“Exploring organizational culture at a state-owned enterprise in South Africa: a process approach”

ARTICLE INFO Patrick J. Lekgothoane, Molefe Jonathan Maleka and Zeleke Worku (2020). Exploring organizational culture at a state-owned enterprise in South Africa: a process approach. Problems and Perspectives in Management, 18(2), 431-440. doi:10.21511/ppm.18(2).2020.35 DOI http://dx.doi.org/10.21511/ppm.18(2).2020.35 RELEASED ON Friday, 03 July 2020 RECEIVED ON Wednesday, 20 May 2020 ACCEPTED ON Friday, 29 May 2020


INTRODUCTION
Organizational culture is a construct of interest to organizational psychology (Naidoo & Martins, 2014) and management sciences scholars (Demir, 2015). It can adversely or positively affect the implementation of the organization strategy. Drucker opines that "culture eats strategy for breakfast" (as cited in Teasdale, 2002, p. 195). It can adversely affect any organization's performance (Kamaamia, 2017), including stateowned enterprises (Chavunduka, Fauzi, Muranda, Sifile, Mabvure, & Dandira, 2014). Internationally, particularly in Greece, it was found that, at a public enterprise with a hostile culture, there was a great deal of politics, jealousy, and negative emotions, there was high attrition of skilled workers, and junior employees were sexually harassed and bullied by managers (Rovithis, Liardakis, Merkouris, Patiraki, Nassilaki, & Philithis, 2017).
A culture with a hostile relationship instead of cooperation is known as bureaucratic (Raaphorst, 2017). A study at a Chinese state-owned enterprise (SOE) showed that, where there was no trust or affective commitment, employees were more likely to resign, and their organizational citizenship was adversely affected (Wong, 2017). Conversely, at an SOE in India, the data showed that a positive organizational culture enhanced job satisfaction and affective commitment (Saha & Kumar, 2018). Besides, the studies found that, where a positive culture existed, employees shared information (Tseng, 2017), and managers treated employees with respect and dignity (Saks & Gruman, 2011).
It can be deduced that there are two types of organizational cultures: positive, and bellicose or hostile. A hostile culture is perpetuated by organizational politics and hostility (O'Donnell & Boyle, 2008), while a positive culture is about helping others and employees (Confeld, 2016). In some instances, it has been found that a culture of nepotism during recruitment and selection is rife (Mahlangu, 2001, p. 46), where employees with appropriate qualifications and experience are overlooked and not appointed in critical and managerial positions (Spector, 1985).
Based on the above discussion, it is clear that both individual (i.e., job satisfaction and organizational citizenship behavior) and organizational (i.e., recruitment and selection) factors predict organizational culture. Previous research has shown that in South Africa, a hostile or bellicose culture prevails in SOEs (Maleka & Rankhumise, 2014). In such an organizational culture, employees are bullied during performance appraisals, and those who object are given poor performance appraisals and do not receive career development opportunities (Maleka, 2012). Furthermore, there is evidence that employees do not go beyond the call of duty in South African SOEs, meaning that a lack of organizational citizenship behavior is rife (Garg & Ramjee, 2013). It was against this background that the researchers embarked on this study to quantify the predictors of a culture of respect or a hostile culture.
Thus, the study aimed to explore organizational culture at the SOE. This was achieved by measuring how respondents rated the organization culture and its predictors, i.e., organizational citizenship behavior and job satisfaction. In the next section, the literature to determine the direct relationship between organizational citizenship behavior and organizational culture is reviewed. Besides, the discussion is on how job satisfaction mediates the relationship between organizational citizenship behavior and organizational culture. The hypotheses are developed.

LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
Scholars internationally have developed organizational culture theoretical frameworks (Fleury, 2009 The unintended consequences of a hostile culture in SOEs in South Africa ultimately compromise parastatals' performance (Mekwe, 2015). It has been found that, in SOEs where the organizational cul-ture is hostile, there is a great deal of politics (Mbele, 2015), jealousy (Barlow, 2016), and negative emotions (Phale, 2008), there is high attrition of skilled workers, and junior employees are sexually harassed (Merkin, 2008) and bullied by managers or supervisors (Fapohunda, 2013). Bullying managers are protected by top managers and are not disciplined (Killoren, 2014 (2000) found that ratings consisted of non-prescribed behaviors, rendering the rating employee to attach his/her personal views on what constitutes organizational citizenship behavior. Thus, high-performing employees rated the relationship between organizational citizenship behavior and performance higher than managers (Turnipseed & Rassuli, 2005).
Mitonga-Monga, Flotman, and Cilliers (2016) revealed a significant positive relationship between organizational citizenship behavior and job satisfaction. Berber and Rofcanin (2012) and Kaur (2013) define organizational citizenship behavior as going beyond call of duty, being present at the workplace, and not abusing leave. In a recent study by Zeyada (2018), organizational culture proved to predict organizational citizenship behavior. However, in a study conducted in India, organizational culture did not predict organizational citizenship behavior (Jain, 2015). In another study, organizational citizenship behavior and organization culture were found to contribute to increasing turnover intentions (Jacobs & Roodt, 2008). It is hypothesized that: H1: There is a positive relationship between organizational citizenship behavior and organizational culture.
In order for mediation to take place, Leech, Barrett, and Morgan (2015) argue that these conditions should be met: 1) the independent variable must significantly predict the dependent variable; 2) the independent variable must significantly predict the mediating variable; and 3) the mediating variable must significantly predict the dependent variable.
In this study, the mediating variable was job satisfaction, which is an affective stage of workers that is influenced by the relationship between workers and their supervisors ( H2: There is a positive relationship between organizational citizenship behavior and job satisfaction.
H3: There is a positive relationship between job satisfaction and organizational culture.
In the next section, the study methodology is discussed.

Participants and setting
The research design was cross-sectional, and respondents were selected using a stratified sampling technique. Respondents were stratified by gender.

Measuring instrument
Five recruitment and selection items and five organizational culture items (three positive and two bellicose) were taken from Maleka (2012). Some of the sample items were: "Organizational politics are rife in the workplace" and "The culture of hostility is rife in the organization." Five organizational citizenship behavior items were taken from Williams and Anderson's (1991, p. 607) scale. Some of the sample items were: "I volunteer to do things I am not required to do" and "My attendance at work is above the norm." Five job satisfaction items were taken from Spector's (1985, p. 713) scale. Some of the sample items were: "My supervisor provides justification for my performance appraisal score" and "My manager attends to my complaints speedily." Similar to Reis et al.'s (2015, p. 1096) organizational culture study and Armstrong's (2015, p. 811) employee attitude studies (i.e., job satisfaction), respondents rated items on a four-point Likert scale, where 1 was strongly agree and 4 was strongly disagree.

Research procedure
The researchers were given ethical clearance by the university before data collection, after which one researcher went to the SOE to collect the data. Respondents were informed that their participation was voluntary and not coerced to participate in the study. The respondents were informed about the benefits of participating, and confidentiality and anonymity were assured by not asking respondents to write their names on the completed questionnaires. Each respondent was given an informed consent form to complete before completing the questionnaire.

Statistical analysis
The measuring instrument items were subjected to principal component analysis using

