Journal of Applied Psychology Copyright 1998 by the American Psychological Association, Inc.

1998. \tol. 83, No. 2, 247-260 0021-9010/98/$3.00

The Effects of Organizational Citizenship Behavior on Performance

Judgments: A Field Study and a Laboratory Experiment

Tammy D. Allen Michael C. Rush

University of South Florida University of Tennessee, Knoxville

The process linking organizational citizenship behavior (OCB) with performance judgments was
investigated in a field and a laboratory study. In the field study, managers rated the task performance
and OCB of 148 subordinates. In the laboratory research, 136 students viewed and rated videotaped
segments of teaching performance that demonstrated either high or low task performance and high
or low OCB. In both studies, liking and perceived affective commitment mediated the relationship
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

between OCB and overall evaluation. Liking also mediated the relationship between OCB and reward
This document is copyrighted by the American Psychological Association or one of its allied publishers.

recommendations. Further, the field study indicated that the causal motive attributed by the manager
for the employee's OCB mediated the relationship between OCB and overall evaluation.

It is now commonly accepted that an essential condition
of organizational effectiveness is a willingness of partici-
pants to "go above and beyond" the formal specifications
of prescribed roles (Barnard, 1938; Katz & Kahn, 1978;
Organ, 1990). Discretionary organizational citizenship
behavior (OCB) (Bateman & Organ, 1983; Smith, Or-
gan, & Near, 1983) includes behaviors such as construc-
tive or cooperative gestures that are neither mandatory
inrole behavior nor directly or contractually compensated
for by formal organizational reward systems (Organ &
Konovsky, 1989). Aggregated over time and persons, citi-
zenship behaviors become important because they facili-
tate the accomplishment of organizational goals and en-
hance organizational performance (Organ, 1990; Podsa-
koff, Ahearne, & MacKenzie, 1997; Podsakoff &
MacKenzie, 1997).
Recent research indicates that managers attend to the
OCB of employees with ensuing consequences of import.
Tammy D. Allen, Department of Psychology, University of
South Florida; Michael C. Rush, Department of Management,
University of Tennessee, Knoxville.
This research was supported through a dissertation fellowship
funded by the family of Walter Melville Bonham, University of
Tennessee, and an American Psychological Association Science
Directorate grant. This article is based on Tammy D. Allen's
dissertation completed under the guidance of Michael C. Rush.
We thank the other committee members, Joyce Russell, Gregory
Dobbins, Dudley Dewhirst, and John Lounsbury. We also thank
Mark Poteet, Walter Borman, and Robert Cardy for their helpful
comments at various stages of this project.
Correspondence concerning this article should be ad-
dressed to Tammy D. Allen, Department of Psychology, Uni-
versity of South Florida, 4202 E. Fowler Avenue, BEH 339,
Tampa, Florida 33620-8200. Electronic mail may be sent to
tallen @ luna.cas.usf.edu.
These studies indicate that both task-related performance
and OCB contribute independently to overall performance
evaluations and that supervisors perceive OCBs as con-
tributing to the value of an employee's performance (e.g.,
Borman, White, & Dorsey, 1995; Motowidlo & Van Scot-
ter, 1994; Orr, Sackett, & Mercer, 1989; Werner, 1994).
Although the existing research has convincingly demon-
strated that OCB does influence performance judgments
(see also Avila, Fern, & Mann, 1988; Jackson, Keith, &
Schlacter, 1983; MacKenzie, Podsakoff, & Fetter, 1991,
1993; Podsakoff & MacKenzie, 1994), there has been a
surprising lack of systematic empirical research designed
to investigate how and why these effects occur. Simply
demonstrating that OCB does influence performance eval-
uations does not further an understanding of the psycho-
logical mechanisms that may underlie this effect. Because
overall evaluations of performance have a variety of im-
portant uses within organizations and can have serious
consequences for employees, it is important to better un-
derstand the process by which OCB influences perfor-
mance judgments. Therefore, die purpose of the present
research was to investigate the underlying processes in-
volved in linking OCB with global performance judg-
ments by focusing on the social-cognitive processes of
raters. Specifically, we tested several variables (i.e., liking,
perceived affective commitment, job expectations, and at-
tributed motive) thought to be implicated in the effect of
OCB on overall evaluation and reward recommendations
in a field study and a laboratory experiment.
Theoretical Development
MacKenzie, Podsakoff, and their colleagues proposed
several reasons why managers may consider OCBs in their
performance evaluations. These reasons include norms of
reciprocity and fairness, schema-triggered affect, informa-
 248 ALLEN AND RUSH
tional distinctiveness and accessibility, illusory correla-
tions, and implicit performance theories (MacKenzie et
al., 1991, 1993; Podsakoff, MacKenzie, & Hui, 1993).
Most of these theoretical explanations have been drawn
from social information-processing theory. The social in-
formation-processing model stipulates that humans have
limited information-processing capabilities, hence many
shortcuts are taken when attaching meaning to and form-
ing expectations about the individual traits and behaviors
of others (e.g., DeNisi, Cafferty, & Meglino, 1984; Feld-
man, 1981; Lord, 1985). Effects in information pro-
cessing can be traced to the use of these cognitive simpli-
fications as evaluations of others are based on only part
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
of the available information. Moreover, these expectations
This document is copyrighted by the American Psychological Association or one of its allied publishers.
form a cognitive context that affects future interactions
with individuals (Lord & Maher, 1991). Terms such as
prototypes, schemata, implicit theories, and cognitive cat-
egories have been adopted to describe the processes that
individuals use to simplify and organize the multitude of
information received. A number of models to explain these
processes have been proposed; however, in general, the
primary stages in the processing of information consist
of attention, categorization, recall, and information
integration.
Extrapolating from the concepts of social information-
processing theory, it seems likely that OCB influences
how a ratee is categorized. The information-processing
model proposes that raters are most likely to attend to
information that is salient and that raters mentally assign
employees to categories on the basis of salient characteris-
tics of the employee (e.g., DeNisi et al., 1984). As sug-
gested by Podsakoff et al. (1993), because OCB is not
required by an organization, it is likely to be a particularly
distinctive and salient behavioral cue. Consequently, in-
stances of OCB may be used to categorize an employee.
Categorization is also based on the extent to which the
ratee matches the characteristics of a prototype (an ab-
stract set of representative features that define members
of a category). One prototype that may be held by raters is
that of the "good employee" (Cardy & Dobbins, 1994a;
Murphy & Cleveland, 1991). Because OCBs consist of
performance that is above and beyond the call of duty
and is geared toward the effective functioning of the orga-
nization, employees who engage in OCB are likely to
match the prototypical features of the good employee.
That is, the social information-processing model suggests
that OCB elicits a cognitive response in the rater as he
or she recognizes the prototype of the good employee.
In sum, we hypothesized that acts of OCB influence
rater cognitive processes and impressions of subordinates,
which in turn affect performance judgments. Once an
individual is categorized, subsequent retrieval of informa-
tion and corresponding judgments tend to be based on
the prototypical characteristics of the category (Feldman,
1981). Further, the categorization can influence what is
noticed about the employee's performance and the attribu-
tions made for observed behavior (Feldman, 1986). Fiske
and Pavelchak (1986) suggested that behaviors easily en-
coded as consistent with an existing cognitive category
can result in category-based affect (e.g., liking) and
strong category-based memory. Moreover, other affective
characteristics associated with the category (e.g., pre-
sumed commitment to the organization) are likely to be
attached to the cognitive representation of the employee
and directly associated with performance judgments. The
specific variables thought to be implicated in the relation-
ship between OCB and performance judgments are further
discussed below.
Affect and Liking
Because employees who perform OCBs make a manag-
er' s job easier, instances of OCB may produce an affective
response and hence enhance a manager's liking for a sub-
ordinate. Fiske's (1981, 1982) theory of schema-triggered
affect may help explain this process (see also Fiske &
Pavelchak, 1986). The theory suggests that affective val-
ues are stored at the top level of a schema, accessible
immediately on categorization of an event matching the
affect-laden schema. It seems likely that individuals who
