Journal of Business and Psychology

, Volume 26, Issue 3, pp 385–396

Individual Values, Organizational Commitment, and Participation in a Change: Israeli Teachers’ Approach to an Optional Educational Reform

Authors

    • School of Political Science, Division of Public AdministrationUniversity of Haifa
  • Lilach Caspary
    • School of Political Science, Division of Public AdministrationUniversity of Haifa
Article

DOI: 10.1007/s10869-010-9186-1

Cite this article as:
Cohen, A. & Caspary, L. J Bus Psychol (2011) 26: 385. doi:10.1007/s10869-010-9186-1

Abstract

Purpose

The purpose of this study was to examine how individual values and organizational commitment are related to teachers’ participation in an optional change program in the Israeli educational system.

Design/Methodology/Approach

Data were obtained from 214 Israeli teachers employed in 25 secular Jewish schools during a time when teachers had the option of joining a reform plan initiated by the government and one of several Israeli teachers’ unions.

Findings

The results, using analysis of variance and logistic regression, showed that teachers who joined the reform valued conservation more than those who did not. These teachers also scored higher on normative organizational commitment in comparison to teachers who did not join the reform. The logistic regression showed that organizational commitment had a stronger effect on participation in the reform than individual values.

Implications

Little data exists on top-down change processes in organizations. Research on issues such as the one discussed here can help academics better understand such change processes, and can help practitioners and decision makers in planning and implementing large-scale change.

Originality/Value

Few studies currently offer data on the relationship between values and commitment, on the one hand, and participation in a top-down change program on the other. Additionally, most studies of attitudes and values have examined them in a stable environment. This study examines how the two concepts are related to participation in change in a turbulent environment and can deepen our understanding of values and commitment.

Keywords

Participation in a changeOrganizational commitmentIndividual values

Introduction

Values are thought to play a functional role in all sorts of work-related processes and outcomes (Lam et al. 2002). In the past decade, researchers have begun to reexamine the effect of values on attitudes (Cohen 2007; Kirkman and Shapiro 2001; Pearson and Chong 1997; Wasti 2003) and on organizational citizenship behavior (OCB) and performance (Ang et al. 2003; Farh et al. 2007). Most studies on values in the workplace have focused on the national level of analysis, in that they have compared aggregated scales of values across countries (key examples are Hofstede 1980; Schwartz 1999). Few studies have examined the effect of values in the workplace at the individual level (Clugston et al. 2000; Fischer and Smith 2006; Kirkman and Shapiro 2001; Wasti 2003). Yet individuals both within and across societies may have quite different value priorities that reflect their heritage, personal experiences, socio-economic level, and acculturation (Schwartz and Bardi 2001).

Values can influence how an individual perceives and interprets a given situation and the importance he or she gives it (Schwartz et al. 2000), as well as how he or she reacts and behaves in given circumstances (Schwartz 1996). Further, values play a central role in determining the fit between individuals and the employment organization (Berings et al. 2004). These facts suggest that an understanding of individual-level differences in values may offer insights into better ways of managing different employees (Francesco and Chen 2004).

The current study takes advantage of a singular opportunity to study values under conditions not often seen in the working world: namely, a top-down program of change, in which employees can choose to participate or not with no ramifications for their job security (Israel Ministry of Education 2009). Since 2007, Israeli teachers have been given the option of joining a national educational reform initiated by Israel’s Ministry of Education and the country’s national union of elementary-school teachers. The reform affects work conditions and compensation for those who join (for instance, those teachers earn higher salaries in return for working more hours), but not for those who choose not to join (Israel Ministry of Education 2009). This study examines individual values in a sample of teachers who had made a decision one way or the other—that is, to participate in the reform or not. By examining the relationship between values and participation, this study investigates the role of values at the individual level of analysis and under conditions of change.

The reform also offers an opportunity to examine another concept worthy of attention: organizational commitment. Despite the growing body of research on organizational commitment, very little research has examined commitment in times of change (Fedor et al. 2006; Kondratuk et al. 2004). Moreover, among researchers who have examined this subject, most have studied the role of organizational commitment in shaping individuals’ reactions to organizational change (Lau and Woodman 1995; Madsen et al. 2005; Meyer et la. 2007). Few have empirically examined commitment among employees who had the privilege of deciding whether or not to participate in a change.

In this study, we examine differences in individual values and organizational commitment between teachers who chose to take part in the reform and others who did not. Our findings will examine how individual values and organizational commitment are related to the decision to participate in a change program when such participation is voluntary. Moreover, because the reform elicited strong emotions (both positive and negative) in many teachers, our study takes place in the context of what can be defined as a turbulent environment for employees. As suggested above, most studies on values and commitment have examined these concepts in stable environments, limiting the generalizability of previous findings. Looking at how these two important variables operate in more unstable conditions will deepen our understanding regarding both of them.

