Human resource management practices , employee engagement and organizational citizenship behaviours ( ocb ) in selected firms in Uganda

The purpose of this study was to investigate the relationship between human resource (HR) practices, employee engagement and organizational citizenship behaviours (OCB) in selected firms in Uganda. Adopting a pragmatic philosophy with a quantitative methods strategy using quantitative crosssectional survey design (N = 210) the researcher examined the relationship between nine antecedents, employee engagement and one outcome – OCB. Self-administered questionnaire of four scales were administered to sampled employees of soft drink and sugar firms Kampala, Mukono and Buikwe districts of Uganda. Hypotheses were tested through correlation and hierarchical regression. All the nine antecedent variables studied were significantly related to employee engagement and employee engagement was significantly related to OCB. The hierarchical regression analysis results shows that five antecedent variables demonstrated a significant relation with OCB role clarity, collaboration, job security, compensation fairness and development. This research has validated the organizational citizenship behaviour model, extended the engagement model and social exchange theory and established that all the nine antecedents studied were related to engagement but only five were significantly related to OCB.


INTRODUCTION
Researchers within human resource management (HRM) field have been concerned with how HRM can lead to improved organizational performance (Huselid, 1995;Delbridge and Keenoy, 2010).While early HRM studies tended to propose a direct link between HR practices and firm performance, recent evidence suggests that the relationship is most likely mediated by a range of attitudinal and behavioral variables at the individual levelemployee engagement, task performance, and organizational citizenship behaviours (OCB) (Guest et al., 2004;Kuvaas, 2008;Snape and Redman, 2010).Prior HRMperformance-studies tended to focus on the macro level and proposed a direct relationship between the two, but lacked convincing theoretical explanations (Guest, 2011).
Recent research has adopted a social exchange framework and quantitative methodologies at the micro level to suggest that the relationship is most likely mediated by attitudinal variables, especially employee engagement which has ushered in a very recent interest in the parallel stream of research linking employee engagement and employee outcomes.Bringing the two streams together suggests that engagement may constitute the mechanism through which HR practices impact individual and organizational outcomes within the framework of social exchange theory.
However, the findings within current stream of HRM research have so far also been largely inconclusive (Alfes, et al., 2013) but evidence has accumulated on the basis of quantitative studies to suggest that high levels of engagement are associated with high levels of performance, citizenship behaviour and individual wellbeing (Christian et al., 2011).In parallel with the development of the Strategic HRM field, researchers in psychology and social psychology have been concerned with exploring how the attitudinal construct of employee engagement could help explain individual performance outcomes.Given the shared focus of these two streams of research, it is perhaps surprising that it is only now that researchers within the HRM field are starting to recognize the relevance of engagement for their endeavours, and studies linking the two are just emerging (Truss, Alfes, Delbridge, Shantz & Soane, 2013;Shuck, Reio & Rocco, 2011;Shuck and Rocco 2013).
Most recent HRM practices-engagement-performance studies situate their analyses within the framework of social exchange theory, arguing that organizational HRM practices send overt and covert signals to employees about the extent to which they are valued and trusted, giving rise to feelings of obligation on the part of employees, who then reciprocate through high levels of performance (Allen et al., 2003;Gould-Williams, 2007;Purcell and Hutchinson, 2007).Studies of mediation often draw on social exchange theory to provide an explanatory framework.Social exchange theory is based on norms of reciprocity within social relationships (Blau, 1964;Emerson, 1976).It is argued that employees are motivated within the employment relationship to demonstrate positive attitudes and behaviors when they perceive that their employer values them and their contribution (Cropanzano et al., 2003;Kuvaas and Dysvik, 2010).
Uganda"s industrial sector, is dominated by Micro, Small and Medium enterprises (MSEs), and contributes 25% to the gross domestic product (GDP).From early 1990s government of Uganda embarked on the country"s Economic Recovery Programme, aimed at removing structural distortions and imbalances in the economy by providing an enabling environment for industry for sustainable industrial development.Despite the government"s efforts to make Uganda"s private sector more competitive, very little appears to be happening in comparison to other countries (Nyanzi, 2012).
The Global Competitiveness Report 2014/15 has ranked Uganda 122 out of 144 economies, in 2013/14.Uganda still lags behind Rwanda (62) and South Africa (56) which are the top performers in Africa -while Kenya is ranked 90 (World Economic Forum, 2014).This leads us to propose that the macro-economic policy environment is a necessary but not sufficient condition for firm performance.The solution to the performance problem lies more at how the HR practices are handled at the firm level.This is in line with researchers who argue that aggregate outcome variables, such as competitiveness, firm financial performance and effectiveness, are too distal from the micro-level HR interventions, and that more proximal outcome indicators at the individual level would provide a better and more reliable measure of individual HR outcomes (Paauwe, 2004;Purcell and Kinnie, 2007;Wright and Haggerty, 2005).
The carbonated soft drink sub-sector in Uganda has attracted a number of players in the last two years including Riham, Fizzy and Azam soda resulting into stiff competition causing the two giant franchises -Coca-Cola and Pepsi to cut prices to remain competitive in the market.Bottled water, which is emerging as the fastest growing in the beverage market has eaten into the carbonated soft drink market (New Vision, 2015: Focus on Manufacturing, June 2).The leading competitor in the bottled water industry is Rwenzori Bottling Company Ltd, a subsidiary of SABMiller.
Since there is no empirical data on outcomes, HR or individual attitudinal variables on the sampled firms, it is necessary to bridge the gap by studying the firm level HR factors (antecedents) and employee engagement hypothesized to predict outcomes like turnover intent which has a bearing on performance and competitiveness.Firms often compete by lowering prices, cutting costs, redesigning business processes and downsizing but all these have limits.New approaches to HRM are supported by numerous scholars who call for a more positive HR approaches, for example, engaging employees rather than focusing on problem-coping strategies as vital for employee retention (Luthans and Avolio, 2009;Bakker et al., 2008;Luthans and Youssef, 2007;Avey et al., 2008;Seligman et al., 2005).

