Social Engagement, Gender Diversity and Employee Performance of the Tanzanian Textile Industry

Businesses operate in an uncertain environment and their survival depends on employee performance. However, the underperformance of employees in some organizations can be due to several important factors, including social engagement and gender diversity. This study investigated the influence of social engagement on employee performance, considering the mediation effect of gender diversity. Questionnaires were distributed to a sample size of 618 employees, and the response rate was 89.6%, that is, 554 employees from textile firms in four Tanzanian regions (Dar es Salaam, Mwanza, Shinyanga, Simiyu). Participants were sampled using a multistage sampling technique. Descriptive and structural equation model were used for data analysis. The findings revealed that social engagement had a significant positive influence on employees’ adaptive, contextual and task performance, while gender diversity partially mediated the influence of social engagement on contextual and task performance. Among others, the article recommends that managers should encourage employees to show concern and care for one another’s opinions and wellbeing and respect for individual values. The promotion of gender diversity among engaged employees can also potentially boost performance.


Background
Today's world's businesses operates in a competitive environment that is more volatile and characterized by uncertainties (Hanaysha, 2016). It is crucial for organizations to understand that "the success of any business highly depends on the performance of their employees" (Othman & Mahmood, 2019). Sriviboon and Jermsittiparsert (2019) argue that employees are critical for the survival of any organization. As such, researchers including Kumar, Duhan and Haleem (2016) have been studying how to improve employee performance within the organization.
The existing knowledge on the antecedents of employee performance indicates that most organizations still suffer from low employee morale and commitment towards their work and organization in general. For instance, Shaban, Al-zubi, Ali, and Alqotaish (2017) reported low employee morale results in poor employee performance, which negatively affects business productivity. To understand employee performance better, Pradhan and Jena (2016) categorized employee performance into three dimensions: task, contextual, and adaptive performance. Task performance plays a critical role in organizational success. Kehoe, Lepak and Bentley (2018) defined task (or in-role) performance as behaviors exhibited by employees, which translate into the achievement of organizational objectives. Contextual performance refers to behaviors exhibited by employees through going out of their way to perform their tasks, including being proactive and cooperating with colleagues to enhance organizational productivity (Borman & Motowidlo, 1993;Motowidlo, Borman, & Schmit, 1997).
Moreover, Pulakos, Arad, Donovan, and Plamondon (2000) explained adaptive performance refers to the capacity for employees to modify and adapt their behavior to suit a new environment or work demands. Low employee performance may be caused by several factors including a poor working environment, low job security, poor employee-employer relationship and, more generally, the company's procedures and policies for rewarding Improving employee engagement practices could enhance employee engagement. Scholars such as Garg, Dar and Mishra (2018) posited that employee engagement is a key factor in enhancing employee performance in organizations. One such dimension is social engagement, which is considered critical for improving employee performance. For instance, Muhammad and Hamdy (2005) found that employees tend to exert more effort when they experience satisfying collegial relationships. Yousef (2017) also argued that supportive colleagues, social interactions and collaboration are all important for enhancing employee performance. Likewise, Kwon, Farndale and Park (2016) reported that the quality of social relationships within an organization influences employee performance.
However, prior empirical literature has mainly focused on investigating the association between employee performance and social engagement without considering the three dimensions (adaptive, contextual, and task) of employee performance. Moreover, the social engagement -employee performance relationship is not a mere direct relationship as previously considered by most peer-reviewed studies, since performance is a multi-dimensional variable that can be influenced by several factors (Ahmad & Ghani, 2010). Other factors may mediate or moderate the association between employee performance and social engagement.
There have also been contradictory findings on the influence of social engagement on employee performance. For instance, Sittar (2020) found a weak, positive correlation between engagement and employee performance. Conversely, Riyanto, Endri and Herlisha (2021) reported no direct relationship between employee engagement and employee performance, except through the mediation effect of motivation and job satisfaction. Prior empirical literature has also mainly focused on sectors such as healthcare (Nasurdin, Ling, & Khan, 2018), education (Koech & Cheboi, 2018 horticulture (Otieno, Waiganjo, & Njeru, 2015), and banking (Dajani, 2015). This study, on the other hand, focused on the textile sector in Tanzania.
Previous contradictory findings could be attributed to the conceptualization of the social engagement -employee performance relationship as a mere direct relationship. However, employee performance is a multidimensional variable influenced by several factors, including gender diversity. Kyalo and Gachunga (2015), Krishnan (2020), and Odhiambo, Gachoka, and Rambo (2018) have previously reported on the influence of gender diversity on employee performance. Moreover, Panicker and Balu (2018), and Onwuchekwa, Onwuzuligbo, and Ifeanyi (2019) found that employee engagement influences gender diversity. With reference to Baron and Kenny (1986), if employee engagement dimensions such as social engagement influence gender diversity, and gender diversity influences employee performance, there is a possible mediation effect. However, there is a dearth of literature on the mediation effect of gender diversity on the influence of social engagement on employee performance. Therefore, this study examined the influence of social engagement on employee performance dimensions, considering the mediation effect of gender diversity.

