Abstract: Background: Despite the cultural sensitivity of organizational citizenship behavior measures, researchers have been using them across cultures. This study filled this gap by properly re-validating a measure developed in Nigeria but not properly validated. It also established that the measure explained work outcomes better than measures developed in other cultures. Materials and Methods: The sample came from employees in three organizations in Lagos, Nigeria. The measure developed in Nigeria and that from advanced countries were administered on the same sample. In this way, the effectiveness of each measure was ascertained. Three statistical methodologies were used. They are principal factor analysis, multiple regression analysis and structural equation modeling technique. Two separate samples were used for the principal factor analysis and confirmatory factor analysis. Results: A 2-factor model for organizational citizenship behavior fully explained the enactment of citizenship behavior in Nigerian context. The measure developed in Nigeria, explained additional variance in outcome variables when the measures from other culture were controlled. Conclusion: The re-validation exercise showed that a 2-factor model of organizational citizenship behavior fully captured the enactment of the behavior by Nigerian sample. The measure also explained outcome variables better than other measures from other cultures. The implication is that since there are cultural effects on the measurement of the behavior, using measure from other culture will under-estimate the level of the behavior in Nigeria and policy developed from such result will be inaccurate. The current study re-validated a measure developed in Nigeria using proper statistical methodology. The use of two separate samples for the exploratory and confirmatory factor analyses, ensures that accurate factor structure was developed for the Nigerian measure. By administering the Nigerian and foreign measures on the same participants, the study established empirically that the Nigerian measures are more effective in the Nigerian context, hence policies based on the results will be more accurate and representative of what is happening locally. This result will encourage future researchers to seek to take seriously the cultural effect of any measure before using in their culture.
INTRODUCTION
Organizations that develop sustainable competitive advantage will out-perform other organizations in a constantly changing environment. Organizational survival also depends on how quickly an organization can build and sustain such competitive advantage that gives it an edge over other competitors. Highly motivated employees provide lasting competitive advantage to organizations. This is because highly motivated employees enhance individual and organizational performance through the combination of positive in-role and extra-role behaviors. Discretionary, extra-role behaviors enable employees to perform beyond their stipulated responsibilities to unleash their total potentials. Organizations that rely on only the prescribed in-role behaviors will not be effective in today’s rapid changing operational environment1,2. The importance of extra-role performance in organizational performance is highlighted by the comment of Katz1 that organizations which depend on only in-role performance are very fragile social system. Consequently, the importance of combining positive in-role and extra-role behaviors cannot be over-emphasized. The combination of in-role and extra-role behaviors will unleash the whole organizational capabilities possessed by the entire workforce. For example, mentoring and willingness to document knowledge for use by future organizational members are two extra-role behaviors that enhance the continuous conversion of individual tacit knowledge to explicit knowledge needed for organizational learning and productivity3.
Organizational Citizenship Behavior (OCB) has been recognized as one of the extra-role behaviors capable of harnessing the total capabilities of organizational members4. The OCB are behaviors that are not part of the organization’s prescribed key job responsibilities of employees but are necessary to enable organizations build competitive edge in human capital. They are discretionary behaviors which employees must be motivated to engage in for the good of the organization. Consequently, organizations will derive high benefits from understanding how OCB is enacted in various contexts of business and in various cultures. Many studies have been done in the developed economies on the role of OCB in organizational performance1,5-8. However, with the advent of globalization, researchers have advocated that existing studies should be re-examined using samples from other cultures9-11. As was rightly put by Bachrach et al.11, ‘National culture influences how members of groups think about what is proper, civilized behavior and influences how one acts toward strangers and colleagues, how one addresses others and how one interacts socially’. Thus, OCB is likely to be influenced by the language, norms, thoughts and values of groups. Nigeria is a high-power distance and collectivist culture12,13. The enactment and understanding of OCB will likely vary from that of developed, low power distance and individualistic cultures6,14-16. Other researchers5,6,17,18 have also advocated for more cross-cultural studies because of the social and cultural sensitivity of OCB behaviors. Specifically, Farh et al.18 stated that little is known about OCB ‘in a global context’. Cross-cultural studies have also identified that OCB could vary across cultures in its dimensions, how much of OCB is demonstrated by employees, its relationship with antecedents and outcomes and the way OCB is demonstrated Podsakoff et al.19. Of interest in the current study is the cultural variation in OCB based on the way OCB is demonstrated and its relationship with outcomes19,20. Other studies have also identified that though OCB is prevalent in most cultures, it will however, differ in the reason and ‘perceived expectancy of behavior’14,16,19-21.
