How Changes in Job Resources and Personal Resources Predict Employee Engagement

This study examines how job resources (i.e., job autonomy, supervisor support, and reward and recognition) and personal resources (i.e., self-efficacy and optimism) affect employee engagement with special reference to managerial employees apparel manufacturing industry in Sri Lanka. This explanatory study is positivistic and deductive, which adopted the survey strategy. The present study uses the mono method in data collection and analysis. A standard questionnaire was used to collect data, and analysis was done using SPSS. The hypothesized model of the study was tested using simple linear regression. Results revealed a positive impact of job resources and personal resources on employee engagement. This study contributes to the existing knowledge while providing many managerial implications.

Open Journal of Business and Management attitudes, absenteeism, turnover, individual, group and organizational performance, quality of customer experience, and customer loyalty. Maynard (2016) mentioned that an engaged employee is productive by two times than a disengaged employee, and that will ensure a better service for customers while leading the organization towards more profits. Even though work engagement has many positive consequences, there is an engagement crisis all over the world.
Gallup studies (http://www.gallup.com/, 2013) have shown that only 13% of employees are engaged in their job, while 63% of employees are not engaged.
The remaining is actively disengaged. In the United States, only 34% of employees are engaged (http://www.gallup.com/, 2018). Further, the estimated cost of the lost productivity of actively disengaged employees is USD 483 billion to 605 billion (http://www.gallup.com/, 2017). In Sri Lanka, 14% of employees are engaged, and 62% of the workforce is not engaged. It showed that 23% of employees were actively disengaged (Crabtree, 2013). With this evidence, few discussions on employee engagement were conducted with managerial personnel attached to the apparel industry in Sri Lanka. These discussions revealed that many employees are not engaged fully in their jobs. Besides, a pilot survey was conducted using managerial level employees. Its findings show that 11% of employees are highly disengaged, and 66% of employees are moderately engaged.
Only 23% of employees are highly engaged. Accordingly, the majority of employees are not fully engaged in their job. Hence, this study examines why managerial employees are not engaged in their jobs with special reference to the apparel sector in Sri Lanka?

Literature Review and Hypotheses Development
One of the common definitions for work engagement is a positive, fulfilling, work-related state of mind that is characterized by vigor, dedication, and absorption (Schaufeli, Salanova, Gonzalez-Roma, & Bakker, 2002). As stated, there are many positive consequences of engagement towards individuals, groups, and organizations. For example, engagement increases the innovative work behavior of employees (De Spiegelaere, Van Gyes, De Witte, Niesen, & Van Hootegem, 2014), better job performance , organizational commitment (Geldenhuys, Laba, & Venter, 2014). Different researchers have identified many factors that affect employee engagement. For example, Saks (2006) showed job characteristics, perceived organizational and supervisory support, reward and recognition, procedural and distributive justice as influential factors over engagement. Moreover, training and development practices (Thavakumar & Nawaratne, 2015), transformational leadership (Thisera & Sewwandi, 2018), career development and work content are some other antecedents of employee engagement. Moreover, some researchers (e.g., Carter, Nesbit, Badham, Parker, & Sung, 2018;Kotze, 2018) examined how job and personal resources affect work engagement. Especially, Demerouti, Bakker, De Jonge, Open Journal of Business and Management ent job characteristics as influential factors in explaining various aspects of employee work lives (e.g., job strain, burnout, and engagement). This model categorized job demands and resources as the main working conditions of a job while emphasizing the underlying assumption indicating that job resources are positively related to work engagement . Accordingly, job resources may include social support, growth opportunities, organizational support, job security, and opportunity for advancement (Coetzee & De Villiers, 2010). The JD-R model suggested that personal resources have an equal motivational impact on work engagement (Xanthopoulou, Bakker, Demerouti, & Schaufeli, 2007).