Purpose of the Study
The purpose of the study is to establish the relationship between HR practices (antecedents),employee engagement and OCB in selected industry in Uganda in the context of the social exchange theory.

Hypothesis
H1: HR practices (antecedents) are positively correlated with employee engagement; H2: Employee engagement is significantly positively correlated with OCB; H3: After controlling for HR practices (antecedents), employee engagement will predict unique variance in OCB.

LITERATURE REVIEW
The classical approaches to strategic HRM imply that the role of good HR practices is to maximize the contribution of human assets in order to achieve corporate goals (Huselid, 1995).Discussion of HR outcomes suggests that good HR practices act to improve the self confidence and flexibility of the workforce and contribute to increased motivation, morale and OCB, which in turn are related to individual and business performance (Becker andHuselid, 1998, 2006;Appelbaum, 2002;Tamkin, 2004).Because no exact number of practices in the good HR practices "bundle" are agreed upon (Boxall andMacky, 2007, Delery, 1998;Becker and Huselid, 1997;MacDuffie, 1996;Thompson, 2000;Batt, 2002), this study looks at nine very common set of good HR practices (antecedents) -Job design characteristics, role clarity, material supplies, collaboration/teamwork, reward and recognition, perceived social support, compensation fairness, job security, employee development, which have been found to have impact on HR outcomes (Rich et al., 2010;Hackman and Oldham, 1980;Xanthopoulou et al., 2009;Seigts and Crim, 2006;Harter et al., 2002;Demerouti and Bakker, 2011;Kahn 1990;Rhoades and Eisenberger, 2002;Ryan and Deci, 2000) and hence to test the model in the context soft drink industry in Uganda.
Good HR practices represent the conceptual dimensions of social exchange that manifest not only the norms of exchange from an employer"s point of view, but also specify the resources of exchange between employers and employees.Certain HR practices (for example, proper job design, training, pay level, benefits level, and job security) reflect higher levels of rewards offered to employees (Shaw et al., 2009;Tsui et al., 1997).On the other hand, certain HR practices, for example, individual pay-for-performance systems, employee monitoring, and formal performance appraisals reciprocate employer expectations about employee performance levels (Shaw et al., 2009).This presents a fair exchange relationship.Thus, from an exchange theory view, good HR practices will likely increase employees" perceived obligation to the employer, elicit engagement, as well as high in-role performance (Shaw et al., 2009).
The concept of organizational citizenship behaviours (OCBs) was introduced to the research literature by Organ and Ryan (1995).OCBs are voluntary contributions Owor 3 at work that include altruistic helping behaviours, compliance with work norms and requirements, courtesy to others to ensure smooth working relationships, sportsmanship to maintain performance under adversity, and civic virtue to contribute constructively to issues that arise in the workplace.OCB has been shown to have significant relations with employee engagement (for example Saks, 2006).Schaufeli and Bakker (2004) suggest that job resources (good HR practices) are related to engagement and engagement is in turn, related to employee outcomes like turnover intention, OCB, and organizational commitment.
Previous research has demonstrated that perceptions of the environment are important in predicting work engagement (Alarcon et al., 2008;Saks, 2006).Given that the job resources are expected to predict engagement and engagement predicts outcomes, it is possible that engagement mediates the relationship between the job resources and the outcomes such as job satisfaction, turnover intentions and OCB.This is consistent with the Maslach et al. ( 2001) model and is more likely given that most of the job resources (for example, job characteristics, POS,) have been associated with various outcomes.Furthermore, several studies have found that engagement mediates the relationship between antecedent variables and outcomes (Schaufeli and Bakker, 2004;Sonnentag, 2003).

