Teachers' organizational citizenship behaviour: Considering the roles of their work engagement, autonomy and leader–member exchange

lable at ScienceDirect

Teaching and Teacher Education 26 (2010) 914e923

Contents lists avai

Teaching and Teacher Education

journal homepage: www.elsevier .com/locate/ tate

Teachers’ organizational citizenship behavior: Examining the boundarybetween in-role behavior and extra-role behavior from the perspectiveof teachers, principals and parents

Elena Belogolovsky a, Anit Somech b,*

a Technion e Israel Institute of Technology, The William M. Davidson Faculty of Industrial Engineering and Management, Technion City, Haifa 32000, Israelb Faculty of Education, University of Haifa, Mount Carmel, Haifa 31905, Israel

a r t i c l e i n f o

Article history:Received 26 March 2009Received in revised form24 June 2009Accepted 27 October 2009

Keywords:Organizational citizenship behaviorIn-role behaviorRole perceptionsTeachers

* Corresponding author.E-mail addresses: [email protected] (E. Be

haifa.ac.il (A. Somech).

0742-051X/$ e see front matter � 2009 Elsevier Ltd.doi:10.1016/j.tate.2009.10.032

a b s t r a c t

This study examined how different stakeholders in school (principals, teachers, parents) conceptualizeteachers' role breadth, i.e., whether they define given behaviors commonly assumed to be OCBs as in-roleor extra-role behaviors. Drawing on a survey of 29 principals, 245 teachers and 345 parents, the resultsshowed that principals and teachers defined more teachers' OCBs in general and toward the school as in-role behaviors than parents did; parents defined more teachers' OCBs toward the student as in-rolebehaviors than teachers did. A wide variety of possible school interventions is indicated, each with thepotential for far broader improvements in school effectiveness.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Today, as educational systems move into an era of reorganiza-tion and are required to work in a competitive and complex envi-ronment (Miller, 2002), success of schools fundamentally dependson teachers who are committed to school goals and values (Oplatka,2006; Somech & Ron, 2007) and more willing to go above andbeyond the call of duty to contribute to successful change, that is, toengage in such organizational citizenship behaviors (OCBs). OCBs,or extra-role behaviors (ERBs), refer to all helping behaviorsextended to colleagues, supervisors, and students, such as lendinga colleague a hand with work overload or preparing specialassignments for higher and lower level students; and extended tothe school at large, such as suggesting improvements in pedagog-ical issues or talking favorably about the school to outsiders (e.g.,Penner, Midili, & Kegelmeyer, 1997; Van Dyne, Cummings, & Parks,1995). OCBs are essential because schools cannot anticipatethrough formally stated in-role job descriptions the entire array ofbehaviors needed for achieving goals (George & Brief, 1992).

Yet despite evidence of the contribution of teachers' OCB toschool effectiveness, critics have questioned the extent to which

logolovsky), anits@construct.

All rights reserved.

OCBs, or extra-role behaviors, may be regarded as discretionary,arguing that the distinction between required behavior, i.e., in-rolebehavior (IRB), and behavior that exceeds one's job requirements,i.e., OCB, is ambiguous (Tepper, Lockhart, & Hoobler, 2001)., andmay differ among persons (Coyle-Shapiro, Kessler, & Purcell, 2004;Morrison, 1994), contexts, and over time (Van Dyne, Cummings, &Parks, 1995). Previous research on OCB took for granted theunderlying nature of OCB (McAllister, Kamdar, Morrison, & Turban,2007), which has been defined as organizationally functionalemployee behavior that is discretionary, beyond the strictdescription of job requirements, and not directly rewarded (Organ,Podsakoff, & MacKenzie, 2006). Thus OCB research has tended tosidestep the potential ambiguity of the OCB construct (Morrison,1994; McAllister et al., 2007), while studies showing thatemployees frequently perceive presumed OCBs as role-prescribed,nondiscretionary, and/or rewarded imply that the commonlyaccepted definition does not accurately characterize employees'perceptions of OCB (e.g., Haworth & Levy, 2001; Tepper et al., 2001).OCB research has also sidestepped the potential subjectivity of theOCB construct by adopting a single perspective (Morrison, 1994),namely that of the supervisor (e.g., Moorman, 1991; Niehoff &Moorman, 1993). Yet there is no agreement in the literature as towhat is the best source for evaluating OCBs, because these behav-iors are not necessarily directed to the supervisor (Allen, Bernard,Rush, & Russell, 2000) and might be directed to co-workers orclients.

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923 915

The purpose of this study was to try to bring these issues intosharper focus. Specifically, it addressed taken-for-granted beliefsabout the underlying nature of teachers' OCB. To clarify thepotential ambiguity and subjectivity of the teachers' OCB construct,we examined how different stakeholders in school (principals,teachers and parents) conceptualize teachers' role breadth, that is,whether they define given behaviors commonly assumed to beteachers' OCBs as their in-role (greater role breadth) or extra-rolebehaviors. Although in the school context teachers' OCBs have beenexamined from principals' and teachers' perspectives (e.g., Somech& Ron, 2007; Vigoda-Gadot, 2007), no systematic work has exam-ined whether different role-holders perceive differently theboundary between in-role and extra-role behaviors. Moreover, tothe best of our knowledge no study has incorporated the client'sperspective, to obtain a more comprehensive picture of the in-rolee extra-role boundary.

We argue that if researchers wish to understand what foretellsteachers' OCB they must first understand how different stake-holders in school conceptualize it because a weighty factor drivingteachers' behavior, and principals' and parents' evaluation of it, iswhether teachers, principals and parents define a given activity asin-role or extra-role. Moreover, an understanding of differentschool stakeholders' perceptions of teachers' role breadth mightfurthermore help minimize teachers' role conflict derived fromambiguity and uncertainty about role expectations (Oplatka, 2006)and it might also point the way to a wide variety of school inter-ventions and changes, each with the potential for far broaderimprovements in school effectiveness.

2. Conceptual background

2.1. OCB: definition and construct

OCBs are those behaviors that go beyond specified role require-ments, and are directed toward the organization as a unit, the team,and the individual, in order topromoteorganizational goals (Somech&Drach-Zahavy, 2000). This definition of teachers' OCB stresses threemain features of extra-role behavior. First, the behavior must bevoluntary, that is, neither role-prescribed nor part of formal job duties.Second, the focus is on behaviors that do not simply happen in anorganization but are directed toward or seen as benefiting the orga-nization (Van Dyne et al., 1995). Third, by this definition OCB ismultidimensional by nature. Although most scholars agree on themultidimensionality of this construct, a reviewof the literature revealsa lack of consensus about its dimensionality. In this study we adoptedthe three-dimension construct of Somech and Drach-Zahavy (2000),which was developed specifically for the context of school. The threedimensions are (a) OCB toward the student, pertaining to behaviorsdirectly and intentionally aimed at improving the quality of teaching(e.g., acquiring expertise in new subjects that contribute to teaching,enhancing the ability to deal with students' special needs); (b) OCBtoward the team, that is, behaviors intentionally directed at helpinga specific teacher (e.g., helping other teachers who have heavy work-loads, orienting new teachers); and (c) OCB toward the organization asa whole, that is, a more impersonal form of behaviors that do notprovide immediate aid to any one specific person but are directed tothe benefit of the whole organization (e.g., volunteering for unpaidtasks, making innovative suggestions to improve the school).

2.2. OCB: in-role vs. extra-role behavior

For better grasp of what OCB is, it must be distinguished fromother concepts close to it. The important distinction should bemade between OCB and in-role behavior. This distinction is meantto draw a line between the types of “behaviors which are required

or expected as part of performing the duties and responsibilities ofthe assigned role” (Van Dyne et al., 1995, p. 222) that is, in-rolebehavior, and the types of behavior that go beyond the formalcontract, that is, extra-role behavior. By Organ's (1988) definition,that OCB consists of positive types of behavior that are not part ofthe formal job description, OCB should be limited to extra-rolebehavior.

