1993) model if explanation is shown to be a moderator of the relationship

between procedural rule violation and fairness perceptions. In terms of practical

28

purposes it would also be useﬁrl, for example, in cases where procedural rules are
somehow violated either purposeﬁrlly or by mistake. Explanations could be used in order
to mitigate the negative consequences on fairness of these violations. This, of course,
would be beneﬁcial to organizations, since explanations are easy and economical to use,
and due to the fact that outcomes such as organizational attractiveness, recommendation
intentions, and job acceptance intentions are related to fairness (Bauer et al., 2001;
Macon et al., 1994; Smither et al., 1993; Truxillo & Bauer, 2001), it is in an
organization’s best interest to preserve positive fairness perceptions whenever possible.

In addition, although interpersonal sensitivity in delivering explanations has been
examined in numerous studies within the organizational justice literature (Bies et al.,
1988; Gilliland & Beckstein, 1996; Greenberg, 1993; Greenberg, 1994; Shapiro et al.,
1994), it has only been examined in one study in the applicant reactions literature
(Ployhart et al., 1999). In fact, Gilliland and Beckstein (1995) remark that “a consistent
theme that has emerged is the importance of interpersonal and explanation dimensions of
procedural justice” (pg. 685). Shaw et al.’s (2003) meta-analysis did not examine the
effects of interpersonal sensitivity, however, it did ﬁnd explanation adequacy to be a
determining factor in the beneﬁcial effects of explanations on fairness and other
outcomes. Therefore, because the adequacy of an explanation has been found to be a
determining factor in the effectiveness of an explanation, and interpersonal sensitivity has
been found to be a main factor contributing to the perceived adequacy of an explanation,
the interpersonal sensitivity of an explanation is a very important aspect to study (Shapiro
et al., 1994). In addition, no study has examined the effects of interpersonal sensitivity

on explanations for rule violations within a selection context. It is interesting to explore

29

whether interpersonal sensitivity will mitigate negative fairness perceptions when used in
explanations for procedures in selection, and not just outcomes, which has been studied
in the past (Ployhart et al., 1999). For instance, interpersonal sensitivity of explanations
could function differently when being used to explain violation of a procedural rule
versus a deﬁnite distributive outcome. This may be the case because according to Shaw
et al.’s (2003) meta-analysis, explanations function differently depending upon the
implications of the context in which the explanation is being used. For example,
instrumental impact (e.g., impact on economic circumstances) might be higher under
outcome contexts than procedural contexts. This is because in the case of an outcome
distribution, the outcome decision has already been made and is binding, and therefore,
its instrumental impact is clear (e. g. not getting a job) and very likely to be high.
However, the violation of a procedural rule (e. g. getting less time to complete a test) will
not deﬁnitively lead to a negative or positive outcome, and thus, is less likely to have as
great of a direct instrumental impact. Therefore, it may be likely that interpersonal
sensitivity of explanations will ﬁrnction differently when related to procedures versus the
distribution of outcomes.

Another aspect of explanations, which has not been very extensively explored in
the justice literature, is the receiver’s characteristics. Receivers’ perceptions of various
facets of the explanation as well as the explainer have been explored extensively, but
actual individual differences between receivers has very rarely been explored. The test-
taking self-efﬁcacy of the receiver was explored in one study (i.e., Horvath et al., 2000),
and that study did ﬁnd that self-efﬁcacy moderated the relationship between type of

explanation and fairness perceptions. In addition, Gilliland and Beckstein (1996)

30

Surveyed authors on their fairness perceptions who had sent in submissions to journals
and received notiﬁcation of acceptance or rejection of their manuscript. The results
showed that explanation for rejection or acceptance was strongly related to intentions to
subnrit another manuscript to the journal for experienced authors, but not for
inexperienced authors. Both these studies are important because they may be an
indication that there are other “boundary characteristics” (Shapiro, 1991) that determine
the effectiveness of an explanation under certain circumstances. For example, it is
possible that in addition to self-efﬁcacy and experience, other individual difference
variables, such as cultural orientation, could determine how explanations are received,
and this paper will be exploring that possibility.
Wm

In summary, past research has linked applicants’ fairness perceptions of an
organization’s selection process to various pertinent outcomes such as organizational
attractiveness, intention to recommend the organization to others, satisfaction with the
organization, intentions to accept a job offer from the organization, intentions to buy
products manufactured by the organization, applicant’s self-perceptions, and perceptions
of organizational climate (Bauer et al., 2001; Fiaschetti, 1998; Gilliland, 1994;
Hausknecht et al., 2004; Macon et al., 1994; Smither et al., 1993; Stinglharnber,
Vandenberghe, & Brancart, 1999; Truxillo & Bauer, 1999, Truxillo et al., 2001). In fact,
researchers have suggested that perceptions of process fairness are more inﬂuential in
determining “institutional attitudes” such as recommendation intentions, than are
perceptions of distributive fairness (Gilliland & Beckstein, 1996; Lind & Tyler, 1988). In

addition, research has shown that the satisfaction and violation of procedural rules during

31

the selection process inﬂuence fairness perceptions of that process. One rule in
particular, consistency of administration, has repeatedly been shown to have relations to
process fairness perceptions (Avery & Quinones, 2002; Bauer et al., 1998; Cohn, White,
& Sanders, 2000; Douthitt & Aiello, 2001; Oliver, 1998; Ployhart & Ryan, 1998; Ryan &
Chan, 1999; Ryan et al., 1996; Truxillo et al., 2001). In spite of the fact that such a
strong link has been made between consistency and fairness perceptions, the literature
indicates that many organizations are still using inconsistent selection practices (N euman,
1996; Novey, Kopel, & Swank, 1996; Novick & Ellis, 1977). In addition, sometimes
inconsistencies are inherent in a selection process (i.e., unstructured interviews).
Therefore, it is important to explore, in the interest of both science and practice, whether
individualism and collectivism could moderate the relationship between the violation and
satisfaction of the consistency rule and fairness perceptions.

Explanations have been shown to be powerﬁrl tools for organizations to use in
order to increase perceptions of fairness in situations where outcomes may not be
desirable. Although the selection context is certainly one of these situations, not very
much research has looked at how explanations can be used in a selection context. Even
more importantly, only a handful of studies have looked at how explanations used during
the actual selection process can affect fairness perceptions, even though Gilliland (1993)
has proposed, and the literature has supported, the notion that what occurs during the
selection process inﬂuences both procedural and distributive justice judgments (Bauer et
al., 1998; Brockner & Wiesenfeld, 1996; Gilliland, 1994; Gilliland & Beckstein, 1996;
Ployhart & Ryan, 1998). Furthermore, although interpersonal sensitivity has been noted

as an important factor inﬂuencing procedural justice perceptions (Gilliland & Beckstein,

32

1995), only one study in the applicant reactions literature has explored this variable
(Ployhart et al., 1999). The Ployhart et al. (1999) study, however, only examined
interpersonal sensitivity when used to explain aspects of the outcome, but it is possible
that interpersonal sensitivity functions differently when used to explain aspects of the
procedure, and this should be explored. Additionally, with the exception of one study
(Horvath et al., 2000), no studies in the selection context have looked at what receiver
characteristics could inﬂuence the relationship between explanations and fairness
perceptions. This is important to study because it will help us better understand the
“boundary” conditions under which explanations operate most effectively.

The individual difference factor that has hitherto been mentioned, and will be of
importance to this paper is cultural orientation, and speciﬁcally, the cultural factors of
individualism (IND) and collectivism (COL). Culture is an integral variable to study in
the context of applicant reactions because of a few reasons. One reason is that as
businesses became globalized and expand and establish ofﬁces overseas, it is in their
vested interest to be aware of the cultural dynamics of their local environments. There
are countless stories about businesses that have not succeeded because they failed to take -
cultural differences into account (Martinko & Douglas, 1999; Soochan, 2003). In fact,
one study found that the degree of similarity between national cultures is one factor that
inﬂuences how well multinational companies cooperate, with greater similarity leading to
better cooperation (Van Oudenhoven & Van Der Zee, 2002). In addition, not only are
cultural differences between countries important, but in some countries that have diverse
populations, such has the United States, even cultural differences between groups within

a country are important to study. According to the latest reports of the US. Census

33

Bureau Latinos are one of the fastest growing minority groups within the United States,
and the Aﬁican-American population is also growing steadily. Although cultural
differences at the community level may not seem as obvious as those at the national level,
they still exist and affect group members’ perceptions.

Furthermore, culture is important to study not only in the general sense of how
companies adapt to their environment, but also speciﬁcally as it relates to reactions of
applicants to the selection process. One aspect in Gilliland’s model of fairness during the
selection process that he believed would moderate the effect of the violation of
procedural rules on fairness perceptions was the expectations individuals brought into the
selection context based on prior application experiences. Bell, Ryan, and Wiechmann
(2004) deve10ped a model of the antecedents and outcomes of justice expectations.
Among the antecedents they proposed affect expectations are existing beliefs, direct
experiences, and indirect experiences, all of which are affected by personal and
environmental characteristics. IND and COL are cultural orientations related to beliefs
about how an individual relates to the collective in terms of deﬁnition of self, goals, and
afﬁliations. Consequently, it is very possible that these prior-held beliefs can either
directly affect expectations, or affect individuals’ experiences, which in turn affect their
justice expectations, and ultimately, their justice perceptions and behavior. Steiner and
Gilliland (2001) also argue for the importance of culture when considering applicant
reactions. They state that organizational justice is concerned with what people think is
fair, and what is fair is based on one’s expectations. Culture determines these
expectations in part by its affect on interpretations of events, and its role in deﬁning

normative and appropriate behavior. Therefore, based on these arguments, it is probable

34

that culture has a powerful inﬂuence on individuals’ expectations when entering a
selection context, and'consequently, reactions to the context.

Although several cultural variables could conceivably affect individuals’ reactions
to selection procedures, individualism (IND) and collectivism (COL) appear to be the
most promising cultural factors to examine for a variety of reasons. One reason is that
IND and COL have a strong base in the literature. An individual’s relation to the
collective and how it affects human interaction within organizations has been studied
since the 1950’s (Barley & Gibson, 1998). Scales of IND and COL have been reﬁned for
the past 20 years (Oyserrnan et al., 2002), and the dimensions have emerged consistently
in various studies across diverse samples and contexts (Smith et al., 1996). Furthermore,
Hofstede’s initial identiﬁcation of IND-COL came ﬁ‘om a multinational survey of work
values, so it has a basis in the work domain, unlike other cultural variables such as
Confucian dynamism or view of the goodness of human nature. Morever, IND-COL, in
addition to power distance, accounted for most of the variance in an ecological factor
analysis of Hofstede’s (1980) value data. In an individual factor analysis of work values
Mezei (1974) found that IND-COL was the most important dimension.

In an examination of cultural variables that seemed most applicable to applicant
justice perceptions during the selection context, Steiner and Gilliland (2001) chose IND-
COL as one of the cultural variables likely to affect perceptions of several structural
aspects of the selection procedure, as well as the distributive process. They also state that
“it is likely that cultural dimensions that are important for social relations and relations to
authority will contribute most to determining the salience of these [information-sharing

and interpersonal treatment procedural] rules” (pg.l32), and IND and COL fall under this

35

categorization. Finally, Lytle et al. (1995) identiﬁed six categories under which cultural
dimensions relevant to organizational contexts and used by organizational researchers
fall. “Deﬁnitions of Self and Others” and “Relations between Societal Members”
composed two of these six categories, both of which are central aspects of IND-COL.
Thus, IND-COL has a history linking it to work-related contexts, was identiﬁed early on
as a variable relevant to work processes, has shown to emerge across samples and
contexts, and was posited to directly relate to formal, information-sharing, and
interpersonal aspects of the selection environment.

Individualism and collectivism were originally conceptualized as two ends of one
cultural dimension by Hofstede (1980). He labeled this dimension “individualism” and
the characteristics that are thought of presently as making up the dimension of
collectivism, were at that time included as characteristics that made up the low end of the
individualism scale. Hofstede’s dimension of individualism was one of four cultural
dimensions he introduced (the others being power distance, uncertainty avoidance, and
masculinity/femininity). Hofstede (1980) originally deﬁned individualism as a country-
level phenomenon, however, subsequent studies of IND and COL have examined these
dimensions at the individual level (Carpenter & Radhakrishnan, 2000; Choi, 2001;
Rarnamoorthy & Carroll, 1998; Triandis & Bhawuk, 1997; Triandis & Singelis, 1998)
and a scale that measures IND and COL in individuals has been developed (Triandis,
1995)

This paper will be treating IND and COL as an individual-level variable
inﬂuenced by culture, rather than how it was originally conceptualized, as a cultural-level

variable. Culture has been deﬁned by Hofstede (1980) as “the collective programming of

36

the mind which distinguishes the members of one human group ﬁom another” or “the
interactive aggregrate of common characteristics that inﬂuence a human group’s response
to its environment” (pg. 25). In other words, culture, which is shaped through
geographic, ecological, historical, genetic, economic, demographic and other inﬂuences,
creates a pattern of societal norms which consist of value systems shared by the group
that exists within the culture. Culture therefore affects the molding of the “mental
model” each individual uses to interpret the world and guide behavior. It does this to a
certain extent by inﬂuencing learning such that individuals in the same culture often go
through a similar learning process. As Triandis (1995) states “people who have been
raised in collectivist cultures tend to ‘cognitively construct’ situations into collectivist
settings; people who have been raised in individualistic cultures tend to convert situations
into individualistic settings.” Thus, there is a common rubric of expectations,
assumptions, interpretations, and associations that individuals of a particular culture share
due to its inﬂuence that affect how these individuals interact with the world.

Although culture generally inﬂuences how individuals brought up in that culture
view and inﬂuence the world, it does not inﬂuence all individuals to the same extent, and
this is why it is important to measure some cultural variables at the individual level. For
instance, although many people in Japan are of a high collectivist orientation, this is by
no means the orientation that every Japanese person holds. For IND and COL in
particular, Triandis (1980) has asserted that there are factors other than national culture
that inﬂuence whether an individual will have a high individualistic and/or collectivistic
orientation. These include the person’s age, their social class, and the way they have

been reared as a child. Morever, Schwartz (1992) states that individuals within a given

37

'culture can deviate substantially in the personal values they hold compared to those
endorsed by their culture and Earley and Gibson (1998) state that “individualistic and
collectivistic tendencies can be found within any given culture at different levels of
analysis” (pg. 270). Singelis and Brown’s (1995) model of the effect of cultural
collectivism on an individual’s behavior operates through the mediating effects of that
individuals’ self-concept. That is, national collectivism is positively related to an
interdependent self-construal and negatively related to an independent self-construal.
Ultimately, however, he found that it was self-construal that was directly linked to
communication behavior, not the individual’s culture’s collectivism. Thus it was the
individual-level IND and COL that was directly related to behavior (interdependent and
independent self-construals are treated as synonymous with individual-level [ND and
COL and will be discussed at length later on in this paper). Empirical analyses also
support the measurement of IND and COL at the individual level. Several studies have
found signiﬁcant within-nation cultural variation in IND and COL (e.g., Coon &
Kemmelmeir, 2001; Mastumoto et al, 1997; Oyserman et al, 2002) indicating that even
within countries, individuals signiﬁcantly differ on their degree of IND and COL. To
assume, then, that because individuals originate from a particular country, they have the
level of IND or COL of that country is erroneous, and to make behavioral predictions
based on that assumption is additionally ﬂawed.

Therefore, this paper will be looking at IND and COL as individual difference
variables. The following section will review the deﬁnitions of IND and COL, and the

work that has been done on these dimensions thus far. Speciﬁc hypotheses in reference

38

to how IND and COL affect relationships between procedural rules, explanations and
fairness will then be presented.
IND and COL

Individualism and collectivism were ﬁrst “discovered” by Hofstede (1980) in his
landmark study of culture entitled Cultures Consequences. The concept of
“individualism” was developed out of a study of employees’ responses to work goal
questions in 40 countries. The degree of individualism of a certain country was
determined by how employees responded to the importance they give to six work goals in
particular: personal time, freedom, challenge, use of skills, physical conditions, and
training. Giving a great deal of importance to the ﬁrst three goals indicated higher
individualism, while ﬁnding the last three goals to be more important indicated a higher
degree of collectivism. Western countries such as the United States, Great Britain, and
Canada generally scored high on this index, while eastern and South American countries
such as Pakistan, Taiwan, Colombia, and Venezuela scored low on this Individualism
Index.

Although Hofstede (1980) was the ﬁrst to coin the terms individualism and
collectivism to describe the constructs involving the individuals’ relation to the collective
(he introduced individualism in his original work, and later added the term collectivism to
represent the low end of the individualism continutun, see Hofstede and Spangenberg, .
1987), other scholars had been studying the same phenomena under different labels years
earlier. Emile Durkheim, also a social scientist, studied the characteristics of ingroups
and classiﬁed the mechanisms by which they exist (Earley & Gibson, 1998). One of

these was the concept of mechanical solidarity versus organic solidarity. Mechanical

39

solidarity, which is often found in very tight groups, such as clans, is existence based on
common bonds among group members, and a sense of obligation to the group. The goal
of existence was to fulﬁll the needs of the group and minimize conﬂicts among group
members. Organic solidarity was used to describe existence of an ingroup based on a
structured system, such as a hierarchy, which represents increasing levels of specialized
labor, and binds peOple together in order to utilize the system to pursue personal goals.
The former was representative of a collectivist orientation, and the latter of an
individualistic orientation. Other scholars such as Max Weber, Verlin F. Tonnies, and
Confucius have also remarked on the nature of the struggle of inﬂuence between the
individual and the collective (Oyserman, Coon, & Kemmelrneier, 2002).

After Hofstede labeled the cultural dimensions of IND and COL, many attempts
were made to appropriately deﬁne the constructs. Hofstede (1980) himself saw
individualism as the construct of interest and collectivism as its polar opposite. He
deﬁned it in terms of how the individual relates to the collective in any given society. He
also stated that its inﬂuence can be noted in the values people hold in a'particular society,
and the way people live in that society (i.e. in nuclear families, extended families,
kibbutzes, etc. . .). Schwartz and Bilsky (1987) extended this classiﬁcation to values in
general, not just work values as Hofstede had done, and indicated which values they
thought were representative of individualistic tendencies and which were representative
of collectivist tendencies. Those values which Schwartz and Bilsky (1987) felt primarily
served the interests of the individual such as sense of accomplishment and social
recognition fell under the motivational domains (i.e. broader value categories) of

enjoyment, achievement, and self-direction, while those values which reﬂected the needs

40

of the collective such as obedient and helpful, fell under the prosocial and restrictive
conformity motivational domains. He found support for this classiﬁcation from both an
Israeli sample and a German sample.

Triandis became very interested in exploring the dichotomy of what Hofstede
referred to as individualism and collectivism at the individual level. Because
individualism and collectivism were terms used to describe these cultural phenomena at
the national level, Triandis adopted the terms idiocentrism and allocentrism to describe
the same phenomena at the individual level. Idiocentrists were primarily deﬁned by
tendencies to give priority to personal goals over ingroup goals, while allocentrists tended
to see ingroup goals as superordinate to personal goals. Therefore, idiocentrists were
likely to give importance to those values that serve themselves solely, such as
independence and personal recognition, while allocentrists gave more importance to
values that promoted the well-being of the ingroup, such as harmony and
interdependence (Triandis et al., 1985; Triandis ct al., 1988).

Triandis (1995), in his landmark book, Individualism and Collectivism, came to
more clearly deﬁne what the dimensions of idiocentrism and allocentrism meant, as well
as their cultural counterparts, individualism and collectivism. He stipulated that there
were four dimensions on which individualists and collectivists differed. These were 1)
the deﬁnition of self, 2) alignment of goals, 3) focus on norms vs. attitudes, and 4)
emphasis on rationality versus relatedness. Collectivists deﬁne themselves in terms of
the groups they belong to, so they have an interdependent deﬁnition of self, while
individualists deﬁne themselves as individuals separate from the groups to which they

belong, indicating an independent deﬁnition of self. For collectivists, individual and

41

group goals are closely aligned, and group goals often take precedence over individual
goals. In the case of individualists, individual goals are disconnected and often
incompatible with group goals, and individual goals take priority over group goals.
Norms are very important for collectivists, and behavior and cognitions are guided by
obligations and duties stipulated by the culture. In contrast, behavior in individualist
cultures is guided by individual needs, attitudes, and rights, and norms are not given
much importance. Finally, for collectivists, maintaining relationships and having a
connection with others regardless of costs or beneﬁts is very important, while
individualists tend to weigh the advantages and disadvantages of having a relationship
before they decide to pursue one.

Triandis (1985) also argued early on for a ﬁrrther distinction of IND and COL that
reﬂected how individuals viewed themselves, similar to the interdependent-independent
view of self discussed earlier, but this time referring to sameness versus distinctiveness.
He termed this dimension of IND and COL as horizontal versus vertical individualism
and collectivism. Horizontal IND and COL refers to the view that everyone is similar
and should be treated similarly. Horizontal collectivists and individualists do not tolerate
being treated differentially based on status or other attributes. In contrast, vertical
collectivists and individualists do recognize that there are differences between individuals
and do see rank and status as a legitimate means of making distinctions between people.
People who support the horizontal view do not like to be viewed as different, or separate
from everyone else, while those who support the vertical view, do accept that they will be

treated differently and do recognize and often wish to attain the privileges of higher rank.

42

Other researchers such as Schwartz (1990, 1992) and Erez and Barley (1993) have
also done work pertaining to the use of IND and COL as cultural dimensions. Schwartz
(1992) created a list of 56 values which he deemed represented the values of most
cultures. Each of these values was then rated according to how important it was in the
lives of individuals. From these ratings, Schwartz found that the values can be organized
into 10 separate clusters. Three of these clusters reﬂected hedonism, stimulation, and
self-direction reﬂected individualist values, while another three—traditionalism,
conformity, and security—reﬂected collectivist values. Triandis, McCusker,.and Hui
(1990) found that out of the six clusters, self-direction was most central to IND, while
conformity was most strongly related to COL. Erez and Barley (1993) developed a
model of cognitive functioning in a cross-cultural, organizational context. This model
included aspects of culture, self motives, and management practices, which they
postulated interacted to inﬂuence work behaviors. Based on their review of the extant
literature, they argued that IND and COL were core characteristics of culture that
inﬂuenced people’s motives for action, their self-identity, and consequently, the way they
responded to situations at work.

Singelis and Brown (1995) developed a theory on the nature of culture, the self,
and communication involving IND and COL. They proposed that IND and COL
orientation affect an individual’s self-construals such that high individualists have an
independent self-construal, while high collectivists have a more interdependent self-
construal. Thus, in a conceptualization similar to Triandis’ (1995), those who have an
independent self-construal (i.e. individualists) deﬁne themselves in terms of their

individual attributes and unique characteristics. These internal attributes are seen as

43

stable, and they are viewed as the primary forces that govern an individual’s behavior.
Those with independent self-construals also see themselves as autonomous and
independent of those who surround them in their social environment. Their individual
entity, although connected with those in their family and social circle, is not intertwined
or greatly shaped by those inﬂuences. In contrast, those with an interdependent self-
construal (i.e. collectivists) see themselves embedded in a larger social network and
deﬁne themselves in terms of their social relationships. They do not view themselves as
clearly separated or differentiated, as do those with independent self-construals, but as
more connected with others. It is thus, very important for them to ﬁnd ways to ﬁt in with
others and form meaningful interpersonal relationships with those surrounding them.
Although they do realize that they have individual personality characteristics and ‘
attributes, they perceive these as being more malleable and situation-speciﬁc than do
those with independent self-construals. Thus, for those with interdependent self-
construals, these personality characteristics can and should be controlled and regulated in
order to express the proper behavior called for in a particular situation, instead of
themselves being the primary cause of the behavior. Consequently, to assert oneself,
which is considered natural and self-afﬁnning to those with independent self-construals,
is considered weak and immature to those with interdependent self-construals.
Independent self-construal is considered synonymous with the terms individualist,
autonomous, and idiocentric, while interdependent self-construal is synonymous with the
terms collective, contextualist, and allocentric (Markus & Kitayama, 1991).

Despite the extensive work that has been done investigating the nature of the

cultural dimensions of IND and COL, there is still a great deal of inconsistency in how

the constructs are currently being conceptualized and studied. In Hofstede’s (1980)
original work, IND and COL were envisioned as value systems that reﬂected polar ends
of one continuum. Therefore, IND and COL at the cultural level were considered a
unidimensional construct. Consequently, if a culture-was high on IND that meant that it
had to be low on COL. This view persisted throughout the 1980’s, and researchers,
instead of measuring IND and COL directly, instead employed Hofstede’s country
classiﬁcation in order to group people by nationality into either an “individualist” or
“collectivist” category. When cultural researchers became more interested in how this
cultural dimension of IND/COL was expressed at the individual level, they realized that
they needed to deve10p instruments to measure this construct. Triandis, Hui, and
colleagues (Hui, 1984; Hui, 1988; Hui & Triandis, 1985; Triandis et al., 1986; Triandis et
al., 1988; Triandis, 1995) undertook this endeavour and interviewed over 50 social
scientists in several countries in order to develop a list of values that spanned various
cultures. The result of further analysis and reﬁnement of this list was a 63-item measure
labeled INDCOL that measured individuals’ levels of IND/COL. The assumption in
creating the scale, however, was still that this was a unidimensional construct, and that
IND and COL were opposite ends of the dimension.

Upon ﬁrrther examination and reﬁnement of the scale, Triandis and colleagues
(1988) discovered that the IND/COL construct at the individual level was in fact,
multidimensional. This was a revolutionary ﬁnding and led Triandis (1988) to remark
“one point can be made quite generally: in both the ecological and the within-culture
analyses and those done in all the cultures so far, the emphases on individualistic and

collectivist themes are orthogonal. . .if one studies a broad range of values (as did

45

Hofstede, 1980) then in that context individualism-collectivism is one dimension. If one
focuses only on self-ingroup relationships (as we did in the present studies), then a
multidimensional structure emerges [emphases added]” (pg. 336).. He also pointed out
that because [ND and COL were independent, treating them as polar opposites was not
only oversimpliﬁed but also inaccurate.

Triandis’ ﬁndings were replicated by Bontempo (1993) who used IRT techniques
to conduct a modiﬁed parallel analysis on the INDCOL measure. He also asserted that
IND and COL should be conceptualized as orthogonal factors and consequently,
measured separately. He cautioned against the common practice of reverse scoring
responses to collectivist items and assuming that they were psychologically equivalent to
individualist items. Yamaguchi, Kuhlman, and Sugimori (1992), using a separate scale
of IND/COL, also found a two factor model ﬁt their items the best. In addition, their
results indicated that in their US. sample, the subscales were not correlated (r=.04, n.s.),
and that IND and COL related differently to different characteristics. For example,
afﬁliative tendency was positively correlated with collectivism but not correlated with
individualism, while need for uniqueness was positively related to individualism, but not
to collectivism. Traﬁmow, Triandis, and Goto (1991), in several experiments designed to
explore the cognitive structure of the self, demonstrated that cognitions related to the
individual and private aspects of the self and those related to the collective and connected
aspects are encoded separately in memory. Their results seem to indicate that the IND
and COL aspects of ourselves are compartmentalized and stored as distinct, unique

factors in our mental schemas.

46

Empirical ﬁndings on the multidimensional nature of IND and COL at the
individual level led to further conceptual and theoretical development on the
dimensionality and orthogonal characteristics of the constructs. Schwartz (1990)
criticized the then dominant view of IND/COL as a dichotomy on several bases. One
argument he presented was that the dichotomy did not account for values that inherently
were of interest to both individualists and collectivists such as maturity values that
promoted the appreciation, understanding, and acceptance of oneself, of others, and of the
surrounding world. Another ﬂaw he saw was that the dichotomy wrongly assumed that
individualist and collectivist values each grouped onto separate poles that were
diametrically opposed to each other. He argued that the subtypes of individualist and
collectivist values such as prosocial values and self-direction sometimes did not vary
together and were sometimes not opposed.

Triandis (1995) put forth a similar argument in his major theoretical work,
Individualism and Collectivism. When he introduced idiocentrism and allocentrism into
the lexicon as representations of IND and COL at the individual level, he described them
as orthogonal constructs that could be simultaneously present in one individual. He
stated that “in short, which elements are sampled is an important determinant of social
behavior. Individuals have both individualistic and collectivist cognitive elements.
Therefore, a person considered allocentric is one who is simply more likely than a person
considered idiocentric to sample collectivist elements and use them to construct the
meaning of a social situation.” (pg. 8). Thus, a person can be both high or low on
allocentrism and hi or low on idiocentrism at the same time. However, across most

settings, being brought up or living in a collectivist culture will make one more likely to

47

access one’s allocentric self while being brought up or living in an individualistic culture
will make one’s individualistic self more salient. Coon and Kemmelrneier (2001) state
that ample evidence from experimental and observational studies as well as analyses of
individual-level measures of IND and COL supports this orthogonal, two factor view of
IND and COL.