A major assumption of past cross-cultural studies in OCB is that OCB scale developed in one culture is usable across cultures. Without comparing the effects of local scale and scale developed outside, it becomes impossible to empirically determine if this assumption is correct. For example, Coyne and Ong14 discovered that Malaysian sample scored higher in OCB when compared to participants from Germany and England. However, since they utilized same scale developed outside Malaysia, it was not possible to identify if the correct level of Malaysian employee’s OCB was captured bearing in mind the assertion of Podsakoff et al.19. Also problematic is the unorganized nature of OCB studies carried out in non-USA settings, such as Nigeria21-24. Two studies were identified that developed OCB models in Nigeria, prior to the study by Olowookere and Adejuwon25. Ehigie and Otukoya26 studied the effect of perception of organizational support (POS) and interpersonal fairness on OCB using workers in one government owned company. Ladebo27 developed an exploratory factor of the three dimensions of OCB used by Bettencourt et al.28 and thereafter studied the personal motives that make agricultural workers engage in OCB. These studies contain the same inadequacy identified in other cross-cultural OCB studies, since they all used OCB scales from other cultures in their analysis. The inadequacy provides justification for an empirical study to explore if the OCB developed in Nigeria will provide better variance in work outcomes. When the OCB measures from Nigeria and developed countries are administered to the same participants, the amount of variance in work and individual outcomes contributed by each can be ascertained.
Consequently, it was a welcomed development when Olowookere and Adejuwon25 developed OCB scale based on Nigerian context. The researchers identified three dimensions of OCB. These they labeled as Organizational Involvement (OI), Interpersonal Relationship (IR) and Dutifulness (D). According to Olowookere and Adejuwon25, OI corresponds to organizational citizenship behavior directed to the organization (OCBO), which is labeled in this study as OCBOL, while IR corresponds to organizational citizenship behavior directed to individuals (OCBI), also labeled in this study as OCBIL. The third dimension, dutifulness, labeled as OCBDUT, is identified as unique to the Nigerian environment. The behaviors listed in the dutifulness dimension are beneficial to both the organization and the individual and as such could not be categorized as either OCBOL or OCBIL. These behaviors represent what workers identified as OCB that contribute directly or indirectly to organizational productivity and make the organization a better place to work. For example, workers in Nigeria participate in funeral or other social ceremonies organized by their coworkers to enhance comradeship in the workplace. This action is not prescribed by organizational rule but is an extra-role behavior aimed at putting the organization in good book of people and confirming to the individual that he/she is highly valued by co-workers.
Despite the development of this scale, the validation exercise carried out by Olowookere and Adejuwon25 has many flaws which must be corrected to have a usable OCB scale. The sample size of 100 used is not adequate if they were to follow all the steps required for proper validation of a new scale29-31. The researchers carried out only Exploratory Factor Analysis (EFA) and did not attempt Confirmatory Factor Analysis (CFA). Predictive validation was based on correlation and the study omitted other recommended statistical analysis techniques needed to validate a new scale29-31.
Statistical means should have been used to determine if the dutifulness dimension contributed additional variance in outcome variables when the OCBOL and OCBIL are controlled. Bearing in mind that the dutifulness dimension contains items that fall under OCBOL or OCBIL, such statistical test is imperative. Furthermore, since there are already existing scales, the study should have compared the predictive power of the scale developed with that of existing scales using relationships already established that are not culture dependent14. In this way, the overarching need to use OCB developed in Nigeria for the Nigerian context will be empirically established.