Yet as Van Dyne et al. (1995) argued, defining behavior as in-roleor extra-role is theoretically important, but empirically it is difficultto distinguish because of OCB's dynamic and relative nature.According to these researchers, there are threemain factors impededifferentiation of in-role and extra-role behaviors. First, thefeatures deemed to make a certain behavior in-role or extra-rolemight be a function of the expectation of the observer who does thelabeling. This is a typical example of role conflict (Katz & Kahn,1978) that might develop because two observers have differentstandards and expectations. The second factor concerns the char-acteristics of the employees being observed: one observer mayhave different standards and expectations for different employees,based on their ability, motivation, etc.; as a result, a specificbehavior will be labeled in-role in one employee and extra-role inanother. Third, perceptions can change over time, so behaviororiginally labeled in-role will later be labeled extra-role, and thereverse. In sum, OCB is difficult to follow and measure empiricallybecause it is not stable, hard to quantify, and defined variously bydifferent people (Van Dyne, Grahm, & Dienesh, 1994).

Likewise, recent research on the way a role is understood by therole holder (Ashforth, Kreiner, & Fugate, 2000; Parker, 1998; Tepperet al., 2001) shows that the distinction between in-role and extra-role behaviors is not as clear as it may first seem. Empiricalevidence suggests that role-holders evaluate work roles and rolebehaviors in different ways; hence, an image of the role and OCB isto some degree unique to the individual (Neale & Griffin, 2006) anddepends on the beliefs the individual brings to the role or accordingto which he/she judges another's job role. This approach to roles isparticularly relevant in schools. Schools cannot anticipate throughformally stated in-role job descriptions the entire array of behaviorsneeded for achieving goals (George & Brief, 1992) because teachers'role is a complex and situational activity that requires professionaljudgments (DiPaola & Hoy, 2005b; Rowan, Raudenbush, & Cheong,1993). Moreover, teachers work in large bureaucratic systems, theirduties are often intensely interpersonal, and OCBs are defined ashelping behaviors (Hannam & Jimmieson, 2002), which makes itdifficult to determine which behaviors in this helping professionare “extra-role”. The OCB boundary is not clear and it may varyacross principals, teachers and parents. Research on role making(Graen, 1976) and psychological contracts (Rousseau, 1989)suggests that roles in organizations are seldom fixed and the roleperceptions evolve as employees and supervisors negotiate thescope of work activities, starting from substantially differentunderstandings about employment obligations. Finally, research insocial information processing (Salancik & Pfeffer, 1978) suggeststhat jobs are cognitive constructions created when employees andemployers make sense of social and behavioral cues. So work rolesare likely to be socially constructed with a subjective boundarybetween in-role and extra-role work behavior (Morrison, 1994) andtherefore subject to multiple interpretations by different stake-holders in school.

2.3. IRB vs. ERB: the perspective of teachers, principals and parents

Since teachers' OCB consists of behaviors that are neitherdescribed nor prescribed by the school, a common set of perfor-mance standards for rating these behaviors does not exist (Allenet al., 2000). Additionally, as we noted above, individuals in the

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923916

school other than the co-worker may frequently be the target ofthese behaviors, so acts of OCB may frequently escape the attentionof the principal or the client-parent.

Although to best of our knowledge this issue has not beenexamined in the school context, in industrial organizations a fewstudies have probed the differences between employees' OCB ratingsources (Allen et al., 2000; Becker&Vance,1993; VanDyne& LePine,1998). This research (Chiaburu, 2007; Morrison, 1994; Pond,Nacoste, Mohr, & Rodrigues, 1997; Putka & Vancouver, 1999; Vey &Campbell, 2004) suggested that employees' perceptions of their jobrequirements may differ substantially from their supervisors'perceptions (e.g., Cardona & Espejo, 2002; Lam, Hui, & Law, 1999;Morrison, 1994). Morrison (1994) and Lam et al. (1999) found thatsupervisors had a broader definition than employees of job in-roles.Not surprisingly, these findings matched prior propositions indi-cating that work roles are subject to negotiation and modification(e.g., Ilgen & Hollenbeck, 1991) or are simply artificially and contin-uously enlarged by supervisors, in an attempt to make employeesover-fulfill their work obligations (e.g., Vigoda-Gadot, 2007).

A number of different explanations have emerged to account forwhy the ratings of employees' OCB obtained from different sources(e.g., subordinates, supervisors) generally do not converge. Forexample, the modest correspondence among ratings from differentorganization stakeholders may be due to the different perceptionsof what constitutes effective performance in a particular job(Campbell & Lee, 1988). In addition, disagreements in their ratingsmay be attributed to the fact that different organization stake-holders differ in their opportunity to observe any given individuals'work behavior (Murphy & Cleveland, 1995) and they are exposed toonly moderately overlapping sets of employees' behavior (Lance,Teachout, & Donnelly, 1992). Finally, Campbell and Lee (1988), aswell as Cardy and Dobbins (1994), described a number of processesthat may lead to disagreement in the ratings provided by differenttypes of raters. These processes include well-established attribu-tional tendencies, such as the self-serving attributional bias and theactor-observer effect, as well as motivational and informationaldifferences between rating sources such as self-raters' needs forself-enhancement (Farh & Dobbins, 1989b) and differences in socialcomparison information available to self-raters and their supervi-sors (Farh & Dobbins, 1989a).

Drawing on these attributional tendencies, Becker and Vance(1993) compared OCB ratings among self, supervisor, and peers.They provided evidence that job incumbents, supervisors, andpeers perceived OCB in different ways. Their results formeda pattern similar to those observed by Harris and Schaubroeck(1988), Allen et al. (2000) and Van Dyne and LePine (1998),demonstrating that ratings made by others (subordinates andsuperiors) were more similar than ratings made by self and others.

Together, these studies highlight the importance of capturingthe extent to which different stakeholders in school view teachers'OCB as in-role or extra-role behaviors. From the findings of thesestudies, which confirm attribution theory (Floyd, 2000; Gibson &Schroeder, 2003; Kelley, 1967; Martinko & Gardner, 1987; Sabini,Siepmann, & Stein, 2001; Wilson & Levine, 1997), it seems likelythat teachers' (actors') perception of their OCBs will differ fromthose of principals and parents (others).

However, an alternative perspective might suggest that parents'perceptions of teachers' OCBs will differ from those of teachers andprincipals. Principals' and teachers' perceptions of teachers' roleboundary will be similar as a result of a common professionalculture due to a similar professional and educational background(Oplatka, 2006). In other words, principals and teachers as role-holders who share the same profession might adopt similar jobbehaviors and attitudes as a result of job norms that characterizethis profession (Bidwell, 1965). Moreover, according to Schneider's

(1987) attractioneselectioneattrition (ASA) model, people withsimilar characteristics tend to be attracted to one another, to stay inthe same work environment, and this tendency might explain thesimilarity in behavior of people in the samework setting. Principalsand teachers working together in schools might find themselvesworking with colleagues much like themselves, so there might bemore similarity in attitudes, values, and personalities betweenorganization members-principals and teachers than between themand their clients e the parents (Somech, 1994).

In sum, this study aimed to understand better the boundarybetween teachers' in-role and extra-role behaviors from multipleperspectives. Specifically, we examined how different stakeholderse principals, teachers and parents e conceptualize teachers' rolebreadth, i.e., whether they define given behaviors commonlyassumed to be OCBs as in-role (greater role breadth) or as extra-rolebehaviors.

3. Method

3.1. Sample and procedure

Participants were 29 principals, 245 teachers and 345 parentsfrom 30 elementary schools in Israel. Participants were recruited asfollows: first, researchers called to 35 schools principals in Israel,randomly drawn from a list provided by the Ministry of Educationand asked to participate in this research. Thirty of them agreed toparticipate. Next, with the principals' permission, questionnaireswere distributed in all schools to principals, all homeroom teachers,and 25 parents. The questionnaires were distributed to parents ofchildren from the same class by the homeroom teacher or by theresearcher on parents' days. Participants were asked to take part inthe study on a voluntary basis. Response rates were 96.6% for prin-cipals, 40.8% for teachers and 46% for parents. The purpose of thestudy was outlined in general; respondents were guaranteedanonymity and a commitment to provide a detailed analysis of thedata whenever requested. Participants were also informed that thequestionnaires were coded for research purposes only, and thatanonymitywould bemaintained for all other purposes thatwerenotstrictly scientific. Coding the principals', teachers' and parents'questionnaireswas essential in order tomatch themwith the school.