In spite of ample evidence, both empirical and theoretical, indicating that IND and
COL should be considered as two separate factors, a great deal of research on IND and
COL still treats these dimensions as opposite ends of a unidimensional, bipolar construct
(Matsumoto et al., 1997; Smith, Dugan, & Trompenaars, 1996). Oyserman, Koon, and
Kemmelrneir (2002) conducted a meta-analysis of the IND and COL literature and found
that out of the 170 studies that examined psychological implications of IND and COL,
only 36 studies (21 %) assessed both IND and COL. They noted that the bipolar, single
dimension approach was by far most popular with researchers studying psychological
implications. The NDCOL measure (Hui, 1988) in its entirety continues to be used as
unidimensional measure of IND or COL with its items reverse scored (Goregenli, 1997;
Ramarnoorthy & Carroll, 1998), despite Bontempo’s (1993) warnings. In fact, several
researchers have emphasized the need to measure these cultural factors as two separate
dimensions (Gelfand, Triandis, & Chan, 1996; Kashima et al., 1995; Oyserman, 1993;
Rhee et al., 1996; Singelis, 1994).

Furthermore, even when studies do utilize the two factor structure of IND and
COL in their methods, they still treat these constructs as if they are non-orthogonal.
Comparisons across these constructs are frequently being made, and conclusions are

often labeled in terms of IND versus COL or idiocentrism versus allocentrism (Earley &

48

Gibson, 1998). In fact, the vast majority of research employs comparisons between these
constructs, rather than between levels of each construct (Carpenter & Radhakrishnan,
2000; Triandis et al., 1988; Yetim, 2003). This view perpetuates the notion that if one is
high on one dimension, one must be low on the other dimension. It also assumes that
individuals cannot both be idiocentric and allocentric, because they would then have to
act in opposing ways simultaneously. As Oyserman, Kemmelrneier, and Coon (2002)
aptly state “. . .the initial framing of individualism and collectivism as opposing social
ﬁameworks left a residue of dichotomizing in the ﬁeld’s theorizing about these cultural
models—as if social institutions, practices, and situations created ﬁxed cultural minds
that make sense of the world through either an individualistic or a collectivistic lens” (pg.
111)

Nevertheless, there have been a few studies that have examined these factors
separately. Coon and Kemmelrneier (2001) found that different ethnic groups had
varying levels of IND and COL. While Asian Americans and Aﬁican Americans scored
higher on the COL scale than did Caucasian Americans, only Aﬁican Americans scored
higher than Caucasian Americans on the IND scale. Consequently, Asian Americans and
Latinos scored as high on the 1ND as did Caucasian Americans. In the case of Asian
Americans, this demonstrates that individuals can be both idiocentric and allocentric at
the same time. In addition, the researchers found a nonsubstantive correlation between
IND and COL across most groups (r=. 12 to r=-.01). The only study this author knows of
that actually examined high and low levels of both IND and COL was conducted with a
sample of Hispanic and Caucasian MBA students. The study found a .04 correlation

between COL and IND, and also that the acculturation of Hispanics was negatively

49

related to collectivism, but unrelated to individualism. What was most interesting,
however, was that individuals’ preference for task-related versus contextual job attributes
differed depending on what subgroup of the combination of hi and low IND and COL
they were in. Those with high individualistic values (both low and high on collectivism)
had a signiﬁcantly higher preference for task-related job attributes than those with low
individualist values. However, only those with low collectivist values and high
individualistic values had a Signiﬁcantly lower preference for contextual job attributes
than those with high collectivist values. This demonstrates that examining the constructs
separately, and even pairing them in different combinations can yield unexpected and
interesting results.

In sum, although the literature clearly supports a two-factor, orthogonal structure
to IND and COL, the majority of research in this area is not consistent with this
conceptualization. Nevertheless, due to the extensive empirical and theoretical support of
the two factor model, I will be examining IND and COL in terms of within factor
comparisons. Thus, I will be comparing the behavior of high versus low collectivists and
high versus low individualists. However, the rationale for my hypotheses will be coming
from a large base of literature that views IND and COL as dichotomous, opposing
constructs. Consequently, based on the majority of the research conducted up until the
present, I expect to see low collectivism ﬁrnctioning similarly to individualism and low
individualism functioning similarly to collectivism.

In addition, although Triandis (1995) has argued for a four factor structure of IND
and COL, assessing IND and COL based on both horizontal/vertical and IND/COL

dimensions, and has provided preliminary evidence in support of this conceptualization

50

(cf. Triandis & Gelfand, 1998), this study will not employ Triandis’s conceptual
structure. This is due in part to the fact that others have failed to replicate the ﬁndings of
Triandis and Gelfand (1998) concerning the vertical/horizontal distinction of IND and
COL (see Lewis, 1996), and furthermore, as has been presented, several studies in this
domain have found ample support for the two-factor version of IND and COL (Oyserman
et al., 2002; Wagner, 1995).
IND/COL and Justice

Many researchers have argued that aspects of culture, such as IND and COL, can
largely determine variance in perceptions of organizational justice, or fairness, across
cultures. In their ﬁamework for examining inﬂuences of various variables on the relation
between justice information gathered and reaction to the decision-making event, Colquitt
and Greenberg (2003) present national culture as being a major player. They include
culture as a contextual variable (as opposed to an individual variable) in the model, that
inﬂuences justice expectations, and therefore, moderates the impact of the justice
information gathered on the both the justice judgment reached, and the reaction to the
decisionmaking event. Choi (2001) also proposed that culture, speciﬁcally the cultural
dimensions of IND and COL, would moderate the effect of certain variables on
procedural fairness judgments. He hypothesized that IND and COL would moderate the
effect of different types of participation on justice perceptions, such that collectivists
would see representative participation as more fair, and individualists would see
individual participation as more fair. Choi also suggested that whether group or

individual outcome favorability was seen as more fair would depend on IND and COL,

51

such that collectivists would favor group outcome favorability, while individualists
would favor individual outcome favorability.

Lind, Tyler, and Huo (1997) also suggested that culture could be a moderator of
associations they proposed in their relational model of procedural justice. They argued
that procedural fairness judgments are deﬁned in relational terms, and that there are three
relational variables that people consider when making procedural justice j udgrnents.
These variables were trust in benevolence, status recognition, and neutrality. They
conducted a study of students from the United States and Japan. They found, in fact, that
culture did moderate the relation of status recognition to procedural justice perceptions,
such that the American participants placed much greater emphasis on status recognition
than did the Japanese participants. Other studies have found similar proof of culture as a
moderator of various variables and justice perceptions (Blader, Chang, & Tyler, 2001;
Pillai, Williams, & Tan, 2001; Rahim, Magner, Antonioni, & Rahman, 2001).

In addition, a great deal of empirical work has been done looking at how cultural
orientation affects justice perceptions in many different organizational contexts
(Kabanoff, 1997). Studies have examined how culture affects fairness perceptions in
negotiation contexts (Buchan, 1998; Gelfand, Higgins, & Nishii et al., 2002; Leung &
Lind, 1986; Leung, Au, Femandez-Dols & Iwawaki, 1992; Tyler, Lind & Huo, 2000)..
Gelfand et a1. (2002) found that those in individualist cultures were more likely to view
an offer by their counterpart in a negotiation context as unfair than those in a collectivist
culture. They also found that individualists tend to have a stronger self-serving bias in
fairness perceptions during negotiation, in that they view their offers as more fair than

their counterparts’, and they view feedback that is self-serving as more fair than feedback

52

that is critical, whereas collectivists tend to have the opposite reaction. Leung et al.
(1992) found that collectivists tend to view conﬂict processes that give them process
control and reduce animosity between participants as more favorable than confrontational
procedures, while individualists are more accepting of confrontational procedures than
collectivists. Tyler, Lind, and Huo (2000) found that in conﬂict situations between
authorities and subordinates, interpersonal treatment by authorities is more important in
determining- favorable reactions to the situation in people ﬁ'om low power distance
cultures versus high power distance cultures. Culture has also been shown to affect
justice perceptions in simple allocation contexts (Choi, 2001; Hui, Triandis, & Yee, 1991;
Hundley & Kim, 1997; Krishnan, 1998; Milliman, Nason, Gallagher, et al., 1998;
Murphy-Berman & Berrnan, 2002), performance appraisal contexts (Leung, Su, &
Morris, 1998; Li & Karakowski, 2001; Ramarnoorthy & Carroll, 1998), and contexts in
which work-related attitudes are being ascertained (Farh, Earley, & Lin, 1997; Jackson,
2001; Leung, Smith, Wang, & Sun, 1996; Pillai, Williams, & Tan, 2001).

Despite the fact that culture and justice have been explored extensively in many
other organizational contexts, the effect of cultural orientation on justice perceptions in
the selection context has largely been ignored. There are only a handﬁrl of studies that
have looked at the inﬂuence of cultural orientation on justice perceptions during the
selection process, even though many have argued, as do Lincoln, Hanada, and Olson
(1981) that “. . .cultural factors shape organizations and the responses individuals make to
them...” (pg. 114). Steiner and Gilliland (2001) have created a model of cultural
inﬂuences on procedural justice reactions during the selection process. They posit that

there are certain cultural dimensions which are likely to have more of an inﬂuence on

53

reactions during the selection process than other dimensions. Using Hofstede’s (1980)
classiﬁcation, as well work done by Triandis (1995), Kluckhohn and Strodtbeck (1961),
Parsons and Shils (1951), and others, the researchers developed a list of cultural
dimensions that they felt would be relevant in a selection context. In addition to
Hofstede’s four dimensions of individualism-collectivism, masculinity-femininity, power
distance, and uncertainty avoidance, locus of control, time perspective, Confucian
dynamism, and ﬁve other variables were examined. Although masculinity-femininity,
and beliefs about human nature were hypothesized to inﬂuence perceptions of fairness
across all categories of procedural rules (i.e. structural aspects, information sharing, and
interpersonal treatment), IND/COL was posited to have a strong inﬂuence on structural
aspects of the selection process including job relatedness, opportunity to perform, and
consistency of treatment. The authors address the need for ﬁrrther research to be done in
the area in order to test their model and clarify the role of cultural orientation on applicant
reactions during selection.

Only a few studies have addressed Steiner and Gilliland’s (2001) concern,
however. Nevertheless, those studies that have been done on the topic do ﬁnd that
cultural orientation inﬂuences applicant reactions and justice perceptions. Phillips and
Gully (2002), conducted a study involving working adults in Singapore and the United
States. They found that Singaporeans were more accepting of graphology and
ethnicity/nationality based selection methods than were Americans. They also found that
Singaporeans did not perceive work-sample tests to be as fair as Americans did. They
attributed this ﬁnding to the notion that for people from individualist cultures, it is

important that they get a chance to demonstrate their skills, or “sell” themselves, and that

54

including this aspect into the selection process makes them feel a greater sense of
fairness, while the same is not true for collectivists. Therefore, a method such as a work-
sample, where one’s skills are openly exposed would not be as favorable to a collectivist
as it would to an individualist. Phillips and Gully’s (2002) study was partly an extension
of a study done earlier by Steiner and Gilliland (1996), in which they examined how
undergraduates in France and the US. reacted to ten different selection techniques. They
found that graphology and the use of biodata was found to be fairer and more acceptable
to French students, while the use of personality tests was more accepted by American
students than French students.

Along a similar vein, Ryan, McFarland, Baron, and Page (1999) surveyed a
sample of over 900 organizations in 20 countries in order to determine whether culture or
_ nation affected the use of certain selection practices. They found that nation itself
accounted for a signiﬁcant amount of variability in selection practices across countries,
and that uncertainty avoidance, and to a lesser extent, power distance accounted for
variance in the use of certain selection practices in particular. Speciﬁcally, uncertainty
avoidance explained variance in amount of veriﬁcation methods used, the degree of
veriﬁcation done, and the numbers of interviews conducted, such that the greater the
uncertainty avoidance, the greater the levels of these variables. Ramarnoorthy and
Carroll (1998) also looked at how culture affected preference for selection practices, and
examined how IND/COL speciﬁcally inﬂuenced reactions to alternative human resource
management practices using a sample of senior undergraduate business students. They
found support for their hypothesis that individuals with a higher collectivist orientation

show less preference for selection practices that are based on performance on ability tests.

55

They explained that cognitive ability test based selection only reﬂects ability to perform
well on the job, and this should not be the sole criterion employed, according to
collectivists, in order to be selected for a job. Collectivists see organizations almost as an
extended portion of one’s family, and therefore, such characteristics as employee loyalty
and person-organization ﬁt are considered very important criteria for selection, and they
often cannot be ascertained through the use of simple test-based selection.

Each one of these studies, however, only examines how culture inﬂuences
reactions to one method of selection over another method; they do not speciﬁcally
examine the procedural rules posited in Gilliland’s (1993) model of justice reactions in a
selection context. Therefore, they are not able to truly test Steiner and Gilliland’s (2001)
propositions about how certain cultural dimensions, such as IND and COL, affect the
salience of particular procedural rules over others during the selection process. This
study is an attempt to address this question and bridge the gap in research linking cultural
dimensions to justice issues in the selection context.

HYPOTHESES

Thus far, this paper has described the importance of the study of applicant
reactions to selection for both the organization (e. g. future recommendation intentions)
and the applicant (self-efﬁcacy, job satisfaction). It has detailed a speciﬁc model,
Gilliland’s (1993) model of fairness in selection, that addresses how certain aspects of the
selection process (i.e. “rules”) affect an individual’s judgment of fairness of that process.
From this model, one rule, consistency of administration, has been selected as the
variable of greatest interest. A Shift from procedural rules during the selection process to

the importance of explanations used during the selection process was then made. The

56

role of interpersonal sensitivity in the process of selection was discussed as well. The
lack of research on the inﬂuence of individual differences on both reactions to the
violation of procedural rules as well as reactions to the use of explanations in the
selection context was then demonstrated. Finally, two very salient individual difference
variables, speciﬁcally, the cultural dimensions of IND and COL were then introduced. A
review of the literature linking culture to justice reactions and the selection context was
then conducted. This next section is structured to introduce multiple hypotheses on the
nature of the relations betwieen procedural rules, explanations, and procedural justice
perceptions, and how IND and COL moderate these relations in the selection context.

The hypotheses to be presented are based on the following model presented in Figure l:

 

Individualism Collectivism

. a. II. at. a.
Consrstency \
v\

Explanation ’ Perceived Adequacy

 

 

 

 

 

 

 

 

 

 

 

 

 

Interpersonal Sensitivity

 

 

    

Procedural Fairness

 

 

 

 

 

 

 

 

 

 

 

.jg.
Figure 1 . Model of the Interaction between IND/COL Orientation and Consistency,
Interpersonal Sensitivity, and Explanation Provision on Procedural Fairness Perceptions
HYPOTHESIS l: IND and COL orientation will moderate the effect of consistency of
administration on procedural fairness reactions such that:

la. Those who are high on collectivism will perceive procedures as less [air when

consistency is violated than those who are low on collectivism

57

lb. Those who are lo_w on individualism will perceive procedures as Lem

when consistency is violated than those who are high on individualism.

Consistency of administration, at its heart, has to do with equality of treatment
(Steiner & Gilliland, 2001). When procedures are conducted in a consistent manner,
everyone who receives the procedure will receive it in the same way and by the same
means as everyone else. When this does not occur, and some people are treated
differentially, this is when inconsistency becomes an issue (Ployhart & Ryan, 1998).
There is ample evidence to show that for collectivists, equality of treatment is of great
importance. Therefore, it will have more of an impact on high collectivists if the
consistency rule is violated than it will have low collectivists.

Hui, Triandis and Yee (1991) conducted a study involving Chinese and American
students, where they asked them to make a hypothetical monetary allocation decision for
themselves and their partner. They measured IND/COL orientation for all the students,
and found that cultural variance on IND/COL accounted for use of the equality rule. The
researchers found that those high on collectivism used the equality rule much more than
those high on individualism. Individualists tended to rely more on the equity rule when
making allocation decisions. Therefore, the collectivist participants were much more
likely to distribute the money equally than the individualists were.

In their review of the inﬂuence of culture and personality, Triandis and Sub
(2002) point out many personality characteristics of individuals ﬁom collectivist versus
individualist cultures. They indicate that in terms of ethics, collectivists prefer equality to
equity. They explain that this is because collectivists are very concerned with harmony,

solidarity, and cohesion, and equal distribution promotes these values. They also detail

58

that those in individualist cultures prefer equity because it is more reﬂective of values
such as competition and productivity, which are of more consequence to individualists.

Triandis (1989) in an earlier paper also details how collectivists base their
relations with others much more on communal relations than exchange relations.
Consequently, they rely to a greater extent on characteristics that deﬁne communal
relationships such as 1) lack of clarity about what is to be exchanged, 2) concern for the
other person’s needs over concem for equity, 3) importance of maintaining equality of
affect, and 4) equality of beneﬁts exchanged. Therefore, he goes on to say that the
applicability of exchange theory to collectivist cultures will not be very strong, since
individuals from these cultures are used to communal relations. Other research has
demonstrated similar results, that the equality rule and not the equity rule is much more
salient for collectivists than it is for individualists (Bond, Leung, & Wan, 1982; James,
1993; Miles & Greenberg, 1993; Steiner & Gilliland, 2001).

In addition to the importance of equality of treatment for collectivists, another
reason why consistency of administration will be more important to high collectivists is
because they have a stronger external locus of control (Triandis & Sub, 2002), and
therefore, will give much more importance to elements of the context, such as how a
procedure is administered, than will low collectivists. That is, during the selection
procedure, collectivists will be likely to perceive they have little personal control over the
situation, and that control lies largely in the hands of others, such as those who administer
the procedure. Therefore, they will be more dependent on these others, and the context

(i.e. aspects of the process external to them), in order to derive meaning from the

59

situation. Consequently, they will be much more affected by any changes that occur
during the process.

There is much research to suggest that collectivists feel that control lies more in
aspects of the situation than aspects of themselves, and therefore, pay greater attention to
aspects of the situation. Triandis and Sub (2002) explain that collectivists see aspects of
themselves as variable, pliable to the needs of the situation, and aspects of the situation as
more or less ﬁxed. In contrast, individualists see aspects of the situation as being
variable, while aspects of the self such as personality or attitudes are relatively ﬁxed. So
“situational determinants of behavior are important universally, but more so in
collectivist than in individualist cultures” (pg. 137). Wagner (1995), in his study of the
effect of IND/COL orientation on cooperation in groups, found that collectivists tend to
be much more reliant and dependent on others when determining how to act in a
situation, rather than relying on their own internal volition. Church (2000) created a
model of culture and personality in which he asserted that personal traits account for
behavior much less in collectivist than in individualist cultures. Triandis (1989) also
found that collectivists attribute events much more to external causes than individualists
do.

Singelis and Brown’s (1995) theory on the nature of culture, the self, and
communication proposes that IND and COL orientations affects individuals’ self-
construals, so that collectivists have a much more interdependent self-construal, while
individualists have a more independent self-construal. As a consequence, collectivists
depend much more on aspects of the context or other people in order to regulate their

behavior than they do on themselves. They also attribute more behavior to contextual

6O

'factors than they do to personal factors, and use descriptions that concern contextual
information rather than dispositional information. In addition, Markus and Kitayama
(1991) state that those with an interdependent view of self tend to see their internal
attributes, such as personality and attitudes, as situation-speciﬁc, and “as such, they are
unlikely to assume a powerful role in regulating overt behavior” (pg. 227). Barley and
Gibson (1998) in their review of IND and COL, also found ample support for the notion
that collectivists place a great deal of emphasis on the context of the situation in
determining meaning.

Steiner and Gilliland (2001) put forth a similar argument as to why consistency
will be more important in collectivist cultures. They argued that those who have more of
an external locus of control rely on others to determine their fate, and therefore, will be
more concerned with aspects of the situation, such as receiving equal treatment, than
those who feel more of a sense of control themselves. Markus and Kitayama (1991) also
state that for those with an interdependent construal of self, “a relatively greater
proportion of all inferences will be contingent on the pragmatic implications of a given
situation, such as the perceived demands of the interviewer, the convention of the
situation, and the rules of the conversation” [emphasis added] (pg. 234). Gudykunst,
Gao, Schmidt, Nishida, Bond, Leung, Wang, and Barraclough, (1992) also found that
because collectivists rely so much on external forces to determine their fate, predictability
is very important to them because they know what to expect from the environment.
Similarly, Schwartz (1990) found that restrictive conformity, or following rules and

norms, is a stressed value for collectivists. By following rules and regulations,

61

collectivists are increasing the predictability of the situation, and forming expectations as
to what should occur.

Consequently, consistency of administration will be of more consequence to high
collectivists than to low collectivists for many reasons. One is that high collectivists
value harmony, and therefore prefer to use the equality rule of distribution. When there
is inconsistency of administration, and people are being treated differentially, then the
equality rule is being violated. Secondly, high collectivists have more of an external
locus of control than low collectivists. Therefore, they depend much more on external
factors in determining their fate. Thus, in a selection situation, aspects of the situation,
such as consistency of test administration will be much more salient to them. They will
accordingly be much more affected by inconsistencies in the situation, because they feel
that they have no control over these inconsistencies. In addition, since they are very
aware of the norms and rules of a situation, they will be more likely to notice any
deviation from these norms.

Note that in this study, my primary hypothesis concerns how collectivists of
varying degree will respond differently to rule violations. Thus, Hla, comparing high and
low collectivists is my primary hypothesis, while Hlb, comparing high and low
individualists is meant more to explore whether different levels of individualism operate
in a similar manner. Therefore, the majority of my rationale is in reference to the

behavior of collectivists, not individualists.

HYPOTHESIS 2: IND and COL orientation will moderate the effect of explainer ’s

interpersonal sensitivity on procedural fairness reactions such that:

62

2a. The eﬂect will be stronger for those who are high on collectivism than for
those low on collectivism.
2b. The eﬂect will be stronger for those who are Log on individualism than for

those who are high on individualism.

Interpersonal sensitivity has to do with the relational aspects of a situation
(Greenberg, 1990). Being more sensitive in a situation means acknowledging that one is
taking account of the other person’s cognitions and/or affect during that situation. Being
interpersonally sensitive when delivering an explanation is an attempt to lessen the
negative impact the consequences of the explanation may have on another individual, and
thereby maintaining good relations with that individual. The basis for why this would be
of greater consequence to high collectivists than for low collectivists is that for
collectivists, maintaining good relations and harmony between people is integral, because
their sense of self is deﬁned by those relations (Triandis, 1995). For individualists, good
relations are not as essential because their sense of self is relatively independent from
their relations with others.

Research has repeatedly demonstrated that for collectivists, relations with others
are critical to their sense of being. One of the main attributes of a collectivist has to do
with his/her interconnectedness with others (Triandis, 1995). In fact, placing others’
desires and goals before one’s own goals and desires in order to maintain interpersonal
harmony is a common theme. Identity, for collectivists, is deﬁned by the groups to which
they belong and the relationships that they maintain. In the conﬂict resolution literature,

it has repeatedly been shown that collectivists prefer dispute resolution procedures that

63

minimize confrontation, and promote harmony and good relations (Bond et al., 1985;
Leung, 1997; Leung & Lind, 1986; Leung et al., 1992; Tyler, Lind, & Huo, 2000). While
in individualist countries such as America, an ideal person is considered one who has
many accomplishments, in collectivist countries such as Japan, an ideal person is one
who has many successful relationships with others (Triandis, 1995).

Noesjirwan (1978) surveyed a mixed sample of students, middle—class dwellers,
and laymen from both Australia and Indonesia in order to determine their main value
orientations. She found that there was a marked difference between the two cultures in
that in Indonesia, a collectivist culture, sociability and community were the major value
orientations, while for Australia, these relational value orientations did not hold. The
author deduces from the results that for Indonesians much more than for Australians, it is
important to have good relations with everyone. Schwartz (1990) also conducted a value-
based analysis, and found that for collectivists, or those from communal cultures,
interpersonal prosocial values are some of the main values governing their life. In
addition, research has found that those ﬁom collectivist cultures prefer bosses that are
more caring to those that are more task-oriented, even if the more caring boss is more
demanding (Hayashi, 1988; Lincoln, Hanada, & Olson, 1981).

There are many collectivist cultures in which interpersonal harmony is considered
a key aspect of relations with others. For instance, in the Hispanic culture, the notion of
simpatico, or kindness and sympathy is a crucial one (Triandis, Marin, Lisansky, &
Betancourt, 1984), and in the Filipino culture the value of having smooth relations with
others is stressed, and being sensitive and adjusting oneself in order to help alleviate

difﬁcult situations is also valued (Church, 1987). The Thais relay the import of trying to

64

maintain the peace and not disturbing others, while the Japanese put emphasis on a
similar notion of keeping interpersonal relations agreeable and amicable, and not
straining them in any way (Markus & Kitayama, 1991).

In addition to the fact that high collectivists are more sensitive to disturbing the
feelings of others, and weigh smooth interpersonal relations more heavily than low
collectivists or individualists do, another reason for why interpersonal sensitivity of
explanation would have more of an impact on them is because they are already sensitive
to it due to their ﬁ'equent use of it in communication. Triandis (1995) argues that concern
for others and sensitivity is very important in collectivist communication. Triandis and
Sub (2002) note that collectivists use more face-saving communication, in order to spare
the feelings of others, and remain sensitive to their situation. Other research has shown
that those with interdependent self-construals will communicate emotions that are more
other-focused (such as sympathy) so that the other person’s viewpoint will be taken into
account (Markus & Kitayama, 1991). In addition, when communicating, collectivists
show a great deal of sensitivity to the feelings and responses of the other party (Singelis
& Brown, 1995). When in conﬂict situations speciﬁcally, collectivists are very likely to
consider the feelings of their adversaries, and take them into account while
communicating (Doucet & Jehn, 1997). They are also more sensitive to the expressions
and reactions of others, and often take these into account before communicating
(Gudykunst et al., 1992).

Therefore, there are various reasons why high collectivists will be more affected
by the interpersonal sensitivity of an explainer when delivering an explanation than will

low collectivists. Collectivists are more interconnected with others, and their relations

65

with others are of great signiﬁcance to them, so they will be more receptive to the
interpersonal aspects of situation. Steiner and Gilliland (2001) also state that “it is likely
that cultural dimensions that are important for social relations. . .will contribute most to
determining the salience of these [interpersonal] rules” (pg. 132). In addition, those with
a collectivist orientation will be more receptive to interpersonal sensitivity when an
explanation is given because they are very familiar with the use of sensitive language
during communication and are very receptive to the use of this language.

Similar to Hla, H2a is my primary hypothesis in this set. Consequently, my main
argument concerns how high collectivists versus low collectivists will react to the
interpersonal sensitivity of the explainer in a situation. The second part of the hypothesis,
H2b, concerning how high versus low individualists will react to interpersonal sensitivity
is an exploratory one, meant to assess whether individualism operates in a similar way

(albeit in the opposite direction) to collectivism.

HYPOTHESIS 3:

3a. Those who receive an explanation for the procedural process will perceive

procedures as more fair than those who do not receive an explanation

3b. This effect will be mediated by the perceived adequacy of the explanation.

AS stated previously, research has demonstrated that explanations can mitigate the
effects of negative outcomes on fairness perceptions (Bies & Moag, 1986; Bies &
Shapiro, 1988; Bies, Shapiro, & Cummings, 1988; Greenberg, 1991; Greenberg, 1993).
However, as Shapiro (1991) discovered, the mitigating effect of explanations on negative

reactions can be determined by “boundary conditions”. One of these “boundary

66

conditions” is the perceived adequacy of the explanation. Shapiro (1991) found that the
perceived adequacy of the explanation mediated the effect of explanation provision on
feelings of disapproval and punitiveness for those who received a negative outcome, but
not their feelings of unforgiveness or injustice. Others, however, have found that the
perceived adequacy of an explanation for a negative outcome does affect perceptions of
justice. Bies et al. (1988) found that the perceived adequacy of an explanation was
negatively associated with participants’ negative reactions at having a request denied by
their boss. These included participants’ anger at their boss and disapproval of their boss.
They also found that adequacy was positively related to participants’ perceptions of
procedural justice. Similarly, in their meta-analysis on the effects of explanations, Shaw
et al., (2003), found that perceived adequacy of an explanation for a negative outcome
had beneﬁcial effects on both procedural and distributive justice perceptions, as well as
cooperation. These effects were greater than that of explanation provision, leading the
authors to suggest that inadequate explanations could lead to greater perceptions of
unfairness than providing no explanation at all. Therefore, this hypothesis is an attempt
to re-address the ﬁndings of Shapiro (1991) in order to determine whether simply
providing an explanation is enough to mitigate the negative effects of rule violations on
procedural justice perceptions, or whether the mitigating effects work through
perceptions of explanation adequacy.

Note that in this study, the nature of the interaction between consistency,
sensitivity of explanation, and explanation provision will only be examined for situations
in which outcome favorability is low, that is, when individuals are rejected for a job.

There are both theoretical and practical reasons for why situations of high outcome

67

favorability will not be examined in the current study. Theoretically, it has been posited
that procedural rules will not be of as great import when a favorable outcome is given,
and this has been supported empirically. Gilliland (1993) states “It is predicted that
procedural rules will have the greatest impact on system fairness in situations in which
distributive rules have been violated. Thus, if job applicants receive job offers that they
think they deserve, they will not be as concerned about the fairness of the selection
process as if they did not receive a job offer [emphasis added]” (pg.721). Brockner and
Wiesenfeld (1996), in their review of 45 independent samples, came to a similar
conclusion. Their results demonstrated that level of procedural justice is not strongly
related to individuals’ reactions when outcome fairness is high (which is usually the case
when individuals are hired), but is strongly related when outcome fairness is low (as is
often the case when individuals are rejected). In support of this notion, Leung and Li
(1990) found that process control (procedural justice) effects on procedural and
distributive justice were strong when the outcome was negative, but were eliminated
when the outcome was positive.