The present study made contributions to the cross-cultural study of OCB in many ways. The study re-validated the measure developed for Nigerian culture by carrying out appropriate statistical analysis (EFA and CFA), using larger sample than was used in the first validation. The study also tested the actual uniqueness of the dutifulness dimension developed by Olowookere and Adejuwon25 to statistically prove that it is indeed a separate dimension. The OCBOL and OCBIL scales were administered on the sample along with the organizational citizenship behavior directed to organization (OCBO) and organizational citizenship behavior directed to individuals (OCBI) scales developed in other cultures. Hierarchical regression analysis was utilized in determining if additional variance in work outcome came from the OCBOL and OCBIL after controlling for the effect of OCBO and OCBI. This is the first study that administered scales developed in a culture alongside that developed in another culture. Turnover was selected as the outcome variable because it has been identified empirically as one of the individual variables that are affected by OCB14,32. Thus, the study answers the question: Are OCB scales transferable across cultures?
The concept OCB has been widely studied in organizational behavior research and has been represented in various ways. Katz1 identified two dimensions of individual performance in an organization. One of these behaviors he described as innovative and spontaneous that goes beyond the role specified by organizations. It was Organ7,4 that formally defined the behavior described by Katz1, as ‘individual behavior that is discretionary, not directly or explicitly recognized by formal reward system and that in the aggregate promotes the effective functioning of the organization’ Since the identification of this role, many versions of the dimensions of OCB have been developed using samples from America and other developed countries1,7,8. For example, Williams and Anderson8 developed two dimensions of extra-role behaviors. These are OCB directed at the organization (OCBO) and OCB directed at individuals (OCBI). Organ4,7,33 identified the OCB dimensions of altruism, conscientiousness, sportsmanship, courtesy and peacemaking, cheer leading and civic virtue. Van Dyne et al.34 identified obedience, advocacy participation, social participation and functional participation, while voice and helping dimensions were added by Van Dyne et al.35. Podsakoff et al.19 organized the identified 30 dimensions of OCB into seven broad categories of helping behavior, sportsmanship, organizational loyalty, organizational compliance, individual initiative, civic virtue and self-development. Despite the organization of the dimensions into manageable numbers. Podsakoff et al.19 stated that many of the dimensions of OCB overlap conceptually. For example, the helping dimension encompasses the altruism, courtesy, peacemaking and cheer leading dimensions developed by Organ4,7,33. The current study adopted the dimensions stated by Williams and Anderson8. The importance of OCB has been itemized as enhancing coworker and managerial productivity, freeing resources to be used in other important areas of operation and improving organizational ability to adapt to changes in its operational environment36.
Though, there exist many studies involving OCB, majority of the studies were aimed at establishing significant relationships between OCB and other constructs and only few were aimed at establishing the validity of the OCB concept35. The consequence of this lopsided study of the construct is the confusion as to the dimensionality of the construct. Since there is no consensus yet on the dimensions of OCB, organizational researchers have made recommendations on how best to utilize the identified dimensions of OCB. The suggestions made include using the dimensions of OCB that are relevant to the sample under study27, use as many of the identified dimensions in a study7,34, using typology of OCB namely OCB directed towards individuals (OCBI) and directed towards organization (OCBO)8, job dedication and interpersonal facilitation37 and affiliative prosocial and challenging prosocial behaviors34. The modern trend in the use of the dimensions of OCB is to differentiate OCB based on who benefits: individual or organization and relevant to the sample under consideration8,38. Smith et al.39 found out that OCBO and OCBI accounted for high variance than most of the identified dimensions of OCB.
