The Employee Voice Behaviours in African Context: The Case of Zimbabwe

: This exploratory study aimed to investigate the Employee Voice Behaviours (EVB) in the African context, and Zimbabwe was chosen as a case study. A sample of 30 volunteers was used in this study and data was collected via WhatsApp social media. The structured interviews were employed to collect data. Thus, the data was collected and placed into categories as a framework. The data analysis included descriptive statistics, multiple regression, correlation and framework. The results indicated that the majority of employees in Zimbabwe present constructive voices than destructive behaviours. The findings also established that Zimbabwean workforce remains disciplined in their organisations. The study again found that the majority of people working in organisations in Zimbabwe is neither married nor have children. Also, the majority go to church and do not consume alcohol/smoke. However, besides a small sample that could have affected the outcome, this study concludes that understanding of EVB remains elusive; therefore, more studies are required in this field. Future researchers should consider qualitative methods with larger samples to establish these variations. Besides, the current research presents significant findings that could inform policy-makers in Zimbabwean organisations.


Introduction
The concept of Employee Voice Behaviours (EVB) has gained momentum in African workplaces and among academics.Many researchers, such as Wood and Glaister (2008), Kok et al. (2014), Ganjali and Rezaee (2016), Klerck (2016), Maree (2016), Emelifeonwu and Valk (2019) have shown an appetite to understand the EVB in an African Context.For example, Klerck (2016) environment.However, the studies of EVB in the African context are lagging in literature; more is needed to be discovered, particularly in Zimbabwean workplace settings.
Africa is a vast continent and countries are subdivided by extensive cultural diversity.The concept of EVB could be different in each culture and mean something in others.Hence, the understanding of EVB becomes critical as it could bring a cross-sectional knowledge of how people in each cultural group voice in the workplace.
However, Couldry (2010) suggested that EVB is vital as a factor in these neoliberalism times that drive economic, political, social and cultural matters.With the African context, organisations made up of people with many differences, including cultures, one size would not surely fit all.According to Couldry (2010), giving value to EVB appears like taking away rights from organisations that undermine employee's democracy.Hence, the process of EVB should be balanced as both sides should win.Hirschman's (1970) argument of voice was that people either leave or fight on to change the status quo.Following this concept put forward by Hirschman, the issue of EVB sounds very clear that individuals either exit or voice in a working setting, but in an African context, the case is far from clear.
Zimbabwean working environment has been marred with labour disputes in recent years.For instance, CT Bolts v Workers Committee (2011), Gweru Water Workers Committee v City of Gweru (2013), and Marange Resources v Core Mining & Minerals (2016), were cases that went through the courts recently.Also, ZUPCO v Mashinga (2017), and Pinkstone Mining (PVT) Ltd & 2 Others v Lafarge Cement Zimbabwe Ltd & Another (2018) are some of the cases that challenge the concept of EVB.This argument was also confirmed by Maitireyi and Duve's (2011) study that found that there are challenges with the arbitration system in Zimbabwe.These contemporary labour differences could have undermined the EVB in the country, with prominent workforce in medical and engineering fields leaving the country.The EVB in Zimbabwe could be seen as still developing in practice, despite the availability of the Labour Legislation that could protect both employees and employers (Zimbabwe Labour Law, 2015).
EVB in literature has been described as the action associated with employee involvement in the company's decision-making process (see, e.g., Rees et al. 2013, Kaufman 2015, Barry and Wilkinson 2016, and Mowbray et al. 2019).Again, EVB is related to employee participation (see, e.g., Van Dyne et al. 2003, Hames 2012, Timming and Johnstone 2015, Mathur and Chuhan 2017).With, CIPD (2019) explicitly defines EVB as communication between employees and employers; thus workforce expressing their innovative and creative ideas for the organisational improvement.EVB is also associated with workforce employee commitment (see, e.g., Machokoto, 2019).
It is important to note that there are limited studies on EVB in Zimbabwe, and it is vital to understand the structure of democracy in Zimbabwean workplace.This current study hopes to make a valuable contribution that stimulates future research in EVB in Zimbabwe.Even though EVB has been thoroughly considered in other contexts, the review of literature has indicated that EVB in Zimbabwe lacked attention that may have cause the system to lag behind.The current study hopes to make a contribution to policymaking in human resources management and future research.The objectives of the study are;