Similar results have been found in relation to the effect of explanations on process
fairness perceptions. Shaw, Wild, and Colquitt (2003) in a meta-analysis of the effects of
explanations on justice perceptions, found that explanations were most strongly .
positively related to procedural justice perceptions, and therefore most beneﬁcial, when
the favorability of the outcome was low. Authors whose manuscripts were accepted by
journal review boards did not alter their perceptions of organizational justice when an
explanation was given versus when no explanation was given, but for those whose

manuscripts were rejected, perceptions of justice were greatly increased when an

68

explanation was given (Gilliland & Beckstein, 1996). Similarly, those workers who were
positively affected by a work-site smoking ban in Greenberg’s (1994) study displayed the
same level of acceptance for the ban regardless of whether the explanation for the ban
was given in a sensitive or neutral manner, but those workers who were negatively
affected by the ban reacted much more favorably when the explanation was given in a
sensitive manner. Additionally, in a study by Colquitt and Chertkoff (2002), outcome
favorability was shown to moderate the effects of explanation on fairness perceptions
such that explanation had a negligible effect on perceptions when the outcome was
positive, but had a strong, positive effect when outcome was negative.

In short, based on previous research, it is unlikely that either consistency of
administration or explanation sensitivity would signiﬁcantly inﬂuence procedural justice
perceptions in a context where outcome favorability is high, such as when applicants are
selected for a job. Furthermore, in addition to the aforementioned theoretical and
empirical reasons for excluding a high outcome favorability condition in the present
study, there are practical reasons for doing so as well. The design being employed is a 3
(consistency, inconsistency no explanation, inconsistency explanation) x 2 (sensitive,
insensitive) between subjects design, and therefore, will result in six different conditions.
Individualism and collectivism levels will also be measured. If outcome favorability is
also manipulated, that will add two more levels, which will lead to a design with 12
conditions. Therefore for reasons based on both theory and practicality, this study will be
limited to looking at the relationship between IND and COL, explanation sensitivity,
consistency, and procedural fairness in a situation where the favorability of the outcome

is low.

69

Furthermore, although no speciﬁc hypotheses will be presented in reference to
these outcomes, several other outcome measures such as performance, recommendation
intentions, reapplication intentions will be examined, as these have shown to be affected
by fairness perceptions in the procedural justice and applicant reactions literature (Bauer
et al., 2001; Gilliland, 1994; Hausknecht et al., 2004; Macon et al., 1994; Smither et al.,
1993). Therefore, to remain consistent with the procedural justice literature, those

variables will be measured in this study.

70

METHOD
Sample

Participants were recruited from the undergraduate and graduate population of a
large Midwestern university. In order to obtain enough variance on the IND and COL
measures, attempts were made to recruit as many participants from the international
student population as possible. This was done through multiple methods. The author
contacted several student organizations that had an international or multicultural student
base and announced the purpose of the study (generically) and need for international
student participation. The author attended several student organization meetings and
events in order to recruit students, including movie nights and coffee hours. Emails were
circulated in student organization listservs and a table was set up outside of the cafeteria
of an international student donnitory in order to recruit students.

Three hundred and thirty-nine students participated in the study. However, due to
random and off-scale responding as well as incomplete data, 23 participants had to be
dropped from analyses, resulting in a total sample of 316 students. Despite several
recruiting efforts, international students comprised only 8% of the total experimental
sample, although a greater number, 25.3%, were members of a minority subgroup. Of
this 25.3%, 2.6% were Hispanic, 8.9% were Asian, 8.2% were Aﬁican-American, and
4.7% were other. Participants were also recruited from the introductory psychology
subject pool. The sample therefore had a greater number of females (72.5%) than males,

and was relatively young, with a mean age of 19.5 years.

71

Measures

All measures used are listed in Appendix A. Reliabilities of all measures are presented in
Table 1.

Manipulation Checks. Six items were used to measure perceived consistency. These
include three items from the consistency measure of the Selection Procedural Justice
Scale developed by Bauer et al. (2001). The other three items to be used were developed
speciﬁcally for this study. These items were meant to assess whether participants
acknowledged that they had less time than others to take the test, if this were the case.
Interpersonal sensitivity was measured using ﬁve items, also taken ﬁ'om Bauer et al.’s
(2001) Selection Procedural Justice Scale. In order to detennine whether participants
recognized that an explanation for the procedural violation was given or not, two items
were developed to measure explanation provision. In order for participants to react to the
manipulations, it was necessary that they be at least somewhat motivated to do well on
the test; therefore, a ten-item scale developed by Sanchez, Truxillo, and Bauer (2000)
was used to measure participants’ test-taking motivation.

IND and COL. In order to measure IND and COL at the individual level, Singelis’
(1994) measure of interdependent and independent self-construals was used. This
measure consists of 30 items. Self-construals are central to the concepts of IND and
COL, and interdependent and independent self-construals are often considered
synonymous with idiocentrism and allocentrism (Markus & Kitayama, 1991). In fact, in
Triandis’ (1995) discussion of the deﬁning aspects of IND and COL he states “the.
deﬁnition of the self is interdependent in collectivism and independent in individualism.

This is reﬂected in various aspects of daily life, including the extent to which individuals

72

share resources with group members and conform to the norms of the group. Scales for
the measurement of this aspect have been developed by Singelis (1994)” (pg. 43).
Indeed, research has shown that COL is positively related to interdependent self-
construals and unrelated to independent self-construals (Singelis & Brown, 1995).
Interdependent and independent self-construals are also considered orthogonal (Markus
& Kitayama, 1991). This measure has also been shown to produce results consistent with
other measures of IND and COL (Coon & Kemmelrneier, 2001), and its construct and
predictive validity have been demonstrated (Singelis, 1994). Several studies have
employed this scale to measure IND and COL (Coon & Kemmelrneier, 2001; Okazaki,
2000; Oishi et al., 1998; Sharkey & Singelis, 1995; Singelis & Sharkey, 1995; Singelis et
al., 1999). In addition, although some caution against using self-construal scales to
measure cultural IND and COL, because they are too constrained to the psychological
level of analysis (Earley & Gibson, 1998), this is exactly the level of analysis that is of
interest to this study. Because this study is exploring how IND and COL at the individual
level, or idiocentrism and allocentrism, relate to psychological processes, this measure is
ideal, because it is focused at the psychological level rather than capturing too broad of a
cultural ﬁame. Originally, for validation purposes, a supplemental measure of IND and
COL developed by Triandis et al. (1988) was going to be used as well. However, the
factor structure of the Triandis measure was different than that of the Singelis measure,
and there was ample evidence in support of the Singelis scale as a measure of IND and
COL to warrant its use as the sole indicator of IND and COL (Coon & Kemmelrneier,

2001). Therefore, results from the Triandis measure will not be discussed. For the

73

purpose of simpliﬁcation, heretofore, Singelis’ (1994) independent dimension will be
referred to as IND, and the interdependent dimension will be referred to as COL.

In the current study, all scales had reliabilities greater than .80, with the exception
of the distributive justice and IND and COL scales. Scales of individualism and
collectivism in past studies have not evidenced very high reliabilities in general (Triandis
et al., 1995). In fact, the 95% conﬁdence intervals around the alpha estimates ﬁ'om the
current sample for both the IND (.58 to .70) and COL (.66 to .72) subscales of the
Singelis (1994) measure included alpha values Singelis himself has reported for these
scales in both his original validation study and subsequent studies utilizing the measure
(Singelis, 1994; Singelis et al., 1999). The low reliabilities often witnessed in measures
of IND and COL may be due to the fact that each of these cultural variables by deﬁnition,
is made up of several aspects, including deﬁnition of one’s self in relation to the group,
tendency to follow norms or be governed by attitudes, subordination or superordination
of personal goals to collective goals, emphasis on cost-beneﬁt versus purely relational
approaches to relationships, et cetera (Triandis, 1995). Although these aspects in
combination comprise a general construct, because they are quite varied, this may not
come through in the alpha reliability statistic (see Triandis et al., 1995 for a discussion on
low reliability of IND and COL measures). Nevertheless, as Guion (1998) states
“evidence of internal consistency may be less relevant than evidence of internal
completeness or relevance in domain sampling” (pg. 245).

In terms of the factor structure of the Singelis (1994) scales of IND and COL,
several past studies have veriﬁed its two-factor structure (Sharkey & Singelis, 1995;

Singelis, 1994; Singelis, 1995). One recent study has suggested that the two main factors

74

have subfactors, but this ﬁnding is new, and has not yet been replicated with any other
samples, whereas the two-factor structure has (Hardin, Leong, & Bhagwat, 2004).
Moreover, the low reliabilities evidenced in this sample are similar to reliabilities
evidenced in other samples involving the self-construal scales (Coon & Kemmelrneier,
2001; Okazaki, 2000; Singelis et al., 1991). A certain amount of the discrepancy could
be due to error evidenced in reliability estimates in general. In fact, in a recent paper,
Duhachek and Iacobucci (2004) have suggested that researchers should include
conﬁdence interval or standard error estimates around their alpha coefﬁcient, as was done
for this study, to better inform others of the error around that reliability estimate.
Nevertheless, in order to examine the factor structure of the self-construal scale with the
current sample, a conﬁrmatory factor analysis was conducted.

The CF A was conducted using AMOS version 4.0. The two-factor model of IND
and COL was tested to see how well this model ﬁt data from the current sample.
Listwise deletion resulted in the elimination of six cases. In order to reduce the number
of indicators for each factor to a manageable size, three item parcels were created out of
the 15 indicators for each factor. The items to be included in the parcel were chosen
based on a systematic process, in an attempt to obtain equal item representation in each
parcel. An exploratory factor analysis was conducted for both DID and COL. The item
that loaded highest on the factor was included in the ﬁrst parcel, the next highest item
was included on the second parcel, and the item with the third highest loading was
including in the third parcel. This method was followed until ﬁve items were assigned to
each parcel. The items included in each parcel were then summed to create one indicator.

Thus, each factor had three modeled indicators. The values of a few relevant ﬁt indices

75

were as follows: 1 2 (9) = 19.93, p<.05, RMSEA=.O63, NFI=.929, CFI=.959,
SRMR=.066. Based on the indices, the two-factor model appeared to ﬁt the data well. A
comparison between this model and a proposed alternative model (Hardin et al., 2004)
will be addressed in the section on future research.

Explanation Adequacy. Explanation adequacy was assessed using four items from a
scale employed by Wiechmann and Ryan (2001). The items were altered to reﬂect the
change in experimental context.

Justice. Both procedural and distributive justice were measured using items adopted
ﬁ'om Gilliland’s (1994) measure of process and outcome fairness. Procedural justice was
assessed using eight items, distributive justice was assessed with four items.

Outcome variables. Participants’ intentions to recommend the organization to others, or
recommendation intentions was measured using Gilliland’s (1994) scale composed of
four items. Two scales adopted from Ployhart and Ryan (1998) were used to assess
participants’ job acceptance intentions and future experiment intentions. Job acceptance
intentions, or the likelihood that the participant would accept the job if offered, was
measured both prior to when the decision to hire or reject was made and also after this
decision was made, each time by a one-item scale. The likelihood that participants’
would participate in similar experiments, or future experiment intentions, was determined
by a four-item measure. This measure was created to simulate job re-application
intentions in an experimental context. Due to the high correlations between these
intentions measures (r=.70-.80, p<.05), they were combined to produce one “intention
towards organization” scale. Participants’ test performance during the selection process

was assessed using their scores on the Wonderlic Personnel Selection Test, which is a

76

cognitive ability test composed of 50 items. Test performance was standardized within
consistency condition, due to the fact that individuals in the inconsistent conditions
received less time to complete the test.
Demographics. Participants’ gender, number of years in the United States, race, GPA,
and year in school were the descriptive variables assessed. Self rated English
. proﬁciency, citizenship, whether participants have had work experience in the United
States, and how many times they have gone on an interview or taken a test in order to get
a job were included as well (see Appendix F).
Procedure

Participants entered the experimental room and signed their name on a sign-in
sheet on which they were assigned a random experimental II) number, which they were
asked to remember. They were then instructed to sit one seat apart whenever possible.
Five minutes after the experiment was scheduled to being, participants were given an
informed consent form to ﬁll out (see Appendix B). Signed consent forms were then
collected, and participants were told that they will be taking an established personnel
selection test in order to see how the instrument performs with people of various age
groups (protocol is listed in Appendix C). This excuse was given so that participants
would be more likely to take the test seriously and was a viable explanation for both the
subject pool participants as well as the international participants. They were also told
that this test has been used in the past by various companies, and some of the names of
these companies were listed in order to emphasize the credibility of the test. The
selection test given to participants was the Wonderlic Personnel Selection Test, a

standard cognitive ability test.

77

Participants were given a survey packet (see Appendix D), and located on the ﬁrst
page of this survey packet was a description of four internship positions being offered by
Company XY Z. These positions were selected to represent four different domains
(journalism, business, science, and social services) that would be attractive to
undergraduate students. These domains were selected based on an informal survey of
several undergraduates. Participants were asked to imagine that they were applicants
seeking an internship position with Company XYZ and to choose the internship position
they found most desirable. They were asked to write down the internship they selected
and why they would like to have the internship, in order to increase the salience of the
desirability to them. They were instructed to consider the environment as a simulated job
enviromnent and the test administrator as a representative from the company to which
they are applying. Participants were told that they would have to take a job selection test
and that their performance on the test would determine whether they would or would not
be hired as an intern. They were also informed that those participants who score highest
on the test would be rewarded by having their names entered into a lottery to win free
movie tickets. Therefore, the participant who scored the best in each session would be
entered into this lottery. In addition, everyone had their names entered into a lottery for a
chance to win a cash prize of $25, $50, $75, or $100, and the top six scorers from every
three sessions will have their names entered in two additional times. The reason why two
levels of prizes were included was to motivate both those who aspired to be the top scorer
in each session, as well as those who wished. to do well, but did not feel like they would

be the top scorer, and so would be discouraged if only the top-scorer prize was offered.

78

Q}!

Li

\rIHA

hit

The test administrator notiﬁed the participants that they would be told whether they were
hired for the job or not at the end of the session.

After this explanation was given, the test administrator then instructed
participants to turn the next page in the survey packet. Participants were told that the
questionnaire on the following pages assessed general attitudes and preferences. They
were asked to write their name and assigned ID numbers on the scantrons, so that the test
administrator could keep a record of their answers and “match” their feedback to their
responses (see section on feedback). This questionnaire sheet contained measures of IND
and COL and test-taking motivation. The survey packets were then collected.
Manipulations

Depending on what condition they were in, participants were then given a
combination of the following three manipulations.

C_on_sistencv Manipulation:

The consistency manipulation was a variation of one employed by Ployhart and
Ryan (1998). Participants were told that they were going to be handed test packets and
that they would have 12 minutes to complete the test. The proctor then wrote the total
amount of time participants had left to complete the test on the board, so that participants
would be aware of the amount of time that had passed. In the inconsistent conditions,
when the proctor began to pass out the forms, she appeared confused and began to
whisper to her assistant. She then excused herself and left the room, while her assistant
stayed in the room. If participants asked why the proctor left, the assistant replied with
the standard response—“I don’t know, but she said she’ll be right back.” After exactly

ﬁve minutes, the proctor then re-entered the room and in the inconsistent conditions, she

79

changed the time on the board to reﬂect the remaining amount of time participants had to
complete the test. Depending on the explanation condition, the proctor told the
participants different things (this will be detailed in the explanation section). She
announced that participants now had only 7 minutes to complete the test packet.
Participants were then asked to begin, and were warned when 2 minutes were left in the
testing period and stopped after 7 minutes had passed. In the consistent conditions,

participants were given the full 12 minutes to complete the test packet.

Intggpersonal Sensitivity Manipulation:

 

The sensitivity manipulation involved two parts. In general, the test proctor
attempted to smile more and appear more ﬁiendly in her interactions with the participants
in the sensitive conditions, and attempted to maintain a stoic, unﬁiendly expression in the
insensitive conditions. After the proctor passed out the ﬁrst questionnaire, a confederate
impersonating a participant raised her hand and asked a question. In the sensitive
conditions, the proctor answered the confederate’s question in a polite way, in the
insensitive conditions, the proctor answered the confederate’s question in a rude manner
(see Appendix C). In addition, when the proctor re-entered the room after leaving for
ﬁve minutes, she explained how much time the participants had left in either a sensitive
or insensitive manner.

Explanation Manipulation:

The explanation manipulation involved giving an explanation for aspects of the
procedure. In the consistent conditions, participants were given explanations regarding
the consistency of the procedure (e.g., you will be given the same amount of time as

others were given), and in the inconsistent conditions participants were either given or '

80

not given an explanation for the inconsistency in procedure (see Appendix C). Thus, in
the no explanation condition, after re-entering the room, the test administrator gave no
explanation for the inconsistency. In the explanation condition, the proctor told the
participants that she had mistakenly brought the wrong forms of the test and had left the
room to obtain the proper forms of the test.

Outcome Favorabilitv Manipulation:

After all participants had been given the allotted time to complete the Wonderlic,
the test forms were collected and participants were given an additional survey to
complete. This “Applicant Experiences” survey included measures of IND and COL
(T riandis’s measure), process fairness, pre-decision job acceptance intentions,
recommendation intentions, and future experiment intentions (see Appendix E).
Participants were told that while they were completing the surveys, the research assistant
was going to manually score their responses to the Wonderlic test. The research assistant
then left the room with the tests. In the hall, the assistant assigned the participants’ ID
numbers to pre-ﬁlled feedback sheets. All participants were informed that they had not
been selected to be in the lottery and that they had not been selected for the simulated job.
The message given to them on the feedback sheet was as follows:

Your results indicate that you [I] did Cl did n_ot score well enough on the exam to be included
in the lottery for the increased chance to win the money or the free movie tickets. You also 0
did Cl did n_og score well enough in order to be selected for the job for which you applied.
You Cl would Cl would n_ot be offered the position at this time.

In all conditions, the boxes next to did not and would not were checked. The options

were presented so that participants felt that they could have performed well or received

the prize, although this was not the case.

81

After participants had completed the ﬁrst survey, the proctor called the
administrative assistant back in the room. The proctor called out participants’ ID
numbers when passing out the feedback forms, so that they felt the feedback was
personalized. After they were given some time to read the feedback sheets, the
participants were told to ﬁll out the last portion of the survey which included questions
on their post-feedback perceptions of procedural and distributive justice, explanation
adequacy, post-feedback job acceptance intentions, recommendation intentions, and
future experiment intentions, demographics as well as consistency, sensitivity, and
explanation provision manipulation checks. The manipulation check measures were
placed at the end at the survey so that they would not reveal the nature of the
manipulations and inﬂuence participants’ answers to other questions.

Debrief'ing

When the participants ﬁnished ﬁlling out the second survey, they brought their
survey packet and scantron to the proctor in the front of the room, and were handed a
debrieﬁng form (Appendix F). In the form they were informed that the selection test was
not administered to them in order to see how it applied to individuals of various age
groups, but in order to see how see how the violation of procedural rules affected their
performance on the test. They were told that their performance on the test was not
actually scored on the spot, and the results were therefore not accurate. The debrieﬁng
form also indicated that all participants will have their names entered into the lottery
three times, not just the top six scorers, but that the top scorer in each session will still be

entered into the lottery for the movie tickets (the tests will be scored later).

82

Pilot Testing

Pilot testing was conducted in order to determine appropriate timing of measures,
clarity of measures, and efﬁcacy of manipulations. The main focus of the pilot testing
was to compare two different manipulations of consistency, in order to see which one
functioned better. The two manipulations that were piloted consisted of one involving a
time consistency manipulation and another one involving a question consistency
manipulation. For the ﬁrst consistency manipulation, participants were told that the
administrator had the wrong version of a cognitive ability test that they had to take, and
so had to wait ﬁve minutes for the administrator to return with the “correct” versions of
the test, and therefore, were given less time than others to complete the test. The second
consistency manipulation involved the participation of two confederates pretending to be
test-takers. During the course of the test administration, one confederate asked a question
and was answered while the other confederate asked a question that the administrator
refused to answer, after which the confederate points out that the administrator answered
the other participant’s question.

Subjects were recruited during summer semester from various psychology classes.
About 15 students participated in each manipulation, resulting in a total N of 3 1. Two
participants’ data had to be removed because of random responding. Results from
analyses showed that the mean for the time manipulation (M =3.06, SD =88) was about a
quarter of a standard deviation lower than the mean for the question manipulation
(M =3.26, SD=.80). The consistency scale consisted of values ranging from 1 to 5, with l
signifying that the participant strongly disagreed that the procedures were consistent, and

5 signifying that the participant strongly agreed that the procedures were consistent.

83

Thus,.participants in the time manipulation felt the procedures were more inconsistent
than participants in the question manipulation. Consequently, the time manipulation,
which was stronger, was employed as the manipulation of consistency for further data
collection. The sensitivity manipulation was also tested, and the mean for the sensitive
condition (M =2.92, SD=1.13) was about one standard deviation higher than the mean for
the insensitive condition (M =1 .84, SD=.99). In addition, when the sensitivity
manipulation was employed during the pilot session, and the administrator answered a
confederate’s question insensitively, there was an audible “gasp” heard, and one
participant was heard to mumble “that’s harsh”. Although in an absolute sense, even in
the sensitive condition, the mean rating for sensitivity was not that high (3=“neither agree
nor disagree”), this could have been due to a possible confound with the consistency
manipulation, so that participants could have felt the administrator was being insensitive
by not treating all participants equally, and this could have affected their ratings of
sensitivity. Unfortunately, the interplay of these rules cannot be avoided, and therefore, it
is important to point out that even despite this, the sensitivity manipulation seemed to

have a strong effect.

84

RESULTS

Scale descriptives, intercorrelations, and reliabilities are presented in Table l.

The means of the fairness scales were all below the midpoint of the scale. The means for
fairness and intention scales post-feedback were lower than means pre-feedback, which is
likely the effect of negative feedback. The ranges of the COL and IND scales are
relatively small. For example, the mean of COL, is 4.86, and the standard deviation is
.57. Therefore, the majority of individuals in this study lie between values of 4.29 and
5.43 on the COL scale, which is a seven point scale. Therefore, this sample is restricted
to mid to high levels of COL, and consequently, some of the predicted interaction effects
may be attenuated due to a lack of variance on the COL measure. The same is true for
the IND measure, which has an even higher mean, and a similar standard deviation. This
will be explored further in the discussion section.

Although the IND and COL scales are supposed to be independent, they were
negatively correlated, however, this correlation was small (r=-.13, p<.05). The measures
of IND and COL were generally only signiﬁcantly related to themselves and test-taking
motivation, with the exception of COL which was also related, albeit a small correlation,
to intentions pre-feedback (r=.12, p<.05). This may reﬂect collectivists need to maintain
good relations and harmony with others, in that they are more likely to display good
intentions towards others to gain their favor, as long as they themselves are not being
treated unjustly. In contrast to the IND and COL measures, the procedural justice
measures were signiﬁcantly related to almost all variables, with the exception of the IND
and COL measures and standardized performance. This is to be expected, because past

studies have shown justice perceptions to be related to each other and the intention

85

.Aoocwe bmeobmv N. 2 Gunmen-6 bweobmv 2 Set emcee :oEB mus—mom Engtoozoo
-EmzesEzcﬁ new 28:2:on mung-58 2: no 53388 2: 53, 82mm bmcobmv m 8 Acouwmmac bweobmv _ Bot owned 338
=< .mo. :2: m8. 20>".— ooemomemm oars: 28:22.80 con—em .Em 8 new Bot emcee 92 $805623 5 one EVE—526M .802

 

8.- 8. 3.- 8. 2. 8.- 2.- 8.- 8.. oo. 8. 8.- 82 8e Savagescoaaéeaaw2
. . . . . . . . . . . . . . cussed-sod ceases 5
:3 S S 2. mm 2 em 2 2 3 we so 2: RN masseuseaas .2
. . . . . . . . , . . . . . cameos-2-2a ceanaawco
63 8 :- 2 8 an mm 2 2 2 S .8 N2 messageaasi
5c 8. 3.. 3. an. E. 2. 8. 8. 8.- 5. EN 005333233292

. . . . . . . . .- . . assess-mom

63 a. 2 a. 2 2 so 8 8 of a; BESZESQSE a
xown—ﬁoomuem

0039.:- 3.560095 .w
. . . .- .- . .- . . 28% 83353
am v 3 2 S 2 8 S 2 _ S. N ceases session .5
63 2. E. 8. S. S.- 82 New 5833. caesium e

33 em. 3. 3. 8. :2 8m 98% ceazegé

seesaw .m

goose conﬁsﬁﬁwev
3522280 .v

$3 2. cm. 2. 2. :. so. we. a: Ed

63 E. S. S. :2. 3m

83 2. 2. 2. 5. SEE: masseuse
63 2.- cm. 2:. assesses
$8 a. a? 822222: ._

 

2 Z S m w n o m w m N _ dd .2

 

mecwaﬁmtocumuﬁ ~28 .nm.:.-.5e2m- $33.26an £8” .~ mﬁﬁ

86

measures included in this study. Procedural justice pre-feedback was related to test-
taking motivation (r=.11, p<.05), although this correlation was very small, while
procedural justice post-feedback was not. Ratings of procedural justice pre and post-
feedback were highly correlated (r=.69, p<.05), although the pattern of correlations of
these variables with other variables was different.

In general, procedural justice perceptions pre-feedback were more highly
correlated with intentions pre-feedback, while procedural justice perceptions post-
feedback were more highly correlated with intentions post-feedback, which is to be
expected. Additionally, procedural justice perceptions both pre and post feedback were
signiﬁcantly related to participants’ perceptions of consistency, interpersonal sensitivity,
and explanation provisions. Distributive justice perceptions were related to procedural
justice perceptions both pre (r=.44, p<.05), and post (r=.60, p<.05) feedback.
Interestingly, post-feedback measures of intentions were more strongly related to
procedural justice perceptions post-feedback than distributive justice perceptions. Test
performance standardized, however, was unrelated to all of the measures. This is
consistent with other ﬁndings; for example, Smither et al. (1993) found a very weak
relation between actual test scores and procedural fairness, and no relation to
recommendation intentions or organizational attractiveness. The meta-analysis by
Hausknecht et al. (2004) also showed a correlation of only .13 between procedural justice
perceptions and actual procedural performance.

Correlations between scale scores and several demographic variables as well as

variables related to selection experience are presented in Table 2.

87

.moweﬁ

Boom no.2 m 528%? 2 8.82 885 .3. :2: m8— m8>2 8585288 88: mecca—oboe pop—om .Em 8 8m 80¢ emcee 8.2 882

 

 

.- . .- . . .- .- .- .- .- 888858
2 8 8 8 2 2 on : 8 2 858858
8388983
2. 8. :. o_.- 2.- :. 2. .2. 2.- 8. 88585
mac-833 2203202.:
8388-2-85
8. 8. 2. 8.- 2.- :. 2. o_. 2.- 2. 8885...ch
"6.838 2203202.:
2. 8. o_. 8.- 2.- 8. 2. 8.- o_ .- 8. 882 0225.85
. . . .- .- . . . .- . 8388-88
2 so 2 8 om 2 8 2 2 2 822 33888
. . . . .- . . . .- . 8888-2-28
2 oo 2 oo 2 8 2 8 2 8 882 88888
. . . .- . . . . .- . 385 ooosaeosev
8 oo oo 8 8 8 oo so 8 8 Eases 88588
oo. 8.- 2. :.- 2.- 2. 2. 8. 2.- 8. 8283. 88588
.- .- . .- .- . . .- .- .- 38;... 83888
8 o_ 2 so 2 oo 2 : 8 8 82:88
.- .- . . .- . . .- . .- 288 882858
8 8 8 oo 2 8 2 8 8 2 85288
so. 8.- so. 8.- 8.- 8.- 8. 8.- 8.- 8.- 88282 823-88
8.- 2.- oo. 8.- 8.- oo. 2. 2.- 8. 2.. 538280
8. _ _. 8.- 8.- 8.- so. 8. 8. 8.- 8. 5882285
oceanomxm cocoComwm cocoComxm . . owmstW-H
. . . «£0 m D 5 8.528220 .028m
89-. 3838:: x83 . . 82m . . . . 8.280 ow<
:50 28> . m D S :8.»
Sumac—om SE com-E smzwem .

 

8.89m. 88% ~28 noSQ-exonmQ 582.8% 88.238800 .N mace-m

88

Relationships in this correlation matrix are similar to the previous one in that IND
was not related to any variables, while COL was signiﬁcantly related to only a few of the
demographic variables. COL was moderately negatively related to prior interview
experience (r=.28, p< .05), indicating that those who scored higher on COL were less
likely to have participated in a job interview previously. The other variables however,
with the exception of the explanation provision and consistency manipulation checks
were signiﬁcantly related to several of the demographic variables.

Age was related to both procedural justice and intentions post-feedback, as well
as intentions pre-feedback. All these relations were positive, indicating that older
individuals were more likely to have higher scores on these scales than younger
individuals. Age was negatively related, however, to standardized test performance (r=-
.14, p<.05), indicating that older people were also likely to score worse on the cognitive
ability test. Gender was related to procedural justice and intentions pre and post-
feedback, and explanation adequacy post-feedback. All these correlations were negative,
indicating that men were more likely to rate these items higher than women.