School size (number of classes) ranged from 7 to 22, with anaverage of 14.38 (SD ¼ 4.13). Regarding socioeconomic level, 17.2%(N ¼ 5) were schools on a low socioeconomic level, 48.3% (N ¼ 14)wereaverageand34.5% (N¼10)wereonahigh socioeconomic level.

Of the principals, 65.5% (N ¼ 19) were female, average age was48.8 years (SD ¼ 7.4); average tenure in managing a specific schoolwas 7 years (SD ¼ 6), and average tenure in teaching was 24 years(SD ¼ 8.8). As for higher education, 37.8% (N ¼ 11) held a Bachelor'sdegree, 58.6% (N ¼ 17) a Master's degree and 3.4% (N ¼ 1)a “professional” degree (equivalent to a junior college diploma,with teaching credentials).

Regarding the teachers, 89.8% were women; the teachers'average age was 41.55 years (SD ¼ 9). Average number of years ofteaching in the current school was 10.7 (SD ¼ 7.5), and the averagenumber of years as teachers was 16.7 (SD ¼ 9.7). In educationalbackground, 60.8% (N ¼ 149) of the respondents held a Bachelor'sdegree, 15.1% (N ¼ 37) a Master's degree and 21.3% (N ¼ 50) had theequivalent of a junior college diploma with teaching credentials;3.6% (N ¼ 9) did not answer this question. There were 179 home-room teachers and 191 teachers of specific subjects (the sumexceeds 100% because some teachers had both duties). Thesedemographic characteristics are similar to those found in compa-rable studies on teachers and principals in Israel (Bogler & Somech,2004; Somech, 2002). The average number of teachers who filledout the questionnaire per school was 5 (SD ¼ 3).

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923 917

Of the parents, 75.9% (N ¼ 262) were women, with average age39.8 years (SD ¼ 5). Regarding educational background, 35.6%(N ¼ 123) completed high school or less, 30.1% (N ¼ 104) helda Bachelor's degree, 16.6% (N ¼ 57) held a Master's degree, 2.8%(N ¼ 10) held a PhD degree or were studying for it, and the rest didnot answer this question. 22.3% (N ¼ 77) of the respondents wereon parents' committees. The average number of parents who filledout the questionnaire per school was 8 (SD ¼ 5).

3.2. Measures

3.2.1. Organizational citizenship behaviorTeachers' organizational citizenship behavior was assessed by

the 24-item questionnaire developed and validated specifically inthe context of schools by Somech and Drach-Zahavy (2000). Weremoved from the original questionnaire two items formulatedunambiguously as extra-role behaviors (i.e., “Volunteer for rolesand tasks that are not required” and “Assume responsibilities thatare not a prescribed part of my job”), which wewould therefore notbe able to test for the extent they could be interpreted as in-rolebehaviors. Respondents were asked to indicate the extent each itemwas a teacher's in-role behavior on a 5-point Likert-type scale from5 (strongly agree) to 1 (strongly disagree). Thus, high scores repre-sent in-role behaviors, and low scores represent extra-rolebehaviors.

In addition, because Somech and Drach-Zahavy (2000) devel-oped their OCB questionnaire from interviews with teachers only,we conducted a pretest to examine what items in it were relevantto principals and parents as well. In a semi-structured interview 3principals (females), 5 teachers (females) and 8 parents (4 malesand 4 females) were asked to list all kinds of behaviors that teachersmight perform to benefit the student, the team, or the organizationas a unit. Note that these respondents were not part of the samplethat completed the final questionnaire. The focus was on behaviorsexceeding specific role expectations and requirements. In theirinterviews, all eight parents said that questions about behaviorsthat teachers may perform to benefit the team were irrelevant tothem, so we decided to omit these items from the questionnairethat was distributed to parents.

The final questionnaire (see Appendix 1) for principals andteachers consisted of 20 items, and that for parents consisted of 12items (omission of the items concerning teacher's OCB toward theteam), including three subscales: (a) 5 items of teachers' OCBstoward the school as a unit (e.g., “Make innovative suggestions toimprove the school”); (b) 8 items of teachers' OCBs toward the teamin the principals' and teachers' questionnaires (e.g., “Offercolleagues work sheets that the teacher has prepared for his/herclass”); (c) 7 items of teachers' OCBs toward student (e.g., “Stay afterschool hours to help students with class materials”); two itemswere deleted because of low reliability. Each subscale in the ques-tionnaire was measured by the mean response to the relevantitems. The reliability level of alpha of the whole questionnaire was.84 (according to the teachers' sample).

3.3. Level of analysis

In order to decide the level of analysis, namely individual(N¼ 619) or school (N¼ 30), we ran two tests: inter-rater reliability(rwg) and intra-class correlation (ICC). First we computed theagreement index (rwg). James, Demaree, andWolf (1984) developedthe rwg within-group similarity or agreement coefficient to assessinter-rater reliability among judgments by a single group of“judges” (respondents) on a single variable (e.g., perception ofteachers' OCBs boundary). A value of .70 or above is suggested asa “good” amount of within-group inter-rater agreement (James,

Demaree, & Wolf, 1993). The findings showed that the mean rwg forteachers' perception of their OCBs in general was .87, toward theschool .78, toward the team .85, and toward the student .83; forparents' perception of teachers' OCBs in general it was .82, towardthe school .74 and toward the student .74. Consequently, all scalesdemonstrated high agreement among organization respondents.The mean rwg values were higher than the generally acceptablelevel of good agreement of .70 (George, 1990).

Next we performed the ICC test. Within-group agreement wasestimated by two measures: ICC(1) provides an estimate of thereliability of an individual respondent's rating and the ICC(2) esti-mates the reliability of mean differences across schools (Bliese &Halverson, 1996). As indicated by James (1982), ICC(1) generallyranges from 0 to .50 with a median of .12. There are, however, nodefinite guidelines for determining acceptable values. Our resultsdemonstrated the following ICC(1) and ICC(2) values: for teachers'perception of their OCBs in general .00 and .01; toward the school.04 and .28; toward the team .02 and .13; toward the student �.02and �.23. Note that the low ICC(2) values obtained in this studywere largely a function of sampling similar groups of teachers inschools, which resulted in less reliable differences between schoolsand thus lower ICC(2) values (cf. Bliese, 2000; James et al., 1984).For parents' perception of teachers' OCBs in general the resultswere .15 and .68; toward the school .12 and .62; toward the student.12 and .63. All of these (other than ICC(1) values of parents'perception of teachers' OCBs in general, toward the school, andtoward the student) were below the median or recommended ICCvalues reported in the literature (Knight, Durham, & Locke, 2001;Liao & Chuang, 2004).

We thus concluded that the variables did not show an appro-priate level of agreement among respondents in different schools,so treating the data on the individual level of analysis was justified.However, a multi-level analysis was necessary since individualswere nested in 30 different schools. For this purpose we used theMixed Linear Models procedure, which is appropriate for analyzingdata at the individual level, when individuals are nested in natu-rally occurring hierarchies. This process is performed throughcalculation of the random effect, which refers to factors that reflecta set of levels representing a sample out of a population of levels(Singer, 1998). In the present study the school level was treated asa random effect factor, so the analysis was assessed at the indi-vidual level, with account taken of the random effect of the schoollevel.

4. Results

Table 1 provides means, standard deviations, reliability coeffi-cients and intercorrelations (Pearson) of the variables calculated forprincipals, teachers and parents at the individual level (N ¼ 619)and at the organizational level (N ¼ 30). All the correlations at theindividual level among teachers' OCB in general and its threedimensions (toward the school, the team and the student) from theprincipals', teachers' and parents' perspective were positive andsignificant.