Literature Review
2.1.EVB Hirschman (1970) was the first writer to pronounce the concept of Voice when he suggested that individuals in an organisation can either Voice to change the status quo or exit the organisation.In his idea, an individual can choose to stand up and voice their concerns and help the organisation's operation or remain quiet and leave the organisation (Hirschman, 1970).The Voice or Exit Theory was developed with politics and economic development in mind (Hirschman, 1970).Therefore these two extreme options in his theory were further developed in the organisational management studies.However, the Hirschman's (1970) initial thought of Voice was expressed in the customer relationship, which could be a different circumstance to the employee-employer relationship.However, Hirschman's proclamation of voice was found to be broad and left no valid measurements of EVB (Maynes and Podsakoff, 2014).Consequently, Hirschman's work triggered interest in the field of people management.
Many scholars emerged, for example, Gorden (1988), Van Dyne and LePine (1998), Van Dyne et al. (2003), Liang et al. (2012), and Maynes and Podsakoff (2014).All these studies took the perspective that Voice is a specific behaviour.Thus, Gorden (1988) found that there were some behaviours among employees; which were either active or passive and either constructive or destructive.Besides, Gorden (1988) is credited in literature for being the first scholar to apply the Voice concept in the employee-employer relationship.However, the Voice concept was established in the contest of customer and manufacturer relationship.Therefore, the relevance of either to Voice or Exit concept in employment relationship was always questionable.Maynes and Podsakoff (2014) contested the Gorden's contribution saying also lacked valid measurements of EVB, like Hirschman's earlier contribution.More so, the submission that EVB is 'active' was found to be inconsistent with Hirschman's original thoughts (Maynes and Podsakoff, 2014).Van Dyne and LePine (1998) argued that all EVB are displayed to nurture the organisations' activities rather than to disrupt.However, Van Dyne and LePine's contribution was criticised, as it was biased (Maynes and Podsakoff, 2014).Also, many previous scholars, for example, Grant et al. (2009) were critical to Van Dyne and LePine's (1998) contribution, they claimed that Van Dyne and LePine's findings were contaminated.Hence, Van Dyne and LePine's results pointed out at one side of EVB.Therefore, the Van Dyne and LePine's viewpoint of EVB may not be reliable to consider in critical studies.Van Dyne et al. (2003) submitted that EVB is prosocial, defensive or acquiescent.They found that workers could show EVB that is related to businesses' visions and employees relate to others (Van Dyne et al., 2003).With Van Dyne et al.'s view of prosocial EVB, there are fewer conflicts within the workplace and their suggestion tally with Van Dyne and LePine's (1998) findings.Also, Van Dyne et al. (2003) established that the workforce could be defensive in their practice to avoid change with intention to protect their interests.Maynes and Podsakoff (2014) suggested the destructive activities at workplaces could be driven by fear of uncertainty, especially when the businesses are embarking on changing mode.
Moreover, Van Dyne et al. (2003) found that the workforce may completely withdraw from the organisation's activities; thus being acquiescent to the business activities.Maynes and Podsakoff (2014) also commended that employees may demonstrate low self-efficacy to influence any change.Perhaps a sense of acquiescent could be correlated to Hirschman's exit claim while being defensive seen as destructive and passive EVB.Liang et al. (2012) presented that employees could express two EVB, which are either promotive or prohibitive voices during the business operation.Liang et al.'s mention of promotive as a particular EVB confirmed the findings by previous scholars, such as Gorden (1988), who found that active and constructive were EVB that was promotive to firms.Van Dyne and LePine (1998) claimed that employees are promotive to their organisations.Also, Van Dyne et al. (2003) also alluded that the EVB of the workforce could be promotive to the organisation's social structure.Liang et al. (2012) suggested that the workforce could support all business activities to see improvements.However, Liang et al. (2012) also highlighted that these employees could also present EVB that are not acceptable in the organisation; the EVB that they named prohibitive voice.In Van Dyne et al.'s (2003) position, prohibitive behaviours could also be observed as defensive or acquiescent, while Gorden (1988) could have called the behaviour as destructive to the business.Maynes and Podsakoff (2014) found that previous scholars' effort could be credited in many ways.For instance, Hirschman's (1970) voice behaviour was meant to challenge the status quo.Likewise, Gorden's (1988) active behaviours were also crucial to the firms as they were promoting improvements, preserving or positively challenging the status quo.Van Dyne and LePine's (1998) voice behaviours also intended to improve business performance and again, positively challenge the status quo.Van Dyne et al. (2003) and Liang et al. (2012) were not left out in the line for credit as their work asserted that employees could have a positive impact on the firms' performances and also positive intentions when challenging status quo.Besides, all these researchers' findings were found not to be specific to what measures EVB.