The current study will thus make a twofold contribution to the literature. First, it is one of the few studies thus far to examine how individual values and attitudes are related to participation in a top-down change. More specifically, it will examine how values and commitment are related to the decision to participate or not in a voluntary top-down change program. Second, it is one of the few to examine change at the individual rather than the organizational level (Elias 2009). The study will thus enhance our understanding of values, attitudes and behavior in the workplace and offer insights both for researchers and for individuals charged with initiating and managing change, especially voluntary change.

Conceptual Framework and Research Hypotheses

Schwartz’s Individual Values Model

Schwartz and Sagiv (1995) defined human values as desirable, trans-situational goals, varying in importance, that serve as guiding principles in people’s lives. The crucial content aspect that distinguishes these values from one another is the type of motivational goal they express. Schwartz (1992, 1996) derived a typology of the different content of values by reasoning that values represent, in the form of conscious goals, three universal requirements of human existence: biological needs, requisites of coordinated social interaction, and demands of group functioning. Values occupy a central position in a person’s cognitive system, and for this reason values influence our attitudes, decision-making processes, and in general all human behaviors.

Schwartz and Sagiv (1995) derived 10 distinct motivational types of values from the three universal requirements. They argued that the 10 value types are organized in two dimensions composed of higher-order value types. The first dimension—Openness to Change versus Conservation—juxtaposes values emphasizing independent thought and action and favoring change (self-direction, stimulation, and hedonism) with those emphasizing submissive self-restriction, preservation of traditional practices, and protection of stability (security, conformity, and tradition). The second dimension—Self-Enhancement versus Self-Transcendence—juxtaposes values emphasizing pursuit of one’s own relative success and dominance over others (power, achievement, and hedonism) with those emphasizing acceptance of others as equals and concern for their welfare (universalism and benevolence). (Note that hedonism fits in both dimensions, relating to both Openness to Change and Self-Enhancement.) Evidence for this theoretical structure has been found in samples from 67 nations (Schwartz 1992, 2005; Schwartz and Sagiv 1995) as well as in recent data from 38 countries (Fontaine et al. 2008). The findings provide substantial support for both the content and structure postulates of the theory, and specifically for the claim that 10 motivationally distinct value types are recognized across cultures and are used to express value priorities.

Organizational Commitment

Researchers today recognize many objects or foci of commitment in the workplace—the organization, the workgroup, the union, the occupation, the specific job, and even work itself (Cohen 2007). While all these foci (except perhaps the workgroup) are legitimate targets for study among teachers, the current study deals with organizational commitment, where the organization is defined as the school. We made this decision because we believe that the school is the key object of commitment for teachers, at least in the Israeli system, where teachers’ unions operate nationally and are therefore, for most teachers, somewhat abstract. Our thinking here relies on the logic of proximity (Gregersen 1993; Mueller and Lawler 1999). Proximal variables are thought to exert a greater influence on employees because proximity provides more opportunities for exchange relationships—an idea that will be discussed later in this paper.

Arguing that OC can be better understood as a multidimensional concept, Meyer and Allen (1984) proposed a two-dimensional model of OC. They called their first dimension affective commitment, defined as “positive feelings of identification with, attachment to, and involvement in, the work organization” (Meyer and Allen, p. 375). They termed the second continuance commitment, defined as “the extent to which employees feel committed to their organizations by virtue of the costs that they feel are associated with leaving (e.g., investments or lack of attractive alternatives)” (p. 375). Later, Allen and Meyer (1990) added a third dimension, normative commitment, defined as employees’ feelings of obligation to remain with the organization. The three-component model has dominated the study of organizational commitment, as demonstrated in the extensive meta-analysis conducted by Meyer et al. (2002).

The Setting

The Values Setting

Israel, originally an ascetic, collectivistic, closed, and relatively homogeneous society, has over the six decades since its founding become materialistic, individualistic, open and pluralist (Sagie and Weisberg 2001). Collectivism was a dominant value in Israeli society and, naturally, in the workforce in the early years of the state, when the Labor party was dominant and socialism was its leading ideology. Israel’s mandatory military service, with its emphasis on group cohesion, reinforced the country’s collectivist ideology. A major shift in values occurred in the last quarter-century of the 1900s, alongside structural changes in the Israeli workforce, including increased privatization and a decline in the size and role of the General Federation of Labor (GFL)—Israel’s national trade union, formerly one of the country’s most powerful institutions and even one of its largest employers. Globalization, the strong effect of American culture, and the collapse of communism all contributed to a reduced level of collectivism in Israeli culture, and toward individual achievement and better quality of life as legitimate personal goals. While this shift appeared sudden at the time, the collectivist ethos was in fact already in decline when Hofstede (1980) collected his data: Hofstede found that Israeli culture at the time leaned slightly more towards individualism than collectivism.