Some other correlations of note are between US. citizenship and procedural
justice post-feedback (r=.22, p<.05), explanation adequacy (r=.l9, p<.05), and test
performance (r=-.20, p<.05). The directions of these relationships suggested that those
who were non US. citizens gave higher ratings of procedural justice post feedback, and
explanation adequacy, but scored lower on the test of cognitive ability. Those individuals
who were non US. citizens were most likely international students, and thus, may not
have been as proﬁcient with English, causing them to score lower on the test. In fact,

there is a moderate negative correlation between English proﬁciency and test

performance (r=-.26, p<.05), indicating that those who did not identify English as their
primary language performed worse. The relation between citizenship and procedural
justice post-feedback and explanation adequacy could be that those who have not lived in
the US. for very long could not be familiar with common selection practices here, and
could therefore have lower expectations for fairness than those were are used to the
customs of the United States. Further support is evidenced in the moderate, positive
correlation between prior work experience in the US. and procedural justice perceptions
post-feedback (r=.23, p<.05), which indicates that those who have not had work
experience in the US. had higher procedural justice perceptions post-feedback than those
who had obtained work experience. In addition, the longer individuals had been in the
US, the lower their procedural justice perceptions after feedback (r=-.20, p<.05).
Manipulation Checks

In order to determine whether the three manipulations conducted in the present
study were effective, separate t-tests were conducted on each of the consistency,
interpersonal sensitivity, and explanation provision measures. Results of these t-tests are
located in Table 3 below. The results indicate that the manipulations worked as intended.
All t-values were negative and signiﬁcant, indicating that in the consistent conditions
individuals gave higher consistency ratings, in the sensitive conditions, individuals gave
higher sensitivity ratings, and in the conditions in which an explanation was given,
individuals were more likely to rate that an explanation was given to them. Test-taking
motivation was also included as a manipulation check, in order to determine whether the
reward offered to participants was sufﬁcient to motivate them to perform well on the

selection test. Mean motivation was 4.51 on a seven-point scale, indicating that mean

90

levels of motivation were greater than the midpoint on the scale, however, motivation to

perform well was not high. This will be discussed further later in the paper.

Table 3. Mean Diﬂ’erences on Manipulation Checks.

 

 

 

Manipulation Check Condition M SD df T
Consistency Consistent 4.25 .59 310 -.4.65*
Inconsistent 3.79 .85
Sensitivity Sensitive 3 .96 .66 308 -l 1.80*
Not Sensitive 2.72 1.1 1
Explanation Explanation 2.58 , 1.20 309 -7.96*
No Explanation 1.57 .79
Note. *p<.01
Hypothesis Testing

Hierarchical multiple linear regression was used to test hypotheses 1 and 2. The
categorical variables of consistency and interpersonal sensitivity were dummy coded,
with the consistent and interpersonally sensitive conditions being coded 1 and the
inconsistent and interpersonally insensitive conditions being coded 0. Explanation was
coded in a similar manner; when‘an explanation was given, the condition was coded 1,
when an explanation was not given, the condition was coded 0. Thus higher codes reﬂect
more “procedurally fair” conditions.

In the current study, several outcomes, including procedural justice perceptions
were investigated at two time points: before participants received information about their
performance on the selection test and their selection for the simulated job, and after they
received this feedback. Justice perceptions post-feedback are of greatest interest, because
these are the justice perceptions applicants leave the selection context holding, and the
perceptions which presumably have the greatest effect on their subsequent behavior.

However, both time periods were examined because fairness of procedural consistency,

91

and thus other variables related to fairness, could be viewed differently depending on i
when in the selection process the fairness measure is given. It is possible that participants
scrutinize the procedural process much more after they learn their outcome than when the
selection process is still underway (Ployhart & Ryan, 1998). Therefore, the relationship
between consistency, sensitivity, explanation, and procedural justice pre and post-
feedback could differ. Consequently, the results for the hypotheses for dependent «
variables both pre and post feedback are presented in one table.
Hypothesis 1
Hypothesized Effects of Con_sistencv

Hypothesis 1 involves the relationship between consistency and procedural justice
perceptions. Results are displayed in Table 4. There were no signiﬁcant relationships to
procedural justice perceptions pre-feedback. There was a main effect of consistency on
procedural justice post-feedback (b=.33, p<.05). Collectivism did not Show a main effect
on procedural justice perceptions post-feedback (b =-.10, n.s.). Thus, it seems that neither
consistency of procedural processes nor collectivist orientation by itself was enough to
cause a change in procedural justice perceptions pre-feedback. However, consistency did
affect notions of procedural justice post-feedback, such that those individuals in the
consistent conditions rated the selection procedure as being fairer than those in the
inconsistent conditions. Nevertheless, this main effect should be interpreted in light of
the interaction presented in the following paragraph.

Hypothesis la states that collectivism will be a moderator of the consistency-

procedural justice relationship. This hypothesis was not supported for procedural justice

92

Table 4. Moderated Regressions of Procedural Justice Perceptions on Consistency,
Collectivism, and Individualism

 

 

 

 

Model b . B R2 ARZ
Procedural Justice Percep
m
Step 1
Consistency .23 .10
Collectivism .1 7 .09 .02
Step 2
Consis x Collect -.41 -.90 .03 .011
Step 1
Consistency .24 .1 l
Individualism .14 .08 .02
Step 2
Consis x Indiv -.29 -.63 .02 .005
Procedural Justice Percep
.1325}.
Step 1 -
Consistency .33 * . 1 5 **
Collectivism .02 .01 .02
Step 2
Consis x Collect -.23 -.52 .03 .004
Step 1
Consistency .33** .15**
Individualism —. 1 O -.06 .03
Step 2
Consis x Indiv .47* 1.06* .04 .014*

 

Note. N= 303. 0=Inconsistent, 1=Consistent. * p<.05. **p<.01.

93

 

 

 

 

 

 

 

. t " . i

8 2.5 ‘#——C—ZH\. -—-——_‘ﬁ
o.-
8
'5 2 . ' ‘
.3 —I— Inconsistent
_. 1.5 .
E ----A~ Comment
I:
"U 1
0
8
I-
n 0.5 - m

o -

Low High

Individual'sm Individualism

Figure 2. Interaction Between Consistency and Individualism on Procedural Justice Post

94

perceptions either pre or post-feedback. Part b of the ﬁrst hypothesis involving the
relationship between consistency and procedural justice perceptions states that
individualism will be a moderator of this relationship. Individualism was a moderator
for procedural justice perceptions post feedback (b=.47, p<.05), but was not for
procedural justice perceptions pre-feedback (b=-.29, n.s.). However, the nature of the
interaction on post-feedback perceptions, which is presented in Figure 2, was not
consistent with the predicted direction of the hypothesis. Therefore, Hypothesis 1b was
not supported. In fact, this interaction shows that in inconsistent conditions, those who
are high on individualism are more likely to have lower procedural justice perceptions
than those who are low on individualism, while the hypothesis predicted the opposite
effect. By contrast, in consistent conditions, the opposite effect is found in that those
who are high on individualism actual have higher procedural justice perceptions post-
feedback than those who are lower on individualism. These effects were relatively small,
however.

One explanation for this ﬁnding may be that those who are high on IND are more
attune to the need for equity in distribution than those low on IND. Because they were
given less time in the inconsistent condition, they knew they were at a disadvantage in
terms of performing on

the test, and thus were not likely to get scores that were in line with their true ability.
Therefore, in the inconsistent conditions, they may have felt “cheated out” of their true
score by not being treated equitably. Those low on IND, in contrast, may be more aware
of the need for equality, and thus they felt that as long as they were being treated the

same as everyone else in their experimental group, it did not matter to them whether they

95

received less time. In essence, they may not have been as concerned about whether their
score reﬂected their ability, as long as they were treated the same as others in their
condition.

Non-hypothesized Effects of Conﬁtencv

There were several other relationships examined in this study for which there
were no a priori hypotheses, but were of interest due to their importance both in previous
research and to those in applied settings. These variables included distributive justice
perceptions, future experiment intentions, recommendation intentions, and job acceptance
intentions (which were combined) and selection test performance. Results from these
analyses are presented in Tables 5 and 6. Note that results for test performance are not
included in these tables, because none of the relationships between the independent
variables or their interactions and test performance were signiﬁcant.

Although there was no main effect of consistency, IND, or COL on distributive
justice perceptions, the interaction between consistency and COL was signiﬁcant (b=-
.26, p<.01). The nature of the interaction is represented in Figure 3. The graph indicates
that in inconsistent conditions, those individuals with a higher collectivist orientation
actually had higher distributive justice perceptions than those with lower collectivist
orientations, yet in the consistent conditions, they had lower distributive justice
perceptions. This is the opposite of what one would expect based on the hypotheses
involving procedural justice, because distributive and
procedural justice are often correlated, as they are in this case as well (pre-feedback,

r=.44, p<.05; post-feedback, r=.60, p<.05). The individuals who were high on COL

96

(ﬂ) ‘1’ (P'r (J)

‘) L/ r
k I

I:

Table 5. Moderated Regressions of Distributive Justice Perceptions on Consistency,
Collectivism, and Individualism

 

 

 

Model b [3 R7 ARZ
Distributive Justice Percep
Step 1
Consistency .06 .05
Collectivism .02 .01 .00
Step 2
Consis x Collect -.26* -.95* .02 .012*
Step 1
Consistency .06 .05
Individualism -.08 -.08 .01
Step 2
Consis x Indiv -.07 -.25 .01 .001

 

Note. N=303. 0=Inconsistent, 1=Consistent. * p<.05. **p<.01.

97

Table 6. Moderated Regressions of Intentions towards the Organization on Consistency,
Collectivism, and Individualism

 

 

 

 

Model b p R2 ARZ
Intentions Pre
Step 1
Consistency .29* . 14*
Collectivism . 19* . l 1 * .04
Step 2
Consis x Collect -.09 -.22 .04 .001
Step 1
Consistency .30** . 15 * *
Individualism .11 .07 ' .03
Step 2
Consis x Indiv -.28 -.68 .03 .006
Intention_§ Post
Step 1
Consistency .32* . 14*
Collectivism . l 3 .07 .03
Step 2
Consis x Collect -.32 -.68 .03 .006
Step 1
Consistency .33 * . 14*
Individualism .08 .04 .02
Step 2
Consis x Indiv -.13 -.27 .02 .001

 

Note. N= 303. 0=Inconsistent, 1=Consistent. * p<.05. **p<.01.

98

2.9 s

2.85

2.75

2.8

 

2.7

 

 

2.65

 

 

Distributive Justice

2.6

 

 

 

2.55 .

Low High

Collectivism Collectivism

—I— Inconsistent
- - -A- -- Consistent

Figure 3. Interaction Between Consistency and Collectivism on Distributive Justice

99

may have responded in this way because they are more aware of external factors that can
affect situations than are those who are lower on COL, and they are also less concerned
about the need for equity in distribution. Thus, they may have been more understanding
in the inconsistent conditions and attributed the inconsistency to uncontrollable external
factors rather than blame it on the administrator, and thus, been more lenient when it
came to their distributive justice ratings. Whereas, for those with a low COL orientation,
the lack of equity in the distribution of rewards could have more salient, resulting in
lower distributive justice ratings from this group. It may be also that those who are high
on COL are driven less by pure competitiveness than those low on COL, because they are
less concerned about “being the best” and so consistency did not matter to them as much
because they did not care about the outcome. This may have been enhanced by the fact
that this was not a high-stakes situation where the outcome was important to the
individuals.
Hypothesis 2
vaothesized Effects of Sensitivity

The second hypothesis refers to the relationship between the interpersonal
sensitivity of the selection process administrator, individuals’ levels of IND and COL,
and their perceptions of procedural justice. As Table 7 shows, there was no main effect
of sensitivity on procedural justice either pre or‘post-feedback. Levels of IND and COL
were also unrelated to procedural justice perceptions at both measurement periods.
Therefore, the sensitivity of the administrator and participants’ levels of IND and COL

had no direct main effect on procedural justice perceptions.

100

Table 7. Moderated Regressions of Procedural Justice Perceptions on Sensitivity,

Collectivism, and Individualism

 

 

 

 

Model b g R2 ARZ
Procedqu Justice Percep
he
Step 1
Sensitivity .06 .03
Collectivism .18 .10 .01
Step 2
Sensit x Collect .04 .09 .02 .008
Step 1
Sensitivity .05 .03
Individualism . 14 .08 .01
Step 2
Sensit x Indiv .32 .78 .01 .008
Procedural Justice Percep
139.81
Step 1
Sensitivity -. 1 O - .05
Collectivism .03 .01 .00
Step 2
Sensit x Collect .O7** .15** .03 .022“
Step 1
Sensitivity -.1 l -.05
Individualism -. 10 -.O6 .01
Step 2
Sensit x Indiv .14 .34 .01 .001

 

Note. N=303. O=Insensitive,1=Sensitive.

*p<.05. **p<.01.

101

2.65 ~

 

2.6 ———————————— ._

2.55 —_._.__A_'l -______

 

2.5

 

—I— Insensitive
---A- -- Sensitive

2.45

 

 

2.35

 

Procedural Justice Post

2.3

 

 

 

2.25

Low High
Collectivism Collectivism

Figure 4. Interaction Between Sensitivity and Collectivism on Procedural Justice Post

102

Hypothesis 2a states that individuals’ levels of COL will moderate the
relationship between sensitivity and procedural justice perceptions. While this was not
true of procedural justice perceptions pre-feedback (b=.04, n.s.), it was true of procedural
justice perceptions post-feedback (b=.07, p<.05). Figure 4 displays the interaction
between these variables. The graphs shows that in the insensitive conditions, those with
higher collectivist orientations had slightly lower procedural justice perceptions than did
those with lower collectivist orientations. In the sensitive conditions, the opposite pattern
is seen, in that those who report higher procedural justice perceptions are higher on
collectivism than those who report lower procedural justice perceptions. Thus,
Hypothesis 2a, which stated that collectivism would moderate the relationship between
sensitivity and procedural justice perceptions was supported for perceptions post-
feedback.

Hypothesis 2b states that IND will moderate the relationship between sensitivity
and procedural justice perceptions. As Table 7 shows, there was no significant
relationship between the interaction of sensitivity and IND and procedural justice
perceptions.

Non-hypothesized Effects of Sensitivity

Regarding the relationship between sensitivity, IND and COL, and the non-
hypothesized variables, there were no main effects of sensitivity or IND on any of the
dependent variables, however there was a main effect of COL on intentions toward the
organization pre-feedback (b=.20, p<.05). This main effect indicates that those

individuals that were higher on collectivism were more likely to indicate that they would

103

Table 8. Moderated Regressions of Distributive Justice Perceptions on Sensitivity,

Collectivism, and Individualism

 

 

 

Model b [3 RI ARI
Distributive Justice Percep
Step 1
Sensitivity -.04 —.03
Collectivism .02 .02 .00
Step 2
Sensit x Collect .01 .04 .00 .001
Step 1
Sensitivity ' -.04 -.03
Individualism -.08 -.08 .01
Step 2
Sensit x Indiv -.04 -.15 .01 .000

 

Note. N=303. O=Insensitive, 1=Sensitive. * p<.05. **p<.01.

104

Table 9. Moderated Regressions of Intentions towards the Organization on Sensitivity,
Collectivism and Individualism

 

 

 

 

Model b a R2 AR’
Intentions Pre
Step 1
Sensitivity . 10 .05
Collectivism .20* .12* - .02
Step 2
Sensit x Collect .06* .13* .04 .018*
Step 1
Sensitivity .10 .05
Individualism .11 .07 .01
Step 2
Sensit x Indiv -.11 -.29 .01 .001
Intention_s Post
Step 1
Sensitivity .02 .01
Collectivism .14 .08 .01
Step 2
Sensit x Collect .06* .13* .01 .016*
Step 1
Sensitivity .01 .01
Individualism .08 .04 .00
Step 2
Sensit x Indiv -.30 -.70 .01 .006

 

Note. N=303. O=Insensitive, 1=Sensitive. * p<.05. **p<.01.

105

3.3 1

3.2

3.1

2.9

Intentions Post

2.8

2.7

2.6

 

 

 

/-
/

 

 

 

 

I

Low Collectivism High Collectiv'nm

——I— Insensitive
— - 'A' - - Sensitive

Figure 5. Interaction Between Sensitivity and Collectivism on Intentions Post

106

accept the job if offered to them, participate in a similar experiment to this one in the
future, and recommend the experiment to others.

In addition to the main effect of COL, there were a few interactions involving
COL and sensitivity that were signiﬁcant. All the signiﬁcant interactions are presented in
Table 9, as they all involved effects on intentions. Sensitivity and COL interacted to
affect intentions both pre (b=.06, p<.05), and post (b=.07, p<.05) feedback. Because the
nature of these interactions are nearly identical, Figure 5, which presents the interaction
on intentions toward the organization post-feedback, is also representing the interaction
on intentions pre-feedback. In Figure 5, the sensitive and insensitive lines look almost
parallel, but this is due to the small range of responses on COL for this sample. One is
able to see that the difference in recommendation intentions post-feedback for sensitive
versus insensitive conditions is greater for those who are high on COL versus those low
on COL. Although in both conditions, those high on COL have higher recommendation
intentions than those low on COL, those high on COL are more likely to have a greater
drop in scores when conditions are insensitive, than those low on COL. This result
follows a similar trend to results for procedural justice perceptions, in that it shows those
high on COL are more sensitive to changes in interpersonal treatment. They are more
aware of the condition of their relationship with the test administrator and respond to
differences in this relationship more strongly than those low on COL, who do not care as

much.

107

Hypothesis 3
Hypothesized Relationships of Explanation

The third hypothesis deals with how giving an explanation for the procedural
process can increase procedural justice perceptions, and how this effect occurs through
perceived explanation adequacy.

In order to test for the mediation, the Baron and Kenny (1986) method was used.
According to this method, the mediator should be regressed on the independent variable,
the dependent variable should be regressed on the independent variable, and the
dependent variable should be regressed on both the mediator and the independent
variable. To conclude that there is mediation, the independent variable must affect the
mediator (ﬁrst regression), the independent variable must affect the dependent variable
(second regression), and, in the third regression, the mediator must affect the dependent
variable. If the independent variable has no effect on the dependent variable when the
mediator is controlled for, then complete mediation can be established (Baron & Kenny,
1986)

Results of these analyses are presented in Table 10. The ﬁrst regression,
regression 1, involved the regression of explanation adequacy on explanation, and this
regression was

signiﬁcant (b= .51, p<.05). The second regression, regression 2, involved the regression
of the IV, explanation, on the DV, in this case procedural justice both pre (regression 23)
and post (regression 2b) feedback. Both these regressions were signiﬁcant (b= .31,

p<.05; b=.27, p<.05, respectively), and had positive regression slopes, indicating that

108

Table 10. Test of Mediation of Explanation Adequacy

 

Model B B RI

 

1. Explanation Adequacy
Explanation .51** g .22“ .05

2a. Procedural Justice Pre
Explanation .31 * . 14* .02

3a. Procedural Justice Pre
Explanation .09 .04
Explanation Adequacy .32 .35" .13

2b. Procedural Justice Post
Explanation .27* .13* .02

3b. Procedural Justice Post
Explanation .05 .02
Explanation Adequgy .37** .40“ .17

 

Note. N= 303 *p<.05, **p<.01
hypothesis 3a, which stated that those individuals who received explanations for the
selection procedure would have higher procedural justice perceptions, was supported. In
order to test for mediation, the third regression, involving the addition of the mediator
into the equation, had to show that when the mediator was added into the equation, the IV
no longer was signiﬁcant. This, in fact, was what occurred for both procedural justice
perceptions pre- (explanation, b=.09, n.s.; explanation adequacy, b=.32, p<.05), and
post-feedback (explanation, b=.05, n.s.; explanation adequacy, b=.37, p<.05).
Therefore, explanation adequacy was a full mediator of the effect of explanation on
procedural justice perceptions, and hypothesis 3b was also supported.
Non-hypothesized Effects of Explanation

Although the moderating effects of IND and COL on the relationship between

explanation and process outcomes were not hypothesized, there is reason to investigate

109

these relationships. Explanation may function similarly to sensitivity, in that it is a way
for employers to show applicants that they care about applicants and are aware of their
concerns. Giving an explanation for the selection process may relay a message to
applicants that the organization is trying to make applicants feel as comfortable as
possible with the process that they are undergoing. This may be especially important for
those with a high collectivist orientation, because they are very aware of external cues
and situational factors that may convey messages (Triandis, 1995). They are also very
cognizant of the need to “save face” in ﬁont of others and the importance of maintaining
good relations, and therefore, may be more receptive to organizations’ attempts to explain
their selection processes. Those who are high on individualistic orientation may be more
attuned to internal cues, and care more about the effect the procedural processes have on
them, rather than any messages the explanation for the procedural processes may be
relaying. Consequently, based on the above rationale, one would expect explanation to
function similarly to sensitivity in how it relates to IND, COL, and selection process
outcomes such as justice perceptions and recommendation intentions.

Thus, for exploratory purposes, the same relationships that were examined for
interpersonal sensitivity were tested for explanation. Thus, the effects of explanation and
the cultural orientation variables on distributive justice and intentions toward the
organization are presented in Tables 12 and 13, respectively. Aside from procedural
justice perceptions (Table 11), explanation had no main effect on any selection process
outcome variables. The main effects of IND and COL on these variables have been
explained in earlier sections as well, therefore only results from the interactions will be

presented.

110

Table I 1. Moderated Regressions of Procedural Justice Perceptions on Explanation,
Collectivism, and Individualism

 

 

 

 

Model b B R2 ARZ
Procedural Justice Percep
m
Step 1
Explanation .30* .14*
Collectivism .1 7 .09 .03
Step 2
Explan x Collect -.18 -.41 .03 .002
Step 1
Explanation .30* . 14*
Individualism .1 3 .07 .03
Step 2
Explan x Indiv -.27 -.63 .03 .005
Procedural Justice Percep
m
Step 1
Explanation .27* . 13 *
Collectivism .03 .01 .02
Step 2
Explan x Collect -.075 -.18 .02 .001
Step 1
Explanation .27* .13*
Individualism -.1 1 -.06 .02
Step 2
Explan x Indiv -.04 -.10 .02 .000

 

Note. N=303. 0=No Explanation, 1=Explanation. * p<.05. **p<.01.

111

Table 12. Moderated Regressions of Distributive Justice Perceptions on Explanation,
Collectivism, and Individualism

 

 

 

Model b B R2 AR]
Distributive Justice Percep
Step 1
Explanation .03 .03
Collectivism .02 .02 .01
Step 2
Explan x Collect -.14 -.53 .04 .003
Step 1
Explanation .04 .03
Individualism -.08 -.08 .01
Step 2
Explan x Indiv -.24 -.95 .02 .012

 

Note. N=303. 0=No Explanation, 1=Explanation. * p<.05. **p<.01.

112

Table 13. Moderated Regressions of Intentions towards the Organization on
Explanation, Collectivism, and Individualism

 

 

 

 

Model b B R7 AR?
mtmtionihe
Step 1
Explanation .12 .06
Collectivism .20* .12* .02
Step 2
Explan x Collect .14 .35 .02 .001
Step 1
Explanation . 1 2 .06
Individualism .1 1 .07 .01
Step 2
Explan x Indiv -.40* -1.02* .02 .014*
Intentions Post
Step 1
Explanation .06 .03
Collectivism . 14 .08 .00
Step 2
Explan x Collect .06 .15 .00 .000
Step 1
Explanation .06 .03
Individualism .08 .04 .00
Step 2
Explan x Indiv —.52* -1.21* .02 .02*

 

Note. N=303. 0=No Explanation, 1=Explanation. * p<.05. **p<.01.

113

 

 

+ No Explanation
-----A Explanation

 

Intentions Pre

 

 

23

 

 

2]

 

Low High
Individualism Individualism

Figure 6. Interaction Between Explanation and Individualism on Intentions Pre

3.2 ,
3.1 .

3.

 

29-

23

 

 

 

2.7 —

Intentions Post

2.6 «en—-5 «-3 . - -

 

25»—u~“——«— a“-.. HW__.,_IEL

 

2A

 

Individualism Individualism

Figure 7. Interaction Between Explanation and Individualism on Intentions Post

114

Although there were no signiﬁcant interactions involving COL, IND interacted
with explanation to affect intentions, both pre-feedback (b =-.39, p<.01), and post-
feedback (b=-.53, p<.05). These interactions are displayed in Figures 6 and 7. In the
interaction on intentions pre-feedback, the difference in intentions between those low and
high on IND was greater in the no explanation condition than in the explanation
condition. In the no explanation conditions, those low on IND were less likely to intend
to support the organization than those higher on IND, and in the explanation conditions,
they were slightly more likely to support the organization. In the interaction on
intentions post-feedback, the differences were much more equal. Those low on IND may
behave similarly in some ways to those high on COL in that they may be more aware of
needs of others around them, and be more empathetic to the importance for others to
“save face”. They also may be more appreciative of the attempt of the organization to
inform its applicants, and the signals that that sends. In the no explanation conditions, it
is possible that those low on IND behaved more like those high on COL, in that they saw
the lack of explanation as a signal to them that the organization was not trying to
maintain good relations, and therefore, reacted more negatively than those who were high
on IND, who were not reading into the meaning of the lack of explanation in the same
way.

By contrast, individuals high on IND were more likely to accept the job offer, _
recommend the experiment to others, and participate in a similar experiment when no
explanation was given for the selection process versus when there was an explanation
given for the selection process. It is possible that for those high on IND, the attempt at

giving an explanation may have backﬁred. They could have seen the explanation as a

115

way to attempt to cover up problems and impression manage, rather than a real attempt to
inform applicants. In addition, they may have seen the explanation as an attempt by the
organization to advance its own objectives. This may be true because those high on IND
may have a view of others’ motivations as similar to their own, in that they may feel that
others’ action are guided by their need to get ahead, not by their concern for others.

Thus, they were in a sense penalizing the organization for “faking”.

116

DISCUSSION

The main purpose of the current study was to explore how an applicant’s cultural
orientation can affect how he/she responds to different aspects of the selection process.
Two procedural justice rules from Gilliland’s (1993) model were selected for their
relevance to both procedural fairness perceptions during the selection process and for the
potential differences in reaction due to individualism and collectivism orientation the
violation of these rules could elicit. In addition, a third variable, explanation, was
examined for its relevance in regards to increasing procedural justice perceptions both in
cases where procedural rules were being violated, and in cases in which procedural rules
were being followed. A few other variables that were relevant to intentions following the
selection process were also examined as outcomes.
Consistency Effects

Results for all hypotheses are summarized in Table 14. Results indicated that
consistency, the ﬁrst rule contained in Gilliland’s (1993) model that was examined, had
differing main effects depending on the time at which procedural justice ratings were
obtained. Although there was no main effect on consistency for procedural justice
perceptions pre-feedback, there was an effect for these perceptions post-feedback. One
of the possible reasons for this difference is that participants could have become more
critical and analytical in general when responding to the procedural fairness questions
post-feedback because at that time, unlike pre-feedback, they knew that the outcome was
negative. Thus, they may just not have been as concerned about justice pre-feedback.
This notion is lent credence by ﬁndings from Bauer et al. (1998) that procedural justice

perceptions both pre and post-feedback were related to consistency. In the Bauer et al.

117

Table 14. Summary of Results for Hypotheses
Hypothesis Result

Hla: Collectivism will moderate the relationship between consistency

and procedural fairness perceptions such that the relationship will be

stronger for those who are high on collectivism than those who are low

on collectivism Not Supported

Hlb: Individualism will moderate the relationship between consistency

and procedural fairness perceptions such that the relationship will be

stronger for those who are low on individualism than those who are high

on individualism Not Supported

H2a: Collectivism will moderate the relationship between sensitivity and

procedural fairness perceptions such that the relationship will be

stronger for those who are high on collectivism than those who are low

on collectivism Supported

H2b: Individualism will moderate the relationship between sensitivity

and procedural fairness perceptions such that the relationship will be

stronger for those who are low on individualism than those who are high

on individualism Not Supported

H3a: Those who receive an explanation for the procedural process will
perceive procedures as more fair than those who do not receive an
explanation Supported

H3b: The explanation effect will be mediated by the perceived adequacy
of the explanation Supported

 

118

(1998) study, participants were actual applicants for an ofﬁce position at an accounting
department, and consequently, had reason to care about fairness throughout the process,
whereas in this study, participants did not have much to lose if the process was unfair,
and thus had less of a reason to care throughout the process.

It also may have been the case that the participants were attending to different
aspects of the selection process when evaluating fairness pre versus post-feedback.
Before learning their outcome, they could have been focusing on the more consistent
aspects of the process when making their ratings, while after the outcome, in part to
explain their rejection, they may have focused on the more inconsistent aspects. Indeed,
the mean procedural justice rating pre-feedback (M=2.74) was a third of a standard
deviation higher than the mean rating post-feedback (M=2.43). In addition, Gilliland
(1993) states that procedural rules become much more vital in determining justice
perceptions when the outcome is negative than when positive. Pre-feedback, the outcome
was unknown, and so could have been assumed by participants to be positive at that
point, whereas post-feedback, the outcome was known to be negative.