Maynes
and Podsakoff (2014) conducted their study that expanded the view of EVB to explore valid measures of EVB.They managed to place all previous forms of EVB into categories using what the EVB was meant to achieve in the business.In their framework, Maynes and Podsakoff (2014) illustrated that EVB is either promotive or prohibitive as dimensions.Under the preservation section, they also pointed out that EVB is again, either supportive or defensive (Maynes and Podkoff, 2014).On the last part, Maynes and Podsakoff (2014) claimed that EVB present self as a challenge that is constructive or destructive.However, the Maynes and Podsakoff's (2014) provided a direction to follow in the study of EVB, such as the need to understand the challenges of destructive EVB and the importance of constructive EVB.

As
Gorden (1988) mentioned, destructive EVB are expressions that are against the business' operations; therefore, present challenges to the management.This EVB include resigning from the organisation and also calculated verbal bad-mouthing against the company (p.285).Furthermore, destructive EVB can be observed as silence to withdraw from the organisation's activities (Gorden 1988, p. 285).Van Dyne et al. (2003) suggested that destructive EVB as the activities that show employees blocking changes in the organisation for selfish reasons; thus being defensive.The interpretation of such defensive behaviours is motivated by fear of the future (p.1372).Again, Van Dyne et al. 2003 mentioned that destructive EVB could be expressed by loss of self-efficacy to effect any meaningful change (p.1373).Destructive EVB affects organisational performance, and destructive EVB is reported caused by psychological recession (Poisat and Theron, 2014).Therefore the management should engage the employees to reverse the psychological recession (Poisat and Theron 2014).

Importance of Constructive EVB
The constructive EVB emerged from Gorden's (1988) writings, where he categorised some behaviours as constructive.For example, employees may be passive but supporting the business activities and also actively coming up with solutions for the business.In other words, employees can display activities that can be a positive contribution to the business.The constructive view was echoed by Van Dyne and LePine (1998) who only saw promotive employees' activities.Also, in Van Dyne et al.'s (2003) work, employees were viewed as prosocial, meaning that their connection supports the social side of the organisation.Hence, Liang et al. (2012) argued that some EVB are promotive.In literature, constructive EVB are linked to increased competitive advantages (see, for example, Albrecht et al. 2015 andPousa andMathieu 2015).Consequently, constructive EVB means more production in the business.Therefore, it is essential to understand the EVB in any business context.However, with African's EVB context lagging, there is a need to cover this knowledge in the body of literature.

Method
This project was an exploratory study to examine the EVB in the African context, and Zimbabwean workplace was taken as a case study for this research.The concept of EVB from Hirschman (1970), Gorden (1988), Van Dyne and LePine (1998), Van Dyne et al. (2003), and to Liang et al. (2012) was used as a framework in this study.The study was conducted between August 2019 and January 2020 via WhatsApp social media.The study targeted individuals who were in some sort of employment in the public and private sector in Zimbabwe.An advertisement for the study was sent to individuals and groups on WhatsApp, and potential participants responded and volunteered to take part in this study.

Sample
The study sent the invitation to many people who use WhatsApp in Zimbabwe, and 37 individuals came forth and volunteered to be interviewed.Among potential participants, there were 23 females and 14 males, who were all in employment.The sample was composed of 31 people working in private sector and 6 in the public sector.Further analysis of the sample highlighted that, for those who mentioned that they work in the public sector; five were in teaching and one in the police force.For those in the private sector, seven mentioned that they are self-employed, and they were rejected from the sample because they were not legible for this study.The remaining 24 people; 13 said they work in private companies, including OK Zimbabwe, TM Zimbabwe, Chicken Inn and Holiday Inn, 8 stated that they were employed in banks, such as FBC, Steward Bank, and ZB Bank.The remaining three mentioned that they work as domestic workers.After all the analysis, therefore, 30 individuals took part in this research.