Descriptive analysis of frequencies of principals', teachers' andparents' definitions of teachers' OCBs as in-role vs. extra-rolebehaviors is shown in Table 2. To understand how principals,teachers and parents differed in their definition of teachers' OCB,we coded answers above the midpoint (¼3) as in-role behaviorsand answers below the midpoint as extra-role behaviors. Thefrequencies yielded several insights. Teachers' OCBs in general,toward the school, and toward the team were defined by all prin-cipals (100%, 100%, and 100%, respectively), the majority of teachers(91%, 89.8%, and 94.7%, respectively) and the majority of parents(71.2% and 73%, respectively) as in-role behaviors. However,

Table 1Means, standard deviations, reliability coefficients and intercorrelations (Pearson) of the variables calculated for principals, teachers and parents at the individual level(N ¼ 619) and at the organizational level (N ¼ 30).

Variable Principals Teachers Parents

M (SD) 1 2 3 4 M (SD) 1 2 3 4 M (SD) 1 2 3 4

1 OCB 3.77 (.36) 1.00 .80*** .81*** .82*** 3.65 (.47) 1.00 .84*** .83*** .77*** 3.41 (.66) 1.00 .82*** .53*** .87***2 OCB toward the school 4.05 (.47) .86 1.00 .54** .51** 3.84 (.66) .89 1.00 .61*** .49*** 3.59 (.80) .78 1.00 .39** .45***3 OCB toward the team 4.27 (.39) .78 .53 1.00 .41* 4.10 (.53) .76 .49 1.00 .37***4 OCB toward the student 3.01 (.47) .82 .59 .41 1.00 3.0 (.57) .74 .63 .23 1.00 3.18 (.77) .89 ,42. .56 1.00

*p < .05; **p < .01; ***p < .001.NOTE: Individual-level correlations are above the diagonal. Organizational-level correlations are below the diagonal. Means and standard deviations are reported for theindividual level.N principals ¼ 29; N teachers ¼ 245; N parents ¼ 345.

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923918

a majority of principals (51.7%) and teachers (58%) defined teachers'OCBs toward the student as extra-role behaviors, while a majorityof parents (52.9%) defined these behaviors as in-role behaviors.

Moreover, across the different stakeholders in school, the find-ings revealed variance for the three teachers' OCB dimensions(toward the student, the team, and the school), but also for eachbehavior and across behaviors. As we see in Table 3, the distinctionbetween teachers' OCB as required behavior, i.e., in-role behavior,and as behavior that exceeds one's job requirements, i.e., extra-rolebehavior, differs among principals, teachers and parents.

We conducted Mixed Models Analyses for the differencesamong teachers, principals and parents in their definition ofteachers' OCB in general and its three dimensions as in-role vs.extra-role behavior. In line with Becker's (2005) recommendationfor treating control variables, we ran the Mixed Model Analyseswith and without the control variables (i.e., school socioeconomiclevel, school size (number of classes) and school type). In theformer case, these variables were entered in the model as a controlfor its effects, while in the latter case theywere eliminated from theanalysis. This procedure was conducted to rule out the controlvariables as a potential explanation for the findings. Because theresults did not differ, we report only the results without the controlvariables (Becker, 2005).

4.1. Teachers' OCB in general as IRB vs. ERB: the perspectiveof teachers, principals and parents

Principals, teachers and parents differed significantly in theirdefinition of teachers' OCB in general (F (2, 586) ¼ 14.16, p < .001).Specifically, contrast analysis revealed that teachers defined theirOCBs in general as in-role behaviors significantly more (M ¼ 3.65;SD ¼ .47) than parents did (M ¼ 3.41; SD ¼ .66), t (586) ¼ 4.70,p < .001. Also, principals defined teachers' OCBs in general as in-role behaviors significantly more (M ¼ 3.77; SD ¼ .36) than parentsdid (M ¼ 3.41; SD ¼ .66), t (586) ¼ 3.35, p < .001. However, we didnot find a significant difference between principals and teachers intheir definition of teachers' OCBs in general (p > .05). So our find-ings suggest less difference between organizational members (i.e.,principals and teachers) than between them and organizationalclients (i.e., parents) in how broadly they define teachers' role

Table 2Principals', teachers' and parents' definitions of teachers' OCBs as in-role vs. extra-role b

Variable In-role behavior (%)

Principals Teachers

1 OCB 100 912 OCB toward the school 100 89.83 OCB toward the team 100 94.74 OCB toward the student 48.3 42

responsibilities, and consequently in whether they define so-calledteachers' OCBs as in-role or extra-role behaviors.

4.2. Teachers' OCB toward the student as IRB vs. ERB:the perspective of teachers, principals and parents

Principals, teachers and parents differed significantly in theirdefinition of teachers' OCBs toward the student (F (2, 586) ¼ 5.82,p < .001). Contrast analysis indicated that teachers definedteachers' OCBs toward the student as in-role behaviors significantlyless (M ¼ 3.0; SD ¼ .57) than parents did (M ¼ 3.18; SD ¼ .77),t (586) ¼ �3.34, p < .001. However, we did not find significantdifferences between principals and teachers, or between principalsand parents, in their definition of teachers' OCBs toward thestudent (p > .05).

4.3. Teachers' OCB toward the school as IRB vs. ERB: the perspectiveof teachers, principals and parents

Principals, teachers and parents differed significantly in theirdefinition of teachers' OCBs toward the school (F (2, 586) ¼ 9.67,p < .001). Contrast analysis showed that teachers defined teachers'OCBs toward the school as in-role behaviors significantly more(M ¼ 3.84; SD ¼ .66) than parents did (M ¼ 3.59; SD ¼ .80),t (586) ¼ 3.53, p < .001. Also, principals defined teachers' OCBstoward the school as in-role behaviors significantly more(M ¼ 4.05; SD ¼ .47) than parents did (M ¼ 3.59; SD ¼ .80),t (586) ¼ 3.24, p < .001. We found no significant differencesbetween principals and teachers in their definition of teachers'OCBs toward the school (p > .05).

4.4. Teachers' OCB toward the team as IRB vs. ERB:the perspective of teachers and principals

Principals and teachers evinced no significant difference in theirperception of teachers' role breadth toward the team (p > .05),suggesting that teachers and principals have similar conceptions ofteachers' job responsibilities in this regard and define the majorityof teachers' OCBs toward the team as in-role behaviors.

ehaviors: frequencies (%).

Extra-role behavior (%)

Parents Principals Teachers Parents

71.2 9 28.873 10.2 27

5.352.9 51.7 58 47.1

Table 3Principals', teachers' and parents' definitions of teachers' OCBs as in-role vs. extra-role behaviors: frequencies (%) and descriptive statistics for all scale items.

Items M (SD) M (SD)

Principals Teachers Parents

OCB toward the schoolOrganize social activities for school 3.48 (1.12) 3.59 (.98) 3.54 (1.07) 3.43 (1.14)Make innovative suggestions to improve the school 4.14 (.89) 4.34 (.67) 4.15 (.84) 4.12 (.94)Attend functions which help the school's image 4.09 (.95) 4.52 (.51) 4.27 (.83) 3.93 (1.02)Organize joint activities with parents 3.57 (1.06) 4.21 (.63) 3.82 (.88) 3.33 (1.14)Decorate the school 3.26 (1.11) 3.62 (.98) 3.40 (1.03) 3.12 (1.16)

OCB toward the teamVolunteer for school committee 3.34 (1.0) 3.41 (1.02) 3.34 (1.05)Work collaboratively with others (planning assignments, joint projects, etc.) 4.77 (.52) 4.86 (.35) 4.76 (.54)Orient new teachers 4.54 (.72) 4.55 (.57) 4.54 (.73)Offer the colleagues work sheets that the teacher prepared for his/her class 4.39 (.81) 4.55 (.63) 4.38 (.83)Participate actively in teachers' meetings 4.62 (.64) 4.97 (.19) 4.58 (.66)Prepare learning programs for substitute teachers 4.13 (1.02) 4.52 (.69) 4.21 (.98)Help other teachers who have heavy workloads 3.43 (1.09) 3.69 (.85) 3.40 (1.12)Help an absent colleague by assigning learning tasks to the class 3.62 (1.1) 3.62 (1.11) 3.62 (1.09)