Social exchange theory
Social exchange theory (Homans, 1961;Blau, 1964;Ekeh, 1974) and employee-organization relationship frameworks (Coyle-Shapiro et al., 2004;Tsui et al., 1995) have suggested that organizations create systems that offer different forms of exchange relationships.Social exchange relationship involves unspecified obligations in which there are "favors that create diffuse future obligations, not precisely defined ones, and the nature of the return cannot be bargained but must be left to the discretion of the one who makes it" (Blau, 1964).The exchanged resources can be impersonal (such as financial) or socio-emotional such as care, respect and loyalty (Aselage and Eisenberger, 2003).It is the socioemotional resources that tend "to engender feelings of personal obligations, gratitude, and trust" (Blau, 1964).A social exchange relationship rests on the norm of reciprocity (Gouldner, 1960).
Employment relationship is seen as an exchange where the employer offers certain returns (for example, pay, benefits, and job security) in exchange for employee contributions (for example, effort, commitment, productivity) and the level of exchange depends on expectations from both sides.Schein (1970) argued that in order for individuals to generate commitment, loyalty and enthusiasm for their organization and its goals, and to obtain satisfaction from their work, there should be a An exchange starts with one party giving a benefit (an inducement) to another.If the recipient reciprocates, a series of beneficial exchanges occur and feelings of mutual obligation between the parties are created (Coyle-Shapiro and Shore, 2007).A broad notion of reciprocity encompasses a feeling of an obligation to repay favorable treatment and for an employee it includes a belief whether one should care about the organization"s well-being and should help the organization reach its goals (Eisenberger et al., 2001).Studies of engagement, like those of high performance HRM practices, draw on social exchange theory to suggest that employees will become engaged with their work when antecedents are in place that signal to them that they are valued and trusted (Rich et al., 2010;Saks, 2006).Although no research has examined the link between HRM practices and engagement, empirical studies have demonstrated a link between high levels of engagement and the same outcomes as the highperformance HRM practices literature.Engaged employees invest themselves fully in their roles (Rothbard, 2001), which may lead to the enactment of active in-role performances (Ho et al., 2011;Macey and Schneider, 2008).Engaged employees may achieve higher performance because they focus their efforts on work-related goals, are cognitively vigilant, and are emotionally and socially connected to their work (Kahn, 1990).
Combining formal HR practices and employee engagement to understand HR outcomes within the context of social exchange theory is the major task of this study.As long as managers provide a favourable work environment, employees will engage their minds, emotions and energies with positive impact on various HR outcomes.
As an alternative to studying the impact of HR practices on the usual individual performance, this study investigates the interim outcomes, OCB.