Data Collection
The 30 individuals were each interviewed for between 15 to 30 minutes using a structured questions interview format.The questions were set to determine whether participants were displaying Hirschman's (1970) exit or voice behaviour (first question).The second question was to determine which category of voice the participants present from Gorden's (1988) theory.The fourth category was to establish whether participants were promoting the business activities, based on Van Dyne and LePine's (1998) suggestion.The fifth question was to find out which behaviours the participants were displaying from the Van Dyne et al.'s (2003) categories.Lastly, the sixth question was to classify participants' behaviour, according to Liang et al.'s (2012) theory.

Results
Figure 1 represents the descriptive statistics that summaries all nine categories of data collected in this study.

Figure 2 Multiple Regression
The data types included are the age of participants, the location where they live, marital status, number of children, whether participants drink alcohol or smoke, and whether they go to church.Also, included was years in employment, the number of disciplines within the organisation, and whether participants could exit or voice when they are not happy about anything.The average age of participants was 40.1 years, with the oldest was 61 years, and the minimum age was 20 years, and the age range was 41 years.Majority of participants were living in Harare as indicated by mean and mode 1, also, the majority were not married as shown by mean and mode 0, so.As a result, the majority of participants had no children.The data also highlights that the majority of participants were not taking alcohol/smoke as the mean and mode of 0. Instead, the majority of participants were going to church; this is confirmed by the mean and mode of 1.
Furthermore, maximum years in employment were 31 years and minimum a year, and years in employment range of 30 years, with average years of service of 9.4 years.Majority of participants said they were never disciplined in their organisation; this is indicated by mean and mode of 0. However, the data suggests that the majority of participants express that they could voice to challenge the status quo rather than leave the organisation.
Figure 2 shows a multiple regression outcome of 8 variables.The researcher wanted to understand why there were variations among the variables and degrees of relationships.The above regression model is weak to explain the variations between Voice/Exit and other variables because it explains only 35% of the variations as indicated by R-Square = 0.35, not significant given that (p-value =.258) and weak coefficients -0.195.Therefore this regression model informs us that it is difficult to determine the variations in the EVB in Zimbabwe.However, the researcher went further to explore each type of EVB displayed by participants.
Figure 3 shows the results from each voice category measured.From the descriptive statistics in Figure 1, the majority of participants indicated that they had intentions to voice that concerns with the organisation's management.However, the types of voice behaviours vary among the participants, as shown in Figure 4. Considering Gorden's theory: 9 participants indicate that they were active and constructive to the organisation, compared to those four who said they were active and destructive.However, 6 participants mentioned that they were constructive but passive, and seven more said they were also passive but destructive to the organisation's activities.Van Dyne and LePine's Theory: the findings show that 20 participants were generally promotive to the business, which was the majority.Van Dyne et al.'s theory: the results show that half of the participants were presenting a defensive voice and did not allow changes to take place in the organisation.Only 7 participants highlighted that they were prosocial, and 8 participants said that they were feeling like resigning from the organisation.Liang et al.'s theory: 18 out of 30 participants presented prohibitive voice, while only 12 participants expressed promotive voice behaviours.However, for further understanding of the data, the research employed Maynes and Podsakoff's (2014) framework to categorise the data from Figure 4. Therefore, the result was a supportive voice (23 participants), constructive voice (37 participants), defensive voice (23 participants) and destructive voice (21 participants).
The Maynes and Podsakoff's framework was useful to see the distribution of the EVB patterns.The bar chart shows that more participants expressed constructive views (37) than being defensive (23).Although participants with supportive voices were few (23), the participants who showed destructive behaviours were the fewest (21).