Another area of concern in the development of OCB is the use of OCB scale developed in USA in a non-USA context in total disregard of the cultural sensitivity of OCB recognized by Ladebo27 and Liu et al.40. To remedy this anomaly, Al-Sharafi and Rajiani5 suggested that OCB should be developed in non-USA contexts and the results used accordingly. Two of the studies in Nigeria, Ehigie and Otukoya26 and Ladebo27 utilized OCB scales developed in the USA setting for their studies. Ladebo27 carried out exploratory factor analysis and identified three factors labeled as loyalty, employee participation and conscientiousness. The only study that has considered the suggestion to develop a non-USA context OCB scale is that by Olowookere and Adejuwon25 using Nigerian sample. This study concluded that OCB in Nigeria has three dimensions, namely extra-role behavior directed to the organization and the individual and a third dimension that contains behaviors identified as extra-role but cannot be categorized exclusively as benefiting the organization or the individual. The dimensions were named as organizational involvement (corresponding to OCBO) and will be designated as OCBOL in this study, interpersonal relationship (corresponding to OCBI) and designated as OCBIL in this study and dutifulness which is unique to the Nigerian context and designated as OCBDUT in this study.
The OCB is discretionary behavior not covered by organizational job requirement. Since they are discretionary, they cannot be rewarded using the formal organizational reward process. The OCB is either a reactive or a proactive behavior. It is reactive when it is the result of favorable organizational climate that encourages individuals to go outside the prescribed organizational role to benefit the organization or the individual. Using the reactive concept, positive relationship has been found among OCB, job attitudes and work climates19,39,41. The OCB can also be conceptualized as proactive behavior, which employees perceive as investment which will be noticed by the organization and the individual rewarded accordingly. This approach is generally used to explain the relationship between OCB and its organizational outcomes. Using the functional approach of human behavior, Rioux and Penner2 adduced that individuals engage in OCB due to personal needs and aspirations. If their expectation is met, employees will be favorably disposed to the organization. If not, the employees will lower their organizational citizenship behavior and distance themselves from the organization. Thus, the level of OCB will imply how an employee distances self from the organization. The lower the level of OCB, the greater the distance and this may lead to greater propensity to leave the organization by way of voluntary turnover. Evidence of a negative relationship between actual turnover and OCB was found by Chen et al.32. Turnover intention has direct effect on actual turnover42. Consequently, based on the study of Chen et al.32, OCB scale derived from developed world will have negative relationship with turnover intention. Since this relationship is not culturally dependent14, the following hypotheses are stated:
| • | Hypothesis 1: The OCBOL dimension developed in Nigeria is negatively related to turnover intention |
| • | Hypothesis 2: The OCBIL dimension developed in Nigeria is negatively related to turnover intention |
| • | Hypothesis 3: The OCBDUT dimension developed in Nigeria is negatively related to turnover intention |
Olowookere and Adejuwon25 stated that the dutifulness dimension is different from the OCBIL and OCBOL dimensions. To ascertain the accuracy of this statement, the following hypothesis is tested:
| • | Hypothesis 4: The OCBDUT will contribute additional variance in turnover intention after controlling for the effects of OCBOL and OCBIL dimensions |
To properly revalidate OCBOL and OCBIL, it is necessary to demonstrate empirically the distinct nature of these dimensions when compared with OCBO and OCBI. If OCBOL and OCBIL predict the organizational citizenship behavior of Nigerian employees better, they should account for additional variance in the value of turnover intention after controlling for OCBO and OCBI. Thus, the following hypotheses are postulated:
| • | Hypothesis 5: OCBOL will explain additional variance in turnover intention when OCBO is controlled for |
| • | Hypothesis 6: OCBIL will explain additional variance in turnover intention when OCBI is controlled for |
According to DeVellis29, new scale validation should pass through very rigorous process which includes scale development, factor analysis and statistical validation. Olowookere and Adejuwon25 did very detail scale development to obtain the items in the three dimensions of OCB for the Nigerian context. However, they carried out only exploratory factor analysis. Exploratory factor analysis determines the number of latent variables that underline the items in the developed scale. It also yields Cronbach’s alpha, which explains the variance shared by the items in each latent variable. Confirmatory factor analysis is used to determine the best factor structure that explains the variance in the items of the scale43. This is carried out using advanced statistical methods such as analysis of moments of structures (AMOS). Confirmatory factor analysis allows the comparison of various factor structures that explain the variance in the items of the scale to determine the best structure to be used for representing the items. For example, confirmatory factor analysis may determine that two or three factor structure fit the items in the scale; factor comparison will determine the superiority of one factor structure over the other. Another aspect of validation of the new scale omitted by Olowookere and Adejuwon25 is the predictive validity of the new scale. The relationship between the OCB developed in the developed world and turnover intention does not depend on the context. Consequently, regression analysis is used to ascertain if the new scale predicts turnover intention of Nigerian workers.