engage in acts of OCB will trigger positive affect from
their managers. Consequently, individuals who display
high levels of OCB will be better liked.
Research has demonstrated that performance evalua-
tions are influenced by rater affect or liking for the ratee
(see Cardy & Dobbins, 1994b, for a review). Generally,
the findings have suggested that positive affect toward a
subordinate has a positive effect on supervisory perfor-
mance ratings and the allocation of organizational rewards
and results in a reduced likelihood of disciplinary action
(e.g., Cardy & Dobbins, 1986; Dobbins & Russell, 1986;
Fandt, Labig, & Urich, 1990; Judge & Ferris, 1993; Tsui &
Barry, 1986). Accordingly, we hypothesized that the rela-
tionship between OCB and performance judgments would
be mediated by liking for the ratee.
Perceived Affective Commitment
Engaging in OCB may also influence rater perceptions
regarding how committed an employee is to the organiza-
tion. Acts of citizenship are characteristic of the behaviors
also associated with highly committed employees. For
example, employees considered committed to an organi-
zation are expected to work long hours and to place a
greater importance on the interest of the organization than
on personal interests when the two interests are in conflict.
Moreover, because OCB is volitional in nature, engaging
in these behaviors provides favorable information regard-
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR
ing an employee's motivation (Borman & Motowidlo,
1993). Consequently, employee acts of OCB may serve
as behavioral cues on which managerial presumptions
of employee commitment to the organization are based.
Indeed, Shore, Barksdale, and Shore (1995) found that
OCB predicted perceived affective commitment.
There is also preliminary evidence suggesting that per-
ceptions of commitment are positively related to other
managerial judgments. Shore et al. (1995) found that per-
ceived affective commitment was positively related to
managerial potential and promotability ratings. Similarly,
Allen, Russell, and Rush (1994) found a positive relation-
ship between perceived organizational commitment and
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
reward recommendations. Because highly committed em-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
ployees are seen as vital to an organization, employers
may find it desirable to offer more rewards to their com-
mitted employees or managers (Powell, 1993). Thus, we
proposed that the relationship between OCB and perfor-
mance judgments would be mediated by perceived af-
fective commitment.
Job Expectations
Several researchers have commented on the degree of
ambiguity regarding the boundary between task perfor-
mance and OCB (e.g., Graham, 1991; Morrison, 1994;
Schnake, 1991; Van Dyne, Cummings, & Parks, 1995).
Although some research has found that managers can and
do distinguish between inrole behaviors and OCB (e.g.,
Werner, 1994; Williams & Anderson, 1991), there is also
evidence to indicate that behavior assumed to be and oper-
ationalized as discretionary in nature may actually be con-
sidered an expected part of the job (Morrison, 1994).
Hence, there are differences between individuals with re-
gard to work behaviors believed to be discretionary versus
behaviors believed to be nondiscretionary.
Research has yet to examine this issue from the per-
spective of the rater. As Morrison (1994) stated, if a be-
havior is thought of as an inrole requirement, the behavior
will be conceptualized differently than if the behavior is
perceived as beyond requirements. Differences in beliefs
regarding expected job behaviors may affect the degree
of influence that OCB has on the evaluation process. That
is, OCB may not have the same positive effect on liking
and perceived affective commitment when the rater views
the behavior as an ordinary part of the individual's job
versus when the rater views the behavior as above and
beyond the call of duty. Accordingly, it is proposed that
the relationships of OCB with liking and with perceived
affective commitment should be moderated by the extent
that OCB is viewed as exceeding the ratee's job require-
ments. More specifically, the enhancing effect of OCB
on liking and perceived affective commitment should be
greater when the behaviors are viewed as beyond job
249
expectations than when the behaviors are viewed as an
expected part of the job.
Attribution and Motive
Attribution theory has been useful for attempting to
understand how perceptions of performance are inter-
preted. The central tenet of attribution theory is that hu-
mans construct causal explanations for their own and oth-
ers' behavior in a need to predict and understand the
environment (Heider, 1958). To date, the research examin-
ing causal attributions in organizational contexts has fo-
cused on traditional task performance (e.g., Green &
Mitchell, 1979; Mitchell, Green, & Wood, 1981; Pence,
Pendleton, Dobbins, & Sgro, 1982). It seems likely that
individuals also search for causal explanations of OCB.
Indeed, Eastman (1994) found that when raters were pre-
sented with the same set of extrarole behaviors, they
attached different causal labels (i.e., good citizenship ver-
sus ingratiation) to the behaviors. Further, employees
whose behavior was believed to be good citizenship rather
than ingratiation received higher overall performance
evaluations and greater pay raises.
Eastman's (1994) findings suggest that one of two
causal motives may be associated with instances of OCB;
altruistic motives and instrumental motives. Elements of
altruistic motives would include personal values, loyalty
to the organization, and a sense of moral standards,
whereas instrumental motives would include a desire to
impress the boss, to obtain recognition or other organiza-
tional rewards, or to build up favors for later exchange.
Behaviors associated with altruistic motives are likely to
positively influence performance judgments, whereas
those associated with instrumental motives may be deval-
ued or discounted (Podsakoff et al., 1993). It is assumed
that an altruistic motive is dominant in the conception of
a good employee, as it would be consistent with other
positively valued schema-triggered affect (cf. Fiske, 1981,
1982). This view is also consistent with Organ's (1990)
view that OCB is essentially self-sacrificing as opposed
to controlled and self-serving, at least until some threshold
of fairness is violated. Accordingly, it is proposed that the
relationship between OCB and performance judgments
will be mediated by the motive attributed to the behavior.
In summary, the specific hypotheses guiding the study
were as follows:
Hypothesis I. The relationship between OCB and the de-
pendent variables, overall evaluation and reward recom-
mendations, will be mediated by liking.
Hypothesis 2. The relationship between OCB and the de-
pendent variables, overall evaluation and reward recom-
mendations, will be mediated by perceived affective
commitment.
Hypothesis 3. The relationship of OCB with liking and
 250 ALLEN AND RUSH
OCB with perceived affective commitment will be moder-
ated by the extent that OCB is viewed as discretionary,
such that the relationships will be stronger when the behav-
iors are viewed as exceeding job expectations.
Hypothesis 4. The relationship between OCB and the de-
pendent variables, overall evaluation and reward recom-
mendations, will be mediated by the motive attributed to
the OCBs.
To provide a rigorous and comprehensive investigation
of the hypotheses, we conducted two studies so that issues
of both internal and external validity could be addressed.
Furthermore, because we were interested in isolating the
effects of OCB, we measured task performance and in-
cluded it as a control variable in all analyses. We tested
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
all of the hypotheses in the field study. We tested all but
Hypothesis 4 in the lab study.
Study 1: Field Study
Method
Participants
The participants were managers with ongoing supervisory
responsibilities for one or more subordinates. Eighty different
managers provided a total of 148 subordinate ratings. The sam-
ple included managers employed by a manufacturing firm, man-
agers of participants in a university-based management develop-
ment program, and managers from multiple organizations who
were contacted via other university sources (e.g., faculty, ad-
vanced graduate students). The overall sample of managers con-
tained 76.6% men; average age was 47.3 years (SD = 9.8);
96.1% were Caucasian; average organizational tenure was 13.7
years (SD = 10.2); average job tenure was 4.4 years (SD —
4.3). Demographics by sample source are available on request.
Procedure
We asked managers to complete questionnaires for up to four
subordinates. Instructions were provided on how to determine
which four subordinates to rate in situations where a manager
had more than four. Participants mailed surveys directly back
to the primary researcher. Ten out of 10 survey packets were
returned from the managers solicited from the manufacturing
firm for a response rate of 100%. We contacted 70 managers to
provide ratings of the participants in the management develop-
ment program. Of those 70, a total of 54 surveys were returned
for a response rate of 77.1%. For the remaining sample, we
distributed a total of 35 survey packets to various managers,
and 16 were returned (45.7%).
Measures
We used established, validated scales when available. We