The dwindling significance of collectivist values was evidenced by the diminishing popularity of the kibbutz (Sagie and Koslowsky 2000). The traditional kibbutz was a collective community whose members handled many services locally and held most property in common. It was, at one time, a successful Israeli fulfillment of the communist ideal. Clear erosion of the participative principles behind the kibbutz occurred during the 1980s and 1990s. The transformation involved privatization of shared facilities and the introduction of a differential wage system to members (Sagie and Weisberg 2001). Economic difficulties and an ideological crisis, as well as a significant reduction in government financial support by the non-Labor Likud party, were blamed for the retreat from egalitarian values.

The shift in values has had political and economic aspects as well. The socialist ideology that previously led the government to an over-involvement in Israel’s economy has been replaced in the past two decades by a more dynamic and competitive market-economy approach (Sagie and Weisberg 2001). As part of this trend, significant state-owned and GFL-owned companies have been privatized or ceased to operate. Currently, the decline of collectivism in Israel has had a decisive influence on the tendency of workers to join trade unions, particularly in the private sector. Employees in the private sector are today inclined to adopt free-market norms of competition, including employment contracts with no trade-union protection. However, it should be noted that this shift has been less sharp in the public sector. A measurably smaller decline in the unionization rate has been observed in the public compared with the private sector (Sagie and Weisberg 2001).

The Educational Setting

Israel’s 7 million citizens include 1.5 million students in the educational system. Israeli schools are divided into four tracks: state, state-religious, Orthodox religious (Haredi), and Arab. Regular state schools offer a secular education with a minimum of religious content; most Israeli children attend such schools. State-religious schools, catering to youngsters from the Orthodox Jewish sector, offer intensive Jewish studies programs as well as secular studies. The Haredi schools operate independently but must adhere to a core curriculum determined by the Ministry of Education to receive funding (Education in Israel 2009). Schools in the Arab sector teach in Arabic and offer a curriculum that emphasizes Arab history, religion, and culture. This study focuses on the secular state educational system.

The education system consists of three tiers: primary education (grades 1–6, approximately ages 6–12), middle school (grades 7–9, ages 12–15), and high school (grades 10–12, ages 15–18). Compulsory education is from kindergarten through 10th grade, though most students complete the 12th grade.

Despite Israel’s top-quality education system, government budget cuts and underpaid teachers have taken their toll. Israel was amongst the top-ranked nations internationally in science and mathematics in the 1960s, but dropped to 33 out of 41 nations in the 2002 Organization for Economic Cooperation and Development (OECD) survey (Education in Israel 2009). Wages for Israeli teachers are low compared to other industrialized countries, according to the OECD, and many teachers have opted to leave education for better-paying jobs or work abroad, causing a “brain drain” (Education in Israel 2009).

The Reform

To combat these problems, in 2007 Israel’s Ministry of Education advanced a reform program for elementary- and secondary-school teachers. The main features of the reform (entitled “New Horizon,” or “Ofek Hadash” in Hebrew) include a new salary ladder that raised salaries for new teachers and allowed veteran teachers to earn more for working additional hours; greater autonomy and authority for principals, who would be paid according to a separate salary ladder; and a redefinition of working hours to include meetings, extra classes, and time for preparation and marking papers, as long as teachers performed these activities on the school grounds and clocked in. In addition, the reform changed the promotion ladder to limit the numbers in the higher ranks (Israel Ministry of Education 2009).

The reform was based on a collective agreement between the union of elementary-school teachers and the government signed in 2007. (It should be noted that elementary- and secondary-school teachers in Israel belong to different unions, with more than one union at some levels; teachers can move from one union to another.) To date the reform has been implemented in some schools, including many elementary schools, but is far from universally accepted. First, the main high school teachers union refused to take part in the reform and did not sign the agreement. The reform has therefore not been implemented in any high school and many middle schools. Moreover, even in schools which have adopted the reform, individual teachers have the option of opting out, and many have done so. In short, the teachers’ perspective toward the reform is one of ambivalence ranging through resistance, as reported in the Israeli media (Kashti 2008; Zelikovitz 2009), with some teachers even arguing that the reform not only fails to improve their working conditions and salaries, but even worsens them. In its early stages of implementation, the reform thus faces an uncertain future and is far from being seen as a ‘new horizon’.

The reform described here fits Ferlie et al. (1996) description of how change is often managed in the public sector, defined by them as “top-down radical shock strategies and the exercise of political clout” (p. 86). This is what public sector management theorists call the strategic management approach to change, where strategic change is achieved by way of top-down implementation after the content of the new strategy has been formulated (Sminia and Nistelrooij 2006). We framed our research questions and hypotheses taking this top-down implementation into account. Our goal was to examine and compare values and commitment among teachers who joined the reform plan and those who did not.

Research Hypotheses

Values and Participation in the Reform

Because employees typically are not given the option of choosing whether to participate in organizational change (at least without the risk of ramifications), very few studies have examined the relationship between values and the decision to participate in a change. Moreover, studies that have looked at values and change have tended to focus on values at the level of the organization, not the individual (Elias 2009; Kabanoff et al. 1995). Given the diversity of personalities found in members of any occupation, it is not surprising that people respond differently to organizational change in general, and to the opportunity to accept or reject some given change in particular (Armenakis and Bedeian 1999; Elias 2009). Some personal differences will be reflected in variations in the values held by the teachers in this study. The first question we seek to answer, therefore, is how the values in the different dimensions differ in magnitude between individuals who chose to participate in the change and those who did not.