Conceptual framework
Basing on the various theories, models and concepts reviewed in the literature, a conceptual framework linking employee engagement, antecedents and OCB have been developed and tested in this study (Figure 1).Empirical studies in this research are discussed in line with the reciprocity framework of the social exchange theory (SET), which proposes that good HR practices (antecedents) influence engagement through the reciprocal exchange of engagement for motivational HR practices which then influences attitudinal and performance outcomes.The nine engagement antecedents studied in this research are job characteristics, role clarity, material adequacy, teamwork, rewards and recognition, perceived social support, compensation fairness, job security and employee development.These antecedents are assumed to separately but also collectively drive employee engagement and engagement influences OCB.

METHODOLOGY Target population
The target population was a total of 1,773 employees from the four purposively sampled firms from Kampala, Mukono, and Buikwe districts.The four firms were purposively sampled because they had well defined organization structure and human resource management systems which are considered to be appropriate for study of an emerging construct of employee engagement (Shuck, 2011;Creswell, 2003).The choice was also justified by the centrality of soft drinks to the healthcare sector, the general community and the industry"s contribution to the economy as a source of revenue and employment (UBOS, 2011).Sugar Corporation of Uganda Limited (SCOUL) was selected because it produces two important ingredients of the Soft drink industrysugar and carbondioxide.The sugar industry is a very vital extension of the soft drink industry and is similar to the sampled soft drink firms in terms of size, structure, employment and environment (UBOS, 2011).
The choice was influenced by the fact that of all the regions of Uganda, these three districts have the highest concentration of firms in Uganda.The 2010/11 census of business establishments in Uganda done by Uganda Bureau of Statistics (UBOS) revealed that 60% of the Businesses were in the central region.The Manufacturing sector, which includes Food Processing and other Manufacturing, employed an average of 4 persons per business.
Because the sampled firms have over 200 employees on average they qualify among the biggest firms in Uganda given 93% of all the businesses in Uganda employed less than 5 persons (UBOS, 2010/11).
This study using a quantitative cross-sectional survey design (N = 210) examined the relationship between nine antecedents, employee engagement and OCB.Because of the three-step nature of the study variables, a regression analysis of the quantitative data was used.The target population was 1,773 employees from four purposively sampled soft drink firms in Kampala, Mukono, and Buikwe districts and a sample of 317 different categories of employees were drawn using Krejcie and Morgan"s (1970) table.Out of the 317 questionnaires distributed, 210 usable ones were collected, giving a response rate of 66% (Table 1).
The instruments used to measure the three variables were standardized instruments used in several empirical studies elsewhere.The antecedents were determined using 58 short questions measuring 9 different antecedents.All antecedents had computed (overall scale) reliability coefficient alpha of 0.91.Employee engagement was determined using the 9-scale University of Utrecht Work engagement scales (UWES-9, -Cronbach alpha, 0.84).

Hypothesis 1
Pearson"s correlation analysis showed that employee engagement is significantly positively related to each of the nine antecedentsthe coefficients ranging from r = 0.48 to r = 0.67 (Table 2).The three dimensions of engagement (vigor, absorption and dedication) are also found to be significantly positively related to the antecedents with the correlation coefficient ranging from r = 0.37 to r= 0.65.Eight of the nine antecedents showed  (Cohen, 1988), providing empirical support for H1.Employees who scored highly on employee engagement also reported = 0.37 to r= 0.65.Eight of the nine antecedents showed large effects correlation coefficients (Cohen, 1988), providing empirical support for H1.Employees who scored highly on employee engagement also reported high levels of all the tested antecedents.

Hypothesis 2: Employee engagement and OCB
As seen in Table 3 the study showed that engagement is significantly positively correlated with OCB (r=0.61,p=0.00), hence providing empirical support for H2.The three factor measure of engagementvigor (r=0.49,p=0.00), absorption (r=0.58,p=0.00) and dedication (r=0.50, p=0.00) are also significantly positively correlated with OCB.