grouped items by construct but did not identify them as such.
Except where noted, responses were based on a 5-point scale
ranging from strongly disagree (1) to strongly agree (5), and
we averaged scores across the items to form each scale score.
Task performance. We measured task performance with the
7-item scale developed by Williams and Anderson (1991; e.g.,
' 'Fulfilling responsibilities specified in job description''). We
used a 5-point scale with responses ranging from ineffective
(1) to highly effective (5). Those authors reported a Cronbach
reliability estimate of .91; in this study, the estimate was .95.
OCB. We measured OCB using the 24-item scale developed
by Podsakoff, MacKenzie, Moorman, and Fetter (1990; e.g.,
' "This employee is always ready to lend a helping hand to those
around him/her''). These items reflect the five dimensions of
OCB proposed by Organ (1988). Podsakoff et al. (1990) re-
ported reliabilities ranging from .70 to .85 for each of the five
subscales. Using a variation of the same scale, MacKenzie et
al. (1991) reported similar reliabilities (.70 to .84). We used
the overall composite score in the present study, with a reliability
estimate of .93.
Liking. We adapted the 4-item measure developed by Wayne
and Ferris (1990) for use in the present study (e.g., "I like
this subordinate"). Those authors reported a Cronbach alpha
estimate of .94 (laboratory study) and .86 (field study); in this
study, the reliability estimate was .87.
Perceived affective commitment. We assessed perceived af-
fective commitment using the 4-item affective commitment sub-
scale of Shore et al.'s (1995) Manager-rated Commitment Scale
(e.g., "This employee appears to be highly committed to the
organization"). Shore et al. reported a Cronbach alpha estimate
of .87; in this study, the estimate was .88.
Job expectations. Similar to the process used by Morrison
(1994), respondents were asked to classify each of the 24 items
that were used to measure OCB into one of two categories: (a)
This behavior is viewed as an expected part of the job; or (b)
This behavior is viewed as somewhat above and beyond what
is expected for the job (negatively worded items were positively
reworded). This categorization process was completed solely
for the purpose of creating the job expectations measure. Behav-
iors that were considered "expected" were coded 0, and behav-
iors considered beyond expectations were coded 1. We com-
puted a scale mean by summing the responses and then dividing
by the total number of behaviors. Hence, scale responses could
range from .00 to 1.0, with a score of .00 indicating that every
behavior was expected and a score of 1.0 indicating that every
behavior was beyond expectations. The reliability estimate for
the scale was .86.
Attribution of motive. We measured perceived motive with
a 12-item scale developed for this study. The items were de-
signed to measure attribution of motive to altruistic causes (six
items; e.g., "sense of moral standards'') and attribution of mo-
tive to instrumental causes (six items; e.g., "desire to seek the
spotlight''). These items immediately followed the items used
to measure OCB and were preceded by the following instruc-
tions: "After reading the list below, please indicate the extent
to which you agree that each of the following may be the reason
for or cause of this employee's actions on occasions when he/
she exhibits behaviors considered 'above and beyond the call
of duty'." The Cronbach alpha reliability estimate in a pilot
study was .84 for the altruistic subscale and .90 for the instru-
mental subscale; in the present study, the estimate was .83 for
the altruistic subscale and .89 for the instrumental subscale.
Overall evaluation. We developed five items to assess the
manager's overall evaluation of the ratee (e.g., "This employee
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR
makes an important contribution to the organization"). The
Cronbach alpha reliability estimate in a pilot study was .74; in
the present study, the estimate was .91.
Reward recommendations. We used five items to assess the
extent to which the rater would recommend the ratee for five
common organizational rewards (salary increase, promotion,
high profile project, public recognition [e.g., company award],
and opportunities for professional development). Responses
were based on a 5-point scale ranging from would definitely
not recommend ( 1 ) to would recommend with confidence and
without reservation (5), alpha = .90.
Demographic variables. We asked respondents to provide
information regarding their age, sex, race, job tenure, and orga-
nizational tenure.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Results
General Analytic Strategy
We tested hypotheses through the use of hierarchical
multiple regression. We used the criteria established by
James and Brett (1984) to determine if mediation effects
existed: (a) The independent variable must be signifi-
cantly correlated with the mediator variable, (b) the medi-
ator variable must be significantly related to the dependent
variable, and (c) when the influence of the mediator vari-
able is held constant, the effect of the independent variable
on the dependent variable should be nonsignificant. First-
order correlations controlling for the effect of task perfor-
mance were computed and indicated that Conditions 1
and 2 were met for each hypothesis proposing mediation
(a copy of the first-order correlations is available on re-
quest). Condition 3 was tested by (a) entering task perfor-
mance into the regression equation as a control variable,
(b) adding OCB to the equation, and then (c) adding the
proposed mediator variable to the equation. We used the
significance of the beta weight associated with OCB at
each step to determine mediation. This procedure was
followed for both dependent variables. We also used the
procedures described by James and Brett (1984) as a
basis for testing moderator effects. Moderation is sup-
ported when the addition of the interaction term (the prod-
uct of the moderator and the predictor) results in a sig-
nificant increment in variance associated with the depen-
dent variable beyond the variance accounted for by the
main effects. Intercorrelations are presented in Table 1.
Hypotheses Testing
The results regarding Hypotheses 1, 2, and 4 are shown
in Table 2. Hypothesis 1 proposed that the relationship
between OCB and the dependent variables would be medi-
ated by liking. As shown in Table 2, the addition of liking
into the equation predicting overall evaluation resulted in
a significant increment in variance (ff 2 A = .07, p <
.001), and the beta weight associated with OCB became
251
nonsignificant. Similarly, the addition of liking in the
equation predicting reward recommendations resulted in
a significant increment in variance (X 2 A = .05, p <
.001), and the beta weight associated with OCB became
marginally significant, indicating partial mediation. To-
gether, these results provide reasonable support for Hy-
pothesis 1.
Table 2 also shows that the addition of perceived af-
fective commitment in the equations predicting overall
evaluation resulted in a significant increment in variance
(S 2 A = .08, p < .001) and that the beta weight associated
with OCB became nonsignificant. In contrast, the addition
of perceived affective commitment in the equation pre-
dicting reward recommendations resulted in a significant
increment in variance (S 2 A = .04, p < .001); however,
the beta weight associated with OCB remained significant.
Hence, it appears that both OCB and perceived affective
commitment have significant direct relationships with re-
ward recommendations. Thus, only partial support was
found for Hypothesis 2.
Hypothesis 3 proposed that job expectations would
moderate the relationship of OCB with liking and OCB
with perceived affective commitment. Analyses failed to
show any significant effects for the interaction term re-
gardless of whether liking was the dependent variable
(^?2A = .00, ns') or perceived affective commitment was
the dependent variable (/? 2 A = .00, ns). Thus, no support
was found for Hypothesis 3.
Last, Hypothesis 4 stated that the relationship between
OCB and the dependent variables would be mediated by
the causal motive attributed to OCB. We entered the two
causal dimensions, altruistic and instrumental, as a set in
the regression equation. Results indicated that the addition
of the causal motive variables in the equation predicting
overall evaluation resulted in a significant increment in
variance (R2A = .07, p < .001), and the beta weight
associated with OCB became nonsignificant. Inspection
of the beta weights for the two causal motive variables
revealed that the mediation effect occurred solely through
the altruistic motive variable. Instrumental attributions ap-
peared to have little effect. By contrast, the addition of the
causal motive variables in the equation predicting reward
recommendations resulted in a significant increment in
variance (« 2 A = .02, p < .01), but the beta weight
associated with extrarole performance remained signifi-
cant, indicating that both OCB and causal motives had
direct effects. It was again noted that only altruistic mo-
tives appeared to be related to the dependent variable.
Thus, partial support was found for Hypothesis 4.
Discussion
The results of the field study provide evidence indicat-
ing that liking mediates the relationship between OCB
 252 ALLEN AND RUSH