In a participative change process, we would expect that those who choose to participate would score higher in values associated with openness to change. However, the reform here was a top-down enterprise, and those who chose to participate in the change had very little to do with initiating or designing it. The reform was initiated by the Israeli Ministry of Education, and its terms and conditions were determined in negotiations with the leadership of the country’s elementary-school teachers union. The reform came into being as a collective agreement in which individual teachers had little say.

Following the above arguments, we can expect teachers who joined the reform to place greater weight on values associated with conservation and self-transcendence. These values suggest a sense of belonging to the collective (self-transcendence) and a focus on stability and submission to the system, union, and government (conservation).

Hypothesis 1a

Those who chose to participate in the reform—which was a non-participative change process—will have higher levels of values associated with conservation.

Hypothesis 1b

Those who chose to participate in the reform—a non-participative change process—will have higher levels of values associated with self-transcendence.

Commitment and Participation in the Reform

Resistance to organizational change can be prompted by several factors, including fears that the change will mean new job requirements, less economic security, disrupted social arrangements, or lower status, along with a general sense of psychological threat (Yousef 2000). Naturally, employees’ commitment to the organization may influence how they react to change; equally so, the prospect of change (and the fears that go along with it) can affect employees’ commitment. As Fedor et al. (2006) pointed out, an individual’s sense of commitment under conditions of change is likely to be shaped by three factors: how fairly the change is carried out; how favorable its outcomes are likely to be; and job-level change, namely the extent to which individuals’ jobs were impacted by the change. The first dimension, fairness, involves the manner in which management treats and involves employees during change, such as providing advance notice of new requirements, routines or procedures and showing respect for affected individuals. Fairness has been shown to be a powerful determinant of individuals’ reactions to organizational change (Fedor et al. 2006), with greater fairness positively affecting employees’ commitment. The second dimension is the outcome valence of the change. Change that contributes to a “better life” for members of the work unit may lead individuals to conclude that the organization holds values and goals consistent with their own, and is acting in their best interests. Therefore, changes whose outcomes are favorable to the unit’s members should be positively related to commitment (Fedor et al. 2006).

The third dimension, job-level change, produced the most interesting finding of the Fedor et al. (2006) study, according to the authors. The least positive response in organizational commitment was reported for changes that were seen as favorable for the unit, but that were characterized by low degrees of work unit change and high degrees of individual job-level change. In other words, it appears—according to Fedor et al.—that individuals respond poorly to changes in which they perceive themselves to be carrying a disproportionate adaptation burden relative to others in their work units. This finding suggests that adaptation demands may be the critical factor in determining individual change reactions. Even “good” change is not always good for certain employees if they need to do most of the adjusting. This issue deserves additional research attention to more fully explicate the degree to which the positivity or negativity of outcomes versus adaptation demands drives attitudes and other change reactions.

The 2007 educational reform in Israel was not unambiguously fair or positive in its expected outcomes. Teachers saw the change as imposed from above (Kashti 2008; Zelikovitz 2009); they were not convinced they would benefit from the change; a major union did not join the reform. Looking at the three-component model of commitment, it might be expected that those who joined the reform would show greater levels of normative commitment. The logic is similar to that presented above for the proposed association between joining the reform and values associated with conservation and self-transcendence. Normative commitment “reflects a sense of obligation on the part of the employee to maintain membership in the organization” (Meyer and Smith 2000, p. 320)—or in this case, to “go with the flow,” following the lead of the reformist union and the employer (formally the education ministry, represented by the principal of the school).

Meyer and Allen (1997) suggested that normative commitment develops through socialization when individuals internalize a set of norms concerning appropriate behavior. Meyer and Allen (1997) specifically wrote that “… [the] presumed process here is one of internalization. Through complex processes involving both conditioning (rewards and punishments) and modeling (observation and imitation of others), individuals learn what is valued and what is expected of them by the family, the culture, or the organization. In the case of normative commitment what gets internalized is the belief about the appropriateness of being loyal to one’s organization” (p. 61).

Those who joined the reform did so in order to conform—an attitude reflected in normative commitment. In contrast, we can expect little difference in affective or continuance commitment between the pro- and anti-reform camps. Continuance commitment should play no role at all here, as the reform has no implications for teachers’ job security. As for affective commitment, is true that the positive relationship found between normative and affective commitment in more than one study suggests that affective commitment might have some relationship to the decision to participate in the reform. Yet we cannot expect as strong a relationship as for normative commitment, given that (1) teachers on both sides of the line are likely to feel ambivalent about aspects of the reform, and (2) too many confounding factors are likely to come into play (friendships and solidarity within the school, how well the teachers and principal work together, and the like). Thus,

Hypothesis 2

Those who chose to participate in the reform will demonstrate higher levels of normative organizational commitment in comparison to those who did not.