Hypothesis 3: Engagement antecedents, engagement and OCB
As seen in Table 4, five of the nine antecedents, role clarity (β = .21,p = .01),collaboration (β = .21,p =.01), compensation fairness (β = -.26,p = .00),job security (β = -.15,p = .03),and development (β = -.31,p = .00)contributed to the prediction of OCB (adj.R 2 = .46,p = .00).After controlling for the antecedents, employee engagement contributed unique variance in OCB (∆R 2 = .037,p = .00);thus, H3 was supported in this model.This means each of the antecedents contributed between 21 to 30% of the variance in OCB and together the five antecedents contributed 46 percent to the variation in OCB.The detailed and complicated model data in table 4 and whose beta values and significance are described above can be summarized in terms of the factors that predict OCB in the figure 2 below.
These findings suggest that OCB can be predicted by the role clarity, collaboration, compensation fairness, job security and development and employee engagement.Overall, the regression model explained 49% (adj.R 2 = .494,p =.00) of the variance in organizational citizenship behaviour (large effect size ;Cohen, 1988).

DISCUSSION
Hypothesis 1 was supported by the Pearson"s correlation analysis which showed that employee engagement is significantly positively related to each of the nine antecedentsthe coefficients ranging from r = .50to r=.67 at 1% level of significance.The three dimensions of engagement (vigor, absorption and dedication) are also found to be significantly positively related to engagement with the correlation coefficient ranging from r =.34 to r=.65.All the antecedents showed large effects correlation coefficients (Cohen, 1988).Discuss of the results of each antecedent variable and its relation to employee engagement now follows.

Job characteristics
Employees who experienced high degree of job characteristics in their jobs were more likely to be engaged (r=0.52,p=0.00).Xanthopoulou et al. (2009) found that job resources, including job autonomy, have a positive effect on daily rates of engagement among fastfood employees (n=42).A study by de Lange et al. (2008), found job autonomy, decision making and job design had potential engagement effects and indeed related to levels of engagement over time.Appraising and giving constructive feedback to employees for good performance helps maintain their motivation (Hackman and Oldham, 1980) and consequently engagement as this signals to them that the employer recognizes them and the principle of reciprocity in social exchange theory is invoked (Cropanzano and Mitchel, 2005).

Role clarity
Employees who experienced high degree of role clarity in Employees who experienced high degree of role clarity in their jobs were more likely to be engaged (r=0.55,p=0.00).Role clarity has been studied by Seigts and Crim (2006) under the idea of "convey" (communicate) where leaders clarify work-related expectations for employees.Similar research appears in Spector"s (1997) Job Satisfaction Survey and House et al. (1983) measure of Role Conflict and Ambiguity.Previous studies found role clarity to be positively related to engagement (Buckingham and Coffman, 1999;Harter et al., 2002).When employees get clear role profiles they feel obliged to respond in kind and "repay" the organization in terms of engagement, hence validating the social exchange theory.

Materials adequacy
Employees who experienced high degree of material adequacy in their jobs were more likely to be engaged (r= 0.62, p=0.00).
The findings of this study are consistent with Seigts and Crim (2006) that inadequate resources are likely to lead to stress, frustration, and lack of engagement.Materials have been found to be positively related to engagement by both Buckingham and Coffman (1999) and Harter et al. (2002).When employees receive adequate resources to work with they feel obliged to respond in kind and "repay" the organization in form of higher engagementhence validating the SET theory.

Collaboration/Team work
Employees who experienced high degree of collaboration in their jobs were more likely to be engaged (r=0.67,p=0.00).Collaborative work environments are often characterized by trust and cooperation and may outperform groups which were lacking in positive relationships (Seigts and Crim, 2003).Employee opinions as cited by Buckingham and Coffman (1999) and Harter et al. (2002) are positively related to engagement.If an employee"s opinion is considered, the employee will feel valued and important and will therefore be willing to emotionally, cognitively and physically engage the self.This finding is in line with the Schaufeli (2002) and Kahn (1990) model and in the spirit of reciprocity expounded by the social exchange theory.