Table 1
Zero-Order Correlation Coefficients Among Study Variables—Field Study

Variable 1 2 • 3 4 5 6 7 8 9

1. Task performance"
2. OCB' .66**
3. Liking' .52** .74**
4. Perceived commitment" .52** .66** .61**
5. Job expectations' .00 .09 .03 .06
6. Altruistic motive" .55** .72** .60** .76** .08
7. Instrumental motive" -.19* -.59** -.49** -.37** -.17* -.46**
8. Overall evaluation" .65** .63** .63** .67** .02 .66** -.27**
9. Reward recommendations' .77** .73** .70** .67** .07 .66** -.30** .78** —
M 4.01 3.90 4.07 3.61 0.25 3.85 2.54 3.52 4.00
SD 0.67 0.56 0.73 0.86 0.21 0.63 0.86 0.86 1.00
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Note. OCB = organizational citizenship behavior. Sample sizes ranged from 134 to 148.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

3
" Based on a 5-point scale, with higher values indicating a greater degree of the construct (e.g., higher OCB). Dummy coded (0 = expected
behavior; 1 = above ami beyond expectations).
* p < .05, two-tailed. **p < .01, two-tailed.

and both dependent variables. Additionally, the results in- and causal motives did not mediate the relationship be-
dicated that perceived affective commitment and causal tween OCB and reward recommendations. When pre-
motives mediate the relationship between OCB and over- dicting reward recommendations, OCB, perceived af-
all evaluation. In contrast, perceived affective commitment fective commitment, and causal motives all had direct

Table 2
Results of Regression Analysis—Field Study

Standardized regression weights

Overall evaluations Reward recommendations

Independent variable Step 1 Step 2 Step 3 Step 1 Step 2 Step 3

Hypothesis 1

Task performance .64*** .43*** .41*** .76*** .51*** .50***

OCB 31*** .03 .39*** .14
Liking .40*** .34***
R2 at each step .40 .46 .53 .58 .67 .72
R2& .06*** .07*** .09*** .05***
Fa 47.83 107.86

Hypothesis 2

Task performance .65*** .41*** .36*** .77*** .51*** 47***

OCB .36*** .14 .39*** 22***
Perceived affective commitment .39*** 27***
R2 at each step .42 .49 .58 .59 .68 .72
/?2A .07*** .08*** .09*** .04***
F' 62.08 113.94

Hypothesis 4

Task performance .65*** .41*** .34*** .77*** .51*** .47***

OCB .36*** .16 .39*** .29**
Altruistic motive .39*** .22**
Instrumental motive .06 .05
R2 at each step .42 .49 .56 .59 .68 .70
ft2A .07*** .07*** .09*** .02**
F 43.54 77.64

Note. OCB = organizational citizenship behavior. /Vs = 127-140.