Although effects on justice perceptions pre-feedback are interesting, it is really
effects on perceptions post-feedback that are of greater import, due to their inﬂuence on
subsequent actions and long-term outcomes. The main effect for consistency was
evidenced on procedural justice post-feedback, however, which replicates several other
ﬁndings (Bauer et al., 1998; Oliver, 1998;Ryan & Chan, 1999), including one by Ployhart
and Ryan (1998) that inconsistency only affects procedural justice ratings when the
applicant is somehow put at a disadvantage by the inconsistency. This ﬁnding is unique,

however, in that it was found that this effect was moderated (moderation effects will be

119

addressed below). There was also a main effect of consistency on intentions toward the
organization both pre-feedback and post-feedback. This is an interesting ﬁnding
considering that previous studies have generally related these types of intention variables
to procedural rules through perceptions of procedural justice (Macan et al., 1994; Truxillo
& Bauer, 1999; Fiaschetti, 1998). The main effect indicated, as expected, that those in
the consistent conditions were more likely to accept a job offer, recommend the
experiment to others, and participate in future experiments than those in the inconsistent
conditions.

Even more interesting than the main effects of consistency on procedural justice
were the interactions. Hypothesis 1a, which stated that COL would be a moderator of the
relationship, and hypothesis 1b, which stated that IND would be a moderator, were not
supported. However, there was an interaction found between IND and consistency, but
not in the hypothesized direction. Why there was no interaction found between COL and
consistency could relate to many factors. One is that it is difﬁcult to ﬁnd signiﬁcant
interactions in general. In a recent article, Aguinis, Beatty, Boik, & Pearce (2005)
reviewed 30 years of moderating effects, and found that the median observed effect size
was only .002. It is also possible that potential interactions could not be detected within
the range of COL that was represented by this sample. As was mentioned earlier, the
mean of COL was 4.86 with a standard deviation of .56, meaning that 68% of the sample
fell between the values of 4.30 and 5.42 on a seven-point scale. This is not a large range,
and therefore, much of the scale is left unrepresented. In addition, because it seems that
it is the middle point of the scale that is represented, and not more of the extreme ends,

this may have made it even more difﬁcult to see how high and low COL contrast,

120

because it was really only “slightly high” and “slightly low” COL that was being
contrasted.

Despite the lack of ﬁnding COL as a moderator, IND was found to be a moderator
of the consistency-procedural justice perception relationship post—feedback. Instead of
those who were high on IND having higher procedural justice perceptions than those
lower on IND in the inconsistent conditions as hypothesis 1b stated, the opposite
occurred. Those who were high on IND actually had lower perceptions of procedural
justice in the inconsistent conditions than did those low on IND. A reason for this could
be that the high individualist’s need for equity in a situation overshadows his/her lack of
a need for convention or reliance on others for fairness. Ample evidence indicates that
those high on IND prefer merit-based hiring to hiring based on equality (e. g., Hui et al.,
1991; Ramarnoorthy & Carroll, 1998; Triandis & Sub, 2002). Moreover, it has also been
found that individualists tend to prefer feedback that is self-serving, and rate it as being
more fair than feedback that is critical (Gelfand et al., 2002). Although there was no
main effect of IND on procedural justice perceptions, this may have been somewhat of a
contributing factor, because all the feedback was negative. In addition, if those lower on
IND function similarly to those higher on COL, then perhaps they were less likely to
criticize the process because they were more understanding of the inconsistency.
Research has shown that those high on COL pay a great deal of attention to external
factors when considering causes for a situation, and are less likely to commit the
fundamental attribution error of blaming circumstances on the internal characteristics of
the person rather than the external characteristics of the circumstance (Earley & Gibson,

1998; Triandis, 1989; Triandis & Sub, 2002; Wagner, 1995). Thus, they may have been

121

TELEIIC

le‘IZfE

more likely to feel that the inconsistency was the result of unforeseen circumstances
rather than the ineptitude of the administrator, and consequently, were less likely to
indicate that their was unfairness in the process, because saying this would be blaming
the administrator.

Although COL was not a moderator for the consistency-procedural justice
relationship, it was a moderator of the relationship between consistency and distributive
justice. The direction of the moderation indicated that in the inconsistent conditions,
those high on COL had higher distributive justice perceptions than those low on COL. In
addition, in the consistent conditions, those high on COL had lower distributive justice
perceptions than those low on COL. This may be because those high on COL are more
likely to focus on blaming themselves for poor performance rather than the distribution
process. Research has shown that those who are high on IND are more likely to attribute
cause to external factors when the outcome is negative, whereas those who are higher on
COL are more likely to attribute negative outcomes to internal factors (Bond, Leung &
Wan, 1982). Their ratings of distributive justice in the consistent conditions could have
been lower because one aspect of the distributive justice questionnaire has to do with
being satisﬁed with the outcome. They may have been even more dissatisﬁed with the
outcome in the consistent conditions because they were given the full time, and they still
did not perform well. Studies have indicated that those who are high on COL are more
likely to show a decrease in self-esteem and feel embarrassed after a failure than those
high on IND (Singelis & Sharkey, 1995; Singelis et al., 1999). Consequently, they could
have been feeling worse about the outcome and themselves after learning they performed

poorly than those low on COL, and thus had lower distributive justice perceptions.

122

 

l'l it
COL

LT;
h-IJ.

precl'

“f‘fl!
M IUE

.
‘1

-o
. A.“

ﬁt":
b-L\

\ 1.11},

Dex:

.

.
lT-l
a,“ *U
.

\l‘
V \-

Overall, it seems that, unlike originally hypothesized, consistency is a rule that
has greater impact for those high on IND than for those high on COL. Taking into
account that someone can, in theory, be high on both characteristics, it is possible that
their high IND would come into play more when reacting to inconsistency than their high
COL. There are several reasons why this may be the case. One is that those who are
high on IND are more likely to be concerned about equity, and lack of consistency often
precludes equity. If one is disadvantaged due to an inconsistent process, one’s
performance will not reﬂect one’s ability or effort, and thus, the outcome will not either.
Another reason is the attributions those high COL are likely to make for inconsistent
processes. Those who are high on COL are very in tune with the power and inﬂuence of
external factors in a situation. Thus, they may be less likely to blame the individuals
conducting the process by stating that the process was unfair, because they realize that
these individuals very likely may not have had any control over the inconsistency.
Consequently, consistency may be less strongly tied to their procedural justice
perceptions. Third, it is possible that those who are high on IND tend to be more
competitive in situations, because achieving personal recognition and accomplishing
personal goals are very important to them, whereas those high on COL are less so,
because they often have to sacriﬁce their personal goals for the goals of their group
(Triandis et al., 1985). Although the effect sizes for the interactions were not large,
(AR2=.012-.014) they did represent a signiﬁcant increase in the variance accounted for.
Sensitivity Effects

Similar to consistency, the effects of sensitivity were signiﬁcant for procedural

justice perceptions post-feedback. Although there was no main effect of sensitivity or

123

IND and COL on procedural justice perceptions, there was a signiﬁcant interaction.
Sensitivity interacted with COL to affect procedural justice perceptions post-feedback.
The nature of the interaction was in support of hypothesis 2a. In other words, in the
insensitive conditions, those who were higher on COL had lower procedural justice
perceptions than those low on COL. In addition, in the sensitive conditions, those high
on COL had higher procedural justice perceptions. This conﬁrms the notion that
interpersonal relations and maintaining harmony is very important to those high on COL,
and therefore, they react negatively when harmony is disrupted due to rudeness and
positively when extra effort is being made to maintain it. Moreover, collectivists are
more familiar with the use of sensitivity in face-saving communication, in order to spare
the feelings of others and to remain considerate of others’ circumstances (Triandis & Sub,
2002). They are also, in fact, much less comfortable with the type of overt hostility
portrayed in the insensitive conditions than they are with more passive hostility (Doucet
& J ehn, 1997). Those who are low on COL, however, may not be as aware of the
interpersonal context and thus, are not as reactive to it. There was no moderating effect
of IND on procedural justice perceptions, however, which meant that hypothesis 2b was
not supported. This may be because those who are high on IND are not necessarily less
aware of the interpersonal context than those who are low on IND, but at both levels, they
are not especially sensitive to the need for interpersonal harmony.

In terms of other outcomes related to procedural justice, there were a few effects
of note. For instance, there was a main effect of COL on intentions toward the
organization pre-feedback. Those who were higher on COL were more likely to intend to

accept the job offer, recommend the experiment to others, and re-participate than those

124

low on COL. This again could be related to collectivists desire to maintain harmony and
avoid conﬂict by endorsing items that refer to recommending the experiment to others,
re-participating in the experiment, and accepting the potential job offer. In addition to the
main effect, COL also interacted with sensitivity to affect intentions both pre and post-
feedback. Similar to the interaction on procedural justice perceptions post-feedback,
these interactions indicated that, although at both levels of COL intentions were higher in
the sensitive than insensitive conditions, those who were higher on COL had a bigger
difference in their ratings between the conditions than those who were lower on COL.
This follows the general pattern of collectivists need for good interpersonal relations and
harmony, and their feelings of dissatisfaction when this need is not fulﬁlled. In addition,
it supports the notion that those who are high on COL are also more sensitive to the
expressions and reactions of others, and often take these into account (Gudykunst et al.,
1992)

In sum, it seems that those who are high on COL respond more positively to
sensitivity than those who low on COL. These interactions provided incremental
variance over main effects ranging from AR2=.016-.022. The heightened responsiveness
of those high on COL may be due to their heightened awareness of nonverbal
interpersonal signals, and their need for interpersonal harmony. It is likely that they view
rudeness as a message from the organization that it is not willing to be considerate of
applicants, and that it is not concerned with maintaining respect and good interpersonal
relations. On the other hand, they are more receptive to politeness and kindness, because
they see this as an implicit sign that the organization is truly considerate, and not just as

an attempt by the organization to make itself appear considerate.

125

Explanation Eﬂects

The effectiveness of explanations in mitigating negative effects of poor outcomes
and the mechanisms by which they do so was also tested in this study. Results were
consistent with past research in that providing an explanation helped to mitigate the
negative effects of poor performance, which was in support of hypothesis 3a. However
this study was unique in that it examined providing explanations both in terms of
explaining the violation of a procedural justice rule, and, when the rule was not violated,
explaining the intent of the rule (in this case consistency). Thus, explanations were
provided for the experimental process, whether that be following a certain procedural
rule, or not following it. In addition, the Shapiro’s “boundary condition” notion of
explanation adequacy was conﬁrmed, supporting hypothesis 3b. That is, explanation
adequacy was a full mediator of the explanation provision-procedural justice relationship.
The more adequate the explanation was perceived to be, the higher the procedural justice
ratings. Consequently, organizations should focus on providing explanations for the
process when something goes wrong, but also when everything goes right, to inform the
participants of the nature of the process. It is also important that they focus on giving
participants explanations that are useful and well-thought out, because how adequate the
explanation is perceived to be makes a huge difference in how fair the process will be
judged. In fact, other studies have shown that giving an inadequate explanation can be
worse than giving no explanation at all (Shapiro, 1991). The results in this study may
have been stronger, however, because the outcome was always negative. Shaw et al.

(2003) and Colquitt and Chertkoff (2001) found that explanations are most effective

126

when outcomes are consistently negative and when they are unexpected. Therefore, it is
possible that when the outcome is positive, this explanation effect may not be as strong.
In addition to the hypothesized effects, other effects involving explanations were
exarrrined in an exploratory capacity. Although there were no main effects of explanation
on any of the other outcome variables, there were two interactions that provided an
increase in variance accounted for (AR2=.014-.02) that was signiﬁcant. Both of these
interactions involved IND. The interaction of IND and explanation on intentions toward
the organization was signiﬁcant both pre and post-feedback. Those higher in IND were
much more likely to participate in a similar experiment, recommend the organization, and
accept a potential job offer than those lower on IND when there was no explanation given
for the process. In addition, the effect was opposite for the explanation conditions—
those lower in IND were more likely to indicate that they would participate in actions
related recommending, re-participating, and accepting an offer from the organization than
those higher in IND. The reasoning behind this effect could involve differences in how
these individuals viewed the explanation. Those who are low on IND may operate
similarly to those high on COL, in that they may be more familiar with the concept of
“saving face”, and see explanations as an attempt to assure others that they are being
considered and informed of any changes or concerns during the selection process. Those
who are high on IND, however, may just view explanations as extra information, that is
good, but not necessarily expected. Thus, when no explanation was given, those low on
IND may have reacted to it much more negatively, because they felt this was a signal to
them by the organization or experimental administrator that she did not care about their

concerns. In essence, they may have taken the lack of explanation much more personally

127

than did those who were high on IND. Accordingly, it has been shown that those who
are high on COL are more familiar with, and are more likely to respond positively to
apologies and excuses, than are those who are high on IND (Tata, 2000; Hui & Au,
2001). Thus, according to Folger and Cropanzano’s (1998) fairness theory, those low on
IND had more of a likelihood of applying the could, should, and would counterfactuals
indicating how the situation could have been conducted differently, than did those high
on IND.
Implications

Taken as a whole, the results from this study present very interesting ﬁndings.
The main ﬁnding was that for every procedural intervention including consistency,
sensitivity, and explanation, there were signiﬁcant interactions with IND and/or COL.
Although these interactions did not always involve effects on procedural fairness, they
did often involve effects on the actions that result from fairness perceptions, such as
recommendation and future participation intentions. Thus, an individual’s cultural
orientation does make a difference when considering their reaction to procedural rules.
Although the effects do not appear to be large as represented by the range of IND and
COL presented in the graphs in this paper (i.e., +1 and -1 SDS from the mean), if the
range is expanded to include within +2 and -2 standard deviations on the IND and COL
variables, the differences would be much greater. In addition, even if not large, the
effects are important in terms of the organization’s standing. These are aspects of the
selection process the organization can actually control, and therefore, even being able to
control 2% of the variance can be beneﬁcial. In addition, the effects found in this study

are probably quite a bit smaller than the effects one would ﬁnd in a real selection

128

'situation, where the stakes are increased ten-fold. Even though participants had relatively
little to gain or lose through this simulated selection situation, there was still an
interactive effect detected on the outcome variables. In addition, the range of responses
on both the IND and COL measures were restricted to the middle of the scale, and thus,
there were not many people representing more of the extremes of each factor. If
replicated with a more diverse sample, effects may be stronger.

This is an indication that it is important not to assume that individuals will react in
generally the same way to actions that take place during the selection process.
Speciﬁcally, in terms of Gilliland’s model, one should be aware that the violation of
procedural rules can produce differing reactions depending on who is a part of the
applicant pool. In several cases, the interactions between cultural orientation and
procedural interventions were signiﬁcant, but there was no main effect. Thus, often
people at one end of IND or COL reacted in the opposite manner to people on the other
end. This indicates that just simply applying the same intervention to everyone uniformly
(for instance, giving an explanation for a rule violation during the selection process) will
not necessarily increase everyone’s intentions to recommend the company to others, and
can be detrimental. Thus, the impact of potential remedies for rule violations, such as the
use of explanations, or general behavior during the process, such as sensitivity needs to
be considered based on the applicant population with which the organization is dealing.

An interesting ﬁnding in terms of the usefulness of the two factors of IND and
COL involves their differential relationship with outcomes. It seems that for some rules,
being high or low on IND is more important, for others, being high or low on COL is

more important, and for still others, one’s levels of both are important. For example,

129

consistency interacted with both COL and IND to affect different outcomes, sensitivity
interacted mainly with COL, and explanations mainly with IND. It is possible, then, that
when considering what rule to emphasize in the selection process, organizations should
consider the mean levels of both factors in their applicant population, and then reference
the levels for the factor that is most important for that rule. Based on these ﬁndings, then,
an individual’s levels on one factor, either IND or COL, may be more important than
their levels of the other factor, depending on what relationship is being moderated.
Because many of the studies on IND and COL do not treat them as separate factors, very
little work has been done thus far on distinguishing which of the two factors are more
important for predicting certain outcomes. This is an area that could yield useﬁrl ﬁndings
and that deserves future research.

Different aspects of the procedure also functioned differently in their relationship
with procedural justice. None of the manipulated variables had main effects on
procedural justice pre-feedback, but both consistency and explanation affected procedural
justice post-feedback. Although procedural justice perceptions pre-feedback are
interesting, it is really procedural justice perceptions post-feedback that are impactful,
because the applicant leaves the selection context with these fairness perceptions. The
lack of any relationship pre-feedback could be due to a lack of criticality in evaluating
procedural fairness pre-feedback, whereas, post-feedback, participants may have been
more attentive due to increased motivation because they were aware of the negative
outcome, and so they evaluated the process more critically. In fact, Gilliland’s (1993)
model states that the stage of the selection process is a possible moderator of the

relationship between the satisfaction of procedural rules and fairness perceptions, and

130

.wU.

'l'r

hl-

Hausknecht et a1. (2004) provide empirical support for this assertion, in that they found
that average correlations were higher between procedural rules and process fairness when
measured simultaneously than when separated in time.

More intriguing, however, is why both consistency and explanation affected
procedural justice perceptions, but sensitivity did not. It could be that sensitivity is more
related to interactional justice than it is to procedural justice, because it is has more to do
with interpersonal issues than the other process variables do. It is also possible that in a
relative sense, sensitivity may be a weaker rule than both consistency and explanations.
In a meta-analysis, Hausknecht et al. (2004) found that certain procedural rules have

stronger effects than others on procedural justice and other outcomes. The effect of
interpersonal treatment on procedural justice perceptions was not exarrrined in their study,
however. This is useful to know for organizations, so they can decide which rule to
emphasize over others in the selection context, if the use of one rule precludes the use of
another.

It may also be that the way in which the sensitivity manipulation was
Operationalized lead to its lack of a main effect. In this study, the main sensitivity
manipulation involved the experimental administrator kindly or rudely responding to a
question from a confederate who represented another participant. Thus, the administrator
Was not outrightly rude to participants themselves as in other experiments (e. g.
Greenberg, 1993; Greenberg, 1994), but she was rude to a fellow participant. Research

has indicated that in some conditions, how others are treated can be as impactful in
inﬂuencing an individual’s procedural justice judgments as how he himself is treated

(Van Den Bos & Lind, 2001). Nevertheless, it is also possible that certain conditions this

131

“bystander” sensitivity effect on procedural justice may be qualitatively or quantitatively
different than a direct sensitivity effect. It would seem that this type of effect would be
most noticeable to those who are more “sensitive” to the interpersonal aspects of a
situation. Indeed, this was conﬁrmed in the signiﬁcant interaction between sensitivity
and COL on procedural justice post-feedback. Therefore, sensitivity could be a more
nuanced rule that can have differing effects based on its operationalization and the
cultural orientation of the individuals responding to it.

The results for explanation were interesting in that explanations seemed to be
valued more for the signal that they gave rather than their actual content. This may be
due to the type of explanation given in this study, which was solely either a two sentence
reason for why the inconsistency occurred or how the procedures were consistent. Very
little was conveyed in terms of actual content, and none of the three types of social
accounts (i.e. justiﬁcations, referential, ideological; Bies, 1987) reﬂected the nature of
this explanation. In addition, this study was unique in that it involved an explanation for
the violation or satisfaction of a procedural rule, not an explanation for why a type of
procedure was used in contrast to other work in applicant reactions (e. g., Gilliland, 1994;
Horvath et al., 2000), and not an explanation for why outcomes were distributed in a
certain way in contrast to other work in the justice literature (Greenberg, 1991;
Greenberg, 1993; Greenberg, 1994). These aspects of the explanation may explain why
those who were low on IND responded more favorably to the use of explanations than
those high on IND. Those high on IND are very focused on their own goals and
opportunities for them to perform well and demonstrate their skills, so when they are

given explanations that are not useful in the sense that they are not able to help them

132

advance their goals, the explanations are meaningless to them. Therefore having an
explanation is more of a distraction than anything else. However, for those low on IND,
who are less self-focused and more aware of interpersonal messages, an explanation
would be a message to them that the organization is attempting to be informative and take
applicants’ needs into consideration. Perhaps if there was more helpful content in the
explanation in Greenberg’s (1991) sense of “self-image enhancing” content, high IND
individuals would have responded to the explanation differently. Consequently,
employers must recognize the cultural value orientation of their applicant pool and if it is
important for them to convey a more symbolic message or a concrete informational
message through the use of an explanation.

Although there were no hypotheses about interactions between the procedural
justice rules, very few studies have examined the violation of multiple rules in
combination, and researcher have stressed the importance of doing so in order to increase
the external validity of ﬁndings (Chan & Schmitt, 2004). Therefore, the interactions
among the procedural rules were examined for their effect on procedural justice pre and
post feedback. Results from the regressions are presented in Tables 15 and 16.
Surprisingly, none of the interactions between the three main procedural rules were
signiﬁcant. The total model with all of the manipulation variables, the 1ND and COL
variables, and their interactions accounted for about 12% of the variance in procedural
fairness pre-feedback, and about 7% of the variance in procedural fairness post-feedback.
The difference in amount of variance accounted for pre versus post-feedback is most
likely due to the effect of outcome favorability. In addition, if other fairness reactions

such as interpersonal and information fairness perceptions were included, perhaps the

133

percent of “fairness” variance accounted for would increase substantially, because some

of the manipulations may have more of a direct relevance to these variables.

134

Table I 5. Moderated Regressions of Procedural Justice Perceptions Pre on Consistency,
Sensitivity, Explanation, Collectivism, and Individualism

 

Model R2 ARF

 

Stgp 1
Consistency

Sensitivity
Explanation .022

Stgp 2
Consistency x Sensitivity

Consistency x Explanation
Sensitivity x Explanation
Consistency x Sensitivity x

Explanation '024 '001

Stgp 3
Individualism

Collectivism .039 .016

Stgp 4
Individualism x Consistency

Collectivism x Consistency

Individualism x Sensitivity

Collectivism x Sensitivity

Individualism x Explanation

Collectivism x Explanation .120 .080**

 

Note. N=303. *p<.05. **p<.01.

135

Table 16. Moderated Regressions of Procedural Justice Perceptions Post on
Consistency, Sensitivity, Explanation, Collectivism, and Individualism

 

Model R2 ART

 

Step 1
Consistency

Sensitivity
Explanation .032

Step 2
Consistency x Sensitivity

Consistency x Explanation
Sensitivity x Explanation
Consistency x Sensitivity x

Explanation '039 '007

Step 3
Individualism

Collectivism .042 .003

Step 4
Individualism x Consistency

Collectivism x Consistency

Individualism x Sensitivity

Collectivism x Sensitivity

Individualism x Explanation

Collectivism x Explanation .068 .026

 

Note. N=303. *p<.05 **p<.01.

136

LIMITATIONS AND FUTURE RESEARCH

Although there were several interesting ﬁndings that emerged from this study,
there were also limitations of the study that should be considered and suggestions for
ﬁlture research to address these issues.
Sample Issues

One limitation involved the sample that was used. It was composed entirely of
college students, mainly undergraduate, but also some graduate students. Thus, this
population may not have been as experienced in participating in a selection process as the
general population may be, although the majority (96.5%) cited having had prior work
experience and only 8% had never participated in a job interview before. Nevertheless,
the majority (65%) stated that they had never had to take a test in order to get a job.
Consequently their expectations for procedural rules such as consistency or sensitivity
during the application process may be different than that of individuals who have
participated in several job selection experiences. However, many organizations also
employ recent college graduates, and this sample may be more representative of that
population. In addition, because the psychology subject pool was utilized to recruit
participants, the percentage of females in the sample (72.5%) was more than double that
of males. Therefore, males were underrepresented in the sample. In this study, gender
was negatively correlated with procedural justice perceptions both pre (r=-.15, p<.05)
and post (r=-.20, p<.05) feedback, however, the meta-analysis on applicant reactions
showed no correlation between gender and procedural justice perceptions (Hausknecht et
al., 2004). Additionally, there were no gender differences on the IND and COL

measures, which was consistent with past studies (Coon & Kemmelrneier, 2001; Triandis

137

et al., 1995). Nevertheless, future research should include a more representative sample
of males.

Despite the fact that there were several attempts made to recruit individuals from
different countries to increase variance on the IND and COL scales, the vast majority of
participants were still from the US. (92.4%). Out of a possible range of responses from
1-7 on the IND and COL scales, 68% of the sample fell between the values of 4.30 and
5.42 on the COL scale, and 4.37 and 5.51 on the IND scale. This may be an indication
of a possible restriction of range of the scale which may have attenuated some of the
moderating relationships between these cultural variables and justice outcomes. This
range is smaller than that of other studies that have utilized Singelis’ (1994) scales of
IND and COL (e. g. Coon & Kemmelmeier, 2001; Okazaki, 2000; Singelis & Sharkey,
1995; Singelis et al., 1999), which seems to indicate that this sample was more
homogeneous culturally than other samples.

However, a broader issue is that the majority of studies utilizing this measure had
scale scores with standard deviations of less than 1.00, indicating that perhaps the
measure itself is not sensitive to differences in IND and COL levels. Nevertheless, it is
important to note that all of the previous studies examined by this author utilized samples
of college students, not broader samples of the general population. Youn (2000) found
differences in levels of IND and COL in an American sample of graduate and
undergraduate students, in that the level of IND of graduate students was signiﬁcantly
greater, and the level of COL was signiﬁcantly less than that of undergraduates. Youn
suggested that this was an indication that American’s students’ educational experiences

are related to a progression of their independent self and a retrogression of their

138

dependent selves, and that this result was in support of Markus and Kitayama’s
theoretical assertion that those in individualistic cultures tend to hold a more independent
world-view. Thus, even by broadening a sample by including graduate students, one can
ﬁnd signiﬁcant differences in levels of IND and COL, which could possibly related to
educational level or age. Although graduate students were also recruited for this study,
the vast majority of participants (98%) were from the undergraduate population.
Therefore, it is very important in future research to attempt to obtain samples that are
more diverse in terms of ethnicity, education, age, and other variables in order to get
better population indices of IND and COL using Singelis’ (1994) measure, and so that
more variance can be obtained. With more variance, effects that were previously
attenuated could emerge.

Another possible avenue to take in the study of IND and COL, is to purposefully
select a particular sample to study, instead of attempting to capture the general
population. For instance, since many of the IND-COL studies to date have utilized
samples that score in the mid-range on IND and COL, it may be useful for the sake of
comparison to select a sample that contains people closer to the extremes of both scales.
This would allow one to best see the differences in behavior that occur between
individuals who are truly “high” or “low” on these measures. Another alternative would
be to select a homogeneous sample in terms of culture (America in general is more
heterogeneous than other countries), and then examine differences in IND and COL
orientation in this culture. The assumption would be that individuals within the culture
would be generally equal in terms of other cultural inﬂuences, so differences would be

due to individual differences in IND and COL. This is the approach Gomez (2003) took

139

when examining IND and COL differences amongst a Hispanic population living in the
US. She was able to ﬁnd differences in preferences according to IND and COL even
amongst this relatively homogenous sample.
Context & Manipulation Issues

The environment in which the study was conducted is another limitation that
should be addressed. This was a lab study using psychology 101 subjects who simulated
being job applicants. Thus, the environment was rather artiﬁcial, though an attempt was
made to make the manipulations in the study as natural as possible. However, the type of
motivation that exists to perform to the best of one’s ability as well as the stakes involved
in the selection outcome were not present to the degree that they would have been in an
actual selection situation. The reward for participants performing well on the test was not
directly tied to their performance in that even if they performed well, their selection to
win the prize money would be random, because it was being determined by a lottery.
There was an attempt to make participants feel that they had more control over their
outcome by telling them that they would have an increased chance of winning if they
performed well, but this may only have increased motivation for some participants. In
addition, there is a possibility that the individual reward may not have been as motivating
for those who were low on IND or high on COL; perhaps a collective or team award may
have been more motivating. Yet, in a true selection context, participants are rewarded
individually.

There was an additional problem concerning group size which was not
anticipated. The number of participants in each session varied substantially from a low

of 4 to a high of 45. It was remarked upon by both this author and her research assistants

140

that it seemed that the manipulations were less effective in the larger sessions. When
there were greater numbers of participants, more comments of incredulity in response to
manipulations were heard than when there were fewer participants. In addition, although
attempts were made to stress the importance to participants of keeping one’s score on
oneself, more participants seemed to audibly discuss their feedback scores in the larger
sessions, which made it more likely that they would discover that every participant
obtained negative feedback. However, out of 19 sessions only 2 had more than 25
participants.

In addition, there were a few issues regarding the effectiveness of the
manipulations. For instance, the effectiveness of the sensitivity manipulation may have
varied depending on the acting ability of the experimental administrator and the
confederate. If the interaction did not appear to be real, then the manipulation would not
have worked as well. In general, though, it seemed that sensitivity manipulation worked
rather well, both anecdotally and based on results from the manipulation check. Only 7%
of the variance in the consistency manipulation check, however, was predicted by the
consistency manipulation. This may be due to the ineffectiveness of the manipulation,
but it also may be due in large part to the location of the measure. The consistency
manipulation check was the last measure in the survey, and participants could have been
fatigued by the time they arrived at this measure, resulting in inaccurate ratings. In
addition, several of the experimental sessions came close to or exceeded slightly the
allotted time for the experiment, and therefore, participants may have been in a rush to

leave, and thus been more careless when answering these items.