MATERIALS AND METHODS
Participants and procedures: Fieldwork was undertaken using questionnaires to obtain data through random sampling technique. Three organizations in the Lagos metropolis agreed to participate in the study. A research assistant worked with a designated individual from the organizations to select random sample of the workers in each organization. The selected participants got the forms directly from the research assistant and submitted the filled questionnaire to the assistant in a sealed envelope provided to each participant. The questionnaire contains a description of the purpose of the study and assurance that information provided would be used strictly for research. The participants were assured that confidentiality of the information provided will be maintained since the organization agreed not to interfere in the process. The sample size was selected in accordance with the requirement of structural equation modeling technique which is the statistical method used for analyzing the data collected44,45. Based on the recommendation of this author, the sample size was determined as 500. The size was rounded up to 600 used for the field work. The employees in the Lagos office of the participating companies are between 1000 and 1500. It was decided to randomly select 400 employees from each organization and questionnaires sent directly to them. The total filled questionnaires received back was 520 and after removing the questionnaires with missing data, the usable questionnaires were 500 (42%). Majority of the participants were between 31 and 40 years (59.5%), 20.9% between 41 and 50 years, 10.6% above 50 years and 9% below 30 years. Over half (65.25%) of the participants had worked in the organizations for more than 15 years, while the rest had worked less than 15 years. Majority of the participants were male (55.5%).
The measures for all the study variables were adapted from existing studies. The items for the OCBOL, OCBIL and OCBDUT were measured using Likert’s scale from 1 (Strongly disagree) to 7 (Strongly agree), while the turnover intention, OCBO and OCBI were measure with Likert’s scale from 1 (Never) to 7 (Always).
The items for OCBOL, OCBIL and OCBDUT dimensions were taken from the study of Olowookere and Adejuwon25. The OCBOL measure contains 9 items. A sample measure is ‘I assist new employees to settle on the job’. The OCBIL measure contains 12 items with sample item as ‘I adjust my work schedule to favor my co-workers’. The OCBDUT measure contains 9 items and sample items are ‘I volunteer to work on assignment during my free time’ and I help co-workers with logistic during official events or presentations’. The Cronbach alpha obtained by Olowookere and Adejuwon25for OCBOL, OCBIL and OCBDUT dimensions are 0.86, 0.75 and 0.74, respectively. The values obtained in the current study are 0.87, 0.82 and 0.77, respectively.
The items for OCBO and OCBI were taken from the study of Lee and Allen46 and each measure contains 8 items. Sample item for OCBO is ‘Keep up with developments in the organization’ while that of OCBI is ‘Willing to give your time to help others who have work-related problems’. The Cronbach alpha for OCBO and OCBI reported by Lee and Allen46 are 0.88 and 0.83 respectively. The current study reported values of 0.81 and 0.72, respectively.
Turnover intention was measured with 3 items scale taken from the study of Aryee et al.47. A sample item for this scale is ‘I think a lot about leaving my organization’. The researchers did not report any Cronbach alpha but the current study had value of 0.77.
Statistical analysis: Prior to conducting factor analysis, the sample was randomly divided into two portions. One portion was used to conduct the exploratory factor analysis, while the other was used to conduct the confirmatory factor analysis and the regression analysis. Exploratory factor analysis was performed using the SPSS 20 software. Confirmatory factor analysis was done using analysis of moments of structures (AMOS) from SPSS. To properly validate the dimensions of the OCB developed in Nigeria, various factor structures were tested. The comparison of the superiority of the various models tested was gauged using the criteria developed by Bentler45 and McDonald and Marsh48. Goodness-of-fit index (GFI) and comparative-fit index must have values close to 0.95 and root mean square error of approximation (RMSEA) must have values less than 0.08 to indicate good estimation of the model. Hypotheses 1-6 were tested using hierarchical regression analysis.