a
All F values were significant at p < .001.
**/? < .01. ***/? < .001.
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR
effects. In concert, these results suggest that much of the
effect that OCB has on overall evaluations of performance
is indirect, via the effect that it engenders on other vari-
ables. These findings suggest that the effect that OCB
has on performance evaluations found in previous studies
largely occurs as a result of the cognitively elicited af-
fective reactions and favorable impressions of the em-
ployee that engaging in OCB produces. That is, employees
who frequently engage in OCB are well-liked, thought to
be committed to the organization, and presumed to engage
in OCB for altruistic reasons.
No support was found for the hypothesis that the rela-
tionship of OCB with liking and OCB with perceived
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
affective commitment would vary to the extent that OCB
This document is copyrighted by the American Psychological Association or one of its allied publishers.
was viewed as discretionary by the rater. This suggests
that regardless of whether or not the behavior is viewed
as expected versus beyond expectations, the positive in-
fluence that the behavior has is the same.
A major limitation to the present study should be noted.
The cross-sectional correlational nature of the data repre-
sents a threat to internal validity in that there is ambiguity
regarding the causal direction of the relationships ob-
served (e.g., does liking influence ratings of OCB or does
OCB influence liking?; Cook, Campbell, & Peracchio,
1990). Further, given that the data consisted of managerial
ratings of task performance and OCB, there is no firm
basis from which to argue that levels of performance func-
tioned as inputs to cognitive processes rather than as out-
puts of the processes. To address these limitations, we
conducted a carefully designed conceptual replication of
the field study, with controlled, objective levels of task
performance and OCB, using an experimental laboratory
methodology. Convergent results from the laboratory ex-
periment will support the assertion that OCB influences
the intervening and outcome variables as hypothesized.
Study 2: Laboratory Study
Method
Participants
The participants were undergraduate psychology and business
students at a large southeastern university. Participants were
asked to sign up for two sessions in exchange for course credit.
A total of 154 students participated in the first session. Of those
154, 136 returned for the second session. The 136 students who
participated in both sessions included 75 men and 61 women,
who ranged in age from 18 to 47 years (M = 23.07, SD =
4.87). Their racial backgrounds were White (89.7%), Black
(3.7%), Asian (4.4%), and other racial background (2.2%).
Procedure
The research was couched in the context of a study of effec-
tive teaching behavior in which students were to complete a
253
university survey and then view and rate videotaped segments
of graduate student teaching performance. Participants were told
that their feedback would be used to develop future training
programs for effective teaching. Participants initially completed
a survey containing the job expectations measure and other filler
items (e.g., satisfaction with the university). During the second
session, held approximately one to two weeks later, participants
viewed a 9-min color videotape of an instructor interacting with
a class and delivering a lecture. We randomly assigned partici-
pants to 1 of 16 conditions (low vs. high task performance x
low vs. high OCB, replicated across two male and two female
instructors). Cell sizes across the four experimental conditions
used for hypothesis testing ranged from 31 to 38. After viewing
the videotape, participants were asked to respond to the depen-
dent measures, manipulation checks, and questions regarding
demographic information.
Development of Materials for the Video
Scripts. We took several steps to develop the behavioral
scripts required for each condition in the design. First, we con-
ducted structured interviews with six undergraduate students
who provided examples of effective and ineffective teaching
behaviors, as well as examples of behaviors considered to be
above and beyond the call of duty. We then presented the exam-
ples considered to be above and beyond the call of duty to an
independent group of 40 students who used a 5-point scale to
indicate the extent to which they considered each behavior to
be above and beyond the call of duty. We included the behaviors
rated as most highly above and beyond the call of duty in the
behavioral script to represent the presence of high levels of
OCB. We also consulted the list of critical incidents of teaching
performance developed by Sauser, Evans, and Champion (1980)
for instances of effective and ineffective teaching and used them
to develop the scripts for the high and low task performance
conditions. Examples of high OCB included taking the class on
a field trip and agreeing to see a student to discuss a personal
problem outside of scheduled office hours. Examples of high
task performance included using examples to help clarify the
lecture material and encouraging class discussion.
Videotapes. Four different actors played the role of instruc-
tor, each performing all of the four roles defined by the manipu-
lations. All actors were advanced doctoral students with prior
experience in teaching and as role-players in training contexts,
assessment center contexts, or both. We had two men and two
women to rule out the possibility that observed effects were due
to gender or other personal characteristics of the individual actor
as opposed to the experimental conditions. To help control for
potential bias, the actors wore the same type of attire and were
trained to display similar mannerisms and type of demeanor. In
addition to the instructor, the videotapes displayed an initial
segment in which another actor, playing the role of a student,
walked to the front of the class and spoke with the instructor.
Though unseen, several other people could be heard on the
videotape interacting with the instructor as students. The final
product was a realistic depiction of a 9-min lecture segment
that included student-teacher exchanges.
Video pilot test. An independent group of 12 doctoral stu-
dents, who were trained assessment-center assessors, viewed
 254 ALLEN AND RUSH
and rated the videotapes. We used a Latin-squares-type design
wherein each of the raters watched four of the teaching segments
and provided ratings of the observed level of task and OCBs.
Judges rated the instructor's task performance on a 5-point scale
ranging from highly ineffective (1) to highly effective (5). The
mean rating under the low condition was 1.67 (SD — .65), and
the mean rating under the high condition was 4.58 (SD = .52,
t — -11.22, p < .001). Judges were also asked to indicate if
the instructor demonstrated behaviors that could be considered
citizenship or extrarole (i.e., above and beyond the call of duty)
on a 3-point scale ranging from there were no examples of
extrarole behavior (1) to there were definitely several examples
of extrarole behavior (3). The mean rating under the low condi-
tion was 1.16 (SD = .39). Ratings under the high condition all
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
converged at 3 (t = -16.32, p < .001).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Measures
The same items used to measure liking, perceived affective
commitment, and overall evaluation in the field study were
slightly modified to fit the present classroom setting (e.g., "This
employee appears highly committed to the organization" was
changed to "This instructor appears highly committed to the
university"; alpha = .89, .91, and .92 respectively). To measure
job expectations, we gave students a list of 20 teaching behav-
iors and asked them to indicate the extent to which they believed
each behavior was an expected part of teaching or if the behavior
was above and beyond expectations of the job. Higher scores
indicated that the behaviors were more likely to be viewed as
beyond expectations while lower scores indicated that the be-
haviors were more likely to be viewed as expected (alpha =
.79). To assess reward recommendations, students indicated the
extent to which they would recommend the instructor for three
rewards that academic institutions commonly offer (e.g., a
teaching award; alpha = .92).
Manipulation Checks
After responding to the items related to the main constructs,
participants were asked to rate the task (a 5-point scale ranging
from highly ineffective, \, to highly effective, 5) and citizenship
behavior (a 3-point scale ranging from no examples of citizen-
ship behavior, 1, to definitely examples of citizenship behavior,
3) of the instructor as manipulation checks. Analyses indicated
that participants in the low task-performance condition rated the
task performance of instructors (M = 1.73) significantly lower
than did participants in the high task-performance condition (M
= 3.33), *(134) = -10.10, p < .001. Participants in the low
OCB condition also rated the OCB of instructors (M = 1.28)
significantly lower than did participants in the high OCB condi-
tion (M = 2.10), r(134) = -7.91, p < .001. However, in
multiple-factor designs it is possible that one manipulation may
affect the independent variable associated with a different ma-
nipulation.' That is, confounding can occur. Consequently, fol-
lowing the guidelines provided in Perdue and Summers (1986),
we conducted a 2 X 2 factorial analysis of variance to examine
the effect of the task and OCB manipulations on each manipula-
tion check. The results of the analyses of variance did suggest
that the manipulations were confounded. Specifically, OCB had
a significant main effect on the task performance manipulation
check, F( 1,132) = 49.33, p < .001. Likewise, task performance
had a significant main effect on the OCB manipulation check,
F(l, 132) = 20.87, p < .001. Given the significant result, we
assessed the effect size associated with each main effect by
calculating omega squared. With regard to the task performance
manipulation check, the effect size associated with the task
performance manipulation was .463 compared with .147 for the
OCB manipulation. With regard to the OCB manipulation
check, the effect size associated with the OCB manipulation
was .336 compared with .088 for the task performance manipu-
lation. As stated by Perdue and Summers, when the effect sizes
for the unintended variables are much smaller than that of the
manipulated variable, then their statistical significance should
not be of great concern. In summary, although not completely
orthogonal, the manipulations did appear to be effective.
We used analysis of variance to determine if there were sig-
nificant differences in the task performance and OCB ratings
given to the four ratees (actors). No significant differences were
detected in task performance ratings, F(3, 132) < 1.0, ns, or
OCB ratings, F(3, 132) < 1.0, ns, across the four ratees. These
results provide evidence that any observed effects were due to
the nature of the performance manipulations and not due to the
gender or other unique characteristics of the actors who por-
trayed the ratees.
Results
Intercorrelations among the variables in the study are
presented in Table 3. The means across each of the experi-
mental conditions are presented in Table 4.
Hypotheses Testing
We tested hypotheses using the same hierarchical multi-
ple regression procedures described for the field study.
We created dummy-coded variables to represent the dis-
tinction between low versus high task performance and
low versus high OCB (low = 0; high = 1 ) .
Table 5 provides results of the regression analyses. Hy-
pothesis 1 proposed that the relationship between OCB
and the dependent variables would be mediated by liking.
As shown in Table 5, the addition of liking in the equation
predicting overall evaluation resulted in a significant in-
crement in variance (fl 2 A = .27, p < .001), and the
beta weight associated with OCB became nonsignificant.
Similarly, the addition of liking to the equation predicting
reward recommendations resulted in a significant incre-
ment in variance (R 2 A = .22, p < .001), and the beta
weight associated with OCB became nonsignificant. Thus,
full support was found for Hypothesis 1.
We also proposed that the relationship between OCB
and the dependent variables would be mediated by per-
' We thank an anonymous reviewer for bringing this to our
attention.
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR 255

Table 3
Zero-Order Correlation Coefficients Among Study Variables—Lab Study

Variable 1 2 3 4 :

1 . Task performance
2. OCB -.07
3. Liking .50** .49**
4. Affective commitment .33** .68** .81**
5. Job expectations -.03 -.10 .01 -.02
6. Overall evaluation .56** .26** .80** .66** .04
7. Combined recommendation .59** .37** .83** .71** -.03 .76** —
M 2.69 2.62 2.01 2.36 2.38
SD 0.98 1.09 0.53 0.96 1.23

Note. OCB = organizational citizenship behavior. Sample sizes for correlations ranged from 135 to 142.
**P < .01, two-tailed.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ceived affective commitment. As shown in Table 5, the
addition of perceived affective commitment in the equa-
tions for overall evaluation and reward recommendations
resulted in significant increments in variance (/? 2 As =
.17 and .13, p < .001, respectively), and the beta weight
associated with OCB in each case became nonsignificant.
These results provide full support for Hypothesis 2.
Hypothesis 3 proposed that job expectations would
moderate the relationship of OCB with liking and with
perceived affective commitment. The test for moderation
failed to show any significant effect either for liking or
for perceived affective commitment (R 2 A = .00, ns).
Thus, no support was found for Hypothesis 3.
Discussion
Consistent with the results of the field study, the results
of the laboratory study indicated that liking mediated the
relationship between OCB and both dependent variables
and that perceived affective commitment mediated the re-
lationship between OCB and overall evaluation. However,
unlike the field study, the lab study also indicated that
perceived affective commitment mediated the relationship
between OCB and reward recommendations. Also consis-
tent with the field study, but contrary to the initial hypothe-
sis, whether or not the OCBs were viewed as expected or
beyond expectations did not moderate the relationship
between OCB and liking or OCB and perceived affective
commitment. Implications for these findings are consid-
ered under the General Discussion.
General Discussion
The results of this research offer several notable find-
ings regarding the effects of OCB on performance judg-
ments. First, consistent with previous research, results of
both the laboratory and the field study indicate that OCB
adds significantly to the prediction of overall evaluation
and reward recommendations beyond the effect of task
performance, thus providing additional evidence that rat-
ers do in fact attend to the OCBs of employees with
ensuing consequences.
Both studies offered strong evidence that the relation-
ship of OCB with overall evaluations and with reward
recommendations was mediated by liking. The findings
suggest that engaging in OCBs triggers positive affect for

Table 4
Means for Measured Variables by Experimental Condition—Lab Study

Low task High task

Task performance performance

High High
Variable Low High Low High OCB OCB OCB OCB

Liking 2.18 3.17 2.22 3.18 1.76 2.56 2.59 3.85

Perceived affective
commitment 2.25 2.96 1.90 3.38 1.51 2.93 2.22 3.85
Job expectations 2.01 1.97 2.04 1.94 2.10 1.93 2.00 1.95
Overall evaluation 1.82 2.88 2.12 2.61 1.70 1.91 2.46 3.38
Reward
recommendations 1.64 3.09 1.94 2.84 1.42 1.84 2.37 3.94

Note, OCB = organizational citizenship behavior.

