Impacts of Perceived Role Demands on Work-Life Balance and Moderating Effects of Work Ethics: Evidence from Public Sector Professionals in Sri Lanka

Work, family and social life are considered to be the most important spheres for an individual’s life. The current study focuses on to determine the level of work-life balance, to analyze the impacts of perceived role demands on work-life balance and to estimate the moderating effects of work ethics between role demands and work-life balance relationship among professionals in Sri Lanka. The sample consists of 386 professionals of Chartered Engineers, Medical Officers and Accountants employed in the public sector organizations in Sri Lanka. Explanatory research design with quantitative research approach of cross sectional survey method was adapted. Primary data was collected using Questionnaire survey with stratified random sampling techniques. Data were analyzed using Structural Equation Modelling approach with Amos 21 and SPSS statistics 23. The Multi-Group Analysis in Amos also has been applied for testing the moderation effect of work ethics. This study found that professionals possess a moderate level of work-life balance and there is significant and negative causal impacts of perceived work and social demands on work-life balance while combined effects of perceived role demands have significantly and negatively impact on work life balance. Moreover, work-life balance is skewed towards works and less in family and social role demand is an emerging issue for professionals. Further, work ethic has significant and partial moderation effect between role demands and work-life balance relationships. This study is significant and beneficial for managers, organizational leaders and researchers to address the needs of employees to develop strategies and policies to address work life balance issues and social implications for employees, family members, societies and researchers. The limitation of the study is that all of the measures were self-reported that common method variance may raise concern. Future studies using longitudinal design would be useful in establishing the temporal causal relationship among the private sector professionals in Sri Lanka.


Introduction
Sri Lanka is a beautiful Island situated in the Indian Ocean with a total population of 21 million. The labour force participation rate of female in Sri Lanka has increased significantly from 25 percent in 1986 and 34.8 percent in 2019 (Dept. of Census, Sri Lanka, 2020). It has been witnessed from literature that major changes have been occurred in work, family and social life domains in terms of economic, demographic, work, social and political contexts in Sri Lanka. Work, family and social domains are vital and inter-connected to people and performing of multiple roles negatively impacts on individuals, families and employing organizations significantly. Mass entry of women into the labour force particularly in service sector leading to increase numbers of dual earners, single parents, dual career couples and elder care related issues. Besides, gender values in terms of male as a breadwinner and female as a home maker ideology is also changing dramatically. Breitenecker and Shah, (2018) stated that work-life balance has indeed become a primary concern to those wishing to have good quality of life. According to Sullivan (2019), research has still indicated that there are significant disparities between men and women pertaining to the work-family balance. Moreover, increase use of information technology and advancement in technology that have made the path to easy access to employer and employee is also caused negative interference. Furthermore, work pressure is imposing burdens on professionals and facing difficult in balancing their work and personal life. Balancing a successful career for professionals with personal and family life is indeed challenging due to the fact that the higher the position, the more time and energy that professional must invest in work and life roles. Duxbury et al. (1993) claimed that professionals work longer hours than other groups of workers, which can amplify work-family conflict. Many researchers have claimed that there are interactions between work and life spheres (Clark, 2000;Greenhaus, Collins and Iason, 2003). Under these circumstances, it is evidence that professionals encounter work-life balance issues. Clark (2000) defines work-life balance as satisfaction and good functioning at work and at home with a minimum role of conflict. Devadoss (2013) concluded that work -life imbalance is result to increase the problems related to employee at work place, decrease the productivity and efficiency and cause to increase social hazards such as increasing number of divorces, increase stress levels. Role demands are structural or psychological claims associated with role requirements, expectations and norms to which individuals must respond or adopt by exerting physical or mental effort (Voydanoff, 2004). Boyar et al., (2008) claimed that despite the volume of research explaining the relationship of role demands and work-family balance, role demand conceptualization is problematic as many of the past studies measure role demands objectively. Hence, to fill this gap, perceived role demands were considered in this study. Work ethics is an important work-related construct. Work ethic is basically the belief that work is a good moral and is defined as a set of beliefs and attitudes reflecting the fundamental value of work (Miller et al., 2001). Thus the purpose of this research is to determine the level of work-life balance, to analyze the impacts of perceived role demands on work-life balance and to estimate the moderating effects of work ethics between role demands and work-life balance relationship among professionals employed in the public sector of Sri Lanka.