Theoretical Literature Review
The current study was grounded by Kahn's theory of employee engagement (Kahn, 1990). This theory is based on two premises. The first premise is that psychological experiences of work drive people's attitudes and behaviors, and that employees get engaged if three conditions (psychological availability, psychological safety, and psychological meaningfulness) are present. Psychological availability refers to mental well-being and feelings that sufficiently drive employees' commitment to given tasks. Psychological availability helps an individual determine how to engage with their role or colleagues (May, Gilson & Harter, 2004). Individuals who experience psychological availability can control and devote their physical, intellectual and emotional energy towards their role performance.
Psychological safety refers to the provision of safety at work, allowing employees to immerse themselves into their roles. It involves the ability of individuals to express their authentic selves at the workplace without the fear of negative repercussions with respect to their career, self-image or status (Kahn, 1990). Psychological meaningfulness refers to employees' perception on whether their roles or tasks are meaningful to the organization and within the context of the wider society.
The second premise states that employee experiences are influenced by the simultaneous interaction of organizational, intra-and interpersonal, group, and intergroup factors. Hence, this study applied Kahn's second premise of employee engagement to explain social engagement as being determined by the psychological availability, psychological safety and psychological meaningfulness of an employee, not as an individual, but as a group member.
Since Kahn's engagement theory only explains the concept of social engagement, it cannot be used to explain the link between social engagement and employee performance. Given this limitation, Social Exchange Theory (SET) was applied to explain the social engagement -employee performance relationship. SET focuses on the state of reciprocal interdependence to explain the influence of employee engagement on employee performance. Robinson, Perryman and Hayday (2004) have reported that engagement is bidirectional between the employer and employee. Hence, the reaction of an employee towards engagement may be determined by the actions of their employer.
The influence of social engagement on employee performance is not a mere direct relationship; it can be influenced by several contingent factors including gender diversity. The contingency theory was used to account for gender diversity as a mediator. Contingency perspectives suggest an interrelationship within and among subsystems, as well as between the organization and its environment (Kast & Rosenzweig, 1973). The theory states that an organization's effectiveness depends on matching internal organizational characteristics with other organizational variables or work demands necessary for achieving the organization's goals (Lorsch & Morse, 1974;Woodward, 1965;Thompson, 1967). Therefore, central to the contingency theory is the importance of contextual factors as determinants of an organization's effectiveness (Birkinshaw, Nobel, & Ridderstråle, 2002). Schwab, Werbel, Hofmann, and Henriques (2015) argue that there is a contingent relationship between social engagement, gender diversity and employee performance. Gender diversity may potentially influence performance when men and women exhibit diverse knowledge, thinking, information, and problem-solving capacities.
For instance, Park (1996) reported that men and women must have different evaluative criteria in assessing alternatives to promoting creativity and problem-solving skills. Gender diversity mediates the relationship between social engagement and employee performance dimensions since this relationship is contingent not only on the number of women and men, but also on the diverse thinking, information, knowledge, and heterogeneity among employee groups that is necessary for effective problem-solving. Following this argument, this study examined the mediation effect of gender diversity from a contingency perspective.