 256 ALLEN AND RUSH

Table 5
Results of Regression Analyses—Lab Study

Standardized regression weights

Overall evaluations Reward recommendations

Independent variable Step 1 Step 2 Step 3 Step 1 Step 2 Step 3

Hypothesis 1

Task performance .56*** .58*** .17** .59*** .62*** .25***

OCB .30*** -.10 .41*** .05
Liking .76*** .68***
R^ at each step .31 .40 .67 .34 .51 .73
S2A .09*** .27*** .17*** .22***
F" 90.34 119.92
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Hypothesis 2
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Task performance .56*** .58*** .33*** .58*** .61*** .40***

OCB -.16 .42*** .02
Perceived affective
communication .66*** .57***
R2 at each step .31 .40 .57 .34 .51 .64
tf'A .09*** .17*** .17*** .13***
F" 58.86 78.79

Note. OCB = organizational citizenship behavior. JVs = 134-135.

fl
All F values were significant at p < .001.
* * p < . 0 1 . *"'p < .001.

the ratee and that this positive affect in turn influences
overall evaluations and reward recommendations. The re-
sults are consistent with Cardy and Dobbins' (1994b)
supposition that one way liking may affect performance
judgments is through schemas. According to Cardy and
Dobbins, the early and intense nature of liking make it a
probable candidate for forming the basis of schematic
processing. The results of the lab study are especially
supportive of this contention. After viewing one segment
of teaching performance, raters had clear positive or nega-
tive affective reactions to the instructor.
Results regarding the role of perceived affective com-
mitment and causal motives were somewhat more com-
plex. Specifically, results of the laboratory study indicated
that the effect of OCB on overall evaluation and on reward
recommendations was mediated by perceived affective
commitment. The results from the field study differed.
Perceived affective commitment mediated the relationship
between OCB and overall evaluation; however, mediation
was not supported when predicting reward recommenda-
tions. Although the effect of causal motives was assessed
in only the field study, we obtained a similar result. Spe-
cifically, the motive attributed to the OCB mediated the
relationship between OCB and overall evaluation; how-
ever, mediation was not supported when predicting reward
recommendations.
One possible explanation for these findings is a consid-
eration of automatic versus controlled cognitive informa-
tion processing. Controlled cognitive processes have been
defined as those occurring under the conscious awareness
of the person, whereas automatic processing is thought
to be in operation when the individual processes informa-
tion and renders judgments without conscious monitoring
or critical deliberation (Ilgen & Feldman, 1983). It seems
likely that providing reward recommendations in the field
setting required more controlled processing than that
which was required in the lab setting. Specifically, the
two settings are different in that for students, making
recommendations regarding instructors entails no subse-
quent consequences. Recommending an instructor for a
teaching award does not typically have a personal impact
on the student rater. On the other hand, for managers,
reward recommendations typically have direct personal
consequences. For example, contemplating whether some-
one should be recommended for a salary increase necessi-
tates budget considerations. Thus, when rendering this
type of judgment, managers may be forced into a more
controlled mode of processing in which they search for
and integrate information from a variety of sources to
determine reward recommendations.
Of additional interest is the finding that managers do
attribute different causal motives to the OCB of their em-
ployees. The motive ascribed appears to covary with the
frequency of OCB reported. That is, individuals who were
reported as engaging in OCB more often were also more
likely to have the cause of the behavior attributed to altru-
istic motives, whereas individuals who were reported as
engaging in OCB less often were more likely to have the
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR
cause of the behavior attributed to instrumental motives.
It appears that managers use consistency information in
determining the employee's motive for engaging in OCB.
This is congruous with previous research, which has iden-
tified consistency as one type of information that supervi-
sors use when evaluating employee motives (Kelley,
1973).
In both studies, job expectations did not moderate the
relationship between OCB and liking or between OCB
and perceived affective commitment. The results indicate
that although there were differences across individuals in
the degree to which they classified behaviors as beyond
expectations, this classification did not influence the effect
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
that OCB had on liking and perceived affective commit-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
ment. It appears that regardless of whether or not the rater
views this class of behavior as an expected part of the
job, the behavior is recognized as positive, and hence the
extent to which it is expected appears to bear no influence
on the effect the behavior has on related judgments. This
finding also has relevance to recent dialogue concerning
the definition underlying terms such as extrarole behavior
and organizational citizenship behavior (Organ, 1997;
Van Dyne et al., 1995). As Organ (1997) suggested, it
may no longer be useful to dichotomize behavior as either
inrole or extrarole. Distinctions such as that of contextual
and task performance as articulated by Borman and Mo-
towidlo (1993), which are less reliant on formal role
expectations, may have more theoretical and practical
meaning.
Overall, results of this research have a number of theo-
retical implications. The results suggest that social infor-
mation-processing theory may be useful for understanding
how OCB influences performance judgments. Although
the present research did not include a direct test of cogni-
tive processes per se, a consideration of this theory helps
shed some light on the method by which OCB exacts an
effect on performance judgments. Specifically, it seems
likely that employees who exhibit OCB match the proto-
type of the good employee. As such, positive affect and
other characteristics (e.g., commitment to the organiza-
tion) associated with a good employee are attached to
the cognitive representation of the ratee. These additional
attributes serve as mediators in transmitting the effect of
OCB on performance judgments. Consistent with previ-
ous research, OCB appears to have a great deal of effect
on performance judgments; however, its influence is pri-
marily through the effect it engenders on other psychologi-
cal factors.
Despite the strong results suggesting that a great deal
of the effect of OCB on performance judgments is trans-
mitted through other variables, additional processes also
appear to be operating. Mediation effects were less perva-
sive when considering reward recommendations. The re-
sults suggest that the process linking OCB with perfor-
257
mance judgments may depend on the type of judgment
considered (i.e., overall evaluation vs. reward recommen-
dations) and the context in which the judgment is
rendered.
The present findings also have theoretical implications
regarding the causal ordering of the liking—performance
rating relationship. Liking effects have generally been as-
sumed to reflect rater biases. Most research has taken the
perspective that supervisors give high performance ratings
to subordinates whom they like rather than they like high
performing employees better (Judge & Ferris, 1993). On
the other hand, Cardy and Dobbins (1994b) speculated
that supervisor liking may be a function of ratee perfor-
mance. Recent field research, using a structured diary-
keeping strategy, indicated support for this conjecture
(Varma, DeNisi, & Peters, 1996). Results of the current
lab study clearly indicated that, given objective levels of
performance, more effective performers were liked more
than were less effective performers. Thus, the present