RESULTS
The mean, Cronbach alpha and correlations between study variables are contained in Table 1. Cronbach alpha for the scales are 0.71 (turnover intention), 0.87 (OCBOL), 0.77 (OCBDUT), 0.82 (OCBIL), 0.81 (OCBO) and 0.72 (OCBI). Except OCB, all the other variables correlated significantly with turnover intention. The correlations values in Table 1 indicate proof of the superiority of the OCB scales developed in Nigeria.
Exploratory and confirmatory factor analysis: Initial rotated exploratory factor analysis extracted three factors with some items for OCBDUT loading in either OCBOL or OCBIL. These items were removed and the factor analysis repeated. Three factors were retained and 56% of the variance in the variables extracted. The final exploratory factor analysis indicated the OCBOL has 6 items; OCBIL has 9 items, while OCBDUT has 6 items (Table 2). The correlation between OCBDUT and OCBOL is 0.70, while that between OCBDUT and OCBIL is 0.51.
| Table 1: | Scale, means, SDs, alphas and intercorrelation for the total sample (N = 500) |
**Correlation is significant at the 0.01 level, *Correlation is significant at the 0.05 level, #Alpha coefficients, TIN: Turnover intention, OCBOL: Organizational citizenship behavior directed at organization local, OCBIL: Organizational citizenship behavior directed at individual local, OCBDUT: Organizational citizenship behavior | |
| Table 2: | Three factor exploratory factor analysis (N = 250) |
OCBOL: Organizational citizenship behavior directed at organization local, OCBIL: Organizational citizenship behavior directed at individual local, OCBDUT: Organizational citizenship behavior dutifulness | |
| Table 3: | Confirmatory factor analysis (N = 250) |
df: Degree of freedom, p: Probability, GFI: Goodness-of-fit index, CFI: Comparative fit index, RMSEA: Root-mean-square-error of approximation, OCBOL: Organizational citizenship behavior directed at organization local, OCBIL: Organizational citizenship behavior directed at individual local, OCBDUT: Organizational citizenship behavior dutifulness | |
Evidence of poor discriminant validity for OCBDUT is shown in the closeness of the variance extracted (0.5) and the square of the correlation it has with OCBOL (0.49)49. The retained items for the various dimensions were subjected to series of confirmatory factor analysis to accurately determine the discriminant validity of the dimensions. Based on the high intercorrelation OCBDUT dimension has with the other dimensions, various models representing combination of OCBDUT and other dimensions were tested. From Table 3, model 2 and 4 had very close estimation indices. However, chi-square difference test indicated that model 2 which combined OCBOL and OCBDUT items in one factor is better. The 1-factor (combining all dimensions in single factor) and 3-factor (OCBOL, OCBOI and OCBDUT as independent factors) models showed very poor model fit (models 5 and 1). Also, a 2-factor model which combined OCBDUT and OCBIL (model 3) also showed poor model fit.
Regression analysis: When OCBOL, OCBIL and OCBDUT were put jointly into a single regressed equation, the dimensions OCBOL and OCBIL explained 13.2 and 2.2% in the variance of turnover respectively. However, the variance explained by OCBDUT was not significant (Table 4). The results indicate support for hypotheses 1 and 2 and rejection of hypothesis 3. To find out which of the two dimensions made OCBDUT redundant, further analysis was done. The OCBDUT was regressed alone and significant variance was obtained. When OCBIL was controlled for, the variance explained by OCBDUT was significant (Table 5). When OCBOL was controlled for, the variance explained by OCBDUT was not significant (Table 6). Thus, in a model containing OCBOL and OCBDUT, the latter dimension became redundant.