141

It also possible that the type of explanation used in this study could have
inﬂuenced the effectiveness of the manipulation as well. The explanation used was very
neutral, and did not contain much information regarding for instance, who was to blame
for the inconsistency (the administrator indicated they had the wrong forms of the test,
but not how that occurred), or how much effort went into making the process a consistent
process. If the explanation had been a type of social account assigning blame to an
outside force (causal), changing the referent standard for the action (referential),
refraining the action (ideological), or even just an apology, effects could have been
stronger (Bies, 1987). However, the advantage of having an ambiguous explanation that
does not ﬁt into any one category is that the ﬁndings are no longer restricted to a
particular type of explanation. Moreover, the fact that the content of the explanation did
not matter so much as the presence of an explanation lends credence to the assertion that
an explanation can be valued as a symbol of the organization’s concern, and not simply
for its informational content. Nevertheless, there has been some research to suggest that
for instance, those high on COL respond more favorably to apologies than those lower on
COL (Hui & Au, 2001 ). Consequently, future research should explore ﬁrrther whether
certain types of explanations are more effective for those high on IND or those high on
COL.

Another area to explore regarding context issues is whether the selection context
itself has an effect on the expression of IND versus COL. In the personality literature,
some have argued for a person by situation interactionist view of personality inﬂuences
on behavior (Endler & Magnusson, 1976). This view holds that situational factors can

affect what aspects of one’s personality are expressed at a give moment in time.

142

Moreover, it is when a situation is trait relevant that it exerts the most “pressure” on the
expression of the trait (Tett & Burnett, 2003). In addition to this situationally-inﬂuenced
view of traits in the personality literature, Triandis has also stressed the importance of
situations in inﬂuencing the expression of IND and/or COL. For instance, Triandis
(1995) states that there are several situational elements that can increase the likelihood
that one’s COL cognitive system will be activated. These include aspects such as if
others in the situation are lcrrown to be collectivists, if the situation calls for the assembly
of a collective (e. g. family, team) at the time, if the situation contains symbols that
emphasize the collective (e. g. uniforms, jerseys, hats), and if the situation involves a
cooperative task.

Thus, it is possible that there are elements in the selection situation that are more
likely to activate the individualistic part of an individual’s cognitive system. In fact, the
selection context possesses characteristics that are the antithesis of many of the
collectivism-inducing characteristics just described. It is generally a competitive
situation in which one is aware that other individuals present are rivals, vying for the
same or a similar position. It is a situation where uniqueness is valued (i.e., in an
interview), and one is encouraged to identify the skills, experience, or characteristics that
make one “stand out”. This has interesting consequences for the utilization of IND and
COL measures during the selection process. It could mean that the applicants’ trait levels
of IND and COL do not reﬂect the levels of IND and COL they express during the
selection context. For example, they could be reporting their general levels of IND and
COL, but the experimental context, because it is a competitive, non-cooperative process,

could be “priming” their IND self to be expressed more strongly. Therefore, the levels of

143

‘4.

3d

 

 

IND and COL affecting reactions during the selection process could be different than the
overall, context-independent levels of IND and COL reported on the measure. Therefore,
the stability IND and COL levels between non-selection and selection contexts should be
explored further, as this could greatly affect the study of these cultural variables across
situations.
IND and COL Measure Issues

Individualism-Collectivism is one of the most widely-studied cultural variables
that emerged from Hofstede’s (1980) categorization of cultural value dimensions. It has
been the subject of several reviews and empirical studies throughout the years (Earley &
Gibson, 1998; Oyserman, Coon, & Kemmelrneier, 2002). Despite the considerable
amount of attention that has been paid to this dimension, a consistently reliable and valid
measure of IND-COL at the individual level has not yet been established. As was
evidenced in this study, the reliability of the subscales of Singelis’ (1994) measure of
IND and COL did not meet standard levels of good reliability, although in general, IND—
COL scales have produced relatively low reliabilities. Even when utilizing his own
measure, Singelis reported levels of reliability ranging frOm .58 to .70 for the IND
subscale, and .53 to .73 for the COL subscale (Singelis, 1994; Singelis et al., 1999;
Singelis & Sharkey, 1995). He argued that the low reliabilities were a result of the
bandwith versus ﬁdelity dilemma (Singelis & Brown, 1995). In other words, in the
interest of capturing the broad nature of the construct as comprehensively as possible, one
sacriﬁces some of the precision or consistency of information obtained.

Although Singelis and Brown (1995) argue that capturing the broader construct is

more useful for prediction, it is possible that deﬁning subfactors within the larger

144

construct may be useful as well. Other research has found value and increased validity in
the use of subdimensions of constructs (e. g., Avery, 2003; Grifﬁn & Hesketh, 2004;
Judge & Bono, 2001). It is possible that subdimensions of IND and COL could be more
internally consistent than the higher order factors, as well as possess differential
predictive validities in relation to various outcomes, such as procedural, interactional, and
distributive justice reactions. Therefore, an alternative model of DID and COL that
included subfactors was tested.

A group of researchers have recently suggested that Singelis’ (1994) scale of IND
and COL is better represented by a six-factor model than a two factor model of IND and
COL (Hardin, Leong, & Bhagwat, 2004). Thus, for exploratory purposes, the six factor
model proposed by Hardin et al. (2004) for the Singelis scale was compared to the two-
factor model of the scale generally accepted. The six factors in the Hardin et al. (2004)
model include four factors that represent the IND construct, and two factors that represent
the COL construct. The four IND factors include Autonomy, Individualism, Behavioral
Consistency, and Primacy of Self, while the two COL factors include Esteem for Group
and Relational Interdependence. The six factor model places two items that were
originally included under COL, item 3 and item 30, under the Autonomy factor, and thus,
the six factor model is not nested within the two-factor model. Therefore, comparisons
between the two-factor and six-factor models were conducted based on overall model ﬁt.
Item parcels were created in the manner described earlier for factors that had greater than
ﬁve indicators. The values of a few relevant ﬁt indices for the six factor model were as
follows: x 2(113) = 282.104, p<.01, RMSEA=.070, NFI=.570, CFI=.675, SRMR=.076.

The values for the two factor model were: X 2 (9) = 19.98, p<.05, RMSEA=.063,

145

NFI=.929, CFI=.959, SRMR=.066. Thus, overall, it seemed that the two-factor model
was a better ﬁt to the data than the six factor model for this sample. However, other
multidimensional models of IND and COL should be explored and evaluated.

Future research should focus on establishing a valid scale of IND and COL with
reliable subdimensions based on a consideration of the deﬁnitional roots of the
constructs. In fact, in T riandis’ (1995) conceptualization of the deﬁnition of IND and
COL at the individual level, he describes four components, or subdimensions, that
comprise the constructs. These include the deﬁnition of self, alignment of goals, focus
on norms and duties versus attitudes and personal needs, and emphasis on relational
versus rational aspects of relationships. Although Triandis (1995) has developed his own
four factor measure of IND and COL at the individual level, the subfactors that comprise
his scale refer to horizontal versus vertical IND and COL orientations, not the
subdimensions just described. Singelis’ (1994) measure, which is one of the most widely
used measures in the literature, was itself developed through an amalgamation of items
from four different scales to capture breadth, and not through an a priori consideration of
focused subcomponent representation. In addition, measures of IND and COL with
subfactors do exist, however they are either not widely used, suffer from low overall or
subdimension reliabilities, refer to the country-level IND-COL constructs, or are not
valid (Coon & Kemmelrneier, 2001; Oyserman et al., 2002; Singelis, 1994). Therefore,
another option for future research is to work on reﬁning or supplementing existing scales
(i.e., Hardin, Leong, & Bhagwat, 2004). In general, future research should concentrate

effort on developing theoretically-based, focused measures of IND and COL, so as to, as

146

Oyserman et al. (2002) so aptly state, “avoid becoming so content packed as to be
theoretically empty” (pg.44).

Another issue to consider is whether IND and COL measures should be tailored to
reﬂect the context in which they are being used. For example, if they are to be used to
predict orgarrizationally relevant outcomes, such as organizational justice perceptions,
perhaps items should reﬂect content related to organizational issues. Robie, Schmit,
Ryan and Zickar (2000) conducted a study in which they examined differences in the
psychometric characteristics of a personality measure when items were contextualized
(i.e. put in the context of work) versus non-contextualized. Although the results of the
study were mixed, the error variances for the contextualized measure were less than that
for the noncontextualized measure. In addition, intuitively it seems plausible that
relationships would be stronger if the context of the items in the predictor matched the
context of the criterion measure (although one has to be reasonable when determining the
level of speciﬁcity of the context). In fact, Schmit, Ryan, Stierwalt, and Powell (1995)
found that the validity of the Conscientious scale on the NEO-PI-R in predicting college
GPA was higher when context or work-speciﬁc items were used instead of the original
items. Another study also concluded that a common behavioral frame of reference
represented in items on a questionnaire measure of personality may have resulted in the
greater validity of it over a bipolar inventory of personality (Cellar, Miller, Doverspike,
& Klawsky, 1996). Thus, perhaps future research should test the utility of using items in
IND-COL scales relating to one’s attitudes towards work structure, work relationships,
work rewards, one’s behavior and preferences at work, et cetera, to predict work-related

outcomes, versus using current broad and general IND-COL measures.

147

Issues Involving Individual-Level IND and COL

A separate but related issue warranting future research attention concerns the
meaning of IND and COL at the individual level. As has been discussed previously in
this paper, IND and COL was ﬁrst conceptualized as a cultural level phenomenon, where
IND and COL were polar ends of a continuum. Thus, being high on one resulted in being
low on the other. Although this cultural level construct was interesting, researchers soon
realized the usefulness of capturing this phenomenon at a more micro-level for the
prediction of individual behaviors, because of variance within country. When they
extrapolated the concept to the individual level, however, the measures they developed
indicated that the variables were independent and that individuals could be high or low on
both variables (Markus & Kitayama, 1991; Triandis, 1986). Yet, there was no clear
indication what the combination of these levels really signiﬁed, and there still is no clear
indication.

The way that the variables are deﬁned currently seems to imply that they are
opposites, nonetheless, people can theoretically be high or low on both. However, there
has been little discussion on what it means theoretically to be high on both factors or low
on both factors and how one would expect an individual to act. Some recent research has
indicated that individuals with different combinations of levels react differently, but this
is still preliminary (Gomez, 2003). What is the difference between being low on IND
versus being high on COL? It is also unclear whether IND and COL are considered trait
variables or state variables. Although most of the literature seems to lean toward
identifying them as stable traits, some recent research has indicated that IND and COL

values can be primed (Oyserman et al., 2001; Traﬁmow, Triandis, & Goto, 1991). The

148

answers to both this questions have strong implications for future research in several
areas, including selection. For example, will one’s reactions during the selection context
based on one’s context-dependent cultural orientation be different than long-term
reactions to the process based on one’s trait-level cultural orientation?

Applicant Reactions Cross-Culturally

Although this study is a step towards understanding the inﬂuence of cultural
factors on reactions to the selection process, there are several avenues still left to explore.
It would be interesting to conduct a similar study to this one examining other rules
relevant to applicant justice perceptions. Rules such as two way communication and
opportunity to perform could have differential impacts on individuals depending on their
IND and COL cultural orientation. If subdimensions of IND and COL are established,
one should examine which subdimension has the greatest moderating effect on reactions
to a speciﬁc procedural rule. In addition, cultural variables other than IND and COL
have been posited to effect individuals’ reactions to procedural rules including confucian
dynamism, masculinity-femininity, and uncertainty avoidance. These variables should be
examined further.

Chan and Schmitt (2004) emphasized the need to match level of speciﬁcity of
predictors to criteria levels of speciﬁcity in research involving applicant reactions in
order to increase predictive validity. Thus, rather than asking for global procedural
fairness reactions, one could possibly ask for more speciﬁc fairness reactions such as
reactions to the test, the interpersonal atmosphere, or the explanation given. Other than
fairness, reactions such as anger, skepticism, and disappointment, which have been

examined in the literature on explanations (e. g. Shapiro, 1991) could also be used as

149

possible outcomes of the violation of procedural rules, as well as predictors of behavioral
outcomes. It would be interesting to see if these reactions mediate the fairness
perception-outcome link, and if so, what individual difference moderators (e. g., perhaps
cultural orientation) could moderate this relationship.

Bell, Ryan, and Wiechmann (2004) recently created a model involving the role of
justice expectations in the applicant reaction process. They argued that the justice
expectations that individuals bring into encounters with the organization (e. g. during the
selection process) in large part shape their reactions, including their subsequent affect,
behavior, and justice perceptions. Three of the antecedents of j ustice expectations in
their model are direct experiences, existing beliefs, and indirect experiences. One could
argue, as Bell et al. do, that an individual’s cultural orientation can inﬂuence all of these
factors. Thus, another avenue for future studies to explore would be to test whether the
effect of cultural orientation on justice expectations is mediated by existing beliefs, direct
experiences, or indirect experiences, and if so, what the relative strength of these
mediators are. In addition, whether individuals’ fairness perceptions are more inﬂuenced
by their expectations versus the events that actually occur during the selection process
could possibly vary depending on their cultural orientation. For instance, if those high on
COL pay more attention to context, perhaps they would be more likely to rely on events
occurring during the actual selection process to form their justice perceptions, rather than
their justice expectations. Consequently, the area of justice expectations and culture

provides intriguing new areas for research to explore.

150

CONCLUSION

The area of applicant reactions during the selection process has recently gained
attention as a valuable stream of research due to the demonstration of its importance for
several organizationally-relevant outcomes. Because it is a relatively new area of
research, however, several aspects regarding how applicants’ perceptions are formed and
how they affect subsequent outcomes are left to be examined and explained. This study
is an attempt to shed light on one of these aspects, namely, how an applicant’s cultural
orientation can affect how he/she perceives the violation of fairness rules during the
selection process. Chan and Schmitt (2004) in their critique of work on applicant
reactions, recommend that future work should examine applicant traits as possible
moderators of the effect of procedures on reactions. They also recommend that rather
than only looking at the inﬂuence of one procedural rule, studies should focus on
examining the effects of multiple procedural rules in concert in order to increase the
external validity of ﬁndings. This study in part addresses both of these concerns. The
results from this study provide some insight on the relationship between applicant
characteristics, procedural justice rules, and perceptions of procedural fairness; in
addition, however, they also spark interesting questions for future study and reveal the
great value of further research in this promising area. In the words of the great John
Steinbeck: “Ideas are like rabbits. You get a couple and learn how to handle them, and

then pretty soon you have a dozen.”

151

APPENDIX A

Measures
Manipulation Checks

NOTE: All statements are scored on a ﬁve point Likert scale except the test-taking
motivation, and IND and COL measures, which are scored on a seven point Likert scale.

Consistenep
Bauer et al. (2001)

l. The test was administered to all applicants in the same way.
2. There were no differences in the way the test was administered to different
applicants.

3. The test administrator made no distinction in how he/she treated applicants.
Additional items

4. I received the same amount of time as other people when taking the test.

5. I received less time than other people when taking the test. (R)

6. I was treated differently than others who have taken this test. (R)

Interpersonal SensitivLQ
Bauer et al. (2001)
1. I was treated politely during the testing process.
2. Test administrators were considerate during the test.
3. Test administrators answered procedural questions in a straightforward and
sincere manner.
4. The testing staff put me at ease when I took the test.
5. I was satisﬁed with my treatment at the test site.

Explanation Provision
l. I received an explanation for why I was given the amount of time I had to
complete the test.
2. It was made very clear to me why I was given the amount of time I had to
complete the test.

T est- T akingMotivation ﬂMotivation Check ")

Sanchez, R. J ., Truxillo, D. M., & Bauer, T. N. (2000)

I would like to be hired for this job.

It would be good to have this type of a job. (altered)

I want to get a job with this company. (altered)

If you do well on this test, you have a good chance of being hired.
I think you will be hired if you get a high test score.

How well you do on this test will affect whether you are hired.
The higher your test score, the better your chance of getting hired.

 

89‘5”?pr

152

8. If you try to do your best on this test, you can get a high score.

9. If you concentrate and try hard you can get a high test score.

10. You can get a good score on this test if you put some effort into it.
Justice Measures

Procedural Justice
Gilliland (1994)
PRE
1. Whether or not I get the job, I feel the selection process was fair.
2. Whether or not I get the job, the procedures used to select people for this job are
fair.
3. Whether or not I get the job, I am satisﬁed with the selection process.
4. Overall, I feel dissatisﬁed with the way people will be selected for the job. (R)
POST (Altered)
l. I feel the selection process was fair.
2. The procedures used to select people for this job are fair.
3. I am satisﬁed with the selection process.
4. Overall, I feel dissatisﬁed with the way people were selected for the job. (R)

Distributive Justice
Gilliland (1994)
' 1. Overall, I feel the results of the selection process were unfair. (R)
2. I feel the hiring decision (accept/reject) was fair.
3. Overall, I am satisﬁed with the hiring decision.
4.

I am dissatisﬁed with the test administrator’s decision about whether or not to hire

me. (R)
Other Measures
Interdependent and Independent Self-Construals

Singelis (1994)
l. I enjoy being unique and different from others in many respects. (IND)

I 2. I feel comfortable using someone’s ﬁrst name soon after I meet them, even when they

are much older than I am. (IND)
Even when I strongly disagree with group members, I avoid an argument. (INTER)
I have respect for the authority ﬁgures with whom I interact. (INTER)
I do my own thing, regardless of what others think. (IND)
I respect people who are modest about themselves. (INTER)
I feel it is important for me to act as an independent person. (IND)
I will sacriﬁce my self interest for the beneﬁt of the group I am in. (INTER)
I’d rather say “No” directly, than risk being misunderstood. (IND)
. Having a lively imagination is important to me. (IND)
. I should take into consideration my parents’ advice when making education/career
plans. (INTER)
12. I feel my fate is intertwined with the fate of those around me. (INTER)
13. I prefer to be direct and forthright when dealing with people I’ve just met. (IND)

PS”

““PWNQ‘P‘

HQ

153

‘1

 

-__._.‘__

r4

14. I feel good when I cooperate with others. (INTER)

15. I am comfortable with being singled out for praise or rewards. (IND)

16. If my brother or sister fails, I feel responsible. (INTER)

17. I often have the feeling that my relationships with others are more important than my
own accomplishments. (INTER)

18. Speaking up during a class (or a meeting) is not a problem for me. (IND)

19. I would offer my seat in a bus to my professor (or my boss). (INTER)

20. I act the same way no matter who I am with. (IND)

21. My happiness depends on the happiness of those around me. (INTER)

22. I value being in good health over everything. (IND)

23. I will stay in a group if they need me, even when I am not happy with the group.
(INTER)

24. I try to do what is best for me, regardless of how that might affect others. (IND)

25. Being able to take care of myself is a primary concern for me. (IND)

26. It is important to me to respect decisions made by the group. (INTER)

27. My person identity, independent of others, is very important to me. (IND)

28. It is important for me to maintain harmony within my group. (INTER)

29. I act the same way at home that I do at school. (IND)

30. I usually go along with what others want to do, even when I would rather do
something different. (INTER)

A dequacv of Explanation
Wiechmann & Ryan (2001); altered for context
1. The explanation given for any problems during the selection process was

adequate.

2. I was satisﬁed receiving this explanation for a problem during the selection
process.

3. I was given an inadequate explanation for a problem during the selection process.
(R)

4. Everyone should at least be given this type of explanation for a problem during
the selection process.

Job Acceptance Intentions

Ployhart & Ryan (1998)

l. Pre-decision: I would accept the job if it was offered to me.

2. Post-decision: Even if I was now offered the job, I would not accept it.

Recommendation Intentions
Gilliland (1994)
l. I would recommend this experiment to my friends.
2. I would tell my friends to participate in this experiment.
3. I would tell others this is a good experiment in which to participate.
4. I think other people should know this is a good experiment.
Additional items
5. I would tell others that I enjoyed participating in this experiment.
6. I would tell others that this was a good experience for me.

154

Future Experiment Intentions
Ployhart & Ryan (1998)
l. I would not like to participate in future experiments like this one.
2. If I was offered the opportunity, I would take part in another experiment like this
one.
3. I will continue to volunteer for experiments similar to this one.
4. If I hear about other projects like this, I would be interested in applying for them.

155

 

APPENDIX B

Informed Consent Form

Informed Consent for Reactions in a Selection Context Study

Primary Researchers:
Srrrriti Shivpuri
Dr. Ann Marie Ryan, Professor of Psychology

Description of Experiment:

In this experiment, you are to take on the role of an applicant during the hiring process. You will
hypothetically be applying for a job that you consider to be highly desirable. In this process, you will be
asked questions about yourself, and various aspects of the process, and you will be asked to take a test that
has been designed to assess your basic skills.

The experiment is meant to look at your performance on the test and general impressions of the application
process.

Estimated Time Required:
45 nrinutes

Risks and Beneﬁts:

You gain exposure to questions that could be used in a real-world setting for selecting applicants for a
position. This can be valuable if you are currently on the job market, or will be in the future. You will also
be introduced to the process of conducting research in an experimental setting, if you have never
participated in an experiment before. In addition, you have the chance of being awarded with multiple
prizes for your participation.

You have been fully informed of the above-described procedure with its possible beneﬁts and risks. The
investigators will be available to answer any questions you may have. If, at any time, you feel your
questions have not been adequately answered or you want to discuss the research, please contact the
investigators ( Smriti Shivpuri, shivpuri@pnsu.edu , #517-355-2171, or Dr. Ann Marie Ryan,
man@msu.edu, #517-353-8855). If you have questions or concerns regarding your rights as a study
participant, or are dissatisﬁed at any time with any aspect of this study, you may contact — anonymously if
you wish — Peter Vasilenko, Ph.D., Chair of the University Committee on Research Involving Human
Subjects (UCRIHS) by phone: 517-355-2180 or email: ucrihs@msu.edu. You are free to withdraw this
consent and discontinue participation in this project at any time without penalty. If you choose not to
participate in this study, alternatives are available (for those participating from the Psychology Subject
Pool) from the subject pool for you to earn credit. If you choose to withdraw from the study prior to its
completion, you will receive credit for the time you have spent in the study.

 

If you agree to participate, please sign your name at the bottom of this form. We also ask for your e-mail
address and phone number so that we can contact you if you win a prize. It is important to recognize that
you will be given a subject number for this experiment. The purpose of this subject number is to keep track
of the various materials you will complete during the experiment. Please be assured that all information
you give us will be kept conﬁdential. Your privacy will be protected to the maximum extent allowable by
law.

Your signature below indicates your voluntary agreement to participate in this study.

156

 

Name: Date:

 

Signature: Email:

Phone:

 

157

 

APPENDIX C

Protocol for Study
PROTOCOL FOR REACTIONS IN A SELECTION CONTEXT STUD_Y_
BEFORE SESSION

Procedure:
Call and email subjects the day before their session is scheduled to remind them of
location and time of session. Determine what condition will be run for that session.

Materials:
Make sure to bring these materials to test site:

Cl List of participants attending session/ sign-in sheet
Cl #2 Pencils

Cl Informed Consent Forms

[:1 Debrieﬁng Forms

El Scantrons (2)

[:1 Experimental Packets (check for correct condition)
1" Survey
Test
2"d Survey
Feedback Form & 3rd Survey

DTimer (if necessary)

158

DURING SESSION

Greet participants as they enter and have them sign the sign-in sheet. Then ask them to
sit one seat apart. Wait till 5 minutes after scheduled session time to begin reading the
protocol sheet.

“Hello, my name is , and I will be your proctor for today’s session. You all
should be here for the study “Reactions in a Selection Context”, if you are not here for
this study please leave at this time.”

LOOK UP AND PAUSE

“You will need a number 2 pencil during this session, please raise your hand if you do f
not have one.” '

LOOK UP AND GIVE PENCILS TO THOSE WHO NEED IT

“We will now be passing out an informed consent form to you. Please read this form and
sign at the bottom. Signing this form indicates that you agree to participate in the study ‘
today. Please look up when you have ﬁnished signing the form and we will collect it I;
from you.”

 

PASS OUT INFORMED CONSENT FORMS AND COLLECT

“We will be passing out the ﬁrst section of the experimental packet and a scantron to you
at this time. The packet will be placed face down on your desk. Please do not turn it
over until I tell you to do so.”

PASS OUT EXPERIMENTAL PACKET AND SCANTRON

“When you signed in for this study, you were given an experimental ID number. Please
ﬁll in this ID number on the scantron, and ﬁll in the corresponding bubbles. If you have
lost the paper that had your ID number on it and do not remember your ID number,
please raise your hand. (PAUSE FOR PARTICIPANTS TO FILL IN INFO). Please turn
to page one of the packet. On this page there is a list of four jobs and their descriptions.
We would like you to now pick the job that you would like to have the most if you had to
choose to take one for one year. Now, on the second page of your experimental packet,
write down on the lines indicated, why you would choose this job over the other jobs.
When you are done, please put your pencil down, and do not turn to other pages in the
experimental packet.”

PAUSE FOR PARTICIPANTS TO FILL IN INFORMATION
“Now, we would like you to imagine that you are an applicant for the job that you just

wrote down, and all the other people in this room are applying for the same job. Imagine
that I am a representative of the company to which you are applying, and you have come

159

here today to take a selection test. Those who score well on this test will be hired, those
who do not will be rejected for the position. This test is an established selection test used
to measure cognitive ability, or intelligence. It is or has been used in the past by
numerous companies including Blue Cross and Blue Shield, Stevenson & Associates,
CEO International, and the National Football League. Therefore, there is a chance you
have seen it before, or will see it again in the future.

After you take the test, you will receive feedback on whether you are hired or rejected
for the position. While everyone will have their names entered into a lottery for the
chance to win cash prizes of $25, $50, $75, or $100, only the top six scorers from every
three sessions will have their names put into the lottery three times, for an increased
chance to win. Therefore, you will be competing against people in other sessions. In
addition, those who are the top scorers from each session will have their names put into a
lottery for a chance to win a pair of free movie tickets, so you will also be competing
against those in your own session. So, if you are not hired, chances are you did not score
well on the exam, and so you are not likely to be entered into the other lottery or have an
increased chance of winning. Does anyone have any questions?”

LOOK UP FOR QUESTIONS

“Before you take the test, please ﬁll out the questionnaire we have passed out to you.
This questionnaire begins on page 3 of the packet we have handed out to you. The
questionnaire asks you questions about your general attitudes and preferences. Your
responses will be kept conﬁdential, so please answer as truthﬁrlly as possible. Mark your
responses on the scantron provided, and raise your hand when you have ﬁnished ﬁlling
out the responses. You may begin.”

COLLECT QUESTIONNAIRES WHEN PARTICIPANTS ARE FINISHED, LEAVE
SCANTRONS

“Now we will pass out the selection test to you. You will have 12 minutes to complete
this test. I will mark the time remaining on the chalkboard, and will warn you when you
have 2 minutes left. We will begin passing out the tests now, please keep them face
down until I tell you to turn them over.”

WRITE 12 MINUTES ON CHALKBOARD. IF RUNNING CONSISTENT
CONDITIONS SKIP NEXT PARAGRAPH.

IF RUNNING INCONSISTENT CONDITIONS:

PICK UP TESTS, LOOK AT TESTS, APPEAR TO BE CONFUSED AND WHISPER
TO ASSISTANT, THEN LEAVE ROOM WITH TESTS. RETURN 5 MINUTES
LATER. IF PARTICIPANTS QUESTION THE ASSISTANT, THE ASSISTANT
MUST ONLY SAY THAT THE EXPERIMENTER SAID SHE’LL BE RIGHT BACK
AND TO STAY IN YOUR SEATS.

160

EXPERIMENTER SHOULD COME BACK WITH TESTS, AND SAY THE
FOLLOWING WITHOUT READING DEPENDING ON WHAT CONDITION IS
BEING RUN:

No Explanation/Consistent:
Pass out the tests. Then state “You will have 12 minutes to complete the test. Please start

marking your answers on question number _ on the scantron. Stop working when you
have reached the end of the test, which will be marked by a STOP on a page in the
packet. I will give you a 2 minute warning and let you know when time is over.”

No Explanation/Inconsistent:
Do not say anything, just erase the 12 minutes written on the chalkboard, write 7 minutes

and tell them “You will now have 7 minutes to complete the test. Please start marking
your answers on question number _ on the scantron. StOp working when you have
reached the end of the test, which will be marked by a STOP on a page in the packet. I
will give you a 2 minute warning and let you know when time is over.” Then pass out test
forms.

Neutral Explanation/Consistent:
“You will be given the 12 minutes. to complete this test like the other groups have been

given. We will let you know when you have 2 minutes left. You all should have the
right form of the test. Please start marking your answers on question number _ on the
scantron. Stop working when you have reached the end of the test, which will be marked
by a STOP on a page in the packet. This is an individual task, so please work on this
independently. You may begin.”

Neutral Explanation/Inconsistent:
Erase the 12 minutes and write 7 minutes on the board. “Although the other groups were

given 12 minutes to complete this test, you will not be. We mistakenly had the wrong
forms for the test, so you will now have 7 minutes to complete the test. We will let you
know when you have 2 minutes left. We cannot allow you to make up for the lost time,
because then we will not have enough time for the rest of the measures. You all should
have the right form of the test now. Please start marking your answers on question
number _ on the scantron. Stop working when you have reached the end of the test,
which will be marked by a STOP on a page in the packet. This is an individual task, so
please work on this independently. You may begin.”

Sensitive Explanation/Consistent:
“You will be given the 12 minutes to complete this test like the other groups have been

given. We will let you know when you have 2 minutes left. We know that you have
taken time out of your busy schedule to help us out, and please know that we really
appreciate that. Thank you. You all should the right form of the test. Please start
marking your answers on question number __ on the scantron. Stop working when you
have reached the end of the test, which will be marked by a STOP on a page in the

161

packet. This is an individual task, so please work on this independently. You may
begin.”