Discussions
The study aimed to investigate the EVB structure in the Zimbabwe context.The findings of the research established that the majority of participants were aged between 20 and 61 years, and were based in Harare, which is a Capital City of Zimbabwe.Again, results show that years of employment ranges from a year to 31 years and many participants expressed that they many participants have intentions to voice to challenge status quo and the majority have never been disciplined within their organisation.Also, the majority of participants indicated that they were not married and had no children.Furthermore, many participants go to church and consume less alcohol or smoke (see Figure 1).Even though the data revealed all these critical knowledge that there are variations in EVB in Zimbabwe, the regression model failed to explain the variations (see Figure 2).Previously, Maynes and Podsakoff (2014) found that there were challenges in understanding EVB in literature, and this study has proven this view.The current research could have supported the point that it remains difficult to conceptualise the nature of EVB constructs.This project is one of the few studies to support the findings by Maynes and Podsakoff (2014).
However, the outcome indicates that there is a positive correlation between Voice/Exit and Marital Status (.35), Number of Children (.33), Age (.32), Years in Employment (.21) and Location (.14) even though the correlation is weak.Again, the results show a weak negative correlation between Voice/Exit and Alcohol/Smoke (-0.12).The results also indicate that there is no correlation between Voice/Exit and going to Church or being disciplined as the results show correlations close to 0. However, the results tell that there is a strong positive correlation between the number of children and marital status, and there is also a positive correlation between being disciplined and alcohol/smoke, even though the correlation is weak (see Figure 5).Besides, it essential to mention that past researchers, for example, Hirschman (1970), Gorden (1988), Van Dyne and LePine (1998), Van Dyne et al. (2003) and Liang et al. (2012) found that EVB are displayed to either challenge/preserve status quo or promote improvements.Even though these previous findings were deemed not consistent (see Maynes and Podsakoff, 2014), the current study also established that many participants expressed that they could challenge the status quo.
Moreover, Maynes and Podsakoff (2014) established a framework that helps to understand the nature of EVB.According to Maynes and Podsakoff, EVB can be put into four categories, which are supportive, constructive, defensive and destructive voices.
The current researcher found Maynes and Podsakoff's framework to interpret the EVB in this study (see Figure 4).The findings illustrated that many participants expressed constructive and supportive voices, and less showed defensive and destructive views.Therefore, at this stage, it is possible to agree with Van Dyne and LePine (1998), who established that employees express more constructive behaviours than destructive.However, more work is needed to develop the variations on Voice/Exit in the African Context.
Also, future studies should use qualitative methods to examine EVB within these employee groups with larger samples.More research need to be carried out on EVB in Zimbabwe.
The limitations in the current study include a sample, 30 people could be a small sample may have affected the reliability of the outcome.Therefore the generalisation of the present findings could be limited.Again, the method of collecting data used in this study could have also affected the validity of the data because the researcher could not confirm whether the potential participants were telling the truth about their employment.Also, another limitation could be associated with processing the data; the regression test failed to explain the variations among the variables.
Therefore the fault could be on the data itself.Nevertheless, testing larger data samples in the same subject could benefit this study's findings.

Conclusion
The outcome of this study suggests that the majority of employees in Zimbabwe present constructive voice, even though few expressed destructive behaviours.Again, the Zimbabwean employees remain disciplined as this was indicated by a few numbers of disciplines in this study.Majority of employees in Zimbabwe are neither married nor have children.Furthermore, few Zimbabwean employees consume alcohol/smoke, while many go to church.The current study also concludes that the understanding of EVB remains elusive as this was proven by the failure of the regression model to explain variations.However, the present study could have been limited by the small sample.Therefore future studies should involve larger samples and employ qualitative methods to establish personal issues that could example EVB.Besides, the current research presents significant findings that could inform policymakers in Zimbabwean organisations.

Figure 1
Figure 1 Descriptive Statistics

Figure 3 Figure 4
Figure 3 Results on Framework

Figure 5
Figure 5 Correlation of Results on Framework These findings set a stage for future research.The subsequent studies should consider investigating if there are causal relationships between Voice/Exit and Age, Location of people, Marital Status, Number of Children, and Years in Employment.Existing literature on EVB, such asHirschman (1970),Gorden (1988), Van Dyne and LePine (1998),Van Dyne et al. (2003) andLiang et al. (2012) even though they established different types of EVB, they all failed to establish validated measures of EVB.