A Comparison of Values and Commitment in their Effect On Participation

Our final hypothesis expects that organizational commitment will have a stronger effect on participation in the change than individual values, based on the logic of proximity. Gregersen (1993) and Mueller and Lawler (1999) argued that proximal variables exert the most significant influence on employees’ actions because proximity provides more opportunities for exchange relationships. Recall that, as discussed above, we can expect teachers who joined the reform to emphasize values associated with conservation (submission of the self to the system or establishment) and self-transcendence (submission of the self to the collective). In the workplace context, the organization—the object of organizational commitment—is a more proximal focus than the system or the collective—the objects of these groups of values. As a more proximal focus, the organization offers more opportunities than the system or collective for exchange relationships. The proximity argument can easily be applied to this study, such that we should expect commitment to have a stronger effect than individual values on participation in the educational reform. This expectation is also based on the argument that greater opportunities for exchange mean greater opportunities for the attainment of motivational goals, as described by Schwartz (1996).

Looking at the three specific commitment forms, we can expect normative organizational commitment to have the strongest effect on the decision to participate, as per the reasoning presented in relation to Hypothesis 2. Similarly, the rationale presented for that hypothesis also explains why continuance commitment is expected to have the weakest effect on participation in the change. Affective commitment, again, falls somewhere in the middle. Because of the ambivalence of teachers and various confounding factors (described above), affective commitment cannot be expected to have as strong an effect as normative commitment. However, affective commitment is regarded as the main form of organizational commitment (Cohen 2003) and as the main representation of the exchange relationship between employer and employees. For these reasons, and because of the strong positive relationship between affective and normative commitment (Cohen 2003; Meyer et al. 2002), we can predict that affective commitment will have some effect on the decision to participate or not in the change.

Hypothesis 3

Organizational commitment will be related significantly more strongly to participation in the change than individual values. Normative commitment will have a significant and strong effect on participation, affective commitment will have a lesser effect, and continuance commitment will have a weak or no effect on participation in the change.

Method

Subjects and Procedure

The population of this study was Israeli elementary-school teachers working in the secular state educational system. We focused on this particular group in order to minimize variations that might be caused by including members of other, culturally different populations, such as Arabs or religious Jews. All teachers in this study were members of the Israeli elementary schools teachers’ union, which, as discussed above, had developed the reform in collaboration with the Ministry of Education.

We distributed questionnaires to 642 teachers in 25 schools. Of these, 214 usable questionnaires were returned, a response rate of 33.3%. Five to 13 teachers returned usable questionnaires from each school. The data were collected during the middle of the 2008–2009 school year (December 2008 to April 2009).

The questionnaires included items on individual values, commitment forms, and demographic characteristics. The questionnaires, which were in Hebrew, were administered on-site and took about 15 min to complete; no compensation was provided. Among the respondents, 31.3% had chosen to participate in the reform, and 68.7% had not. It should be noted that in 2008 about 23,657 teachers joined the reform (Israeli Teachers Union 2008)—nearly 27% of the roughly 88,000 teachers in the secular state system (Data on the Israeli education system 2009). The percentages of all teachers and teachers in the sample who chose to join the reform are therefore quite close.

Demographic characteristics of the final sample were as follows: 92% of the respondents were female, the average age was 36.5, and the average tenure in the occupation and in the school was 14.2 years and 9.4 years, respectively. Most of the respondents, 68%, were married.

Scales

Participation in the change was measured as a dichotomous variable where 0 = not participating in the change and 1 = participating.

Individual values: The Portrait Values Questionnaire (PVQ) was applied to measure the 10 basic values (Schwartz 2005; Schwartz et al. 2001). The PVQ consists of short verbal portraits of 40 different people, gender-matched with the respondent. Each portrait highlights goals and aspirations that point implicitly to the importance of a value. The portraits describe each person in terms of what is important to him or her; thus, they capture the person’s values without explicitly identifying values as the topic of investigation. The score for the importance of each value is the average rating given to these items (6-point scale); all designated a priori as markers of a value. All the value items have demonstrated near equivalence of meaning across cultures in analyses using multi-dimensional scaling (Schwartz 2005). Following Schwartz’s model we aggregated the items into the four higher-order value types: conservation (conformity, tradition, and security; 13 items), self-transcendence (universalism and benevolence; 10 items), self-enhancement (achievement and power; 7 items), and openness to change (self-direction, stimulation, and hedonism; 10 items).

Organizational commitment was measured using Meyer et al. (1993), with six items for each of the three dimensions (affective, continuance, and normative commitment). All the commitment constructs were measured on a 7-point scale (1 = strongly disagree to 7 = strongly agree). While all items were included in the original analysis, we ultimately omitted two items from the affective organizational commitment scale in order to increase reliability: 1. “I would be very happy to spend the rest of my career with this school,” and 2. “I really feel as if this school’s problems are my own”. It should be noted that the term “the school” or “my school” was used in all organizational commitment items.