Sensitive Explanation/Inconsistent:
Erase the 12 minutes and write 7 minutes on the board. “Although the other groups were

given 12 minutes to complete this test, you will not be. We mistakenly had the wrong
forms for the test, so you will now have 7 minutes to complete the test. We will let you
know when you have 2 minutes left. We cannot allow you to make up for the lost time,
because then we will not have enough time for the rest of the measures. We understand
that this may be frustrating to you, and we sincerely hope this will not be a problem for
you. We know that you have taken time out of your busy schedule to help us out, and
please know that we really appreciate that. Thank you. You all should have the right
form of the test now. Please start marking your answers on question number _ on the
scantron. Stop working when you have reached the end of the test, which will be marked
by a STOP on a page in the packet. This is an individual task, so please work on this
independently. You may begin.”

GIVE SUBJECTS A TWO MINUTE WARNING, AND TELL THEM TO STOP ONCE
TIME IS OVER.

“Please stop. Make sure your name is written on the ﬁrst page of the exam, and pass
your exams to the front of the row. We now have some additional short questionnaires for

you to ﬁll out while we grade your exams. Please start marking your answers on scantron
2.99

PASS OUT QUESTIONNAIRES. FEIGN TO GRADE EXAMS WITH ASSISTANT.
WRITE PARTICIPANTS NAMES ON FEEDBACK FORMS. WAIT UNTIL
PARTICIPANTS FINISH FILLING OUT QUESTIONNAIRE.

“We have marked your exams, and will now pass you back feedback sheets letting you
know how you did. After the feedback statement, there are a ﬁnal set of questions to
answer. You should begin answering these question on question number _ on the
scantron sheet. Once you are ﬁnished answering these questions, bring your scantron and
your feedback form up to the front. PASS OUT FEEDBACK FORMS, AND CALL
OUT NAMES WHEN HANDING BACK.

COLLECT SCANTRONS FROM THOSE WHO HAVE FINISHED AND THANK
THEM FOR THEIR TIME. GIVE THEM DEBRIEFING SHEET.

162

APPENDIX D

Survey One

Company XYZ Application Form
Dear Applicant:

We have been informed that you are interested in applying for a position as an intern in
our company. Because we have several partner organizations that are actively involved
in our internship program, we have a variety of internship opportunities available for you
if you are hired as an intern. Currently, there are four areas that have open positions
available and we would like you to specify exactly which you would like to be
considered for, so that we can best suit your preferences if you are hired. The four
positions along with their descriptions are listed below.

Position 1:

Intern at WKYM Channel 5 News:

WKYM is a well-known news station that has had a successful internship program for
many years. As an intern at this station, you would be involved in researching news
developments, drafting news stories, and producing various segments in the program.
Some interns have even had the Opportunity of being a guest anchor for the morning
show! Interns are currently needed in the entertainment and local events areas.

Intern at Magnetika Inc.:

Magnetika Inc. is a major Fortune 500 corporation that is well-respected nationwide.
Working as an intern in this organization would be a great introduction into the business
world. This company will give you the ﬂexibility of choosing from a variety of business
areas to work in including sales, marketing, human resources, accounting, etc. You
would also have the Opportunity to periodically sit in on executive board meetings.

Intern at Zephyr Org.:

Zephyr Org. is a research and development ﬁrm specializing in pharmaceutical and
environmental work. As an intern at this organization, you would be an active participant
in various activities, from researching the impact of various drugs, to traveling on-site to
investigate environmental issues, drafting up reports of research results, and presenting
information with your research group to top management. You will be exposed to
various aspects of the research process as well as cutting-edge research that is being
conducted in the areas of medicine and the environment.

Intern at GACW (Group for the Advocacy of Children’s Welfare):

GACW is a non-proﬁt organization that focuses on promoting the welfare of children in
inner-city neighborhoods. As an intern with GACW you will gain experience in

163

community outreach programs, child welfare workshops, abuse counseling and
educational initiatives. You will also learn how to draft reports and referendums to alter
and improve current child welfare policy, and even have the chance to meet with some of
the major political players in Congress who are actively involved in child welfare policy
formation.

In order to better gauge your ﬁt for the job, we would like you to list 1) the internship
position you are seeking out of the list of four and 2) the reasons why you would like this
job.

Position for which you are applying:

 

Why would you like to have this job over the other jobs?

 

 

 

 

 

 

 

 

 

 

 

 

STOP: PLEASE WAIT FOR INSTRUCTIONS TO BEGIN NEXT SECTION.

/—\
STOP

LJ

 

 

 

 

ATTITUDE AND PREFERENCES SURVEY

This survey is designed to measure your attitudes and preferences concerning
various aspects of your life. All your answers on this survey will be kept strictly
conﬁdential, so please answer the following questions as truthfully as possible.
If you have any questions, raise your hand and the test administrator will help
you.

Mark your answers on the scantron provided. Please read the following
statements as if they referred to you. Indicate your level of agreement with the
statements listed below according to the following scale:

 

 

1 2 3 4 5 6 7
Strongly Disagree Somewhat Don’t Somewhat Agree Strongly
Disagree Disagree Agree or Agree Agree

Dis—agree

 

I enjoy being unique and different from others in many respects.

I feel comfortable using someone’s ﬁrst name soon after I meet them, even when they

are much older than I am.

Even when I strongly disagree with group members, I avoid an argument.

I have respect for the authority ﬁgures with whom I interact.

I do my own thing, regardless of what others think.

I respect people who are modest about themselves.

I feel it is important for me to act as an independent person.

I will sacriﬁce my self interest for the beneﬁt of the group I am in.

I’d rather say “No” directly, than risk being misunderstood.

10. Having a lively imagination is important to me.

11. I should take into consideration my parents’ advice when making education/career
plans.

12. I feel my fate is intertwined with the fate of those around me.

13. I prefer to be direct and forthright when dealing with people I’ve just met.

14. I feel good when I cooperate with others.

15. I am comfortable with being singled out for praise or rewards.

16. If my brother or sister fails, I feel responsible.

IWRNTONEWUMGE

Nl—o

cwsewew

165

 

 

 

l 2 3 4 5 6 7
Strongly Disagree Somewhat Don’t Somewhat Agree Strongly
Disagree Disagree Agree or Agree Agree

Disagree

 

17. I often have the feeling that my relationships with others are more important than my
own accomplishments.

l8. Speaking up during a class (or a meeting) is not a problem for me.

19. I would offer my seat in a bus to my professor (or my boss).

20. I act the same way no matter who I am with.

21. My happiness depends on the happiness of those around me.

22. I value being in good health over everything.

23. I will stay in a group if they need me, even when I am not happy with the group.

24. I try to do what is best for me, regardless of how that might affect others

25. Being able to take care of myself is a primary concern for me.

26. It is important to me to respect decisions made by the group.

27. My personal identity, independent of others, is very important to me.

28. It is important for me to maintain harmony within my group.

29. I act the same way at home as I do at school.

30. I usually go along with what others want to do, even when I would rather do
something different.

TURN TO NEXT PAGE

166

 

 

The following statements are related to the ability test you will be taking Shortly.
The “job” that is being referred to in these statements is the job that you listed on
the ﬁrst page of this packet, in other words, the internship position that you are
applying for. Again, all your answers will be kept conﬁdential, so please answer
as truthfully as possible. Please indicate your agreement with the statements
listed below according to the following scale:

 

 

l 2 3 4 5 6 7
Strongly Disagree Don’t Agree Strongly Agree Strongly
Disagree Agree or Agree Agree

Disaggee

 

 

31. I would like to be hired for this job.

32. It would be good to have this type of a job.

33. I want to get a job with this company.

34. If I do well on this test, I have a good chance of being hired.
35. I think I will be hired if I get a high test score.

36. How well I do on this test will affect whether I am hired.

37. The higher my test score, the better my chance of getting hired.
38. If I try to do my best on this test, I can get a high score.

39. If I concentrate and try hard I can get a high test score.

40. I can get a good score on this test if I put some effort into it.

/——\
STOP

\__/

 

 

 

 

167

APPENDIX E

Survey Two

APPLICANT EXPERIENCES SURVEY

Now that you’ve taken the test, we would like to know about your application
experiences. Please indicate your agreement with the statements listed below
according to the following scales. Please be as honest as possible.

 

 

1 2 3 4 5 6
Strongly Disagree Don’t Agree Strongly Not
Disagree Agree or Agree Applicable

Disagee

 

 

IMPORTANT: PLEASE BEGIN MARKING ON QUESTION # 41 UN SGANTRON

41

. Whether or not I get the job, I feel the selection process was fair.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.

Whether or not I get the job, the procedures used to select people for this job are fair.
Whether or not I get the job, I am satisﬁed with the selection process.

Overall, I feel dissatisﬁed with the w_ay people will be selected for the job.

I would accept the job if it was offered to me.

I would not like to participate in future experiments like this one.

If I was offered the opportunity, I would take part in another experiment like this one.
I will continue to volunteer for experiments similar to this one.

If I hear about other projects like this, I would be interested in applying for them.

I would recommend this experiment to my friends.

I would tell my ﬁiends to participate in this experiment.

I would tell others this is a good experiment in which to participate.

I think other people should know this is a good experiment.

I would tell others that I enjoyed participating in this experiment.

I would tell others that this was a good experience for me.

TURN TO NEXT PAGE

168

 

Before you receive feedback on your performance on the selection test, we have a
few more questions concerning your background, personality and preferences.
As always, all your answers on this survey will be kept strictly conﬁdential, so
please answer the following questions as truthfully as possible. If you have any
questions, raise your hand and the test administrator will help you.

Mark your answers on the scantron provided. Please read the following
statements as if they referred to you. Indicate your level of agreement with the
statements listed below according to the following scale:

 

 

l 2 3 4 5 6 7
Strongly Disagree Somewhat . Don’t Somewhat Agree Strongly
Disagree Disagree Agree or Agree Agree

Diggree

 

 

31 . When faced with a difﬁcult personal problem, it is better to decide what to do myself,
rather than follow the advice of others.

32. I enjoy meeting and talking to my neighbors everyday.

33. I can count on my relatives for help if I ﬁnd myself in any kind of trouble.

34. What happens to me is my own doing.

35. If the group is slowing me down, it is better to leave it and work alone.

36. Even if their child won the Nobel prize, parents should not feel honored in anyway.

37. In most cases, to cooperate with someone whose ability is lower than oneself is not as
desirable as doing the thing on one’s own.

38. People at higher levels of an organization must look after those below them.

39. People at lower levels in a group or organization should carry out the decisions of
people at higher levels.

40. The hierarchy of groups in a society should remain consistent over time.

41. People at higher levels should expect to have more privileges than those at lower
levels.

42. People at lower levels in an organization should not expect to have much power.

43. One should live one’s life independently of others as much as possible.

44. I would help, within my means, if a relative told me that he (she) is in ﬁnancial
difﬁculty. (In this questionnaire, “relatives” refer to those relatives who are not your
next of kin. Uncles, cousins, grandmother fall into this category).

TURN TO NEXT PAGE

169

 

l 2 3 4 5 6 7
Strongly Disagree Somewhat Don’t Somewhat Agree Strongly
Disagree Disagree Agree or Agree Agree

Diggree

 

 

 

45. I would rather struggle through a personal problem by myself than discuss it with my
ﬁiends.

46. I like to live close to my good friends.

47. The most important thing in my life is to make myself happy.

48. I tend to do my own things, and most people in my family do the same.

49. Aging parents should live at home with their children.

50. What I look for in a job is a friendly group of coworkers.

51. One does better working alone than in a group.

52. Individuals should be judged on their own merits, not on the company they keep.

53. Organizations work best with clear and formal hierarchies.

79. What is your age?

1 8

19

20

21

22

23

24+

whee-99‘?

80. What is your gender?
a. male
b. female

81. What is your year in school?
freshman

sophomore

junior

senior

5''1 year +

graduate

rm 99 9‘!»

82. Which of the following best characterizes you?
a. US. Citizen
b. Non-citizen — Canadian
c. Non—citizen - other

TURNTONEXTPAGE

170

83. Is English your primary language?

a.
b.

yes
no

84. What ethnicity do you consider yourself to be?

r-p'qe rm 9.0 9‘9»

Mexican American

Puerto Rican

Other Hispanic

American Indian or Alaskan native

Asian

Black/African American
Caucasian/Whitc/Not of Hispanic origin
Native Hawaiian or other Paciﬁc Islander
Other

85. How many years have you been in the United States?

9999‘?

<1 year
1-3 years
3-5 years
5-10 years
>10 years

86. What is your current cumulative GPA (grade point average)?

qo .

less than 1.00
1.00 to 1.49

a
b.
c. 1.50 to 1.99
d.
e
f

2.00 to 2.49

. 2.50 to 2.99

3.00 to 3.49
3.50 or greater

87. Have you had any work experience within the United States?

a.

b.

Yes
No

88. How many times have you participated in a job interview?

sup-99‘s»

never
1-2 times
3-4 times
5-6 times
more than 6 times

89. How many times have you had to take a test in order to get a job?

a
b.
c.
d.
6

never
1-2 times

3-4 times

5-6 times

more than 6 times

171

/—\
(STOP)

L/

STOP: PLEASE WAIT FOR INSTRUCTIONS TO BEGIN NEXT SECTION. DO NOT
TURN "IE PAGE.

 

 

 

 

172

Now that you have received feedback on your performance, there are a few more
questions we would like you to answer. You may see some of the same questions
you have answered before, but please answer these according to how you feel
currently. Again, please reply as honestly as possible because your answers will
be kept conﬁdential. Indicate your agreement with the statements below

 

 

 

 

according to the following scale:

1 2 3 4 5 6
Strongly Disagree Don’t Agree Strongly Not
Disagree Agree or Agree Applicable

Disagree
90. I feel the selection process was fair.

91.
92.
93.
94.
95.
96.
97.

98.
99.

The procedures used to select people for this job are fair.

I am satisﬁed with the selection process.

Overall, I feel dissatisﬁed with the way people were selected for the job.

Overall, I feel the results of the selection process were unfair.

I feel the hiring decision (accept/rej ect) was fair.

Overall, I am satisﬁed with the hiring decision.

I am dissatisﬁed with the test administrator’s decision about whether or not to hire
me.

Even if I was now offered the job, I would not accept it.

I would not like to participate in ﬁiture experiments like this one.

100. If I was offered the opportunity, I would take part in another experiment like this

101.
102.
103.
104.
105.
106.
107.
108.
109.

one.

I will continue to volunteer for experiments similar to this one.

If I hear about other projects like this, I would be interested in applying for them.

I would recommend this experiment to my friends.

I would tell my friends to participate in this experiment.

I would tell others this is a good experiment in which to participate.

I think other people should know this is a good experiment.

I would tell others that I enjoyed participating in this experiment.

I would tell others that this was a good experience for me.

The explanation given for any problems during the selection process was adequate.

TURN TO NEXT PAGE

173

 

 

 

 

l 2 ‘ 3 4 5 6
Strongly Disagree Don’t Agree Strongly Not
Disagree Agree or Agree Applicable

Disagree
110. I was satisﬁed with the explanation I received for a problem during the selection
process.

111. I was given an inadequate explanation for a problem during the selection process.

112. I received an explanation for why I was given the amount of time I had to complete
the test. ,

113. It was made very clear to me why I was given the amount of time I had to complete
the test.

114. The test administrator(s) treated everyone politely during the testing process.

115. Test administrator(s) were considerate during the test.

116. Test administrator(s) answered questions in a straightforward and polite manner.

117. The testing administrator put everyone at ease during the test.

118. I was satisﬁed with the way everyone was treated at the test site.

119. The test administrators did not treat any applicants in an impolite manner.

120. The test was administered to all applicants in the same way.

121. There were no dzﬂerences in the way the test was administered to different
applicants.

122. The test administrator made no distinction in how he/she treated applicants.

123. I received the same amount of time as other people when taking the test.

124. I received less time than other people when taking the test.

125. I was treated differently than others who have taken this test.

126. The test administrator treated all applicants equally.

You have now completed the survey. Please give this booklet, your pencil (if

borrowed from administrator) and your scantron to the test
administrator. Thank you for your time!

174

APPENDIX F

Debrieﬁng Form

Debrieﬁng Form for Reactions in a Selection Context Study

Thank you very much for your participation in this study. The main purpose of this study was to examine
how different cultural orientations affect how people perceive a selection process. Research has shown that
there are certain “rules” that people expect to be upheld during a selection process, and if these rules are
violated, people do not react as favorably to the selection process (Gilliland, 1993). This negative reaction
can affect many of the individual’s behavioral intentions, including intentions to accept the job, intentions
to recommend the company to others, and attractiveness of the organization ( Macon, Avedon, Paese, &
Smith, 1994; Smither, Reilly, Millsap, Pearlman, & Stoffey, 1993). What we wished to examine was how
people’s cultural orientation affected whether they perceived the violation of certain rules to be negative or
not.

Therefore, the selection test was given to you, not in order to see how it performed with individuals of
different ages, but to see how your performance changed when rules were violated. Therefore, we gave
individuals in certain conditions less time to take the test than others purposefully. If you were in one of
these conditions, contrary to what we presented, we did not possess the wrong exam initially, this was used
as an excuse to reduce the amount of time you had to complete the exam. We are sorry for this deception,
but we felt it was necessary in order to present a realistic and believable situation in which some
individuals received more time than others. You may have also experienced an instance when the test
administrator was rude to a fellow test taker. This was also done to see how you would react to this
situation, not because the test administrator was naturally short of temper. You also may have been told
that you did not do well on the test you took. We actually did not score individuals’ performance on the
exam during the experimental period, we simply presented negative feedback to all participants. Therefore,
if you received negative feedback on your performance telling you that you would not be hired or entered
into the lottery to win the prize. this was FALSE, and NOT reflective of your TRUE performance on
the test. The reasoning behind giving everyone negative feedback was that we determined both through
research and deduction, that it in situations where people are accepted for a job, they are much more likely
to react positively, and other aspects of the application process will not be as important to them (Ployhart &
Ryan, 1998). Consequently, we felt it would be more interesting and insightful to look at how individuals
would react differently when they were rejected for a position. We are genuinely sorry for any distress this
may have caused you.

Due to the nature of the manipulation, ALL individuals, not just the top six scorers, will have their names
put into the lottery for the monetary prize THREE times. Also, we will score your performance on the test
at a later date, and the top scorers from each session will have their names entered into the lottery for the
free movie tickets. Therefore, you will be awarded for scoring well, and those of you who had less time
will NOT be at a disadvantage, because you will be compared ONLY to those people who were in the
same session as you. You will be informed at the end of data collection if you are a prize recipient or not
through email.

Thank you again for your participation, and we please ask you to PLEASE NOT mention the nature of the
manipulations or deceptions in the experiment to others who may participate in this experiment in the
future. Sharing this information with others can bias they way they respond to the questions, and threaten
the validity of the study we worked so hard to design. In addition, this could lead to the need for us to
collect additional data, thereby reducing your odds of winning the prizes. We appreciate your help, and if
you have further questions about this study in particular, or Industrial/Organizational Psychology in
general, please do not hesitate to contact us (Smriti Shivpuri, shivpur@msu.edu, or Dr. Ann Marie Ryan,

ﬂanan@msu.edu).

175

References:
Gilliland, SW. (1993). The perceived fairness of selection systems: An organizational
justice perspective. Academy of Management Review, 18, 694-734.
Macon, T.l-l., Avedon, M.J., Paese, M., & Smith, DE. (1994) The effects of applicants’
reactions to cognitive ability tests and an assessment center. Personnel Psychology, 47, 715-738
Ployhart, R.E., & Ryan, AM. (1998). Applicants’ reactions to fairness of selection
procedures: The effects of positive rule violations and time of measurement. Journal of Applied Psychology,
83, 3-16.
Smither, J.W., Reilly, R.R., Millsap, R. E., Pearlman, K., & Stoffey, R.W. (1993) Applicant reactions to selection
procedures. Personnel Psychology, 46, 49-76

176

REFERENCES

Aguinis, H., Beaty, J .C., Boik, R.J., & Pierce, CA. (2005). Effect size and power in
assessing moderating effects of categorical variables using multiple regression: A
30-year review. Journal of Applied Psychology, 90, 94-107.

Ambrose, M.L., Kulik, C.T., & Harden, L.K. (1991). Inﬂuence of social comparisons on
perceptions of organizational fairness. Journal of Applied Psychology, 76, 239-
246.

Arvey, R.D. & Sackett, PR. (1993). Fairness in selection: Current developments and
perspectives. In N. Schmitt & W.C. Borman (Eds), Personnel Selection in
Organizations (pp. 171-202). San Francisco: Jossey-Bass.

Arvey, R.D., Strickland, W., Drauden, G., & Martin, G. (1990). Motivational
components of test-taking. Personnel Psychology, 43, 695-716.

Avery, DR. (2003). Personality as a predictor of the value of voice. Journal of
Psychology, 13 7, 435—446.

Avery, D.R., & Quinones, M.A. (2002). Disentangling the effects of voice: The
incremental roles of opportunity, behavior, and instrumentality in predicting
procedural fairness. Journal of Applied Psychology, 87, 81-86.

Baron, R.M, & Kenny, DA. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1 173-1 182.

Bauer, T.N., Truxillo, D.M., Sanchez, R.J., & Craig, J.M. (2001). Applicant reactions to
selection: Development of the selection procedural justice scale. Personnel
Psychology, 54, 387-419.

Bauer, T.N., Dolen, M.R., Maertz, C.P., & Campion, M.A. (1998). Longitudinal
assessment of applicant reactions to employment testing and test outcome
feedback. Journal of Applied Psychology, 83, 892-903.

Bazennan, M.H., Schroth, H.A., Shah, P.P., Diekmann, K.A., & Tenbrunsel, AB. (1994)
The inconsistent role of comparison others and procedural justice in reactions to
hypothetical job descriptions: Implications for job acceptance decisions.
Organizational Behavior and Human Decision Processes, 60, 326-352

Bell, B.S., Ryan, A.M., & Wiechmann, D. (2004). Justice expectations and applicant
perceptions. International Journal of Selection and Assessment, 12, 24-38.

177

Bies, R.J. (1985). Individual reactions to corporate recruiting encounters: The
importance of fairness. Unpublished manuscript.

Bies, R.J. (1987). The predicament of injustice: The management of moral outrage. In
L.L. Cummings and BM. Staw (Eds), Research in Organizational Behavior
(Vol. 9), Greenwich, CT: JAI Press.

Bies, R.J. & Moag, IS. (1986). Interactional justice: Communication criteria of fairness.
In R.J. Lewicki, B.H. Sheppard, and M.H. Bazerrnan (Eds), Research on
negotiation in organizations (pp. 43-55). Greenwich, CT: JAI Press.

Bies, R.J., & Shapiro, BL. (1988). Voice and justiﬁcation: Their inﬂuence on
procedural fairness judgments. Academy of Management Journal, 31, 676-685.

Bies, R.J., Shapiro, D.L., & Cummings, LL. (1988). Causal accounts and managing
organizational conﬂict: Is it enough to say it’s my fault? Communication
Research, 14, 381-399.

Blader, S.L., Chang, C.C., & Tyler, TR. (2001). Procedural justice and retaliation in
organizations: Comparing cross-nationally the importance of fair group processes.
International Journal of Conﬂict Management, 12, 295-311.

Bond, M.H. (2002). Reclaiming the Individual from Hofstede’s Ecological Analysis—A
20-Year Odyssey: Comment on Oyserman et al. (2002). Psychological Bulletin,
128, 73-77.

Bond, M.H., Leung, K., & Wan, K.C. (1982). How does cultural collectivism operate?
The impact of task and maintenance contributions on reward distribution. Journal
of Cross Cultural Psychology, 13, 186-200.

Bontempo, R. (1993). Translation ﬁdelity of psychological scales: An item response
theory analysis of an individualism-collectivism scale. Journal of Cross-Cultural
Psychology, 24, 149-166.

Brockner, J ., & Wiesenfeld, BM. (1996). An integrative framework for explaining
reactions to decisions: Interactive effects of outcomes and procedures.
Psychological Bulletin, I 20, 189-208

Buchan, NR. (1998). Culture, fairness, and trust: Contrasting inﬂuences on negotiation
behavior and outcomes in China, Korea, Japan, and the United States.

Dissertation Abstracts International Section A: Humanities and Social Sciences,
59, 1256.

Burke, M.J., Normand, J ., & Raju, NS. (1987). Examinee attitudes toward computer-
administered ability testing. Computers in Human Behavior, 3, 95-107

178

 

Carpenter, S., & Radhakrishnan, P. (2000). Allocentrism and idiocentrism as predictors
of in-group perceptions: An individual difference extension of cultural patterns.
Journal of Research in Personality, 34, 262-268.

Cellar, D.F., Miller, M.L., Doverspike, D.D., & Klawsky, J .D. (1996). Comparison of
factor structures and criterion-related validity coefﬁcients for two measures of

personality based on the ﬁve factor model. Journal of Applied Psychology, 81,
694-704.

Chan, D. (1997). Racial subgroup differences in predictive validity perceptions on
personality and cognitive ability tests. Journal of Applied Psychology, 82, 311-
320

Chan, D., & Schmitt, N. (2004). An agenda for future research on applicant reactions to
selection procedures: A construct-oriented approach. International Journal of
Selection and Assessment, 12, 9-23.

Chan, D. Schmitt, N., Jennings, D., Clause, C. S., &Delbridge, K. (1998a). Applicant
perceptions of test fairness: Integrating justice and self-serving bias perspectiv.es
International Journal of Selection and Assessment, 6, 232- 239

Chan, D., Schmitt, N., Sacco, J.M., & DeShon, R.P. (1998b). Understanding pretest and
posttest reactions to cognitive ability and personality tests. Journal of Applied
Psychology, 83, 471-485

Chapman, BS. (2001). Modeling job applicant decision processes: Integrating applicant
reactions to selection procedures into the critical contact framework of recruiting.

Dissertation Abstracts International: Section B; The Sciences and Engineering,
61, 5605

Chinese Cultural Connection. (1987). Chinese values and the search for culture-ﬂee
dimensions of culture. Journal of Cross-Cultural Psychology, 18, 143-164

Choi, J. (2001). A cross-cultural study of procedural justice: The effects of outcomes,
procedures, and cultural values on procedural justice perceptions and employee
trust in managers. Dissertation Abstracts International Section A: Humanities
and Social Sciences, 62, 1116.

Church, A.T. (2000). Culture and personality: Toward an integrated cultural trait
psychology. Journal of Personality, 69, 651-703.

Cohn, E.S., White, 8.0., & Sanders, J. (2000). Distributive and procedural justice in
seven nations. Law and Human Behavior, 24, 553-579.

Colquitt, J .A., & Chertkoff, J.M. (2001). Explaining injustice: The interactive effect of

179

 

explanation and outcome on fairness perceptions and task motivation. Journal of
Management, 28, 591-610.

Colquitt, J.A., & Greenberg, J. (2003). Organizational justice: A fair assessment of the
state of the literature. In J. Greenberg (Ed) Organizational Behavior: The State
of the Science (pp. 165-210). Mahwah, NJ: Lawrence Erlbaum Assoc.

Coon, H.M., & Kemmelrneier, M. (2001). Cultural orientations in the United States:
(Re)Examining differences among ethnic groups. Journal of Cross-Cultural
Psychology, 32, 348-364.

Dipboye, R.L. & De Pontbriand, P. (1981). Correlates of employee reactions to
performance appraisals and appraisal systems. Journal of Applied Psychology,
66, 248-251.

Doucet, L., & Jehn, K.A. (1997). Analyzing harsh words in a sensitive setting: American

expatriates in communist China. Journal of Organizational Behavior, 18, 559-
582.

Douthitt, E.A., Aiello, J .R. (2001). The role of participation and control in the effects of
computer monitoring on fairness perceptions, task satisfaction, and performance.
Journal of Applied Psychology, 86, 867-874.

Earley, P.C., & Gibson, C. B. (1998). Taking stock of our progress on individualism-
collectivism: 100 years of solidarity and community. Journal of Management, 24,
265-304.

Endler, N.S., & Magnusson, D. (1976). Toward an interactional psychology of
personality. Psychological Bulletin, 83, 956-974.

Erez, M. & Earley, RC. (1993). Culture, Self-Identity, and Work. New York, NY:
Oxford University Press.

Farh, J .L., Earley, P.C., & Lin, SC. (1997). Impetus for actions: A cultural analysis of
justice and organizational citizenship behavior in Chinese society. Administrative
Science Quarterly, 42, 421-444.

Fiaschetti, SR. (1998). Applicant reactions and intentions toward the company: Before
and aﬂer the hiring decision. Dissertation Abstracts International: Section B: The

Sciences & Engineering, 59, 1398.

Fields, H. (1950). The group interview test: its strength. Public Personnel Review, 11,
139-146.

F olger, R. & Cropanzano, R. (1998). Organizational justice and human resource
management. Thousand Oaks, CA: Sage.

180

Gelfand, M.J., Higgins, M., Nishii, L.H., Raver, J .L., Dominguez, A., Murakami, F .,
Yarnaguchi, S., & Toyama, M. (2002). Culture and egocentric perceptions of
fairness in conﬂict and negotiation. Journal of Applied Psychology, 8 7, 833-845.

Gelfand, M.J., Triandis, H.C, & Chan, D.K.-S. (1996). Individualism versus collectivism
or versus authoritarianism? European Journal of Social Psychology, 26, 397-410.