Empirical Literature Review
Employee performance is a crucial contributor to the success of any business organization. Saeed and Asghar (2012) explain employee performance as the successful completion of tasks by an individual based on the organization's goals. According to Sultana, Irum, Ahmed, and Mehmood (2012), employee performance involves achieving particular goals benchmarked against predetermined standards. Scholars conceptualize employee performance differently. For instance, Parker, Williams, and Turner (2006) explain the aspects of employee performance as including proactive work behavior, problem-solving, and idea implementation. Schepers, Jong, de Ruyter, and Wetzels (2011) argue that employee performance consists of work performance and disciplined effort. Charbonnier-Voirin and Roussel (2012), on the other hand, attest that employee performance includes interpersonal adapatability, reactivity to difficult situations, creativity, and handling work-related stress.
Koopmans, Bernaards, Hildebrandt, de Vet, and van der Beek (2013) argue that employee performance includes task and contextual performance, and behaviors towards counterproductive work behavior. Finally, Pradhan and Jena (2016) argue that employee performance is categorized into three groups: adaptive performance, contextual performance, and task performance. Task performance includes job-explicit behaviors considered fundamental to the role outlined in the job description. Adaptive performance refers to employees' ability to provide support necessary to the job profile in the face of an ever-changing work environment. Contextual performance is a crucial dimension of employee performance that involves one's willingness to exhibit voluntary behaviors useful to the organization, such as volunteering for extra work, and team spirit (Spector & Fox, 2002). Individuals who experience contextual performance perform beyond the call of their specified duties by taking personal initiatives at their workplace (Frese, Kring, Soose, & Zampel, 1996).
The success of any business depends on various practices, including employee engagement practices. Avery and Cameron (1998) posited that employee engagement is important for contextual performance. This is also supported by Garg et al. (2018), who posited that employee engagement is key to improving employee performance within the organization. Jackson, Colquitt, Wesson and Zapata-Phelan (2006) argue that social engagement is of importance at work, and that there is an increasing acknowledgement of employees working collectively. In support of this, Nasurdin and colleagues (2018) reported that peer support enhances employees' contextual performance. Similarly, Rubel and Kee (2013) found that supervisors' and organizational support, as an aspect of social engagement, positively influences employee performance.
The volatile nature of the business environment makes adaptive performance critical for individual employees and the organization in general (Park & Park, 2019). Adaptive performance explains the extent to which individuals are flexible in an ever-changing business environment (Shoss, Witt, & Vera, 2012). Previous empirical studies have established a positive association between social engagement and adaptive performance. Scholars such as Park, Lim, Kim, and Kang (2020), who investigated organizational support as an aspect of social engagement, found that organizational support positively affected adaptive performance. Similarly, Kaltiainen and Hakanen (2020) reported that an increase in social engagement was positively associated with adaptive performance.
Social engagement was also found to be significantly and positively associated with employee task performance. Bakker and Demerouti (2007) assert that employees who are engrossed in their work exhibit better task performance. This is supported by some studies, such as that of Farooq, Zia-Ud-Din, Iram and Nadeem (2018) who reported that employee engagement positively influences task performance. Similarly, Ma'arof and Mat (2019) investigated the effect of employee engagement on nurses' job performance in Malaysian public hospitals, and found a positive, significant relationship between employee engagement and task performance. Other researchers also found a positive association between employee engagement and task performance (Alfes, Shantz, Truss, & Soane, 2013;Yin, 2018;Yongxing, Hongfei, Baoguo, & Lei, 2017). Moreover, Bakker, Demerouti and Brummelhuis (2012) reported positive associations between work engagement and contextual performance, as well as work engagement and task performance. Results from a meta-analytic study by Christian et al. (2011) established a positive association between task performance and engagement. Employee engagement is determined by several interactive factors within the workplace, such as gender diversity. Gender diversity involves a good mix of male and female in organizations and in the allocation of jobs. Existing literature supports that gender diversity is positively related to improved employee performance. Mwatumwa (2016) confirmed that gender diversity significantly influences employee performance, which translates into greater organizational performance. Numerous studies have further supported a positive association between gender diversity and employee performance (Elsaid, 2012;Krishnan, 2020;Odhiambo, Gachoka, & Rambo, 2018;Rizwan, Khan, Nadeem, & Abbas, 2016;Khan, Sohail, Sufyan, Uddin, & Basit, 2019). In a similar vein, Badal and Harter (2013) assert that workforce diversity can have a major impact on employee engagement. Onwuchekwa et al. (2019) also found a direct relationship between gender diversity and employee engagement. Thus, as a dimension of employee engagement, social engagement could exhibit similar outcomes.