Job Resources and Employee Engagement
Many empirical studies (e.g., Bakker, Hakanen, Demerouti, & Xanthopoulou 2007;Salanova & Schaufeli, 2008;Hakanen, Schaufeli, & Ahola, 2008) have shown that job resources impact positively on engagement. Schaufeli and Bakker (2004) described that when employees feel that their jobs are less resourceful, they will leave the organization. If the company provides adequate job resources for their employees, they will remain with the organization. Thus, a lack of job resources increases the disengagement level of employees. Based on this evidence, the first hypothesis is proposed as; H1: There is a positive impact on job resources on work engagement.
This study focuses on three main job resources (i.e., job autonomy, perceived supervisor support, and reward and recognition), which are critical for employees attached to the apparel sector. Job autonomy is the degree of employee sovereignty in making job-related decisions (Hackman & Oldman, 1975 & Bakker, 2008;Saks, 2006) have shown that job autonomy positively influences the engagement. Therefore, the second hypothesis is set as: H1 a : There is a positive impact of job autonomy on work engagement.
Thirdly, supervisory support, which is defined as the degree of employee perception towards their supervisor's enthusiasm for his employees' well-being and the contribution that the supervisor makes (Tuzun & Devrani, 2011) is considered. If the supervisor supports his subordinates to carry out their work effectively, employees will work hard to achieve organizational goals. Further, supervisory support is positively related to work engagement (e.g., Demerouti & Bakker, 2008;Saks, 2006;Xanthopoulou et al., 2007). Based on this empirical evidence, the following hypothesis is suggested. H1 b : There is a positive impact of supervisory support on work engagement.
When looking at the reward and recognition, researchers (e.g., Kahn, 1990) have described that reward and recognition positively impact on work engagement. Hence, the following hypothesis is set. H1 c : There is a positive impact of reward and recognition on work engagement. Open Journal of Business and Management Xanthopoulou, Bakker, Demerouti, & Schufeli (2009) mentioned that personal resources (e.g., self-efficacy, optimism, and organization-based self-esteem) affect the work engagement, next to job resources. Supporting this argument, Prieto, Soria, Martínez, & Schaufeli (2008) found that personal resources control the level of burnout while increasing the level of engagement. Similarly, empirical studies (e.g., Hu, Schaufeli, & Taris, 2011;Xanthopoulou et al., 2007;Xanthopoulou, Bakker, Demerouti, & Schufeli, 2009) have shown that personal resources (e.g., self-efficacy, optimism, organization-based self-esteem) positively impact on work engagement. Accordingly, the following three hypotheses are proposed.
H2: There is a positive impact of personal resources on work engagement H2 a : There is a positive impact of self-efficacy on work engagement H2 b : There is a positive impact of optimism on work engagement Figure 1 shows the conceptual framework of the study.

Research Methodology
The present study is positivistic and deductive research. The purpose of the present study is explanatory, and it explains the causality between job and personal resources and employee engagement. Accordingly, it adopts the survey strategy and follows the mono method. Moreover, this cross-sectional study uses managerial employees attached to the apparel sector in Sri Lanka. As the pilot survey results reveal, executives in the apparel sector show disengaged behaviors at work. Further, the apparel sector is one of the key players in the Sri Lankan economy. Therefore, this sector was chosen for the study purpose. Data collection was done via a self-administered questionnaire included three main parts for demographic information (i.e., age, gender, marital status, highest educational qualifications, and tenure), job resources (job autonomy, supervisor support, reward, and recognition) and personal resources (self-efficacy and optimism). The questionnaire was distributed among 230 managerial level employees, and 174 questionnaires were returned. 167 questionnaires were used for the final analysis after excluding incomplete responses.

Measures
Employee engagement was measured via Utrecht Work Engagement Scale (UWES) developed by Schaufeli, Bakker and Salanova (2006). It consists of seventeen (17) items for measuring three dimensions of engagement (i.e., vigor, absorption, and dedication). This 7-point Likert scale has anchors ranging from never (0) to always (6). Job-Demand Resource questionnaire (Bakker, Demerouti, & Verbeke, 2004) was used to measure job resources and personal resources.
All the items were 5 points Likert scale (Never to very often).

Sample Composition
The sample consisted of a majority (59%) of males and 41% females. The majority of respondents (39%) were in the age group of between 35 -45 years. 64% of the sample had qualified with a bachelor's degree. 20% of respondents have a postgraduate degree. Others have passed the advanced level examination.