Validity and reliability
The researchers followed Christensen, Johnson, and Turner (2015), Babbie (2013) and Kumar (2014), advice and achieved content validity by presenting the measuring instrument to an expert to validate the items. Before data collection, the measuring instrument was pre-tested on 10 respondents. Before principal component analysis, the suitability of the data for factor analysis was assessed. The Kaiser-Meyer-Olkin value was 0.70, which was above the value of 0.5 recommended by Hair, Black, Babin, and Anderson (2014). Bartlett's test of sphericity was statistically significant at p = 0.00. Principal component analysis revealed three factors with eigenvalues exceeding 1, explaining a total variance of 59.80%. As suggested by Hair, Black, Babin, and Anderson (2014), a cut-off of 0.3 was used in this study as inclusion criterion for factor loadings in the model. Three factors emerged after conducting factor analysis. The first factor, labeled as job satisfaction, had five items with factor loadings ranging from 0.57 to 0.82. The second factor, labeled as organizational citizenship behavior, had three items with factor loadings ranging from 0.76 to 0.81. The third factor, labeled as organizational culture, had two bellicose items with factor loadings of 0.838 and 0.842.
The Cronbach's alphas for the organizational psychology scales were as follows: 0.76 for job satisfaction, 0.69 for organizational citizenship behavior, and 0.62 for organizational culture. It is argued that all organizational psychology scales were reliable because they were within the Cronbach's alpha range of 0.6 to 0.9, as suggested by Maree (2016). Table 1 presents the descriptive statistics. The data showed that the respondents agreed with organizational culture (M = 2.00, SD = 0.77) and organizational citizenship items (M = 2.00, SD = 0.69).

RESULTS
On the other hand, the data showed that the respondents disagreed with job satisfaction (M = 3.00, SD = 0.65) and recruitment and selection (M = 3.00, SD = 0.83) items.
Before calculating mediation, certain assumptions were checked, and the data on histogram showed a bell shape. The line on the P-P plot points was on the diagonal "from bottom left to right", and the correlations between the independent variables (i.e., job satisfaction and organizational cit-izenship behavior) were above 0.7. None of the variance inflation factors were above 10, and the tolerance was less than 0.10 (Pallant, 2016). These data suggested that the assumptions of normality, linearity, and multicollinearity were not violated (Tabachnick & Fiddel, 2014). Table 2 presents the mediation results. The data showed that the first path, known as a', had an R-squared of 0.01, while F (1, 197) = 1.23 and p = 0.27. It can be interpreted that organizational citizenship behavior explained 1% of total variance of job satisfaction. H1 was supported because the organizational citizenship behavior significantly predicted organizational culture (β = 0.19, p = 0.02). H2 was not supported as it could be observed that organizational citizenship behavior did not significantly predict job satisfaction (β = 0.07, p = 0.28). H3 was not supported because job satisfaction was not a significant predictor of organizational culture (β = 0.05, p = 0.55). When job satisfaction and organizational citizenship behavior were predictors of organizational culture, the R-squared was 0.03, F(2, 196) = 3.39 and p = 0.04, meaning that the combination of job satisfaction and organizational citizenship behavior explained 3% of total variance of organizational culture. The data also showed that the confidence interval included zero (-0.01,0.27), suggesting that mediation did not take place.