Work-Life Balance (WLB)
The origin of the research of work -life balance was found from the seminal work of Rapport and Rapport in1969 and it spells out that both work and family needs time and energy. In defining work-life balance, Work means paid employment while Life means activities outside of paid employment including activities in the household and with friends, family and community. Numerous authors have explained different definitions, conceptualizations, causes, measures and outcomes of work-life balance. Kelliher, Richardson & Boiarintseva (2018) claimed that academic knowledge around the WLB concept is not as solid and extensive and WLB is a concern mainly for working parents, where caring for dependent children is the relevant load in their life. Further, there has been some debate in the literature about the appropriate terminology to use, with some authors preferring terms such as work-life interface instead (Kelliher, 2016). Wayne et al., (2017) claimed that balance is a different concept from conflict and enrichmen. According to Oludayo et al., (2018), WLB is a concept designed to improve job predictability and career sustainability of employees. Work -life balance is achieving satisfying experiences in all life domains and to do so requires personal resources such as energy, time and commitment to be well distributed across domains (Kirchmeyer, C., 2000). Moreover, based on a critical review of literature, work-life balance conceptualization was classified into seven categories as shown in Table 1.  .
The degree of satisfaction with work and family role. Casper, et al., (2018).

2.Belief
Extent to which employees hold a favorable evaluation regarding their combination of work and non-work roles, arising from the belief that their emotional experiences, involvement, and effectiveness in work and non-work roles are commensurate with the value they attach to the roles. 3. Fit Crooker, et al., (2002).
A fit between the demands of multiple roles and the availability of personal resources. Frone, (2003). Low levels of conflict and high levels of inter-role facilitation. 5.Social Norms 6.Engagement Grzywacz & Carlson,(2007) Marks & MacDermid, (1996.

4.Equilibrium
Accomplishment of role-related expectations that are negotiated and shared between an individual and his/her role partners. The tendency to become fully engaged in the performance of every role in one's total role system, to approach every typical role and role partner with an attitude of attentiveness and care. 7.Miscellaneous (Others) Hayman (2005) evaluated 15 items scale of work-life balance adapted from an instrument reported by Fisher et al. (2003) and validated to capture work-life balance measure that is widely used. Theories and models relevant to work-life balance is explained into four perspectives called as conflict perspective, compensation perspective, enrichment perspectives and integration perspectives. Moreover, the working definition of work -life balance is the extent to which an individual combines, prioritizes, attunes and fulfills work, family and social role obligations that represent with experiences of satisfactions, positive gains by interactions and minimum interferences. Katz and Kahn (1978) characterize a role as the building block of social system and the summation of the requirements with which system confront their members as individuals. Demands of a role is the association with responsibilities, requirements, expectations, duties, and commitments with a given role. According to Boyar et al., (2007), demand is a global perception of the level and intensity of responsibility within the work (or family) domain. Additionally, researchers have defined both work and family demand concepts to include variables such as number of hours worked, schedule, role conflict, role ambiguity, role overload, stress from work, marital status, family work hours, number of children at home, and dependents at home (Carlson & Kacmar,2000;Voydanoff,1988). Moreover, professionals and managers may have more access to greater flexibility relative to other workers enabling them to better cope with the heavy work and family demands (Eaton, 2003;Rau and Hyland, 2002). According to current authors, social demands refer to obligations and commitments of friends and relations, neighbors, community and other relevant parties beyond work, family and self domains. Actually perceived work, family and social role demands are the constructs of interest in this umbrella term of perceived role demands.