findings provide further evidence that liking should not
automatically be assumed as a source of bias and that it
may indeed be a function of ratee performance.
One intent of conducting both a lab and a field study
was to help address the weaknesses inherent in the use
of any single research design. When two designs are com-
bined in one investigation to explore the same issue, the
strengths of one design can help compensate for the weak-
nesses of the other (Wayne & Ferris, 1990). Nevertheless,
limitations to the present research should be considered.
In the field study, the use of data collected at a single point
in time, from a single source, introduces the possibility of
common method variance. Although common method bias
is an unlikely explanation for results that are convergent
across the two studies, the effects of such bias cannot be
completely ruled out.
Another concern might be that in the lab study college
students, not managers, served as raters. However, partici-
pants were put in a rating situation they were very familiar
with, and the purpose of their participation was made
realistic. It should also be kept in mind that the primary
purpose of the laboratory study was to determine if the
findings observed in the field study could be replicated
under objectively defined conditions of task performance
and OCB. With this in mind, concerns about the internal
validity of the study took precedence over concerns about
the generalizability of the results (Mook, 1983; Sackett &
Larson, 1990). It can also be argued that the psychological
processes that were the focal point of interest in the pres-
ent study are common to all individuals, and hence gener-
alizability is less of an issue (Cardy & Dobbins, 1994a).
One final limitation to both studies concerns the identi-
fication of the locus of the effects observed. That is, al-
though the results are suggestive of the cognitive processes
that raters undergo in reaction to OCB, the present investi-
 258 ALLEN AND RUSH
gallon did not include an actual test of the cognitive mech-
anisms involved. As noted by Cardy and Dobbins
(1994b), examining the effects of manipulations on rat-
ings can lead to only an inference regarding the resultant
cognitive operations.
With these limitations in mind, the current results sug-
gest several avenues of future research worthy of pursuit.
In terms of the processes whereby OCBs influence perfor-
mance judgments, results of both the field and lab study
indicate the involvement of a complex system of interre-
lated affective and cognitive variables. Although the pres-
ent study provided a theoretical vantage point for begin-
ning to understand these processes, further theoretical de-
velopment is needed to better understand how OCB
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
uniquely produces performance judgment changes. One
consideration for future research is to examine if these
processes differ across the various dimensions of citizen-
ship (e.g., altruism, conscientiousness). Additional re-
search is also needed to identify other variables that may
be involved in this phenomenon. For example, perceived
affective commitment is likely to be affected by a multi-
tude of factors in addition to observed OCBs (e.g., organi-
zational tenure, willingness to relocate). Moreover, there
may be variables that moderate the relationship between
OCB and performance judgments. Inclusion of additional
variables in a full-information model would help research-
ers better understand the relationship between OCB and
performance judgments.
Future research should also examine the extent to which
managers purposefully reward OCB. For example, Podsa-
koff et al. (1993) speculated that managers may deliber-
ately and consciously recall acts of OCB and intentionally
reward employees out of a desire to reciprocate. More-
over, additional research examining factors such as pro-
motion rates and salary levels of individuals as a function
of level of OCB would be helpful in understanding how
this class of behavior contributes to the actual allocation
of organizational rewards.
In summary, the present research has attempted to ex-
pand the nomological net surrounding the effect of OCB
on performance judgments. The results of the two studies
permit considerable confidence in the conclusion that ad-
ditional psychological factors are involved in the effect
of OCB on performance judgments. Given the effect of
OCB on the outcomes investigated and the significant
impact of other attributes, researchers should continue to
examine the influence of cognitive and affective factors
on performance evaluation and reward outcome variables
in order to gain a greater appreciation of the decision
processes involved. Additional research in this area seems
not only warranted but critical to advancing theory and
practice regarding performance judgments.
References
Allen, T. D., Russell, J. E. A., & Rush, M. C. (1994). The ef-
fects of gender and leave of absence on attributions for high
performance, perceived organizational commitment, and allo-
cation of organizational rewards. Sex Roles, 31, 443-464.
Avila, R. A., Fern, E. F., & Mann, O. K. (1988). Unraveling the
criteria for assessing the performance of salespeople: A causal
analysis. Journal of Personnel Selling and Sales Manage-
ment, 8, 45-54.
Barnard, C.I. (1938). The functions of the executive. Cam-
bridge, MA: Harvard University Press.
Bateman, T. S., & Organ, D. W. (1983). Job satisfaction and the
good soldier: The relationship between affect and employee
"citizenship." Academy of Management Journal, 26, 587-
595.
Borman, W. C., & Motowidlo, S. J. (1993). Expanding the crite-
rion domain to include elements of extrarole performance. In
N. Schmitt & W. C. Borman (Eds.), Personnel selection in
organizations (pp. 71-98). San Francisco: Jossey-Bass.
Borman, W. C., White, L. A., & Dorsey, D. W. (1995). Effects
of ratee task performance and interpersonal factors on super-
visor and peer ratings. Journal of Applied Psychology, 80,
168-177.
Cardy, R. L., & Dobbins, G. H. (1986). Affect and appraisal
accuracy: Liking as an integral dimension in evaluating per-
formance. Journal of Applied Psychology, 71, 672-678.
Cardy, R. L., & Dobbins, G. H. (1994a). Performance ap-
praisal: Alternative perspectives. Cincinnati, OH: Southwest-
ern Publishing.
Cardy, R. L., & Dobbins, G. H. (1994b). Performance ap-
praisal: The influence of liking on cognition. In J. F. Porac,
J. R. Meindl, & C. Stubbart (Eds.), Advances in managerial
cognition and organizational information processing (Vol. 5,
pp. 115-140). Greenwich, CT: JAI Press.
Cook, T. D., Campbell, D. T, & Peracchio, L. (1990). Quasi
experimentation. In M. D. Dunnette & L. M. Hough (Eds.),
Handbook of industrial and organizational psychology (2nd
ed., Vol. 1, pp. 491 -576). Palo Alto, CA: Consulting Psychol-
ogists Press.
DeNisi, A. S., Cafferty, T. P., & Meglino, B. M. (1984). A cog-
nitive view of the performance appraisal process: A model and
research propositions. Organizational Behavior and Human
Performance, 33, 360-396.
Dobbins, G. H., & Russell, J. (1986). The biasing effects of
subordinate likableness on leaders' attributions and corrective
actions. Personnel Psychology, 39, 759-777.
Eastman, K. K. (1994). In the eyes of the beholder: An attribu-
tional approach to ingratiation and organizational citizenship
behavior. Academy of Management Journal, 37, 1379—1391.
Fandt, P.M., Labig, C. E., & Urich, A. L. (1990). Evidence
and the liking bias: Effects on managers' disciplinary action.
Employee Responsibilities and Rights Journal, 3, 253-265.
Feldman, J. M. (1981). Beyond attribution theory: Cognitive
processes in performance appraisal. Journal of Applied Psy-
chology, 66, 127-148.
Feldman, J. M. (1986). A note of the statistical correction of
halo error. Journal of Applied Psychology, 71, 173-176.
Fiske, S. T. (1981). Social cognition and affect. In J. Harvey
(Ed.), Cognition, social behavior, and the environment (pp.
227-264). Hillsdale, NJ: Erlbaum.
Fiske, S. T. (1982). Schema-triggered affect: Applications to