Research Hypotheses
Based on the previous literature this study hypothesized the following: H1: Social engagement positively influences contextual performance.
H3: Social engagement positively influences task performance.
H4: Gender diversity mediates the relationship between social engagement and employee performance.

Participant Characteristics
The scope of the study was based on four regions in Tanzania: Dar es Salaam, Mwanza, Shinyanga, and Simiyu. Region selection was based on the density of manufacturing industries, specifically textile firms. Dar-es-Salaam has the highest density of manufacturing firms, including textile firms, whereas Mwanza and Shinyanga regions' economies are dominated by textile firms, and serve as the main source of income for the livelihoods of people in these regions. Similarly, Simiyu was chosen because it represented regions with relatively lower business density, but whose economy is also dominated by the textile sector (National Bureau of Statistics [NBS], 2014). All employees included in the sample, irrespective of their level within the firms, participated in the study. Additionally, employees were both the unit of inquiry and unit of analysis, since they were in a better position to explain their level of social engagement as they performed their work, towards their colleagues, and how their performance was influenced.

Sample Size, Power, and Precision
The population of the article was 55,778 textile firm employees (NBS, 2016) and the sample size was 618 employees. Out of the 618 employees to whom questionnaires were distributed, 554 returned the questionnaires. Thus, the response rate was 89.6% (see table A1, appendix A). Respondents were selected using a probability sampling technique, which allows for generalization of the findings (Sekaran & Bougie, 2010). The study adopted a cluster sampling technique, where employee selection was based on the four geographical regions.
After identifying the adequate sample size for each region, workplaces were randomly selected for participation, and employees were subsequently randomly selected as respondents. The sample size for Dar es Salaam region was 203 employees, 173 employees for Mwanza, 117 employees for Shinyanga, and 125 employees for Simiyu.
The Taro Yamane formula was used to calculate the sample size of the study: where n = sample size, N = population size, and e = error margin.

Measures Used
Primary data were collected using questionnaires. In order to ensure the internal consistency of the study variables, Cronbach's alpha analysis was conducted. Results revealed that Cronbach's alpha coefficient for all variables was above 0.7, indicating adequate internal consistency as suggested by Nunnally (1967). Cronbach's alpha coefficient for social engagement was 0.939, 0.823 for adaptive performance, 0.922 for contextual performance, 0.843 for task performance, and 0.842 for gender diversity.

Research Design
This study adopted a cross-sectional design, where data were collected once at the same point in time. The cross-sectional design was deemed appropriate for this study since it facilitates rapid and uniform data collection across many respondents at one time (Wakahiu, Ngigi, & Karanja, 2016). Hence, this design was employed to collect data from a sample of the population to estimate the relationship between the variables of interest.