When OCBO was controlled for, OCBOL contributed additional variance to the value of turnover intention (Table 7). Similarly, when OCBI was controlled for, OCBIL contributed additional variance to the value of turnover intention (Table 8). Thus, hypotheses 5 and 6 are supported.
| Table 4: | Hierarchical regression of OCBOL, OCBIL and OCBDUT on turnover intention |
**p<0.001, aMale: 1 and female: 2, bA higher value reflects older age group, cA higher value reflects longer tenure, OCBOL: Organizational citizenship behavior directed at organization local, OCBIL: Organizational citizenship behavior directed at individual local, OCBDUT: Organizational citizenship behavior dutifulness | |
| Table 5: | Hierarchical regression analysis of OCBDUT on turnover intention when the effect of OCBIL is controlled |
**p<0.001, OCBOL: Organizational citizenship behavior directed at organization local, OCBIL: Organizational citizenship behavior directed at individual local, OCBDUT: Organizational citizenship behavior dutifulness | |
| Table 6: | Hierarchical regression analysis of OCBDUT on turnover intention when the effect of OCBOL is controlled |
**p<0.001, OCBOL: Organizational citizenship behavior directed at organization local, OCBDUT: Organizational citizenship behavior dutifulness | |
DISCUSSION
The suggestions of possible cultural sensitivity of OCB scales entail that OCB scale should be developed in various cultures. However, for the research to produce meaningful results, it must be planned in such a way as to answer some pertinent questions.
| Table 7: | Hierarchical regression analysis of OCBOL on turnover intention when the effect of OCBO is controlled |
**p<0.001, OCBOL: Organizational citizenship behavior directed at organization local, OCBO: Organizational citizenship behavior directed at organization | |
| Table 8: | Hierarchical regression analysis of OCBIL on turnover intention when the effect of OCBI is controlled |
*p<0.05, **p<0.001, OCBIL: Organizational citizenship behavior directed at individual local, OCBI: Organizational citizenship behavior directed at individual | |
One of the questions is how effective are OCB scales across culture? Proper answer can only be obtained when the various scales are validated using exploratory and confirmatory factor analysis and determining how far OCB scales predict known work and individual outcomes. This study is unique in that it advanced the usability of OCB scale developed by Olowookere and Adejuwon25 in the following ways:
| • | It utilized multiple statistical method of analysis in the validation exercise to remove all forms of redundancies in the newly developed scale (OCBOL, OCBIL and OCBDUT) |
| • | Increasing the statistical accuracy of the validation exercise by utilizing different samples for the exploratory and confirmatory factor analysis |
| • | Administered the newly developed OCBOL, OCBIL and OCBDUT scales and the OCBO and OCBI scales on the same participants to ascertain the effectiveness of each scale in predicting work outcomes |
The last improvement is important for two reasons. Firstly, past studies assumed that scales are equally effective across cultures and so never made efforts to develop local scales of OCB14,50. Secondly, going by the above assumption, previous studies have not been able to understand the variation of OCB across culture advocated by Blakely et al.51. The results obtained in this study support the notion that developing scales of OCB across cultures will enhance the understanding of OCB across cultures.
The high correlation OCBDUT had with OCBOL and OCBIL is an indication that OCBDUT scale may be redundant. When 3-factor structure was stipulated, exploratory factor analysis yielded three distinct factors. However, when 2-factor structure was stipulated, all the items of the OCBDUT factor loaded on the OCBOL dimension. The Cronbach alpha of the combined factor is similar to that of the single OCBOL factor. This is the initial indication that the OCBDUT dimension may be redundant when OCBOL factor is in the same model. McNeely and Meglino38 and Williams and Anderson8 agreed that the modern trend in OCB study is to differentiate it by those directed at the organization and those directed at individuals. The two dimensions OCBO and OCBI, accounted for high variance than other dimensions39. Hence the finding of the exploratory factor analysis is not surprising.
Table 3 indicates that 2-factor models in numbers 2 and 4 both have acceptable fit indices. Model 2 which is a combination of items in OCBDUT and OCBIL dimensions in a factor and OCBIL items in another factor, is superior based on values of the indices and the chi-square difference test. Confirmatory factor analysis produces a better factorial structure than exploratory factor analysis43. Thus, the results of the confirmatory factor analysis confirm what was observed in the exploratory factor analysis. When regressed separately on turnover intention, the three dimensions had significant relationship with turnover, though the variance explained by OCBDUT was low. The variance explained by OCBDUT was not significant when the dimensions were put in the same step in the regression equation. Hierarchical regression identified that OCBDUT was redundant only when OCBOL was controlled for. The items for OCBDUT fit into items that benefit the organization. Thus, taken together the results of the exploratory, confirmatory factor analysis and hierarchical regression analysis indicate that OCBDUT dimension is redundant when the OCBOL dimension is controlled.