social perception. In M. S. Clark & S. T. Fiske (Eds.), Affect
 EFFECTS OF ORGANIZATIONAL CITIZENSHIP BEHAVIOR
and cognition: The 17th Annual Carnegie Symposium on cog-
nition (pp. 55-78). Hillsdale, NJ: Erlbaum.
Fiske, S. T., & Pavelchak, M. A. (1986). Category-based versus
piecemeal-based affective responses: Developments in
schema-triggered affect. In M. Sorrentino & E. T. Higgins
(Eds.), Handbook of motivation & cognition: Foundations
of social behavior (pp. 167-203). New \ork: Guilford Press.
Graham, J. W. (1991). An essay on organizational citizenship
behavior. Employee Responsibilities and Rights Journal, 4,
249-270.
Green, S., & Mitchell, T. (1979). Attributional process of lead-
ers in leader-member interactions. Organizational Behavior
and Human Performance, 23, 429-459.
Heider, F. (1958). The psychology of interpersonal relations.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
New York: Wiley.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Ilgen, D. R., & Feldman, J. M. (1983). Performance appraisal:
A process focus. In B. Staw & L. Cummings (Eds.), Research
in organizational behavior (Vol. 5, pp. 141-197). Green-
wich, CT JAI Press.
Jackson, D. W., Keith, J. E., & Schlacter, J. L. (1983). Evalua-
tion of selling performance: A study of current practices.
Journal of Personal Selling and Sales Management, 3, 43—
51.
James, L. R., & Brett, J. M. (1984). Mediators, moderators, and
tests for mediation. Journal of Applied Psychology, 69, 307-
321.
Judge, T. A., & Ferris, G. R. (1993). Social context of perfor-
mance evaluation decisions. Academy of Management Jour-
nal, 36, 80-105.
Katz, D., & Kahn, R. L. (1978). The social psychology of orga-
nizations (2nd ed.). New %rk: Wiley.
Kelley, H. H. (1973). The processes of causal attribution. Amer-
ican Psychologist, 28, 107-128.
Lord, R. G. (1985). An information processing approach to
social perceptions, leadership, and behavioral measurement
in organizations. In B. M. Staw & L. L. Cummings (Eds.),
Research in organizational behavior (Vol. 7, pp. 87-128).
Greenwich, CT: JAI Press.
Lord, R. G., & Maher, K. J. (1991). Leadership and information
processing: Linking perceptions and performance. New %rk:
Harper Collins.
MacKenzie, S. B., Podsakoff, P. M., & Fetter, R. (1991). Orga-
nizational citizenship behavior and objective productivity as
determinants of managerial evaluations of salesperson perfor-
mance. Organizational Behavior and Human Decision Pro-
cesses, 50, 123-150.
MacKenzie, S. B., Podsakoff, P. M., & Fetter, R. (1993). The
impact of organizational citizenship behavior on evaluations
of salesperson performance. Journal of Marketing, 57, 70-
80.
Mitchell, T, Green, S., & Wood, R. (1981). An attributional
model of leadership and the poor performing subordinate:
Development and validations. In B. M. Staw & L. L. Cum-
mings (Eds.), Research in organizational behavior (Vol. 3,
pp. 197-234). Greenwich, CT: JAI Press.
Mook, D. G. (1983). In defense of external invalidity. American
Psychologist, 38, 379-387.
Morrison, E. W. (1994). Role definitions and organizational
259
citizenship behavior: The importance of the employee's per-
spective. Academy of Management Journal, 37, 1543-1567.
Motowidlo, S. J., & Van Scotter, J. R. (1994). Evidence that
task performance should be distinguished from extrarole per-
formance. Journal of Applied Psychology, 79, 475—480.
Murphy, K. R., & Cleveland, J. N. (1991). Performance ap-
praisal: An organizational perspective. Boston: Allyn and
Bacon.
Organ, D. W. (1988). Organizational citizenship behavior: The
good soldier syndrome. Lexington, MA: Lexington Books.
Organ, D. W. (1990). The motivational basis of organizational
citizenship behavior. In B. Staw & L. Cummings (Eds.),
Research in organizational behavior (Vol. 12, pp. 43-72).
Greenwich, CT: JAI Press.
Organ, D. W. (1997). Organizational citizenship behavior: It's
construct clean-up time. Human Performance, 10, 85-98.
Organ, D. W, & Konovsky, M. (1989). Cognitive versus af-
fective determinants of organizational citizenship behavior.
Journal of Applied Psychology, 74, 157-164.
Orr, J. M., Sackett, P. R., & Mercer, M. (1989). The role of
prescribed and nonprescribed behaviors in estimating the dol-
lar value of performance. Journal of Applied Psychology, 74,
34-40.
Pence, E. C., Pendleton, W. C., Dobbins, G. H., & Sgro, J. A.
(1982). Effects of causal explanations and sex variables on
recommendations for corrective action following employee
failure. Organizational Behavior and Human Performance,
29, 227-240.
Perdue, B. C., & Summers, J. O. (1986). Checking the success
of manipulations in marketing experiments. Journal of Mar-
keting Research, 23, 317-326.
Podsakoff, P.M., Ahearne, M., & MacKenzie, S. B. (1997).
Organizational citizenship behavior and the quantity and qual-
ity of work group performance. Journal of Applied Psychol-
ogy, 82, 262-270.
Podsakoff, P. M., & MacKenzie, S. B. (1994). Organizational
citizenship behaviors and sales unit effectiveness. Journal of
Marketing Research, 31, 351-363.
Podsakoff, P. M., & MacKenzie, S. B. (1997). The impact of
organizational citizenship behavior on organizational perfor-
mance: A review and suggestions for future research. Human
Performance, 10, 133-152.
Podsakoff, P. M., MacKenzie, S. B., & Hui, C. (1993). Organi-
zational citizenship behaviors and managerial evaluations of
employee performance: A review and suggestions for future
research. In G. R. Ferris (Ed.), Research in personnel and
human resources management (Vol. 11, pp. 1-40). Green-
wich, CT: JAI Press.
Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter,
R. (1990). Transformational leader behaviors and their ef-
fects on followers' trust in leader, satisfaction, and organiza-
tional citizenship behaviors. Leadership Quarterly. 1, 107-
142.
Powell, G. N. (1993). Women and men in management (2nd
ed.). Newbury Park, CA: Sage.
Sackett, P. R., & Larson, J. R. (1990). Research strategies and
tactics in industrial and organizational psychology. In M. D.
Dunnette & L. M. Hough (Eds.), Handbook of industrial and
 260 ALLEN AND RUSH
organizational psychology (2nd ed., Vol. 1, pp. 419-489).
Palo Alto, CA: Consulting Psychologists Press.
Sauser, W. I., Evans, K. L., & Champion, C. C. (1980). Two
hundred fifty scaled incidents of college classroom teaching
behavior. Resources in Education, 15, 30.
Schnake, M. (1991). Organizational citizenship: A review, pro-
posed model, and research agenda. Human Relations, 44,
735-759.
Shore, L. R, Barksdale, K., & Shore, T. H. (1995). Managerial
perceptions of employee commitment to the organization.
Academy of Management Journal, 38, 1593-1615.
Smith, C. A., Organ, D.W., & Near, J. P. (1983). Organiza-
tional citizenship behavior: It's nature and antecedents. Jour-
nal of Applied Psychology, 68, 653-663.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Tsui, A. S., & Barry, B. (1986). Interpersonal affect and rating
This document is copyrighted by the American Psychological Association or one of its allied publishers.
errors. Academy of Management Journal, 29, 586—599.
Van Dyne, L., Cummings, L. L., & Parks, J. (1995). Extra-role
behaviors: In pursuit of construct and definitional clarity (a
bridge over muddied waters). In L. L. Cummings & B. M.
Staw (Eds.), Research in organizational behavior (Vol. 17,
pp. 215-285). Greenwich, CT: JAI Press.
Varma, A., DeNisi, A. S., & Peters, L. H. (1996). Interpersonal
affect and performance appraisal: A field study. Personnel
Psychology, 49, 341-360.
Wayne, S. J., & Ferris, G. R. (1990). Influence tactics, affect,
and exchange quality in supervisor-subordinate interactions:
A laboratory experiment and a field study. Journal of Applied
Psychology, 73, 487-499.
Werner, J. M. (1994). Dimensions that make a difference: Examin-
ing the impact of in-role and extrarole behaviors on supervisory
ratings. Journal of Applied Psychology, 79, 98-107.
Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and
organizational commitment as predictors of organizational
citizenship and in-role behavior. Journal of Management, 17,
601-617.
Received January 7, 1997
Revision received November 10, 1997
Accepted November 11, 1997 •