Data Analysis
For the sake of ranking the variables of this study, mean scores were used for the descriptive analysis. Mean scores were categorized into low, moderate and high scores. According to Oxford (1990) and Oxford and Burry-Stock (1995), the low, medium and high mean score ranges are 1-2.4, 2.5-3.4, and 3.5-5.0, respectively. Structural equation model (SEM) was used for the data analysis to test the hypothetical relationship between social engagement and employee performance dimensions. The study adopted SEM because of its capacity to accommodate many variables in the same model.

Recruitment
Participants were recruited from November, 2020 to April, 2021. Data from Dar es Salaam region were collected from November 2020 to December 2020, from January 2021 to February 2021 in Mwanza region, and from March 2021 to April 2021 in Shinyanga and Simiyu regions. Focusing on education levels, 18 respondents (3.2%) had no formal schooling, 117 (21.1%) were of primary school qualification, and 178 (32.1%) had received an Ordinary-level certificate. Moreover, 50 respondents (9.0%) were Advanced-level certificate holders, 81 (14.6%) held a degree/advanced diploma, and 3 respondents (0.5%) had postgraduate qualifications. Thus, the descriptive findings demonstrated that the highest level of education for the majority of respondents was Ordinary-level education.

Inferential Results
The study hypothesized that social engagement would positively influence employee performance dimensions. Under social engagement, there were hypotheses H 1a , H 1b , and H 1c . H 1a hypothesized that social engagement would positively influence adaptive performance, H 1b hypothesized that social engagement would positively influence contextual performance, and H 1c hypothesized that social engagement would positively influence task performance. Findings of this study revealed that social engagement had a positive influence on adaptive, contextual performance and task performance. The findings showed that a unit change in social engagement resulted in a change in adaptive performance by 0.110, significant at a p-value of 0.018. In addition, a unit change in social engagement led to a change in contextual performance by 0.395, significant at a p-value of 0.001. It was also revealed that a unit change in social engagement resulted in a change in task performance by 0.392, significant at a p-value of 0.001.
It was established that social engagement significantly influenced adaptive performance (p = 0.018), contextual performance (p = 0.001) and task performance (p = 0.001), which means that all hypotheses were accepted. Figure  1 and appendix B provide more details on how employee engagement and the dimensions of employee performance are related. The study examined the mediation effect of gender diversity on the relationship between employee engagement and employee performance. There were three hypotheses, namely H 4a , H 4b , H 4c . H 4a predicted that gender diversity would the relationship between social engagement and adaptive performance. H 4b predicted that gender diversity would mediate the relationship between social engagement and contextual performance. Lastly, H 4c predicted that gender diversity would mediate the relationship between social engagement and task performance.
According to Baron and Kenny (1986), there are several conditions for a mediating variable. These conditions include: first, that the independent variable must significantly influence the dependent variable; second, that the independent variable should have a significant influence on the mediating variable; third, that the mediating variable should have a significant influence on the dependent variable; fourth, that the estimates of the direct model should decrease when the mediating variable is introduced in the model; and finally, that if the direct relationship remains significant when the mediator is introduced, then there is partial mediation. Conversely, if the direct relationship is non-significant when the mediating variable is introduced, then there is full mediation.
Gender diversity had no mediation on the relationship between social engagement and adaptive performance because gender diversity had no significant influence on adaptive performance. According to Baron and Kenny (1986), the mediator must have a significant influence on the dependent variable for the mediation effect to occur.
The results further revealed that gender diversity partially mediated the relationship between social engagement and contextual performance. The estimates on the direct relationship between social engagement and contextual performance decreased from 0.395 to 0.244, but remained significant at p = 0.001. Additionally, the influence of gender diversity on contextual performance was significant at p = 0.001, while the influence of social engagement on gender diversity was significant at p = 0.001. Therefore, the conditions for a mediation effect were met.
Furthermore, the results showed that gender diversity had a partial mediation effect on the relationship between social engagement and task performance. The estimates on the direct relationship between social engagement and task performance decreased from 0.392 to 0.194, but remained significant at p = 0.001. Additionally, the influence of gender diversity on task performance was significant at p = 0.001, while the influence of social engagement on gender diversity was significant at p = 0.001. Thus, the conditions for a mediation effect were met.