Data Analysis

Because the dependent variable was dichotomous (respondents were either participating in the reform or not), we used a nonlinear logistic regression model for the analyses (Morita et al. 1993). We chose a logistic regression for two reasons. First, logistic regression is preferable to linear probability models (multiple regressions with a dichotomous dependent variable) for uncovering the relationship between a dummy variable and several independent variables because logistic regressions constrain estimated probabilities to between 0 and 1. Second, regression techniques which use ordinal least squares are based on the assumption of constant variance of the error term (homoscedasticity). When a dummy variable is used as a dependent variable, this assumption is violated and the t and F statistics used for hypothesis testing are unreliable. Hence, some results obtained in this manner may be misleading (Aldrich and Nelson 1984).

Logistic models are not interpreted in the same way as linear models. Beta coefficients do not directly predict the value of the dependent variable, but rather the probability that the dependent variable will attain a specific value, in this case 0 (not participating) or 1 (participating). In the models that follow, raising the value of variables with significant positive coefficients increased the probability that a given respondent had chosen to participate in the change, while raising the value of those with significant negative coefficients increased the probability that a respondent had chosen not to participate. In this way the coefficients were interpreted in terms of their relationship to the comparison group, in a way roughly analogous to a linear model.

Several measures for assessing the predictive efficacy of logistic regression were applied here (Demaris 1992). First, we employed a likelihood ratio chi-square test statistic, often referred to as a model chi-square; in logistic regression, this is the analogue of the global F test. Second, we checked our classification results of the dependent variables for the percentage of cases correctly predicted. Third, we employed the Wald Chi-Square statistic, which tests the unique contribution of each predictor in the context of the other predictors and so eliminates any overlap between them. Fourth, based on Demaris (1992), we applied RL2, a measure based on the log likelihood 02. Like R2, RL2 tends to range from 0 to 1, but RL2 tends to underestimate the proportion of variation explained.

Finally, we used the odds ratio as an indicator for the relative strength of the independent variables. The odds ratio is a measure of effect size, describing the strength of association or non-independence between two data values. The odds ratio for a predictor is defined as the relative amount by which the odds of a given outcome increase (O.R. greater than 1.0) or decrease (O.R. less than 1.0) when the value of the predictor variable is increased by 1.0 units. An odds ratio greater than 1.0 means the event is more likely to happen than not, while an odds ratio less than 1.0 means the event is less likely to happen than not. For example, in the current study, an odds ratio equal to 1.286 means that there is a 1.286 times greater chance that a given respondent has chosen to participate in the reform than not. An odds ratio of .778 means that there is a .778 times greater chance that the respondent has chosen not to participate. An odds ratio of 1.0 means the two variables is independent.

Findings

Table 1 presents the basic statistics of the variables and the inter-correlations among them. Results show acceptable reliabilities of the research variables following DeVellis’s (1991) criterion of .60 and above. The correlations among the independent variables were all .60 or below and quite preclude the possibility of excessive multicollinearity.
Table 1

Descriptive statistics, reliabilities (in parentheses), and inter-correlations among research variables

Variables:

M

SD

1

2

3

4

5

6

7

8

Individual value types

1. Conservation

3.29

.66

(.74)

       

2. Self-transcendence

3.80

.65

.49***

(.75)

      

3. Openness to change

3.12

.80

.08

.32***

(.82)

     

4. Self-enhancement

2.91

.94

.18**

.13*

.46***

(.82)

    

Organizational commitment

5. Affective

5.46

1.32

.06

.19**

.12

.01

(.78)

   

6. Continuance

3.55

1.24

.00

−.19**

.02

.28***

−.06

(.67)

  

7. Normative

4.54

1.43

.19**

.20**

−.03

−.03

.55***

.14*

(.80)

 

Dependent variable

8. Participation in the change program (1 = participation)

.30

.46

.18**

.06

−.01

.01

−.08

.09

.16**

 

N = 214

p ≤ .05, ** p ≤ .01, *** p ≤ .001

To establish the discriminant validity of the scales applied here, several procedures recommended by Podsakoff et al. (2003) were applied. First, confirmatory factor analysis was performed using the AMOS structural equation modeling program. In this analysis, we compared the fit of a four-factor model for the four individual value types to the alternative fit of a single, one-factor model. The results for the four-factor model revealed the following fit indices: χ2 = 139.84 (DF = 48); χ2/df = 2.91; CFI = .92; IFI = .92; NFI = 88; and RMSEA = .09. In the second model tested, the value items were loaded onto a single factor, producing χ2 = 569.90 (DF = 54); χ2/df = 10.55; CFI = .52; IFI = .53; NFI = .50; and RMSEA = .21. The findings support the superiority of the four-factor model over the one-factor model. To further examine this contention a chi-square difference test was conducted. The chi-square difference test indicated that the four-factor model fit significantly better than the one-factor model (chi-square difference = 430.06; DF = 6; p ≤ .000).