OCB toward the studentStay after school hours to help students with class materials 3.21 (1.26) 3.50 (1.10) 3.06 (1.24) 3.30 (1.31)Arrive early for class 4.03 (1.16) 4.28 (.96) 4.21 (1.04) 3.87 (1.23)Acquire expertise in new subjects that contribute to the work 4.56 (.70) 4.72 (.45) 4.62 (.57) 4.49 (.79)Stay in class during breaks in order to listen to students 3.49 (1.09) 3.34 (.86) 3.64 (1.02) 3.39 (1.15)Go to school on free days to prevent problems in the class 2.1 (1.2) 1.82 (.82) 2.14 (1.18) 2.09 (1.19)Participate in private celebrations of students (e.g., birthdays) 1.95 (1.15) 1.90 (.94) 1.66 (.93) 2.20 (1.27)Invite students to the home 1.59 (.96) 1.48 (.74) 1.60 (.98) 1.83 (1.15)

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923 919

5. Discussion

The objective of this study was to address taken-for-grantedbeliefs about the underlying nature of teachers' OCB, which hasbeen defined as organizationally functional teachers' behavior thatis discretionary, beyond the strict description of job requirements,and not directly rewarded (Organ, 1988; Organ et al., 2006). Toclarify the potential ambiguity and subjectivity of the OCBconstruct, we examined how different stakeholders in school(principals, teachers and parents) conceptualize teachers' rolebreadth, that is, whether they define given behaviors commonlyassumed to be OCBs as in-role (greater role breadth) or extra-rolebehaviors (Bachrach & Jex, 2000; Morrison, 1994).

The results of the present study, in line with prior evidence fromresearch conducted in industrial organizations (e.g., Allen et al.,2000; Cardona & Espejo, 2002), indicate that there is no clear,common understanding of the boundary between teachers' OCBand in-role behavior across different stakeholders in school. In otherwords, our findings suggest that different role-holders perceivedifferently teachers' work roles and responsibilities. Consequently,the definition of so-called OCBs as in-role or extra-role behaviors isto some degree unique to the individual and depends critically onthe beliefs he/she brings to the role or according to which he/shejudges another's work roles. Work roles therefore are likely to be

Table 4Results of Mixed Models Analyses, testing principals', teachers' and parents' definitions o

Dependent variables OCB OCB to

Estimate SE t F Estima

Independent variables14.16***

Differences between teachers and parents .22 .05 4.70*** .21Differences between principals and parents .36 .11 3.35*** .44Differences between principals and teachers .13 .11 1.26 .23

Variance of school level intercept .03 .05Variance of individual-level intercept .30 .50

***p < .001; **p < .01; *p < .05.

socially constructed, with a subjective boundary between in-roleand extra-role work behavior (Morrison, 1994), hence subject tomultiple interpretations by different stakeholders in school. Thesefindings point to the need to take the perceptions of other organi-zation actors into account. Because raters on different levels mightobserve different aspects of teachers' OCB and also use differentstandards when judging it (Mount, Judge, Scullen, Sytsma, &Hezlett, 1998), each source has its bias and it is hard to say that onerating source is more valid than another (Organ, 1990).

Our findings demonstrate a significant difference betweenorganizational members (i.e., principals and teachers) and organi-zational clients (i.e., parents) in their definition of teachers' OCB asin-role vs. extra-role behavior (Table 4). This finding might supportSchneider's (1987) attractioneselectioneattrition (ASA) modelwhich suggests that people with similar characteristics tend to beattracted to one another, to stay in the same work environment,and this tendency might explain the similarity in behavior ofpeople in the samework setting. Therefore, there is more similarityamong organization members-principals and teachers thanbetween them and their clients e the parents (Somech, 1994).Moreover, in the context of school, because principals need to serveat least five years as teachers before their appointment, teachersand principals share a similar professional and educational back-ground, and a common professional culture (Oplatka, 2006).

f teachers' OCBs as in-role vs. extra-role behaviors.

ward the school OCB toward the team OCB toward the student

te SE t F Estimate SE t Estimate SE t F

9.67*** 5.82**.06 3.53*** �.19 .06 �3.34***.13 3.24*** �.17 .13 �1.31.14 1.66 .16 .10 1.64 .02 .13 .16

.001 .03

.26 .43

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923920

Further, we found that teachers and principals defined theteachers' role responsibilities broadly, and consequently theydefined more so-called teachers' OCBs as in-role behaviors thanparents did. This finding is not surprising and is consistent withstudies conducted in industrial organizations (e.g., Morrison, 1994;Van Dyne & Ellis, 2004). These studies showed that whenmanagerswere asked to evaluate their subordinates' performance they tookOCBs into account too (MacKenzie, Podsakoff, & Paine, 1999;Motowildo & Van Scotter, 1994; Podsakoff & MacKenzie, 1994;Van Scotter & Motowidlo, 1996; Werner, 1994). Principals appar-ently tend to define relatively larger teachers' job scopes becausethey are intent on being effective. From their perspective teachers'OCBs should be related to their own effectiveness and efficiency(Podsakoff, MacKenzie, & Hui, 1993), so it is to the principal'sadvantage to define more teachers' OCBs as in-role rather thanextra-role behaviors (Wanxian & Weiwu, 2007). For example,a teacher who is willing to help a new teacher may reduce the timeand effort that the principal must put in to guide and train this newteacher. As Vigoda-Gadot (2007) suggests, sometimes principals re-define behaviors previously perceived as extra-role as in-role, or“required,” without agreement from the teachers. In doing so, theysometimes extend the definition of formal duties into the informalarea of goodwill and put pressure on teachers to work beyond theirformally required duties.

From the teachers' perspective, it might be suggested that byextending the definition of teachers' formal duties, principals causeteachers to define their OCBs as in-role behaviors. In other words,teachers might tend to engage in OCBs as part of their role obliga-tions in response to external pressure by significant others in theworkplace (i.e., principals) (Vigoda-Gadot, 2007). Another possibleexplanation for thefinding suggesting that teachers define their roleresponsibilities broadly relates to the essence of teachers' work inschools. The ethical and humanistic dimensions of teachers' workfrequently act as a source of intrinsic motivation for individualteachers, and inspire them to remain committed to the profession(O'Connor, 2008). Moreover, because teaching is characterized as anambiguous role with vague boundaries and an unclear input-process-outcome connection (Meyer, Scott, & Deal, 1992), teachersmayprefer to includemany behaviors, defined in research literatureas OCBs, in their formal role definition. This process may be asa result of cognitive dissonance (Festinger, Carlsmith, & Bem, 2007)caused by their regularly performing a multitude of unrewardedextra-role behaviors because of external pressure and expectations(from principals, parents, the public, etc.). As a result, teachers soobliged feel dissonance, i.e., disharmony. To assuage it, they modifytheir definition of these behaviors from extra-role to in-rolebehaviors. However, this finding is inconsistent with findings fromstudies conducted in industrial organizations, which showed thatsupervisors have a broader definition of job in-roles than employeesdo (Chiaburu, 2007; Morrison, 1994; Pond et al., 1997; Putka &Vancouver, 1999; Vey & Campbell, 2004). It seems that schoolorganizations create a work environment different from those inother types of organizations, thus leading to different findings.

Moreover, our study for the first time incorporates the clientperspective so as to obtain a more comprehensive picture of the in-roleeextra-role boundaries. Based on this multi-perspectiveapproach it clarifies the potential subjectivity and ambiguity of theOCB construct that has been sidestepped by earlier research.

Concerning the dimensions of teachers' OCB, several insightsemerge. First, the parents defined more so-called teachers' OCBstoward the student as teachers' in-role behaviors than teachers did.The reason, we suggest, is that parents' expectations of the schoolare naturally individual in character and relate greatly to the child'sneeds, abilities and interests. These expectations emphasize his/hercentrality in all that the school and teachers do, ignoring other

aspects of the teacher's role. Interestingly, most principals, teachersand parents defined many teachers' OCBs toward the studentinvolving the social aspect of the teacher's role (e.g., “participate inprivate celebrations of students”; “invite students to the home”) asa teacher's extra-role rather than in-role behaviors. A possibleexplanation for this is that in the public eye the teachers' role isprimarily to promote the student's academic development. Schools,and consequently teachers and principals, are primarily evaluatedaccording to their students' achievements.