Gilliland, SW. (1993). The perceived fairness of selection systems: An organizational
justice perspective. Academy of Management Review, 18, 694-734.

Gilliland, S. W. (1994). Effects of procedural and distributive justice on reactions to a
selection system. Journal of Applied Psychology, 79, 691-701.

Gilliland, S.W. & Beckstein, BA. (1996). Procedural and distributive justice in the
editorial review process. Personnel Psychology, 49, 669-691.

Gilliland, S.W., Benson, L., & Schepers, DH. (1998). A rejection threshold in justice
evaluations: Effects on judgment and decision-making. Organizational Behavior
and Human Decision processes, 7 6, 113-131

Gilliland, S.W., & Cherry, B. (2000). Managing “customers” of selection processes. In
J .F. Kehoe (Ed.), Managing selection in changing organizations: Human
resource strategies (pp. 158-196). San Francisco: Jossey—Bass.

Gilliland, S.W., Groth, M., & Baker, RC. (2001). Improving applicants’ reactions to
rejection letters: An application of fairness theory. Personnel Psychology, 54,
669-703 '

Gomez, C. (2003). The relationship between acculturation, individualism/collectivism,
and job attribute preferences for Hispanic MBAs. Journal of Management
Studies, 40, 1089-1105.

Greenberg, J. (1986). Determinants of perceived fairness of performance evaluations.
Journal of Applied Psychology, 71, 340-342.

Greenberg, J. (1987). Reactions to procedural injustice in payment allocations: Do the
ends justify the means? Journal of Applied Psychology, 72, 55-61.

Greenberg, J. (1990). Organizational performance appraisal procedures: What makes
them fair? In M.H. Bazcrrnan, R.J. Lewicki, & B.H. Sheppard (Eds), Research on

negotiation in organizations. Greenwich, CT: J AI Press

Greenberg, J. (1991 ). Using explanations to manage impressions of performance
appraisal fairness. Employee Responsibilities and Rights Journal, 4, 51-60.

181

Greenberg, J. (1993). Stealing in the name of justice: Informational and interpersonal
moderators of theft reactions to underpayment inequity. Organizational Behavior
and Human Decision Processes, 54, 81-103.

Greenberg, J. (1994). Using socially fair treatment to promote acceptance of a work site
smoking ban. Journal of Applied Psychology, 79, 288-297.

Greenstein, R., J affe, J., & Kayatin, T. (1999). Low unemployment, rising wages fuel
poverty decline. Center on Budget and Policy Priorities. Retrieved ﬁ'om
http://www.cbpp.org/9-30-99pov.pdf - Feb 2, 2004.

Grifﬁn, B., & Hesketh, B. (2004). Why openness to experience is not a good predictor of
job performance. International Journal of Selection and Assessment, 12, 243-
25 1 .

Gross, ML. (1962). The Brain- Watchers. New York: Random House. Retrieved from
www.asa3.org/ASA/topics/Mathematics/PSCF6-90Faw.html - 42k - Nov. 15,
2002

Gudykrmst, W.B., Gao, G., Schmidt, K.L., Nishida, T., Bond, M.H., Leung, K., Wang,
G., & Barraclough, RA. (1992). The inﬂuence of individualism-collectivism,
self-monitoring, and predicted outcome value on communication in ingroup and
outgroup relationships. Journal of Cross-Cultural Psychology, 23, 196-213.

Hardin, E.E., Leong, F .T., Bhagwat, AA. (2004). Factor structure of the self-construal
scale revisited: lrnplications for the multidimensionalilty of self-construal.
Journal of Cross-Cultural Psychology, 35, 327-345.

Hausknecht, J .P., Day, D.V., & Thomas, SC. (2004). Applicant reactions to selection
procedures: An updated model and meta-analysis. Personnel Psychology, 5 7,
639-683.

Hayashi, C. (1988). National character of the Japanese. Tokyo: Statistical Bureau,
Japan.

Heilman, M.E., Rivero, J .C., & Brett, J .G. (1991). Skirting the competency issue: Effects
of sex-based preferential selection on task choices of men and women. Journal of
Applied Psychology, 76, 99-105

Hofstede, G. (1980). Culture’s Consequences: International Differences in Work Related
Values. Newbury Park, CA: Sage Publications.

Hofstede, G., & Spangenberg, J. (1987). Measuring individualism and collectivism at
occupational and organizational levels. In C. Kagitcibasi (Ed.), Growth and
Progress in Cross-Cultural Psychology (pp. 113-122). Berwyn, PA: Swets North
America.

182

Horvath, M., Ryan, A.M., 8L Stierwalt, S. (2000) The inﬂuence of explanations for
selection test use, outcome favorability and self-efﬁcacy on test-taker perceptions.
Organizational Behavior and Human Decision Processes, 83, 310-330

Hui, CH. (1984). Individualism-collectivism: Theory, measurement and its relation to
reward allocation. Unpublished manuscript. Champaign, IL: University of
Illinois.

Hui, CH. (1988). Measurement of individualism-collectivism. Journal of Research in
Personality, 22, 17-36.

Hui, C.H., & Au, K. (2001). Justice perceptions of complaint-handling: A cross-cultural
comparison between PRC and Canadian customers. Journal of Business
Research, 52, 161-173.

Hui, C.H., & Triandis, H.C. (1985). Measurement in cross-cultural psychology. Journal
of Cross-Cultural Psychology, 16, 131-152.

Hui, C.H., Triandis, H.C., & Yee, C. (1991). Cultural differences in reward allocation:
Is collectivism the explanation? British Journal of Social Psychology, 30, 145-
157.

Hundley, G., & Kim, J. (1997). National culture and the factors affecting perceptions of
pay fairness in Korea and the United States. International Journal of
Organizational Analysis, 5, 325-341.

Hunthausen, J.M. (2000) Predictors of task and contextual performance: Frame-of-
reference effects and applicant reaction effects on selection system validity.
Dissertation Abstracts International: Section B: The Sciences and Engineering,

61,1119

James, K. (1993). The social context of organizational justice: Cultural, intergroup, and
structural effects on justice behaviors and perceptions. In R. Cropanzano (Ed.),
Justice in the Workplace: Approaching Fairness in Human Resource
Management. Hillsdale, N.J.: Lawrence Erlbaum AssoCiates.

Janz, T., & Mooney, G. (1993). Interviewer and candidate reactions to patterned
behavior description interviews. International Journal of Selection and

Assessment, 1, 165-169

J ehn, K. & Doucet, L. (1997). Analyzing harSh words in a sensitive setting: American
expatriates in communist China. Journal of Organizational Behavior, 18, 559-
582.

Judge, T.A., & Bono, J .E. (2001). Relationship of core self-evaluations traits—self-
esteem, generalized self-efﬁcacy, locus of control, and emotional stability—with

183

job satisfaction and job performance: A meta-analysis. Journal of Applied
Psychology, 86, 80-92.

Kabanoff, B. (1997). Organizational justice across cultures. In P. C. Early and M. Erez
(Eds) New Perspectives on International Industrial/Organizational Psychology.
San Francisco: New Lexington Press, 676-712.

Kashima, Y., Yarnaguchi, S., Kim, U., Choi, S.-C., Gelfand, M.J., & Yuki, M. (1995).
Culture, gender, and self: A perspective from individualism-collectivism research.
Journal of Personality and Social Psychology, 69, 925-937.

Kluckhohn, F., & Strodtbeck, F. (1961). Variations in Value Orientation. Westport:
Greenwood Press.

Kluger, A.N., & Rothstein, HR. (1993) The inﬂuence of selection test type on applicant
reactions to employment testing. Journal of Business and Psychology, 8, 3-25.

Krishnan, L. (1998). Allocator/recipient role and resource as determinants of allocation
rule preference. Psychological Studies, 43, 21-29.

Kohn, L.S., & Dipboye, R.L. (1998). The effects of interview structure on recruitment
outcomes. Journal of Applied Psychology, 28, 821-843

Latham, G.P., & Ginnegan, B.J. (1993). Perceived practicality of unstructured, patterned,
and situational interviews. In H. Schuler, J .L. Farr & M. Smith (eds). Personnel

selection and assessment: Individual and organizational perspectives (pp. 41-55).
Hillsdale, N.J.: Erlbaum.

Leung, K. (1997). Negotiation and reward allocations across cultures. In P.C. Earley &
M. Erez (Eds) New Perspectives on International Industrial/Organizational
Psychology, (pp. 640-675). San Francisco: New Lexington Press.

Leung, K. & Li, KL. (1990). Psychological mechanisms of process-control effects.
Journal of Applied Psychology, 75, 613-620.

Leung, K. & Lind, EA. (1986). Procedural justice and culture: Effects of culture, gender
and investigator status on procedural preferences. Journal of Personality and
Social Psychology, 50, 1134-1140.

Leung, K., Au, Y., Femandez-Dols, J.M., & Iwawaki, S. (1992). Preference for methods
of conﬂict processing in two collectivist cultures. International Journal of
Psychology, 27, 195-209.

Leung, K., Smith, P.B., Wang, Z.M., & Sun, HF. (1996). Job satisfaction in joint

venture hotels in China: An organizational justice analysis. Journal of
International Business Studies, 2 7, 947-962.

184

Leung, K., Su, 8., & Morris, M. (1998). When is criticism not constructive? The roles of
fairness perceptions and dispositional attributions in employee acceptance of
critical supervisory feedback. Unpublished manuscript, Chinese University of
Hong Kong.

Leventhal, GS. (1980). What should be done with equity theory? In K]. Gergen, M.S.
Greenberg, & R.H. Willis (Eds), Social exchange: Advances in theory and
research (pp. 27-55). New York: Plenum.

Lewis, K. (1996). Do individual values matter? The effect of team member moral
character on work team reputation and performance in a cross-cultural setting.
Unpublished doctoral dissertation, University of California, Irvine.

Lincoln, J .R., Hanada, M., & Olson, J. (1981). Cultural orientation and individual
reactions to organizations: A study of employees of Japanese armed forces.
Administrative Science Quarterly, 26, 93-115.

Lind, E.A., & Tyler, T. R. (1988). The social psychology of procedural justice. New
York: Plenum.

Lind, E.A., Tyler, T.R., & Huo, Y.J. (1997). Procedural context and culture: Variation in
the antecedents of procedural justice judgments. Journal of Personality and
Social Psychology, 73, 767-780.

Lounsbury, J.W., Bobrow, W., & Jensen, J.E. (1989). Attitudes toward employment
testing: Scale development, correlates, and “known-group” validation.
Professional Psychology: Research and Practice, 20, 340-349

Macon, T.H., Avedon, M.J., Paese, M., & Smith, DE. (1994) The effects of applicants’
reactions to cognitive ability tests and an assessment center. Personnel
Psychology, 4 7, 715-738

Markus, H.R., & Kitayama, S. (1991). Culture and the self: Implications for cognition,
emotion, and motivation. Psychological Review, 98, 224-253.

Martin, C.L., & Nagao, D. (1989). Some effects of computerized interviewing on job
applicant responses. Journal of Applied Psychology, 74, 72-80

Matsumoto, D., Weissman, M.D., Preston, K., Brown, B.R., & Kupperbusch, C. (1997).
Context-speciﬁc measurement of individualism-collectivism on the individual
level. Journal of Cross-Cultural Psychology, 28, 743-767.

Maznevski, M.L., DiStefano, J .J ., Gomez, C.B., Noorderhaven, N.G., & Wu, P. (1997).
The cultural orientations framework and international management research.

185

Paper presented at the Academy of International Business annual meeting,
Guadalajara, Mexico.

McFarlin, D.B., & Sweeney, PD. (2001). Cross-cultural applications of organizational
justice. In R. Cropanzano (Ed), Justice in the Workplace, (Vol. 2). Hillsdale,
N.J.: Lawrence Erlbaum Associates.

Miles, J .A. & Greenberg, J. (1993). Cross-national differences in preferences for
distributive justice norms: The challenge of establishing fair resource allocations
in the European community. Research in Personnel and Human Resource
Management, Suppl. 3, 133-156.

Milliman, J., Nason, S., Gallagher, E., Huo, P., Von Glinow, M.A., & Lowe, KB. (1998).
The impact of national culture on human resource management practices: The
case of performance appraisal. In J .L. Cheng and RB. Peterson (Eds) Advances
in International Comparative Management, 12. Stamford, CT: J AI Press.

Moscoso, S. (2000) Selection interview: A review of validity evidence, adverse impact
and applicant reactions. International Journal of Selection and Assessment, 8,
237-247

Mosier, CL (1947) A critical examination of the concepts of face validity. Educational
and Psychological Measurement, 7, 191-205

Murphy, K.R., Thornton, G.C., & Prue, K. (1991). Inﬂuence of job characteristics on the
acceptability of employee drug testing. Journal of Applied Psychology, 76, 447-
453 '

Murphy, K.R., Thornton, G.C., & Reynolds, DH. (1990) College students’ attitudes
toward employee drug testing programs. Personnel Psychology, 43, 615-631

Murphy-Berman, V., & Berman, J .J . (2002). Cross-cultural differences in perceptions of
distributive justice: A comparison of Hong Kong and Indonesia. Journal of Cross
Cultural Psychology, 33, 157-170.

National Institute of Industrial Psychology (193 8) Some recent developments in the
psychology of personnel management in Great Britain. International

Management Congress, Washington Personnel General Management Papers, 84-
86

Neuman, Y.B. (1996). Perceptions of faculty evaluations programs in Israel and the
United States. Dissertation Abstracts International Section A: Humanities and
Social Sciences, 57, 0579.

Noesjirwan, J. (1978). A rule-based analysis of cultural differences in social behaviour:
Indonesia and Austrailia. International Journal of Psychology, 13, 305-316.

186

Novick, M.R. & Ellis, DD. (1977). Equal opportunity in educational and employment
selection. American Psychologist, 32, 306-320.

Novy, D.M., Kopel, K.F., & Swank, PR. (1996). Psychometrics of oral examinations for
psychology licensure: The Texas examination as an example. Professional
Psychology: Research and Practice, 2 7, 415-417.

Oishi, S., Schimmack, U., Diener, E., & Suh, EM. (1998). The measurement of values
and individualism-collectivism. Personality and Social Psychology Bulletin, 24,
1 177-1 189.

Okazaki, S. (2000). Asian American and White American differences in affective
distress symptoms. Journal of Cross-Cultural Psychology, 31, 603-625.

Oliver, DH. (1998) Exploring applicant reactions to selection processes from an
organizational justice perspective. Dissertation Abstracts International: Section
B: The Sciences & Engineering, 59, 902.

Oyserman, D. (1993). The lens of personhood: Viewing the self, others, and conﬂict in a
multicultural society. Journal of Personality and Social Psychology, 65, 993-
1009.

Oyserrnan, D., Coon, H.M., & Kemmelrneier, M. (2002). Rethinking Individualism and
Collectivism: Evaluation of Theoretical Assumptions and Meta-Analyses.
Psychological Bulletin, 128, 3-72.

Oyserman, D., Kemmelrneier, M., & Coon, HM. (2002). Cultural psychology, a new
look: Reply to Bond (2002), F iske (2002), Kitayama (2002), and Miller (2002).
Psychological Bulletin, 128, 1 10-117.

Parsons, T., & Shils, EA. (1951). Toward A General Theory of Action. Cambridge, MA:
Harvard University Press.

Peterson, J .M. (1949) An Illinois group activity test. Public Personnel Review, 10, 222-
225

Phillips, J.M., & Gully, SM. (2002). Fairness reactions to personnel selection techniques
in Singapore and the United States. International Journal of Human Resource
Management, 13, 1186-1205.

Pillai, R., Williams, E.S., & Tan, J .J . (2001). Are the scales tipped in favor of procedural

or distributive justice?: An investigation of the US, India, Germany, and Hong
Kong (China). International Journal of Conﬂict Management, 12, 312-332.

187

Ployhart, R.E., & Ryan, A.M. (1997) Toward an explanation of applicant reactions: An
examination of organizational justice and attribution frameworks. Organizational
Behavior and Human Decision Processes, 72, 308-335.

Ployhart, R.E., & Ryan, A.M. (1998). Applicants’ reactions to fairness of selection
procedures: The effects of positive rule violations and time of measurement.
Journal of Applied Psychology, 83, 3-16.

Ployhart, R.E., Ryan, A.M., & Bennett, M. (1999) Explanations for selection decisions:
Applicants’ reactions to informational and sensitivity features of explanations.
Journal of Applied Psychologi, 84, 87-106

Rahim, M.A., Magner, N.R., Antonioni, D., & Rahman, S. (2001). Do justice
relationships with organization-directed reactions differ across US and
Bangladesh employees? International Journal of Conﬂict Management, 12, 333-
349.

Ramarnoorthy, N., & Carroll, 8.]. (1998) Individualism/Collectivism orientations and
reactions toward alternative human resource management practices. Human
Relations, 51 , 571-588

Rhee, E., Uleman, J.S., & Lee, H.K. (1996). Variations in collectivism and individualism
by ingroup and culture: Conﬁrmatory factor analyses. Journal of Personality and
Social Psychology, 71, 1037-1054.

Robertson, I.T., Iles, P.A., Gratton, L., & Sharpley, D. (1991). The impact of personnel
selection and assessment methods on candidates. Human Relations, 44, 963-982

Robie, C., Schmit, M.J., Ryan, A.M., & Zickar, MJ. (2000). Effects of item context
speciﬁcity on the measurement equivalence of a personality inventory.
Organizational Research Methods, 3, 348-365.

Rosse, J.G., Miller, J .L., & Stecher, MD. (1991) A ﬁeld study of job applicants’
reactions to personality and cognitive ability testing. Journal of Applied
Psychology, 79, 987-992 ' '

Ryan, A.M., & Chan, D. (1999). Perceptions of the EPPP: How do licensure candidates
view the process? Professional Psychology: Research and Practice, 30, 519-530.

Ryan, A.M., Greguras, G., & Ployhart, RE. (1996) Perceived job relatedness of physical
ability testing for ﬁreﬁghters: Exploring variations in reactions. Human
Performance, 9, 219-240

Ryan, A.M., McFarland, L., Baron, H., & Page, R. (1999). An international look at

selection practices: Nation and culture as explanations for variability in practice.
Personnel Psychology, 52, 359-391.

188

Ryan, A.M. & Ployhart, RE. (2000). Applicants’ perceptions of selection procedures
and decisions: A critical review and agenda for the future. Journal of
Management, 26, 565-606

Ryan, A.M. & Sackett, RR. (1987). Pre-employment honesty testing; Fakability,
reactions of test takers, and company image. Journal of Business and
Psychology, 1, 248-256

Rynes, S.L. (1993). Who’s selecting whom? Effects of selection practices on applicant
attitudes and behavior. In N. Schmitt 8:. WC. Borman (Eds) Personnel selection
in organizations (pp. 240-274). San Francisco: Jossey-Bass

Sanchez, R.J., Truxillo, D.M., & Bauer, TN. (2000). Development and examination of

an expectancy-based measure of test-taking motivation. Journal of Applied
Psychology, 85, 739-750.

Schmidley, D.A., & Gibson, C. (1999). US. Census Bureau, Current Population
Reports, Series P23-195, Proﬁle of the F oreign-Born Population in the United
States: 1997. Washington, DC: US. Government Printing Ofﬁce.

Schmidt, F.L., Urry, V.W., Gugel, J .F. (1978). Computer assisted tailored testing:
Examinee reactions and evaluations. Educational and Psychological
Measurement, 38, 265-273

Schmit, M.J., & Ryan, A.M. (1992). Test-taking dispositions: A missing link? Journal
of Applied Psychology, 77, 629-637

Schmit, M.J., Ryan, A.M., Stierwalt, S.L., & Powell, AB. (1995). Frame of reference
effects on personality scale scores and criterion-related validity. Journal of
Applied Psychology, 80, 607-620.

Schmitt, N. & Gilliland, SW. (1992) Beyond differential prediction: Fairness in
selection. In D.M. Saunders (Eds), New approaches to employee management:
Fairness in employee selection, 1, (pp. 21-46). Greenwich, CT: JAI Press

Schuler, H. (1993). Social validity of selection situations: A concept and some empirical
results. In H. Schuler, J .L. Farr, & M. Smith (Eds) Personnel selection and
assessment: Individual and organizational perspectives (pp. 11-26) Hillsdale,
NG: Erlbaum

Schwartz, S. H. (1990). Individualism-collectivism: Critique and proposed reﬁnements.
Journal of Cross-cultural Psychology, 21, 139-157.

Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical
advances and empirical tests in 20 countries. In M. Zarma (Ed.) Advances in

189

Experimental Social Psychology, Vol. 25 (pp. 1-65). New York, NY: Academic
Press.

Schwartz, S. H., & Bilsky, W. (1987). Toward a universal psychological structure of
human values. Journal of Personality and Social Psychology, 53, 550-562.

Seijts, G.H., & Jackson, S. E. (2001). Reactions to employment equity programs and
situational interviews: A laboratory study. Human Performance, 14, 247-265.

Shapiro, D.L. (1991). The effects of explanations on negative reactions to deceit.
Administrative Science Quarterly, 36, 614-630.

Shapiro, D.L., Buttner, E.H., & Barry, B. (1994). Explanations: What factors enhance
their perceived adequacy? Organizational Behavior and Human Decision
Processes, 58, 346-368.

Sharkey, W.F., & Singelis, TM. (1995). Embarrassability and self-construal: A
theoretical integration. Personality and Individual Differences, 19, 919-926.

Shaw, J .C., Wild, E., & Colquitt, J .A. (2003). To justify or excuse? A meta-analytic
review of the effects of explanations. Journal of Applied Psychology, 88, 444-
458.

Singer, M. (1990) Determinants of perceived fairness in selection practices; An
organizational justice perspective. Genetic, Social and General Psychology
Monographs, 116, 477-494

Singelis, TM. (1994). The measurement of independent and interdependent self-
construals. Personality and Social Psychology Bulletin, 20, 580-591 .

Singelis, T.M., Bond, M.H., Sharkey, W.F., & Lai, C.S.Y. (1999). Unpackaging
culture’s inﬂuence on self-esteem and embarrassability: The role of self-
construals. Journal of Cross-Cultural Psychology, 30, 315-341.

Singelis, T.M., & Brown, W.J. (1995). Culture, self, and collectivist communication:
Linking culture to individual behavior. Human Communication Research, 21,
354-389.

Singelis, T.M., & Sharkey, W.G. (1995). Culture, self-construal, and embarrassability.
Journal of Cross-Cultural Psychology, 26, 622-644.

Smith, P.B., Dugan, S., & Trompenaars, F. (1996). National culture and the values of
organizational employees. Journal of Cross-Cultural Psychology, 2 7, 231-265.

Smither, J.W., Reilly, R.R., Millsap, R. E., Pearlman, K., & Stoffey, R.W. (1993).
Applicant reactions to selection procedures. Personnel Psychology, 46, 49-76

190

Steiner, D.D., & Gilliland, SW. (1996). Fairness reactions to personnel selection
techniques in France and the United States. Journal of Applied Psychology, 81,
1 34-141 .

Steiner, D.D. & Gilliland, SW. (2001). Procedural justice in personnel selection:
International and cross-cultural perspectives. International Journal of Selection
and Assessment, 9, 124-137.

Stinglhamber, G., Vandenberghe, C., & Brancart, S. (1999) Reactions of job applicants to
personnel selection techniques: An investigation in a French-speaking context.
Travail Humain, 62, 347-361

Tata, J. (2000). Toward a theoretical framework of intercultural account-giving and
account evaluation. International Journal of Organizational Analysis, 8, 155—
1 78.

Taylor, M.S., & Bergrnann, T]. (1987) Organizational recruitment activities and
applicant reactions at different stages of the recruitment process. Personnel
Psychology, 40, 261-285.

Tett, R.P., & Burnett, DD. (2003). A personality trait-based interactionist model of job
performance. Journal of Applied, Psychology, 88, 500-517.

Thorsteinson, T.J., & Ryan, A.M. (1997) The effect of selection ratio on perceptions of
the fairness of a selection test battery. International Journal of Selection and
Assessment, 5, 159-168

Triandis, H. C. (1989). The self and social behavior in differing cultural contexts.
Psychological Review, 96, 506-520.

Triandis, H.C. (1995). Individualism and Collectivism. Boulder, CO: Westview Press.

Triandis, H.C., & Bhawuk, D.P.S. (1997). Culture theory and the meaning of
relatedness. In C.P. Barley and M. Erez (Eds) New Perspectives on International
Industrial/Organizational Psychology (pp. 13-52). San Francisco, CA: The New
Lexington Press.

Triandis, H.C., Bontempo, R., Betancourt, H., Bond, M., Leung, K., Brenes, A., Georgas,
J ., Hui, C.H., Marin, G., Setiadi, B., Sinha, J .B.P., Verrna, J ., Spagenberg, J.,
Touzard, H., & de Montmollin, G. (1986). The measurement of etic aspects of

individualism and collectivism across cultures. Australian Journal of Psychology,
38, 257-267.

191

Triandis, H.C., Bontempo, R., Villareal, M.J., Asai, M., & Lucca, N. (1988).
Individualism and collectivism: Cross-cultural perspectives on self-ingroup
relationships. Journal of Personality and Social Psychology, 54, 323-338.

Triandis, H.C., Chan, D.K.S., Bhawuk, D.P.S., Iwao, S., & Sinha, J.B.P. (1995).
Multimethod probes of allocentrism and idiocentrism. International Journal of
Psychology, 30, 461-480.

Triandis, H.C., & Gelfand, MJ. (1998). Converging measurement of horizontal and
vertical individualism and collectivism. Journal of Personality and Social
Psychology, 74, 118-128.

Triandis, H.C., Leung, K., Villareal, M.J., & Clack, F. (1985). Allocentric vs. idiocentric
tendencies: Convergent and discriminant validation. Journal of Research in
Personality, 1 9, 395-415.

Triandis, H.C., Marin, G., Lisansky, J. & Betancourt, H. (1984). Simpatia as a cultural
script of Hispanics. Journal of Personality and Social Psychology, 47, 1363-
1 375.

Triandis, H.C., McCusker, C., & Hui, CH. (1990). Multimethod probes of individualism
and collectivism. Journal of Personality and Social Psychology, 59, 1006-1020.

Triandis, H.C., & Singelis, TM. (1998). Training to recognize individual differences in
collectivism and individualism within culture. International Journal of
Intercultural Relations, 22, 35-47.

Triandis, H.C. & Suh, E. (2002). Cultural inﬂuences on personality. Annual Review of
Psychology, 53, 133-160.

Traﬁmow, D., Triandis, H.C., & Goto, S.G. (1991). Some tests of the distinction
between the private self and the collective self. Journal of Personality and Social
Psychology, 60, 649-655.

Trubinsky, P., Ting-Toomey, S., & Lin, S. (1991). The inﬂuence of individualism-
collectivism and self-monitoring on conﬂict styles. International Journal of
Intercultural Relations, 15, 65-84.

Truxillo, D.M., & Bauer, TN. (1999). Applicant reactions to test score banding in entry-
level and promotional contexts. Journal of Applied Psychology, 84, 322-339

Truxillo, D.M., Bauer, T. N., & Sanchez, R.J. (2001) Multiple dimensions of procedural
justice: Longitudinal effects on selection system fairness and test-taking self-
efﬁcacy. International Journal of Selection and Assessment, 9, 336-349.

192

Tyler, T.R., Lind, E.A., & Huo, Y.J. (2000). Cultural values and authority relations: The
psychology of conﬂict resolution across cultures. Psychology, Public Policy, and
Law, 6, 1138-1163.

US. Congress, Ofﬁce of Technology Assessment, The Use of Integrity Tests for Pre-
Employment Screening, OTA-SET-442 (Washington, DC: US. Government
Printing Ofﬁce, September 1990) Retrieved on 11/15/02 from
www.wws.princeton.edu/cgi-binfbyteservprl/ ~ota/disk2/1990/9042/904206.PDF

Van Den Bos, K., & Lind, A.E. (2001). The psychology of own versus others’ treatment:
Self-oriented and other-oriented effects on perceptions of procedural justice.
Personality and Social Psychology Bulletin, 27, 1324-1333.

Wagner, HI, J.A. (1995). Studies of individualism-collectivism: Effects on cooperation
in groups. Academy of Management Journal, 38, 152-172.

Wiechmann, D., 8L Ryan, A.M. (2001). Explanation speciﬁcity in selection contexts:
Effects- on applicant reactions to cognitive ability and personality tests.
Unpublished manuscript. Michigan State University

Yamaguchi, S., Kuhhnan, D.M., & Sugimori, S. (1992, May). Universality of
personality correlates and dimensionality of person ’s collectivistic tendencies.
Paper presented at the Center for Korean Studies Colloquium, University of
Hawaii, Honolulu.

Yetim, U. (2003). The impacts of individualism/collectivism, self-esteem, and feeling of
mastery on life satisfaction among the Turkish university students and
academicians. Social Indicators Research, 61 , 297-317.

Youn, I. (2000). The culture speciﬁcity of epistemological beliefs about learning. Asian
Journal of Social Psychology, 3, 87-105.

193

II" E’f'll‘lt be..--’y.

tel.-. triflilol It‘ll-r ITA.

1.". b! -1

0 .Ti...’ ti. 1' bill.

ll. lull L... :..7 15 IV.-.‘

0\

911.7111 reﬁlkwt.

E

 

11111111113111! ﬁllllllllliﬂllllf