Work Ethics
Work ethic is an important work-related construct and basically it is the belief that work is a good moral. Weber (1958) claimed that people would strive for success and wealth because the values associated with the work ethic would become entrenched in society. Work ethic may be defined as a set of beliefs and attitudes reflecting the fundamental value of work (Miller et al. 2001). For the current study, work centrality, self-reliance and hard work dimensions of work ethics were selected as these factors influence more on role demands and work-life balance relationship. Work centrality is defined as individuals' beliefs regarding the degree of importance that work plays in their lives (Paullay et al., 1994). Self-reliance is the individual ability to depend on self to get things done, to meet the needs and make decisions without need of other people. Hard work is the very important Work Ethics work ethic and assess one's belief in the virtues of hard work.
Thus, the objectives of this current study was to determine the level of perceived work-life balance and to analyze the impacts of perceived work, family and social role demands on work-life balance with the moderating effect of work ethics among the professionals employed in the public sector of Sri Lanka.

Theoretical Model
On the basis of the literature review, the following theoretical model (Figure 1) has been proposed for the study.

Independent Variable
Moderating variable Dependent Variable

Hypotheses
From the literature review, Sheikh et al. (2018) found that work demands and home demands cause significant effects on work-life balance. In the work of Semlali and Hassi, (2016), it was found that balance is achieved when there is harmony between work and life. Aziz and Chang (2013) claimed that work and family demands effect significantly on work-life balance. Based on the conceptual model, the literature review and the finding of the above researchers, the following hypotheses were postulated.
H1a: Perceived work, family and social role demands have a significant and negative impact on work-life balance.
H1b: Combined effects of perceived work, family and social role demands have a significant and negative effects on work-life balance.
H2: Perceived work role demand has the strongest relationship with work-life balance than that of perceived family or social role demands and work-life balance relationships.
H3a: Work Ethics has a moderating effect on the relationship between work-life balance and perceived work, family and social role demands.
H3b: The effects of perceived work role demands on work-life balance will be different between the group with high level of Work-Ethics and the group with low level of Work-Ethics.

Theoretical Lenses for the Overall Model
Theoretical frame work was specified by two theories namely spillover theory and work-family border theory.
2.6.1 Spillover Theory (Pleck, 1977, & Lambert, 1990 Spill-over theory (Pleck, 1977) has been the most popular theory for examining the work-family interface. According to this theory, workers carry the emotions, attitudes, values, moods, skills and behaviors from their work role into their family life and vice versa (Lambert, 1990). This theory states that spillover can occur in both directions -work to personal life and personal life to work and can have positive or negative effects.
2.6.2 Work-Family Border Theory (Clark, 2000) Border theory persists that work and family are separate but mutually influential domains. It is a theory about work/family balance that explains how individuals manage and negotiate the work and family spheres and the borders between them to attain balance and roles are separated and divided by the boundaries which could be physical, psychological and emotional (Clark, 2000).

Perceived social role demand
Perceived work role demand Perceived family role demand Work -Life Balance

Research Design
A cross sectional survey with quantitative data collection approach has been used in this research study. Self-reported structured questionnaires were used to measure the relevant constructs. To meet the research objectives, a structural equation model was developed.

Population and Sample
The population of this research study were twenty four thousand professionals (24,000) of Chartered Engineers, Medical Officers and Accountants employed in the public sector organizations in Sri Lanka. Sekaran (2003) claimed that a population of 10,000 or more, a researcher would probably consider a sample of 384. According to Yamane's formula, Sample size n = N/1+N (e) 2 where N = Total population, e = Margin of error (0.05), n = 24000/ 1+24000 (0.05)2 = 393.
Hence, sample size for this study was taken as 386 using stratified simple random sampling techniques.