Recognition and rewards
Employees who experienced high degree of recognition/ rewards in their jobs were more likely to be engaged (r=0.53,p=0.00).The findings of this study are consistent with Seigts and Crim (2006) who emphasize that good leaders frequently recognize their employees by congratulating and by coaching them.Recognition has been found to have a weaker but positive relationship to engagement by Harter et al. (2002).
This study found a strong positive relation (r=0.53,p=0.00) -(large effect size ;Cohen, 1988).Recognition/ reward given by the employer is interpreted as a gesture that has to be reciprocated in terms of higher engagement (Cropanzano and Mitchel, 2005).

Perceived social support (PSS)
Employees who experienced high degree of PSS in their jobs were more likely to be engaged (r=0.59,p=0.00).An important aspect of psychological safety (Kahn, 1992) stems from the amount of care and support employees" perceive to be provided by their organization as well as their direct supervisor.Supportive work environments allow members to experiment and to try new things and even fail without fear of the consequences (Kahn, 1990;May et al., 2004).Basing on the social exchange theory, Rhoades et al. (2001) and Saks (2006) posit that PSS creates a reciprocal obligation on the part of employees to care about the organization"s welfare and to help the organization reach its objectives.This is in line with the Schaufeli/Kahn model and the social exchange theory.

Compensation fairness
Employees who experienced high degree of compensation fairness in their jobs were more likely to be engaged (r=0.51,p=0.00).Compensation fairness refers to the perceptions that employees have regarding equity in company internal and external compensation and benefits.Researchers have found that when pay is fair in comparison with other"s pay, a worker is more likely to be engaged (Milkovich and Newman, 2005;Vandenberghe and Tremblay, 2008).The perceived fairness will compel them to seek to pay back their employer by engagement.This is consistent with and validates the social exchange theory.

Job security
Employees who experienced high degree of job security in their jobs were more likely to be engaged (r=0.48,p=0.00).Once security is assured employees are more likely to be engaged.This study shows that the relation between job security and engagement is relatively low compared to all the other variables considered hereshowing medium effect size (Cohen, 1988).This is not surprising as job security is a major concern of many employees in Uganda.Assurance of job security given by the employer is interpreted as a goodwill that has to be reciprocated in terms of higher engagement (Cropanzano and Mitchel, 2005).

Employee development
Employees who experienced high degree of employee development in their jobs were more likely to be engaged (r=0.63,p=0.00).Work settings in which employees have opportunities for development provide opportunities for growth and employee motivation and engagement (Hackman and Oldham, 1976;Ryan and Deci, 2000).This study found the development -engagement relation (r=0.63,p=0.00) to be one of large effect size (Cohen, 1988).Opportunities for employee development by the employer are interpreted as a benevolence that has to be reciprocated in terms of higher engagement (Cropanzano and Mitchel, 2005).The study findings validate social exchange theory.

Hypothesis 2
Hypothesis 2 stated that there would be a positive correlation between employee engagement and OCB.The results of this study show that engagement was significantly positively correlated with OCB (r=0.61,p=0.00), hence providing empirical support for H2.The three factor dimensions of engagementvigor (r=0.49,p=0.00), absorption (r=0.58,p=0.00) and dedication (r=0.50, p=0.00) are also significantly positively correlated with OCB.Other studies have also found positive relationships between employee engagement and OCB (Rurkkhum, 2010).Avey et al. (2008) for example, found that employees with psychological capital and positive Owor 9 emotion were likely to have an employee engagement attitude and performed more OCB.In addition, the result of this study is similar to the finding of Wang ( 2009) in which a positive and strong relationship between organizational support and OCB was found.
Research by Ensher et al. ( 2001) also reported that when employees perceived sincere organization support in terms development opportunities, they experience engagement and are more likely to reciprocate by willingly participating in their organization"s non-mandatory activities.

Hypothesis 3
Predictors of organizational citizenship behaviour (OCB): Hierarchical regression analyses were performed to test H3 that stated that after controlling for selected engagement antecedents -employee engagement will predict unique variance in OCB.Testing the first regression model, in the first block, five of the nine engagement antecedents, role clarity (β = .21,p = .01),collaboration (β = .21,p =.01), compensation fairness (β = -.26,p = .00),job security (β = -.15,p = .03)and development (β = -.31,p = .00)contributed unique variance to the prediction of OCB (adj.R 2 = .46,p = .00)in the regression equation.This means each of the antecedents contributed between 21 to 30% of the variance in OCB, and together the five antecedents contributed 46% to the variation in OCB.