Main Findings
The study results are in line with findings from prior empirical studies such as those of Nasurdin et al. (2018), who reported that peer support significantly influenced employee performance. Yousef (2017) found that positive co-worker relationships led to greater employee commitment and lower rates of absenteeism. Strachan (2015) also argued that positive feedback and a supportive environment allow for feelings of social belongingness, which leads to improved performance. This is in line with Yousef (2017) and Jagannathan (2014), who reported that the environment, team and co-worker relationships significantly influenced employee performance.
The aforementioned discussion on various prior literature reveals that most researchers have been attracted to study the influence of social engagement on employee performance. However, their findings differ from the results of the current study, which focused on the three dimensions of employee performance, rather than examining employee performance in its totality. Interestingly, findings of the current study reveal that social engagement has differing influences on individual employee performance dimensions, which is further supported by the social exchange theory.
Study findings revealed that organizations that foster an environment conducive to employees' social engagement also increase the likelihood that their employees will demonstrate better adaptive, task and contextual performance. This is supported by the social exchange theory's main statement, which asserts that employees' reactions are determined by the actions of their employer.
The results on the mediating effect of gender diversity are an extension of prior empirical work conducted on the role of gender diversity on employee performance. In their study, Krishnan (2020) revealed that gender diversity significantly influenced employee performance. Likewise, Kyalo and Gachunga (2015) also found that gender diversity had a significant influence on employee performance. However, the findings of this study differed from those of prior empirical literature since the present study focused on the mediation effect of gender diversity.

Conclusion
This study examined the influence of social engagement on employee performance dimensions. It concluded that social engagement had a positive, significant influence on adaptive, contextual and task performance. In this regard, employee engagement aspects such as information sharing on employees' goals and work attitudes create an environment for employees to demonstrate better contextual and task performance. Additionally, caring about others' well-being and opinions, and consideration of colleagues' goals and values, can all play a part in promoting a team spirit among employees, thereby resulting in better contextual and task performance.
This study also examined whether gender diversity mediates the relationship between employee engagement and employee performance dimensions. Findings revealed that it partially mediated the influence of social engagement on task performance and contextual performance. In this regard, a good mix between men and women; equal opportunities for managerial positions, an optimal mix of the genders in job allocation; equal opportunities in career development, and a gender quota policy within the organization all provide an environment for employee engagement dimensions to influence employee performance. Hence, a conducive internal environment within the organization in terms of its gender diversity is of paramount importance in promoting employee performance, including task and contextual performance.

Recommendations
Social engagement had an influence on adaptive, contextual and task performance. By socially engaging their employees, managers could successfully promote employees' adaptive, contextual and task performance. Hence, managers should ensure that employees share their work and goals with colleagues. Managers should also encourage employees to show concern for each other's opinions, care for one another's well-being, and respect each other's values.
Gender diversity was found to have a partial mediation effect on the influence of social engagement on contextual and task performance. In this case, organizations are encouraged to promote gender diversity in order for engaged employees to demonstrate better performance. Managers should ensure that there is a good mix of men and women at the workplace, that managerial appointment positions are based on merit rather than gender, and that there is a mix of genders in job allocation. In addition, equal opportunities for men and women in their career development should be encouraged, as well as ensuring that a gender quota policy on recruitment and retention is instituted and implemented.
Nevertheless, the study did not categorise its findings across different business sizes, including micro, small, medium and large businesses. It is possible that social engagement has different influences on employee performance dimensions across different business sizes. In addition, future studies may consider using both qualitative and quantitative methods to explore the reasons why gender diversity had diferent medition effects on the relationship between social engagement and employee performance dimensions. Although this limitation does not undermine the existing findings, it focuses on providing more results on the influence of social engagement on employee performance. Future studies should study the effect of social engagement on employee performance dimensions based on differing categories (i.e., micro, small, medium, large) of business sizes.  Source: Field data (2020).