Procedure
Before starting the research study, a pilot study was conducted among 35 professionals of Medical officers, Chartered Engineers and Accountants employed in the public sector organizations in Sri Lanka using simple random sampling techniques. Later, primary data was collected through the use of a written structured questionnaire that was hand delivered and posted by postal mail by using stratified simple random sampling techniques to selected professionals of Chartered Engineers, Medical Officers and Accountants employed in the public sector organizations in Sri Lanka. Each questionnaire was prefaced by a paragraph explaining objectives of the survey. Data collection was taken place during the period of May 2018 to August 2018. A total of 920 questionnaires were distributed and 398 were returned and the response rate was 43.2%. Among them, the questionnaires which contained missing data were deleted, resulting in a total of 386 responses to be used for empirical analyses. Demographic characteristics of participants are given in Table 5.

Measures
The variables used in this study are work-life balance, perceived work, family and social role demands and work ethics and these measures are given below.

Work-Life Balance Measures
In this study, researchers developed and validated work -life balance scale has been used. Each dimension consist of three items using a five point Likert Scale ranging from (1) strongly disagree to (5) strongly agree. Work-Life Balance scale altogether contained 13 items and these items are given in Table 2. Perceived work role demands (PWD) measure was measured using four items developed by Boyar et al. (2007). Perceived family role demands (PFD) measure was measured using three items developed by Boyar et al. (2007) and one item was developed and validated by authors and this item is "I am facing burden on eldercare responsibilities to my family member". Perceived social role demands (PSD) measure was measured using five items developed and validated by current authors and all of these items were anchored using a five point Likert Scale ranging from (1) strongly disagree to (5) strongly agree. These measures are given in Table 3.

Work Ethics Measures
Work ethics measures consist of work centrality, self-reliance and hard work and these were measured using three items developed and validated by Miller, Woehr, & Hudspeth, (2001) as shown in in Table 4. All of these items were anchored using a five point Likert Scale ranging from (1) strongly disagree to (5) strongly agree.

Data Analysis
In this research study, the data collected from research instruments were numerically scored and quantified. Descriptive and Regression analyses were conducted using SPSS 23 to calculate the mean, frequency, standard deviation, correlations and combined effects. Data were also analyzed using Structural Equation Modelling approach with Amos 21(Analysis of Moment structures) to test the fitness of proposed theoretical model and to find the causal effects of perceived role demands on work -life balance. Moreover, Multi-Group Analysis in Amos also has been applied for testing the moderation effect between perceived role demands and work-life balance.

Demographic Characteristics of Participants
In terms of demographic analysis of the 386 respondents' demographic characteristics are illustrated in the following Table 5.

Reliability and Validity Analysis of Measurement Constructs
The measurement constructs consist of five latent variables called as work role demands, family role demands, social role demands, work-life balance and work ethics and containing 35 observed variables.

Reliability Analysis
The internal consistency reliability of the measured variables were analyzed using IBM SPSS statistics 23 for Cronbach's Alpha values (α). The reliability values for all measured variables ranged from 0.792 to 0.912 and all values were within threshold values. The threshold of Cronbach's Alpha value should be more than .7 (Nunnally, 1978). The Composite Reliability (CR) indicates the reliability and internal consistency of a latent constructs ranged from 0.819 to 0.912. The cut off value of composite reliability was 0.6 (Fornell & Larker, 1981). These reliability information are shown in Table 2. Hence, it was concluded that measurement properties for each variable was reliable.

Convergent Validity
Convergent validity was established both by using standardized loadings and Average Variance Extracted (AVE).
If an item loads significantly >0.5 (Field, 2009, p. 648) on the factor, it is ensure the convergent validity is prevalent. The Confirmatory Factor Analysis (CFA) showed that each item loading is greater than 0.5. Calculated Average Variance Extracted values as per Table 6 are ranged from 0.531 to 0.775. AVE > 0.5 (Fornell & Larker, 1981) is required for every construct for ensuring the convergent validity. Accordingly all variables are satisfied to ensure convergent validity as illustrated in Table 6.
AVE= ∑Қ2 /n-Қ= factor loading of every item, n = number of items in a model

