Work-Life Balance: Scale Development and Validation

Work-Life Balance: Scale Development and Validation Chapter 5 in the Book: The Work-Family Balance in Light of Globalization and Technology (pp. 109-130)

Cite as: Agha, K., Azmi, F. T. & Khan, S. A. (2017). Work-Life Balance: Scale Development and Validation. In: Heras, M. L., Chinchilla, N. & Grau, M. (eds). The Work-Family Balance in Light of Globalization and Technology (pp. 109-130). Cambridge Scholars Publishing, Newcastle upon Tyne, UK.

5 WORK-LIFE BALANCE: SCALE DEVELOPMENT AND VALIDATION KAKUL AGHA, SKYLINE UNIVERSITY COLLEGE, UNITED ARAB EMIRATES

FEZA A. TABASSUM ALIGARH MUSLIM UNIVERSITY, INDIA

AND SAMI A. KHAN KING ABDUL AZIZ UNIVERSITY, KINGDOM OF SAUDI ARABIA

Introduction Analyzing the dynamic human resource management (HRM) function, complexities of working relationships and the increasing importance for organizations to retain their workforce, it becomes imperative to study issues that help engagement and motivation of the workforce for optimum performance in the workplace. In that context, work-life balance (WLB) and flexible working practices become highly relevant. It is important from the perspective of the employees that they are able to integrate work and family matters in a balanced way so that their performance does not get hampered. A good WLB entails satisfaction and worthy functioning at work and home with a minimum role conflict (Clark 2000). Employment flexibility and achieving a positive work-life balance have also emerged as important issues for contemporary employers seeking to make best use of their diverse workforce. WLB is currently regarded highly by policy makers, practitioners, consultants, HR managers, and academics. Greater flexibility and balance between work and family has led to tangible benefits for employees as well as organizations (Agha and Azmi 2011; Baltes, Briggs, Huff, Wright and Neuman 1999; Becker 1995; Standen, Daniels and Lamond 1999). Studies in WLB emphasize an increased burden for women in

Work-Life Balance: Scale Development and Validation

3

employment (Frankenhaeuser et al, 1989; Lundberg et al. 1994; Smithson and Stokoe 2005, Gerson 2010). Some of the researchers have defined work-life balance in a broad sense and Byrne (2005) describes work life balance as “juggling five aspects of our lives at any one point in time: work, family, friends, health and spirit (or self)”. The other issues that are important in the workplace include increased levels of stress, competition and insecurities that disrupt life balance (Bonney 2005). Rigid models of work are driving highly qualified workers into jobs with diminished skill level in order for them to have a life outside work (EOC 2007). Unfortunately, the UK distinguishes itself with a “long hour culture” called presentism, known as “face time” in the United States. Research shows that many people work very arduous long hours and that one in five regularly takes work home in the evening, but less than one in two workers has any control over his or her working hours (Doyle and Reeves 2001, Simpson, 1998). European Legislation defines 48 working hours a week as an appropriate maximum and reviews of literature on working hours and health inform that when people work much beyond these hours, their health and performance begin to deteriorate and their overall work-life balance becomes affected (Sparks et al. 1997). A balance at work is important for employees, which has emerged as a vital policy consideration in organizations as an increasing number of families are struggling to juggle between work and their family commitments (Heraty, Morley and Cleveland 2008; Khan and Agha 2013). Therefore, there is a global understanding of the importance of balancing and integrating family and work for all employees in contemporary times. It warrants deeper insight into the phenomenon of WLB. This research investigates and validates the scale on WLB.

Research Gap and Objectives Work-life balance is concerned with people, the most important connecting link between an organization and its success. Workers experience family conflict when the stress felt at work is carried into their family environment. They experience work conflict when they encounter stress at home, which affects their work. Organizations are reducing this conflict by providing flexibility and supportive work environments to foster integration of work and family role (Agha 2014). Thus, considering WLB issues and formulating supportive strategies help integrate work and family issues, and provide tangible benefits to all stakeholders (Friedman and Greenhaus 2000; Pollitt 2004). Work-life balance is a topic related to the entire global workforce.

4

5

Currently, due to diversity amongst workforce and emerging realities of global business, the focus has shifted to maintaining a good WLB for making the workforce productive, happier and engaged. If employees see how the organization makes it easier to balance work and family, the organization may gain a competitive advantage in recruitment and retention. Employees who have access to family responsive policies have greater organizational commitment and have a significantly lower intention to quit (Forsyth and Polzer-Debruyne 2007; Grover and Crooker 1995). Therefore, the perception of employees about the organization’s intention to support WLB policies at a strategic level need to be in a positive direction and should enable the employee to establish a psychological contract with the organization. Despite the increasing importance of WLB, there is still a paucity of empirical research on work-life balance in the context of the Sultanate of Oman. Invariably, a large amount of literature is available on HRM, especially training and development, omanization and human resource development (Bordia and Blau 1998) but insignificant research on WLB. The dynamic business environment of the Sultanate of Oman is a good testing field for this concept. Keeping in mind the fact that Oman is rapidly becoming a busy and vibrant place to live and work, a study on WLB in the Omani context is timely and pertinent. The research objective of this study is scale development and validation of the work-life balance scale using the three constructs of Work Interference with Personal Life (WIPL), Personal Life Interference with Work (PLIW), and Work Personal Life Enhancement (WPLE).