DISCUSSION
One of the first specific objectives was to determine how the respondents rated organizational culture and its predictors. The mean score of the organizational culture items (M = 2.00, SD = 0.69) showed a negative trend, suggesting that respondents agreed that organizational politics and culture of hostility were rife at the SOE. These data were consistent with the study conducted by Maleka and Rankhumise (2014) at another parastatal. The unintended consequences of workplace politics are that employees who do not have the right skills are appointed into managerial positions. Since they do not have managerial competencies, they cannot help the SOE achieve its objectives, and they have adversarial relationships with subordinates (Maleka, 2012). Drucker (as cited in Teasdale, 2002, p. 195) opines that "culture eats strategy for breakfast". Previous research has shown that, when the culture is hostile, employees do not come to work (Magee et al., 2017;Sarafraz & Kia, 2015), and they only do their jobs and do not work beyond the call of duty (Garg & Ramjee, 2013). When such a culture is rife, there are no mentoring, coaching, and development opportunities, as found by Booysen (2007). To reverse these trends, managers at SOEs should foster a culture of respect that rewards and recognizes employees who go beyond call of duty (Berber & Rofcanin, 2012;Kaur, 2013;Mahlangu, 2001;Spector, 1985). Furthermore, Van Rooij and Fine (2018) posit that a positive culture that promotes transparency, honesty, and responsibility should be inculcated or fostered by managers.
Research has found that managers who changed a negative culture to one of respect and inspiration were supported by their subordinates to achieve strategic goals (Phale, 2008).
Similar to organizational culture, the mean score for organizational citizenship behavior was (M = 2.00; SD = 0.69). This suggested that respondents agreed with these statements. The present study results showed that employees rated organizational citizenship behavior positively when responding to the statements: "I volunteer to do things I am not required to do" and "My attendance at work is above the norm." Bachrach, Powell, Bendoly, and Richey (2005, p. 14) found that when employees exhibited organizational citizenship behavior, they worked independently and per- It was found that respondents rated job satisfaction items negatively (M = 3.00, SD = 0.65). It can be claimed that respondents disagreed with statements such as: "My supervisor reviews my performance with me" and "My immediate supervisor articulates a compelling vision of the future." Previous research at another South African SOE showed that managers who were insecure and did not like to be challenged gave employees low performance scores (Maleka, 2012). Cappelli and Conyon (2017) and Idowu (2017) found that supervisors who justified performance appraisal scores enhanced employee satisfaction levels. The results were not consistent with research by the OECD (2014), which found that when supervisors resolved employees' complaints in the parastatal, employees were satisfied.
To test mediation or indirect effects, the Hayes' (2014) analysis was used. The results showed a positive relationship between organizational citizenship behavior and job satisfaction, but the relationship was not significant (β = 0.07, p = 0.28). Similarly, Mitonga-Monga, Flotman, and Cilliers (2016) found a significant positive relationship between organizational citizenship behavior and job satisfaction. In SOEs where such relationships are significant, the organizations perform and reach their strategic objectives (Wiedmer, 2011).
The indirect relationship between job satisfaction and organizational culture was also tested. It was also found that job satisfaction did not significantly predict organizational culture (β = 0.05, p = 0.55), but the relationship was positive. Toumivaara (2015) also found a positive relationship between job satisfaction and organizational culture. Based on the results of the indirect effect, it can be concluded mediation did not take place.

CONCLUSION
The main purpose of the study was to explore organizational culture at the SOE. Notwithstanding the limitations of this exploratory study on two predictors of organizational culture, the study has contributed by indicating that organizational citizenship behavior is a predictor of organizational culture, where such a relationship has not been fully established in the literature in the South African context. Additionally, the study has contributed by finding that, even though the SOE's organizational culture was predominantly bellicose, employees volunteered to go beyond call of duty, and workplace attendance was above the norm.
It can be concluded that a bellicose culture could be changed to a positive culture if SOE's managers: • created campaigns that will foster a culture of respect instead of a bellicose culture (i.e., hostility and politics); • continued encouraging and motivating employees to go beyond call of duty and not abuse the leave system; and • maintained a culture of conducting performance appraisals fairly and resolving employees' complaints speedily.
Unlike in this study, future researchers should have five respect and bellicose items, so that, when factors analysis is conducted, the two cultures are unidimensional and distinct. In the future, a similar study should be conducted in other business units and SOEs, using larger representative samples. It is recommended that in the future, research should focus on engagement as a mediator, and remuneration and gender as predictors of job satisfaction. Previous SOE research has found that, in bellicose cultures, gender pay disparities are immense (Maleka, 2012).
The limitations of the study were, firstly, that the study was conducted at only two business units of the organization in one province. Secondly, the study was based on a cross-sectional design, which gave a once-off snapshot picture of organizational culture at the SOE.
Despite these limitations, this study contributed to the body of knowledge in that it is one of the few that were conducted at a South African SOE that explored whether job satisfaction mediated the relationship between organizational citizenship behavior and organizational culture. This study contributes to the body of knowledge by establishing a predictive path between organizational citizenship behavior and organizational culture, which, according to the literature, is under-researched in the South African