Discriminant Validity
Discriminant validity test was assessed using correlation matrix with square root of the average variance extracted value as shown in Table 7. This shows that square root of average variance extracted for each construct (Diagonals) interactions were greater than the inter construct correlations. Hence discriminant validity has been established.  To test the degree of combined impacts of perceived role demands on work -life balance, stepwise multiple regression analysis using SPSS 23 has been carried out and the results are presented in Table 12.  According to the results of the above Table 12, overall correlation coefficient (R) of the three predictor variables and WLB was .542 and the R Square was .293 with significant at .035. It implies that about 29.3 percent of the variance of WLB has been significantly explained by the three predictor variables with greater predictive power. Furthermore, multiple correlation coefficient (R) of the two independent variables of work role demands and family role demands (Model 2) and WLB was .534 and the R Square was .285 with significance at .000. Therefore, there is statistical evidence to claim to the hypothesis (H1d) that the degree of combined impacts of work role demands, family role demands and social role demands have jointly and significantly impact on worklife balance.

Results of Correlations among Study Variables and Hypothesis Testing
In the next step, Pearson coefficient of correlations were calculated to find out the relationship between perceived role demands and work -life balance relationships. With respect to this correlation matrix as shown in Table 13, perceived work role demand was found to be negatively and significantly correlated with work -life balance (r=-.507, p=.000) and this relationship is strongest. Perceived family demands was found to be negatively and significantly correlated with work -life balance ( r= -.434, p=.000) and perceived social demands was found to be negatively and significantly correlated with work -life balance ( r= -.246, p=.000). This estimation supports the second hypothesis (H2) of perceived work role demand has the strongest relationship with work-life balance than that of perceived family or social role demands and work-life balance relationships.  Business and Management Vol. 15, No. 8; As seen from Table 14 illustrations, the results revealed that moderation is significant for high work-ethics group and low work-ethics group between perceived work, family and social role demands and work -life balance relationship since the differences in Chi-square value between constrained and unconstrained model is more than 3.84 for all three cases for one degree of freedom at 0.000 significance levels. This results support the hypothesis (H3a) of Work Ethics has a moderating effect on the relationship between perceived work, family and social role demands and work-life balance relationships. Since both low work-ethics group and high workethics group moderation estimates are significant, partial moderation occurs in that path (Bae, 2011). As can be seen from the Figure 4, the output of the unconstrained modified structural model with low work ethic group, it was found the direct effect of β =-0.54, p = 0.000 and as can be seen from the Figure 5, the output of the unconstrained modified structural model with high work ethic group, it was also found the direct effect of β =-0.41, p = 0.000. Furthermore, the paths between perceived family and social role demands and WLB were constrained and analyzed using both groups separately. It was found with low work-ethic group, the β =-0.885, p = 0.000 and high work ethic group, the β =-0.857, p = 0.000. These findings shows that low work-ethics group has a higher effects than high work-ethics group on the moderation effects between perceived work role demands and work-life balance relationship. This results supports the Hypothesis (H3b) of the effects of perceived work role demands on work-life balance will be different between the group with high level of Work-Ethics and the group with low level of Work-Ethics.