Concerning the teacher's OCB toward the school, teachers andprincipals defined teachers' role responsibilities toward the schoolmore broadly than parents did, hence they defined more so-calledteachers' OCBs toward the school (e.g., “organize social activities forschool”; “organize joint activities with parents”) as in-role behav-iors. This finding is not surprising, since principals are “school-oriented” in their thinking because they are responsible for thewhole school's effectiveness. They have high work performanceexpectations of teachers e their subordinates, which lead them tobroaden teachers' in-role duties. Similarly, teachers as members ofthe school organization have a variety of responsibilities to theclass, but also to the whole school. They define many of their so-called OCBs toward the school as the part of their job, in contrast toparents, who define these behaviors as teachers' extra-rolebehaviors because they don't necessarily see any direct impact ofthese teachers' behaviors on their children. Moreover, becauseparents are not exposed to all the variety of teachers' behaviorsoutside the class, theymay not see the relevance of those behaviorsto their children's achievements and well-being.

Regarding the teachers' OCB toward the team, a comparison ofteachers' and principals' perceptions only revealed that bothgroups defined the majority of these behaviors as teachers' in-roleactivities. We suggest that although in schools, as loosely coupledorganizations, teachers work in isolation from their peers, theyview their OCBs toward the team as routine and as part of theirprofessional identity (DiPaola & Tschannen-Moran, 2001). From theteacher's perspective, helping his/her colleague is mutuallynecessary behavior that both sides are interested in (Somech & Ron,2007). Teachers generally have high social needs, but find them-selves in a profession that isolates them from their colleagues. Toavoiding feelings of loneliness and dissatisfaction, they tend toengage in strong collegial relations, reducing isolation from peersthrough joint planning and implementation of activities anddefining them as a part of their job (Somech & Drach-Zahavy,2000). We posit that principals likewise broaden teachers' in-roleduties toward the team because it is to their advantage to definethese so-called teachers' OCBs as in-role behaviors since this broaddefinition may ensure teachers' performance of these behaviorsthat are necessary for school effectiveness.

In conclusion, we find that teachers, principals and parentsassign many teachers' OCBs (e.g., innovative and creative instruc-tion, out-of-school activities, helping behaviors, etc.), already dis-cussed in the literature as part of the teachers' moral and ethicalcommitments (e.g., DiPaola & Hoy, 2005a, 2005b; Somech & Drach-Zahavy, 2000), to the prescribed side of teaching. The ideal of serviceis imbedded in the ideal of teaching and includes responsibility forother people's development and commitment tomoral values (Day,1999), such as helping to co-workers or enhancing the ability to dealwith students' special needs. This perception of teacher's role isessential because although principals and teachers do have jobdescriptions covering teachers' main job responsibilities, schoolscannot anticipate through formally stated in-role job descriptionsthe entire array of behaviors needed for achieving goals (George &Brief, 1992). We suggest, therefore, that the teachers' role maycomprise many behaviors commonly assumed to be teachers' OCBsand measured in previous research as extra-role constructs.

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923 921

5.1. Limitations and suggestions for future research

Notwithstanding the merit of this study, several limitationsshould be noted. First, given its exploratory nature its results have tobe replicated. To generalize our findings we recommend that futurestudies replicate our measurement instruments and model inschools in other countries and cultural settings. A second limitationhas to dowith the possible bias thatmight be derived from teachers'fear of critique or social desirability bias. Although we assured theparticipants that the questionnaire was anonymous, teachers maynot have defined some specific behavior as extra-role behavior outof fear that the principal, who pressured them to perform it, mightcome across their responses. Moreover, the findings revealedhomogeneity across parents. Presumably, parents who agreed tospend time on completing the questionnaire are parents who aredeeply involved in their children's education. Accordingly, theyhavehigh expectations from teachers and define their role broadly.

Several avenues for future research should be pursued. First,while the results of the present study provide evidence thatperceptions of teachers' OCB boundary do differ across stakeholdersin school, the reasons underlying these differences need furtherinvestigation. We suggest that organizational factors (e.g., rewardsystem, organizational climate, and principal's leadership style) andindividual factors (e.g., personality, values)maybepossible causesofthese differences. Second, because our findings showed thatteachers' behaviors commonly defined in literature as OCBs, i.e.,discretionary, beyond the strict description of job requirements, andnot directly rewarded (Organ, 1988; Organ et al., 2006), are actuallydefined by different stakeholders in school as in-role behaviors, wepropose in future research to investigate an appropriate measure ofteachers' OCB, drawing a distinction between it andother constructsrelated to it, such as in-role behaviors. It is also important to examinein futurewhat rating source ismost valid to evaluate teachers' OCBs.Finally, future research may examine how different school stake-holders' perceptions of the boundary between teachers' OCBs andin-role behaviors relate to school effectiveness, relationships amongthese stakeholders, and attitudes to school.

5.2. Practical implications

Given our core assumption that it is important to address howdifferent stakeholders in school define teachers' job responsibili-ties, our findings offer practitioners some useful insights. First,mapping teachers' OCB in school and sharpening the deferencebetween it and in-role behavior from various angles may helpminimize teachers' role conflict that may derive from ambiguityand uncertainty of role expectations (Oplatka, 2006). It may helpalso to minimize the conflict among different school stakeholdersthat may arise due to different expectations from teachers asa result of different interests. In light of the present findings, wewould suggest that different stakeholders' perceptions of teachers'role breadth be taken into consideration in management practice.For example, in certain situations it might be preferable for teachersto hold broader job definitions, so that they can perform functionalbehaviors without feeling that they are doing something extra: asMorrison (1994) found, employees are more likely to perform OCBsthat they define as in-role rather than extra-role. To ensure moreconsistent performance of teachers' OCBs, and in an attempt tomake teachers fulfill their work obligations, principals mightinfluence teachers' definitions of these behaviors by artificially andcontinuously enlarging their work roles. In addition, as Werner(2000) suggests, for legal and practical reasons, if teachers' OCB isrecognized as part of their performance domain, this shouldinfluence performance appraisal practices. Specifically, schoolsshould emphasize teachers' job-relevant criteria in their appraisal

systems (Werner & Bolino, 1997) as specifically and behaviorally aspossible. However, if teachers' OCB is explicitly included in theappraisal process, then some of the definitional issues that haveplagued OCB research must be addressed (Morrison, 1994; Organ &Konovsky, 1989) because it would be hard to argue that a givenbehavior is “extra-role” once it appears on the school's rating form.

Further, the social environment in the workplace, as shaped byprincipals, peers and parents, has a strong influence on the defini-tion of a teachers' work responsibility as in-role or extra-role. Thus,understanding how principals, peers and parents define the teach-er's role breadth is critical. While some activities are definitelyconsidered as extra-role behavior by teachers; parentsmayperceivesuch activities as teachers' in-role duties (e.g., staying after school inorder to help a student, or using one's free time to organize a socialevent for the students). Moreover, teachers may prefer to includemanybehaviors, defined in the literature asOCBs, in their formal roledefinition as a result of parents' and principals' expectations. Thisprocess may be beneficial for the short run, but might impactteachers' sense of fairness for the long run. Thus, principals shouldclarify their teachers what are the boundaries between formal in-role behaviors and voluntary, spontaneous behaviors.

Finally, policy makers' knowledge of principals', teachers' andparents' perceptions of the teacher's role boundary is critical, sothat they can use it when constructing avenues for parents'involvement in school, since this homeeschool relationship isbased primarily on the expectations that the parent holds regardingthe teacher's job responsibilities.