To establish the discriminant validity of the organizational commitment scales, we performed additional confirmatory factor analyses. We compared the fit of a three-factor model for the three commitment dimensions to the alternative fit of a single, one-factor model. The results for the three-factor model (incorporating affective, continuance, and normative commitments) revealed the following fit indices: χ2 = 132.50 (DF = 24); χ2/df = 5.52; CFI = .85; IFI = .85; NFI = .82; and RMSEA = .15. In the second model tested, all commitment variables were loaded onto a single factor, producing χ2 = 280.11 (DF = 27); χ2/df = 10.37; CFI = .64; IFI = .65; NFI = .62; and RMSEA = .21. The fit indices for the three-factor model are not high, but they are much better than those for the one-factor model. We then conducted a chi-square difference test, which indicated that the three-factor model fit significantly better than the one-factor model (chi-square difference = 147.60; DF = 3; p ≤ .000). The findings support its superiority over the one-factor model.

To test for common method variance, we performed a Harman’s one-factor test (Harman 1967; Podsakoff and Organ 1986). All the 40 items of the values and organizational commitment variables were entered into a principal components factor analysis with varimax rotation. According to this technique, if a single factor emerges from the factor analysis or one “general” factor accounts for most of the variance, common method variance is deemed present. However, the results of the analysis revealed 14 factors (explaining 66% of the variance) with eigenvalues greater than one, and with only one-factor accounting for more than 10% (13.06%) of the variance. These results are consistent with the absence of common method variance. The findings showed that the respondents were able to differentiate among the different scales applied in this study.

In short, three tests were performed for a deeper examination of the scales applied here. The findings showed that the respondents were able to differentiate among the different dimensions of values and organizational commitment, and that the data are not inflated with common method errors.

Hypothesis 1 proposed that levels of conservation and self-transcendence would be higher for teachers who had joined the reform. The hypothesis was supported for conservation. One-way analysis of variance reveals a significantly higher level of conservation for the reform group (M = 3.47; SD = .65) than the non-reform group (M = 3.21; SD = .66) (F = 7.24; p ≤ .01). No significant difference was found for self-transcendence.

Hypothesis 2 proposed higher levels of normative organizational commitment for those participating in the reform. The analysis of variance supports this hypothesis. Those who joined the reform showed higher levels of normative organizational commitment (M = 4.88; SD = 1.44) in comparison to those who did not (M = 4.39; SD = 1.40) (F = 5.46; p ≤ .05).

As described above, logistic regression was employed to predict the probability that a given respondent had chosen to participate in the change program. The predictor variables were the four sets of individual values and the three organizational commitment dimensions. A test of the full model versus a model with intercept only was statistically significant, χ2 (7, N = 214) = 18.85, p = .009. The model was able to correctly classify 74.5% of the respondents. The −2 log likelihood was 242.49, and the Nagelkerke R2 was .12.

Table 2 shows the logistic regression coefficient, Wald test, and odds ratio for each of the predictors. Employing a .05 criterion of statistical significance, conservation, affective commitment, and normative commitment were significantly related to participation in the change. Hypothesis 3 predicted that the relationship between commitment and participation in the change would be stronger than the relationship between values and participation. This hypothesis was supported by the data. First, two of three commitment dimensions were related to participation in the change, in comparison to one set of values out of four. Second, the two strong odds ratios for the two commitment dimensions also support this hypothesis.
Table 2

Logistic regression predicting participation in the change from individual values and organizational commitment

Predictor:

B

Wald χ2

P

Odds ratio

Individual value types

1. Conservation

.602

4.052

.044

.548

2. Self-transcendence

−.172

.276

.599

1.188

3. Openness to change

.114

.236

.627

.893

4. Self-enhancement

−.136

.450

.502

1.146

Organizational commitment

5. Affective

−.390

6.910

.009

.677

6. Continuance

.091

.404

.525

1.095

7. Normative

.424

7.659

.006

1.528

N = 214

It should be noted that the findings of the logistic regression provide additional support for Hypotheses 1 and 2. First, the significant positive relationship between conservation and participation in the change is in accordance with the prediction of Hypothesis 1. Second, the significant positive relationship between normative commitment and participation is in accordance with Hypothesis 2. The strong negative relationship between affective commitment and participation in the change, and the strength of the odds ratio for this variable, were not expected. It should be noted that the one-way analysis of variance revealed no difference in affective commitment between those who participated in the change and those who did not.

Finally, we performed two additional analyses. First we examined tenure in the occupation and tenure in the school as control variables (each in a separate equation) and found no significant relationship between either one and the dependent variable. Second, though we did not advance hypotheses regarding interactions between the four value types and the three commitment dimensions, we tested for possible interactions between them. The findings showed that of the 12 possible interactions only one was significant, an interaction between normative commitment and self-enhancement. A plot of the interaction showed that for those with low normative commitment, higher levels of self-enhancement will increase the probability of participating in the change program.