Hypotheses Testing
Preliminary analysis (i.e., normality, linearity, homoscedasticity, reliability, validity, multicollinearity, common method variance) was conducted before testing hypotheses. Moreover, simple linear regression was used to test the hypotheses of the study.
As shown in Model 1 (Refer Table 1), R-value is .71 and R 2 .51. It denotes that aggregate job resources explain the variance of employee engagement by 51%.
The p-value is .00, which is less than .05. The Beta value .71, and it is positive. Accordingly, H1 is supported. Hence, it can be concluded that job resources positively affect employee engagement. Job autonomy also positively impacts on employee engagement. The beta value is .73, and the p-value is .00. Therefore, H1 a is supported. Job autonomy explains 32% (R 2 = .32) of variance in employee engagement. Model 3 shows that supervisor support positively affects employee engagement. The beta value is .80. R-value is .65 and R 2 is .43. P-value is .00. Therefore, H1 b is supported. When looking at the reward and recognition, it shows a positive beta value of .88. P-value is .00 and R 2 is .38. 38% of the variance of the dependent variable is explained by reward and recognition. Hence, H1 c is accepted. Table 2 depicts the statistics of regression analysis for personal resources and employee engagement. Model 5 (Refer Table 2) shows a positive beta value of 1.01. P-value is .00. Moreover, R 2 is .34. It means that personal resources explain the variance of employee engagement by 34%. Therefore, H2 is accepted. Model 6 regressed employee engagement by self-efficacy. Beta is .89 and p-value is .00.
Accordingly, H2 a is accepted. H2 b is also statistically supported. P-value is .00, beta value is .89. R 2 is .29, which explained 29% of the variance of employee engagement. Accordingly, self-efficacy positively influences employee engagement.
When considering optimism as a personal resource, results show a beta value Open Journal of Business and Management  of .66, and the p-value is .00. Hence, H2 b is accepted. Further, 23% (R 2 = .23) of the variance of employee engagement is explained by the optimism of a person.

Discussion
This study focused on the impact of job resources (i.e., job autonomy, supervisor support, reward, and recognition) and personal resources (i.e., self-efficacy, optimism) on employee engagement. The findings of the study revealed that job resources and personal resources positively affect the engagement of employees.
These findings are consistent with previous studies (e.g., Hakanen et al., 2008;Salanova & Shaufeli, 2008 It also showed that supervisor support positively affects employee engagement. As shown by many researchers (e.g., Kahn, 1990;Saks, 2006), reward and recognition has a positive impact (Beta = .66, p-value = .00, R 2 = .38) on en-T. J. R. Thisera, G. A. D. S. Wijesundara Open Journal of Business and Management gagement. When looking at personal resources, the study revealed a positive impact of personal resources on engagement. These findings are consistent with previous studies (e.g., Bakker & Woerkom, 2017). Further, supporting previous findings (e.g., Mache et al., 2014;Hu et al., 2011), the present study found that employee engagement is positively affected by self-efficacy and optimism of a person.

Implications
There are limited research studies on employee engagement (e.g., Thisera & Seewandi, 2018;Wickramasinghe & Perera, 2014) and job characteristics (Thisera & Nawaratne, 2018) in Sri Lanka. Iddagoda, Opatha and Gunawardana (2016) described that there is an empirical gap regarding work engagement and personal resources in Sri Lanka. Therefore, this study nurtures the existing literature while fulfilling the knowledge gap. Further, the findings provide many insights for managerial employees, decision-makers, and policymakers in organizations. Based on the results, managers can enhance job resources (e.g., job autonomy, supervisor support, reward, and recognition) to improve the engagement level of managerial employees. Notably, it is necessary to take actions for enhancing job autonomy and creating favorable interpersonal relations. Further, an attractive reward and recognition plans should be used. When looking at personal resources, managers should focus on improving self-efficacy and optimism within employees via different training and counseling programs.

Limitations of the Study
The present study is cross-sectional and used only a single method of data collection and analysis. Furthermore, it considered only executive and above level employees attached to the apparel industry in Sri Lanka. Selected job resources and personal resources were used for the study purpose. Therefore, it is recommended to take the necessary actions to overcome these limitations in future research.