Appendix 1

Organizational citizenship behavior

OCB toward the school

1. Decorate the school2. Organize social activities for school3. Make innovative suggestions to improve the school4. Attend functions which help the school's image5. Organize joint activities with parents

OCB toward the team

1. Help an absent colleague by assigning learning tasks to theclass

2. Volunteer for school committee3. Work collaboratively with others (planning assignments, joint

projects, etc.)4. Orient new teachers5. Offer the colleagues work sheets that the teacher prepared for

his/her class6. Participate actively in teachers' meetings7. Prepare learning programs for substitute teachers8. Help other teachers who have heavy workloads

OCB toward the student

1. Invite students to the home2. Participate in private celebrations of students (e.g., birthdays)3. Stay after school hours to help students with class materials4. Arrive early for class5. Acquire expertise in new subjects that contribute to the work6. Stay in class during breaks in order to listen to students7. Go to school on free days to prevent problems in the class

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923922

References

Allen, T. D., Bernard, S., Rush, M. C., & Russell, J. E. (2000). Ratings of organizationalcitizenship behavior: does the source make a difference? Human ResourceManagement Review, 10(1), 97e114.

Ashforth, B. E., Kreiner, G. E., & Fugate, M. (2000). All in a day's work: boundariesand micro role transitions. Academy of Management Review, 25, 472e491.

Bachrach, D. G., & Jex, S. M. (2000). Organizational citizenship and mood: anexperimental test of perceived job breadth. Journal of Applied Social Psychology,30, 641e663.

Becker, T. E. (2005). Potential problems in the statistical control of variables inorganizational research: a qualitative analysis with recommendations. Organi-zational Research Methods, 8, 274e289.

Becker, T. E., & Vance, R. J. (1993). Construct validity of three types of organizationalcitizenship behavior: an illustration of the direct product model with refine-ments. Journal of Management, 19, 663e682.

Bidwell, C. E. (1965). The school as a formal organization. In J. G. March (Ed.),Handbook of organizations. Chicago: Rand McNally.

Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability. InK. Klein, & S. Kozlowski (Eds.), Multi-level theory, research, and methods inorganizations (pp. 349e381). San Francisco, CA: Jossey-Bass.

Bliese, P. D., & Halverson, R. R. (1996). Individual and nomothetic models of jobstress: an examination of work hours, cohesion and well being. Journal ofApplied Psychology, 26(13), 1171e1189.

Bogler, R., & Somech, A. (2004). Influence of teacher empowerment on teachers'organizational commitment, professional commitment and organizationalcitizenship behavior in schools. Teaching and Teacher Education, 20, 277e289.

Campbell, D. J., & Lee, C. (1988). Self-appraisal in performance evaluation: devel-opment versus evaluation. Academy of Management Review, 13, 302e314.

Cardona, P., & Espejo, A. (2002). The effect of the rating source in organizationalcitizenship behavior: A multitrait-multimethod analysis. Research Paper No. 474.

Cardy, R. L., & Dobbins, G. H. (1994). Performance appraisal: Alternative perspectives.Cincinnati, OH: South-Western Publishing Co.

Chiaburu, D. (2007). From interactional justice to citizenship behaviors: roleenlargement or role discretion? Social Justice Research, 20(2), 207e227.

Coyle-Shapiro, J., Kessler, I., & Purcell, J. (2004). Exploring organizationally directedcitizenship behavior: reciprocity or ‘It's my job’? Journal of Management Studies,41, 85e106.

Day, C. (1999). Developing teachers: The challenge of lifelong learning. London: FalmerPress.

DiPaola, M. F., & Hoy, W. K. (2005a). Organizational citizenship of faculty andachievement of high school students. The High School Journal, 88, 35e44.

DiPaola, M. F., & Hoy, W. K. (2005b). School characteristics that foster organizationalcitizenship behavior. Journal of School Leadership, 15, 308e326.

DiPaola, M. F., & Tschannen-Moran, M. (2001). Organizational citizenship behaviorin school and its relationship to school climate. Journal of School Leadership, 11,424e447.

Farh, J. L., & Dobbins, G. H. (1989a). Effects of comparative performance informationon the accuracy of self-ratings and agreement between self and supervisorratings. Journal of Applied Psychology, 74, 606e610.

Farh, J. L., & Dobbins, G. H. (1989b). Effects of self-esteem on leniency bias in self-reports of performance: a structural equation model analysis. PersonnelPsychology, 42, 835e850.

Festinger, L., Carlsmith, J. M., & Bem, D. J. (2007). Does cognitive dissonance explainwhy behavior can change attitudes? In J. A. Nier (Ed.), Taking sides: Clashingviews in social psychology (2nd ed.). (pp. 74e91) New York: McGraw-Hill.

Floyd, K. (2000). Attributions for nonverbal expressions of liking and disliking: theextended self-serving bias. Western Journal of Communications, 64, 385e404.

George, J. M. (1990). Personality, affect, and behavior in groups. Journal of AppliedPsychology, 75, 107e116.

George, J. M., & Brief, A. P. (1992). Feeling good doing good: a conceptual analysis ofthe mood at work organizational spontaneity relationship. PsychologicalBulletin, 112, 310e329.

Gibson, D., & Schroeder, S. (2003). Who ought to be blamed? The effect of orga-nizational roles on blame and credit attributions. International Journal of ConflictManagement, 14, 95e117.

Graen, G. (1976). Role making processes within complex organizations. InM. D. Dunnette (Ed.), Handbook of industrial and organizational psychology(pp. 1202e1245). Chicago: Rand McNally.

Hannam, R. L., & Jimmieson, N. L. (2002). The relationship between extra-rolebehaviours and job burnout for primary school teachers: A preliminary modeland development of an organisational citizenship behaviour scale. In: Shilton,Wendy, Jeffrey and Ruth 2002 Annual Conference of the Australian Association forResearch, Brisbane, 1e5 December 2002 (pp. 1e17).

Harris, M. M., & Schaubroeck, J. (1988). A meta-analysis of self-supervisor, self-peer,and peer-supervisor ratings. Personnel Psychology, 41, 43e62.

Haworth, C. L., & Levy, P. E. (2001). The importance of instrumentality beliefs in theprediction of organizational citizenship behaviors. Journal of VocationalBehavior, 59, 64e75.

Ilgen, D. R., & Hollenbeck, J. R. (1991). The structure of work: job design and roles. InM. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and organizationalpsychology, Vol. 2 (pp. 165e207). Palo Alto, CA: Consulting Psychologists Press.

James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journalof Applied Psychology, 67, 219e229.

James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interraterreliabilitywith andwithout response bias. Journal of AppliedPsychology, 69, 85e98.

James, L. R., Demaree, R. G., & Wolf, G. (1993). Rwg: an assessment of within-groupinterrater reliability. Journal of Applied Psychology, 78, 306e310.

Katz, D., & Kahn, R. L. (1978). The social psychology of organizations (2nd ed.). NewYork: Wiley.

Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.),Nebraska Symposium of Motivation, 15 (pp. 192e238).

Knight, D., Durham, C. C., & Locke, E. A. (2001). The relationship of team goals,incentives, and efficacy to strategic risk, tactical implementation, and perfor-mance. The Academy of Management Journal, 44, 326e339.

Lam, S. S., Hui, C., & Law, S. L. (1999). Organizational citizenship behavior:comparing perspectives of supervisors and subordinates across four interna-tional samples. Journal of Applied Psychology, 84(4), 584e601.

Lance, C. E., Teachout, M. S., & Donnelly, T. M. (1992). Specification of the criterionconstruct space: an application of hierarchical confirmatory factor analysis.Journal of Applied Psychology, 77, 437e452.

Liao, H., & Chuang, A. (2004). A multilevel investigation of factors influencingemployee service performance and customer outcomes. Academy of Manage-ment Journal, 47, 41e58.

MacKenzie, S. B., Podsakoff, P. M., & Paine, J. B. (1999). Do citizenship behaviorsmatter more for managers than for salespeople? Journal of the Academy ofMarketing Science, 27, 396e410.

Martinko, M., & Gardner, W. (1987). The leader member attribution process.Academy of Management Review, 12(2), 235e249.

McAllister, D. J., Kamdar, D., Morrison, E. W., & Turban, D. B. (2007). Disentanglingrole perceptions: how perceived role breadth, discretion, and instrumentalityand efficacy relate to helping and taking charge. Journal of Applied Psychology,92, 1200e1211.