Contributions to knowledge:
The study contributed to prevailing debate about whether there was a direct link between HR practices (antecedents) and firm performance, or that the relationship is mediated by a range of attitudinal and behavioral variables at the individual level.This study filled that gap by proposing that in the Uganda context, the antecedents-engagement-OCB relationship is mediated by attitudinal variableemployee engagement.This study also contributed to the debate that aggregate outcome variables used in the existing management literature, such as competitiveness, firm financial performance and organizational effectiveness, are too distal from the micro-level HRM interventions, by proposing a more proximal outcome indicator such as OCB to measure individual HRM outcomes (Wright and Haggerty, 2005;Purcell and Kinnie, 2007;Paauwe, 2004).Besides, most of the studies that have been conducted on the predictors of employee engagement and their outcomes in recent years were mostly centered on the Western world such as the United States (Britt, 2003;May et al., 2004), Netherlands (Schaufeli and Bakker, 2004;Xanthopoulou et al., 2007), Spain (Salanova et al., 2005), Finland (Hakanen et al., 2006;Mauno et al., 2007), Greece (Xanthopoulou et al., 2009), Norway (Martinussen et al., 2007).This study filled that research gap in the Uganda among employees of soft drink firms.

CONCLUSIONS
Variations in OCB in Uganda context can be explained by the changes in employee engagement which in itself is influenced by changes in HR practices.Employee engagement is therefore a significant mediator between the HR practices and OCB.This research has validated and extended the engagement and social exchange theory by establishing that role clarity, collaboration, compensation fairness, job security, and development are significant predictors of OCB.

RECOMMENDATIONS
Arising from the findings of this research, the following recommendations are made, hypothesis by hypothesis: Hypothesis 1: To enhance employee engagement, practitioners should ensure that each of the nine antecedents in their businesses are nurtured -job characteristics, role clarity, material resources, collaboration, reward and recognition, perceived social support, compensation fairness, job security and development.Hypothesis 2: To reap the benefits of employee engagement and increase OCB it is recommended that business should use the finding of this study to develop interventions and strategies for improvement of employee engagement and OCB among employees.Hypothesis 3: The recommendation is that since role clarity, collaboration, compensation fairness, job security, and development as well as employee engagement are significant predictors of OCB, managers need to appreciate role of social exchange in liking these antecedents and OCB.

LIMITATIONS OF THE STUDY
This study used a cross-sectional research design as opposed to a longitudinal research design.Longitudinal and experimental studies are required to provide more definitive conclusions about the causal effects of employee engagement and the extent to which social exchange explains these relationships.In addition, much of the data came from a self-reported questionnaire.This could affect the reliability and validity of the data because respondents may have answered the questions to reflect more socially acceptable responses rather than ones that reflect their real opinions.Although the results of this study might have been affected by method bias, statistical results give us confidence in the results.

RECOMMENDATIONS FOR FUTURE RESEARCH
Future research on OCB, engagement and antecedents should not be limited to a quantitative paradigm.Qualitative studies might assist in better understanding the phenomenon under study.To replicate this study, structural equation modelling (SEM) as well as longitudinal studies could be used to capture changes in the variables over periods of time.Future researchers should expand the sample to include the entire food and beverage industry.A comparative study could also be done with the service industry.A study could also concentrate of the five antecedents which emerged significantcollaboration, reward and recognition, perceived organizational support, compensation fairness, and development.Instead of the intermediate outcome -OCB, future research should consider final outcomes like productivity, profitability, customer satisfaction and competitiveness.

Table 1 .
Population and Sample of the sampled firms.

Table 2 .
Correlation -Employee engagement (V, A, D) and the nine engagement antecedents.

Table 3 .
Correlationemployee engagement (V, A and D) and OCB.