The current study is the only study that administered a locally developed scale along with the scale from outside the culture. Other studies have always assumed that OCB measures were effective across culture and concentrated on finding out the level of enactment of OCB behaviours across culture14,50. This assumption is contrary to the stipulation of Blakely et al.51 that understanding of OCB will vary across culture and called for developing scales unique to cultures. Hierarchical regression analysis (Table 7, 8) indicate that locally developed scales (OCBOL and OCBIL) contributed additional variance when the scales from another culture were controlled for. Thus, the locally developed OCB scales are better in measuring the level of organizational citizenship behavior enacted in the local community. Using OCB scales from other cultures to measure the OCB enactment by Nigerian employees will grossly represent the level of OCB and affect other work variables that are outcomes of OCB.
Theoretical and practical implications: Theoretically, the study made valuable contributions to the cross-cultural study of OCB. It demonstrated that a 2-factor structure of OCB described the enactment of OCB by employees in Nigerian context as was also demonstrated in other cultures. However, the study demonstrated that the scales developed in Nigeria captured the OCB level of employees in Nigeria and made better contribution to the variance in turnover intention when compared to OCB scales from other cultures. Thus, utilizing OCB scales developed outside Nigeria to measure the OCB level for employees in Nigeria will underestimate the OCB enactment of the employees and give false impression of individual and organizational outcomes. Practically, the study provides usable OCB scales to all researchers who are interested in determining the OCB levels of employees and how such behaviors affect work related outcomes in Nigeria and similar cultures. The HRM experts now have tool that will provide valuable information in their attempt to understand the enactment of OCB in Nigeria and how such behaviors affect important work outcomes.
Limitation and direction for future study: Since this is the first time such cross-cultural study on OCB is done, future studies should build on the results obtained by replicating the study in other cultures. This is necessary to ascertain the cultural effect on OCB. A more sophisticated statistical methodology should be used in future studies to build a comprehensive model that includes the antecedents and outcomes of OCB. The study did not measure culture in the questionnaire used but assumed the conclusions drawn by Hofstede12,13. Future studies should attempt to obtain cultural dimension values from the participants. The study is based on cross-sectional data collection and so common method variance cannot be completely avoided. However, the first factor in an unrotated exploratory factor analysis extracted little variance compared to the variance extracted by the other factors. This shows that the effect of common method variance is minimal49. A further confirmation of the absence of common method variance is the poor fit of a single factor that combines all the dimensions in a CFA analysis.
CONCLUSION
The two components of OCB that reflect the enactment of the behavior in Nigeria are the OCB directed at organization and OCB directed at individuals. The third dimension, OCBDUTL, was redundant when the two factors were controlled for. The OCB scale developed in Nigeria explained the variance in turnover intention better than scales developed outside Nigeria. Hence, using such scales in Nigeria will underestimate important Organizational variables. The study is thus, an incentive for researchers to the consider cross-cultural effect of OCB scales before use.
SIGNIFICANCE STATEMENTS
| • | Because of the cultural effect on organizational citizenship behavior (OCB), a measure was developed in Nigeria but not properly validated |
| • | The effectiveness of this measure compared to those from other cultures had never been determined |
| • | A 2-factor structure for OCB was adequate for the Nigerian context and explained work outcomes of employees better than those developed elsewhere |
| • | Future researchers in OCB are to be cautious while using measures developed elsewhere especially when there are cultural differences |
| • | Efficient and effective public health service requires employees willing to do OCB, accurate measure of OCB and their effects on work outcomes and employees appropriately reward for their behaviours. These are possible only with accurate measure for OCB |
ACKNOWLEDGMENT
I am grateful to my research assistant, Ms. Sese, Ebisindou Stephanie who helped in the process of administering the questionnaires and collection of same.