Discussions
The results of this study show that professionals possess a moderate level of work -life balance (mean value was 3.01 with a standard deviation of 0.80). It was also identified that majority of the respondents ( which is within the threshold value of less than 3.0. These overall indices show that the model fits the data very well. In the above SEM model, perceived work role demands had a strongest negative impacts on work -life balance (β =-0.503, p = 0.000) while perceived social role demands had a moderate negative impacts on worklife balance (β = -0.111, p = 0.05) but the weakest, negative and no significant effect was observed in perceived family role demand and work -life balance (β =-0.068, p = -0.35) path. This findings of negligible effects of perceived family role demand is still noteworthy because it may reflect the reality of professionals. The findings of the study shows that the hypothesis which states that perceived role demands have a negative impact on work -life balance is evidence only for perceived work and social demands but no significance for family role demand.
Furthermore, multivariate Pearson correlation analysis provides the evidence that the strongest, negative and significant correlations was found between perceived work role demand and work -life balance (r=-.507, p=.000), perceived family role demand was found to be negatively and significantly correlated with work -life balance (r= -.434, p=.000) and perceived social role demand was found to be negatively and significantly correlated with work -life balance (r= -.246, p=.000). This specifies that if perceived role demands are increased, the work-life balance is decreased and vice versa. This finding is consistent with the results of Lewis & Cooper (1987) who reports that role demand is well documented source of pressures experienced by employees when balancing work and family. Moreover, from the output of stepwise multiple regression analysis using SPSS 23, R Square was .293 with significant at .035. It implies that about 29.3 percent of the variance of work-life balance has been significantly explained by perceived work, family and social demands variables with greater predictive power which shows evidence for the combined effects of perceived role demands on work-life balance.
Work-Ethics has a moderating effect on the relationship between work, family and social role demands and work -life balance relationship were supported. According to Multi-Group CFA, the estimates for both the constrained and unconstrained model, Chi-Square value differs by more than 3.84 so that the moderation occurs in that paths in such a way that work-ethics has a moderating effect between perceived role demands and worklife balance relationships. Since both low work-ethics group and high work-ethics group moderation estimates are significant, partial moderation occurs in that path. Finally, the present study also suggest from SEM analysis that effect size for low work-ethics group, the β =-0.885, p = 0.000 and high work-ethic group, the β =-0.857, p = 0.000 and this shows that low work-ethic group has a higher effect than high work-ethics group for moderating the effect between perceived work role demands and work-life balance relationships. Moreover, this findings also suggest the notion that in a higher work -ethics group ethical behavior is inherent and existing in their value system and so role demands causes less effects.

Conclusions
In this study, a model has been tested which posits that perceived role demands affect work-life balance. This investigation demonstrated that professionals involved in this study possess a moderate level of work -life balance and having moderate level for all its dimensions. The overall results of the SEM model fit indices indicates that the data fit the model very well. Furthermore, this study verified that perceived work and social role demands have negative and significant causal impacts on work-life balance while no significant causal effect was found by family demands. This findings elucidated that work-life balance is skewed towards works and less in family and social role demand is also an emerging issue for professionals. Besides, statistical evidence was also found to claim that the degree of combined impacts of perceived role demands have jointly and significantly impact on work -life balance. Moreover, this study provides some further important insights that negative and significant correlations were found between perceived role demands and work-life balance and the strength of the relationship is higher for work demands then family demands and least for social demands. More precisely, this study added to generate the empirical evidence for work ethics has a partial moderation effect between perceived role demands and work -life balance relationships. Furthermore, it is one of the few studies through which moderation analysis investigates how low work-ethics group has higher effects than high work-ethics group on moderating the relationship between perceived work role demands and work-life balance relationship of public sector professionals in Sri Lanka. It serve as a reference landmark for public sector organizations to cultivate high work ethics culture among their employees.

Implications and Recommendations
This study is significant and beneficial for managers, organizational leaders and researchers to address the needs of employees and to initiate creative thinking to develop strategies and policies to combat or address the worklife balance issues in an effective and efficient way. Moreover, this study is proved to be a milestone and crucial to professionals, employees and family members for better understanding how different role demands influence work -life balance. Furthermore, this study helped the professionals to understand the importance and a background for their current status of work -life balance in a context in which professional live.

Limitations and Direction for future Research
The first and foremost limitations of the study are that all of the measures were self-reported that common method bias may raise concern. Another limitation is that the research design is cross sectional survey and it would constraints for the true causality. Hence, further studies using longitudinal design would be useful in establishing the temporal order of relationship and insights of strong causal relationships among variables. Since the focal point of this study is public sector, future study also could be conducted among the private sector professionals in Sri Lanka.