Meyer, J. W., Scott, R., & Deal, T. E. (1992). Institutional and technical sources oforganizational structure: explaining the structure of educational organizations.In J. Meyer, & R. W. Scott (Eds.), Organizational environments (pp. 45e70).Newbury Park, CA: Sage.

Miller, N. (2002). Insider change leadership in schools. International Journal ofLeadership in Education, 5, 343e360.

Moorman, R. H. (1991). Relationship between organizational justice and organiza-tional citizenship behaviors: do fairness perceptions influence employee citi-zenship? Journal of Applied Psychology, 76, 845e855.

Morrison, E. W. (1994). Role definitions and organizational citizenship behavior: theimportance of the employee's perspective. Academy of Management Journal, 37,1543e1567.

Motowildo, S. J., & Van Scotter, J. R. (1994). Evidence that task performance shouldbe distinguished from contextual performance. Journal of Applied Psychology, 79,475e480.

Mount, M. K., Judge, T. A., Scullen, S. E., Sytsma, M. R., & Hezlett, S. A. (1998). Trait,rater and level effects in 360-degree performance ratings. Personnel Psychology,51, 557e576.

Murphy, K. R., & Cleveland, J. N. (1995). Understanding performance appraisal: Social,organizational, and goal based perspectives. Thousand Oaks, CA: Sage.

Neale, M., & Griffin, M. A. (2006). A model of self-held work roles and role transi-tions. Human Performance, 19(1), 23e41.

Niehoff, B. P., & Moorman, R. H. (1993). Justice as a mediator of the relationshipbetween monitoring and organizational citizenship behavior. Academy ofManagement Journal, 36, 527e556.

O'Connor, K. E. (2008). “You choose to care”: teachers, emotions and professionalidentity. Teaching and Teacher Education, 24, 117e126.

Oplatka, I. (2006). Going beyond role expectations: toward an understanding of thedeterminants and components of teacher organizational citizenship behavior.Educational Administration Quarterly, 42, 385e423.

Organ, D. W. (1988). Organizational citizenship behaviors: The good soldier syndrome.Lexington, Massachusetts: Lexington Books.

Organ, D. W. (1990). The motivational basis of organizational citizenship behavior.Research in Organizational Behavior, 12, 43e72.

Organ, D. W., & Konovsky, M. (1989). Cognitive versus affective determinants oforganizational citizenship behavior. Journal of Applied Psychology, 74, 157e164.

Organ, D. W., Podsakoff, P. M., & MacKenzie, S. B. (2006). Organizational citizenshipbehavior: Its nature, antecedents and consequences. Thousand Oaks: SagePublication.

Parker, S. K. (1998). Enhancing role breadth self-efficacy: the roles of job enrichmentand other organizational interventions. Journal of Applied Psychology, 83, 835e853.

Penner, L. A., Midili, A. R., & Kegelmeyer, J. (1997). Beyond job attitudes: a person-ality and social psychology perspective on the causes of organizational citi-zenship behavior. Human Performance, 10(2), 111e131.

Podsakoff, P., & MacKenzie, S. B. (1994). Organizational citizenship behavior andsales unit effectiveness. Journal of Marketing Research, 31, 351e363.

Podsakoff, P. M., MacKenzie, S. B., & Hui, C. (1993). Organizational citizenshipbehaviors and managerial evaluations of employee performance: a review andsuggestions for future research. In G. R. Ferris, & K. M. Rowland (Eds.), Research inpersonnel and human resources management (pp.1e40). Greenwich, CT: JAI Press.

Pond, S. B., Nacoste, R. W., Mohr, M. F., & Rodrigues, C. M. (1997). The measurementof organizational citizenship behavior: are we assuming too much? Journal ofApplied Social Psychology, 27, 1527e1544.

Putka, D. J., & Vancouver, J. B. (1999, August). The moderating effects of employeeperceptions of behavioral discretion on the relationship between OCB and job

E. Belogolovsky, A. Somech / Teaching and Teacher Education 26 (2010) 914e923 923

satisfaction. Paper presented at the annual meeting of the Academy ofManagement, Chicago, IL.

Rousseau, D. M. (1989). Psychological and implied contracts in organizations.Employee Responsibilities and Rights Journal, 2, 121e139.

Rowan, B., Raudenbush, S., & Cheong, Y. (1993). Teaching as a non-routine task:implications for the organizational design of schools. Educational AdministrationQuarterly, 29(4), 479e500.

Sabini, J., Siepmann, M., & Stein, J. (2001). The really fundamental attribution Errorin social psychological research. Psychological Inquiry, 12, 1e14.

Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to jobattitudes and task design. Administrative Science Quarterly, 23, 224e253.

Schneider, B. (1987). The people make the place. Personnel Psychology, 40, 437e453.Singer, J. D. (1998). Using SAS proc mixed to fit multilevel models and individual

growth models. Journal of Educational and Behavioral Statistics, 24, 323e355.Somech, A. (2002). Explicating the complexity of participative management: an

investigation of multiple dimensions. Educational Administration Quarterly, 38,341e371.

Somech, A. (1994). The social loafing phenomenon compared to extra role behavior(pro-social behavior) as a function of culture and goal setting. Unpublisheddissertation, Technion: Israel Institute of Technology.

Somech, A., & Drach-Zahavy, A. (2000). Understanding extra-role behavior inschools: the relationships between job satisfaction, sense of efficacy, andteachers' extra-role behavior. Teaching and Teacher Education, 16, 649e659.

Somech, A., & Ron, I. (2007). Promoting organizational citizenship behavior inschools: the impact of individual and organizational characteristics. EducationalAdministration Quarterly, 5, 1e29.

Tepper, B. J., Lockhart, D., & Hoobler, J. (2001). Justice, citizenship and role definitioneffects. Journal of Applied Psychology, 86, 789e796.

Van Dyne, L., Cummings, L. L., & Parks, J. (1995). Extra-role behaviors: in pursuit ofconstruct and definitional clarity. Research in Organizational Behavior, 17,215e285.

Van Dyne, L., & Ellis, B. J. (2004). Job creep: a reactance theory perspective onorganizational citizenship behavior as over fulfillment of obligations. In J. Coyle-Shapiro, L. Shore, M. S. Taylor, & L. Tetrick (Eds.), The employment relationship:Examining psychological and contextual perspectives (pp. 181e205). Oxford:Oxford University Press.

Van Dyne, L., Grahm, J. W., & Dienesh, R. M. (1994). Organizational citizenshipbehavior: construct redefinition, measurement and validity. Academy ofManagement Journal, 37, 765e802.

Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role behaviors: evidenceof constructive and predictive validity. Academy of Management Journal, 41,108e119.

Van Scotter, J. R., & Motowidlo, S. J. (1996). Interpersonal facilitation and jobdedication as separate facets of contextual performance. Academy of AppliedPsychology, 81, 525e531.

Vey, M. A., & Campbell, J. P. (2004). In-role or extra-role organizational citizenshipbehavior: which are we measuring? Human Performance, 17, 119e135.

Vigoda-Gadot, E. (2007). Redrawing the boundaries of OCB? An empirical exami-nation of compulsory extra role behavior in the workplace. Journal of Businessand Psychology, 21(3), 377e405.

Wanxian, L., & Weiwu, W. (2007). A demographic study on citizenship behavior asin-role orientation. Personality and Individual Differences, 42, 225e234.

Werner, J. M. (1994). Dimensions that make difference: examining the impact of in-role and extra role behaviors on supervisory ratings. Academy of AppliedPsychology, 79, 98e107.

Werner, J. (2000). Implications of OCB and contextual performance for humanresource management. Human Resource Management Review, 10(1), 3e24.

Werner, J. M., & Bolino, M. C. (1997). Explaining U.S. Courts of Appeals decisionsinvolving performance appraisal: accuracy, fairness, and validation. PersonnelPsychology, 50, 1e24.

Wilson, S., & Levine, K. (1997). Attribution complexity and actor-observer bias.Journal of Social Behavior and Personality, 12(3), 709e727.