Discussion

Despite some individual successes, change is difficult to pull off (Beer and Nohria 2000; Waldersee and Griffiths 2009). Yet the reasons for failure or success in change processes, particularly in the public sector, remain unclear (Sminia and Nistelrooij 2006). As we have noted, few studies have examined individual values and organizational commitment in relation to participation in a change program—an unsurprising fact, given that this research trend is relatively new. The examination of this relationship is thus one of the main contributions of this study. The study makes a further important contribution by looking at this relationship specifically among employees faced with a major reform, a reform that is questionable in its advantages and resisted by both many employees and at least one union. Our findings show that both individual values and organizational commitment are concepts that can increase our understanding of employees’ decision-making on whether or not to participate in a change program. The findings offer a number of insights that suggest interesting ideas and directions for future research, which will be elaborated below.

First, the findings show that values are related to the decision to participate in the change and should be considered as one of its determinants—a conclusion that is consistent with previous research. The importance of values arises from the fact that although—or perhaps, because—they are determined in the early socialization process (through family, the education system, etc.; Schwartz and Bardi 2001), they seem to affect employees’ work attitudes and behaviors in the workplace across all career stages. Naturally, participation in change, a situational concept, will also be affected by the work setting, perhaps even more than by values. But values clearly play an important role in determining attitudes at work, including attitudes toward participation in change.

In the current study, the patterns of relationships between values and participation in the reform are interesting, given the unique setting examined here. For example, conservation was positively related to participation in the change, as expected. This finding shows an important relationship between values that reflect the importance of stability, tradition, and the collective, and participation in a change process such as the one described here. In keeping with this finding, our results show no relationship between openness to change or self-enhancement and participation. These values reflect an emphasis on individual interests and a positive outlook toward change. One explanation for this finding is that the change described here was a top-down process (Sminia and Nistelrooij 2006) strongly influenced by political considerations. As such, it was more appealing to those who value conformity and obedience to authority—values reflected in conservation.

The strong positive relationship between normative organizational commitment and participation in the change provides additional support for this contention. Normative commitment is associated with conformity and a sense of general obligation to the system (Meyer and Allen 1997), so it is not surprising that in the current setting it was positively related to participation in the reform. The strong negative relationship between participation and affective organizational commitment, while not expected, offers additional support for the logic developed here. Affective commitment develops as a result of positive exchanges/experiences with the organization, and represents psychological attachment to the organization. The top-down change process described here was applied with little input or personal involvement on the part of employees. The choice to participate in the reform may represent expectations of exchange from the larger system rather than any positive exchange with the organization about which respondents were asked in the survey—that is, the school.

Any consideration of this study should keep in mind that the research was performed in a very unstable setting—namely, during a process of change that had failed to win acceptance from a large portion of those who would be affected by it. The voluntary nature of the reform afforded us the unusual opportunity of comparing employees who participated in the change with those who did not. As we have described, from the point of view of the teachers this change was a top-down process (Sminia and Nistelrooij 2006). While the reform was approved by one teachers’ union, such a collective agreement does not constitute a participative process; and in any case, only one union (of several) was involved. It is possible (though not unequivocally shown by the data) that those teachers who joined the reform did so not because they were convinced the reform would benefit them, but because they were the kind of people who tend to conform to the system.

Sminia and Nistelrooij (2006) argued that from an organizational development point of view, a bottom-up approach with the full participation and active involvement of all employees is seen as essential for ensuring the strategic reorientation actually is realized. Of course, in any really large system, a fully participative process—one in which all individuals are personally involved—is clearly impossible. But employees’ perceptions on this matter are nonetheless important. Based on their analysis of a planned change in the public sector, Sminia and Nistelrooij (2006) found that the role of top management in making employees feel involved appeared to be essential to the project’s successful outcome.

Several limitation of this research should be noted. First, the study is based on data collected from Israeli teachers, and one should be cautious before generalizing the findings here to other cultures or to other occupations. More such studies in other cultures are needed before a firm generalization of the current findings can be obtained. Second, studies that examine other commitments, such as union commitment, should also be performed in future research to examine their effects in comparison or in addition to organizational commitment. Third, we did not collect data specifically on the teachers’ perceptions regarding how favorable they thought the outcomes of the change were likely to be. As described above, such perceptions are one of the three factors cited by Fedor et al. (2006) as likely to shape an individual’s sense of commitment under conditions of change. Measuring perceptions of favorability would have allowed us to better isolate the effects of values and commitment. Fourth, there is always the possibility of common method errors when all independent variables are collected from the same source. Finally, the cross-sectional design of this study does not allow causal conclusions.

However, despite these limitations, this study highlights important findings and suggests some important directions for future research. Most important, this study examined how values and attitudes are related to participation in a large-scale change process. Most studies that have examined attitudes and values have done so in what can be termed a stable environment. By examining how values and attitudes operate in a more turbulent environment, this study provides interesting data that can assist decision makers in planning large-scale change, and can enhance our general understanding of behavior and attitudes in the workplace.

Copyright information

© Springer Science+Business Media, LLC 2010