WLB Constructs Hayman (2005) developed a psychometric instrument to measure WLB in organizations. A 15-item scale had been adapted from a 19-item scale originally developed by Fisher-McAuley et al. (2003) that was designed to capture employee perceptions on WLB. The 15-item scale measured Work Interference with Personal Life (WIPL), Personal Life Interference with Work (PLIW), and Work Personal Life Enhancement (WPLE) as three constructs of WLB. Fisher-McAuley et al. (2003) examined the relation between employees’ beliefs about having a balance between work and personal life and the feeling of job stress, job satisfaction, and reasons why an employee would quit his or her job. They simplified the WLB by placing three measurable dimensions, namely: work interference with personal life; personal life interference with work; and work and personal life


5

enhancement. The dimensions indicate that interference of work in personal life and on the other hand the interference of life issues in work related matters are important to study while understanding work-life balance related issues. Constructs of WLB are: Construct 1: Work Interference with Personal Life (WIPL) According to Hayman (2005), this construct includes work related factors that impact an individual’s personal life. It measures the impact of work on personal life. The seven items measured by this construct are: 1) My personal life suffers because of work 2) My job makes personal life difficult 3) I neglect personal needs because of work 4) I put personal life on hold for work 5) I miss personal activities because of work 6) I struggle to juggle work and non-work and 7) I am unhappy with the amount of time for non-work activities. Construct 2: Personal Life Interference with Work (PLIW) Hayman (2005) explains the second construct as the impact of or the interference of the personal life on work. It measures the reverse phenomenon i.e. the impact of personal life on the work of individuals. The four items measured are: 1) My personal life drains me of energy for work 2 ) I am too tired to be effective at work 3) My work suffers because of my personal life and 4) It is hard to work because of personal matters. Construct 3: Work Personal Life Enhancement (WPLE) The third construct explains how work and personal life enhance each other. The items help understand the support and enhancement provided by work on personal life and vice-versa. The four items measured are: 1) My personal life gives me energy for my job 2) My job gives me energy to pursue personal activities 3) I have a better mood at work because of personal life and 4) I have a better mood because of my job.

Research Methodology A research design details the procedures necessary for obtaining the information needed to structure or solve research problems (Malhotra and Birks 2007). Studies aimed at quantifying relationships are descriptive and experimental. Yin (1994) informs of the use of a rigorous research methodology (RM) in a study. In line with his suggestions, this research

6

5

aims at reducing discrepancies by utilizing a rigorous RM. There are various ways in which research work can be carried out. For this study, single cross-sectional design was used. In order to empirically test the scales, primary data was collected from Higher Education Institutions (HEIs) in the Sultanate of Oman. Teachers were respondents as they are the subject of this study. Authors like Arthur and Boyles (2007) have supported the aptness of the deployment of one respondent as it helps providing valid and reliable data in comparison to data from multiple sources. In order to collect data, a research instrument was compiled based on an extensive literature review. The questionnaire was pilot tested and after necessary modifications, it was administered to the study sample. It was translated into Arabic, the mother tongue of the Sultanate of Oman. The reliability and validity of the research instrument were ascertained. Data generated was then subject to analysis.

Development of Research Instrument The final survey instrument based on the above constructs contained the following items: Independent Variables: Measures of Work-Life Balance (WLB) • Work Interference with Personal Life (WIPL) • Personal Life Interference with Work (PLIW) • Work Personal Life Enhancement (WPLE) WLB was measured using a 5-point Likert scale anchored with the end points 1=strongly disagree to 5=strongly agree. These descriptors were chosen to balance or neutralize any tendency to over-report difficult or unacceptable behaviours and conditions faced by the respondents at their workplace (Fairbrother and Warn, 2003). Five-point scale has been commonly used in WLB research (e.g. Boyar et al. 2003; Fairbrother and Warn 2003; Forsyth and Polzer-Debruyne 2007). Fairbrother and Warn (2003) further noted that ambiguity in questions lead to confusion amongst respondents, thereby altering the meaning of the question. This lead to error in the data collected through the questionnaires. Arthur and Boyles (2007) point out that researchers should design questionnaire items that capture the specific substantive focus of the component being assessed. Hence, efforts were made to keep the items as simple, specific and objective as possible.


7

The research instrument was developed in three stages: Stage 1: Identification of measures/constructs from literature and adaptation of the questionnaire Extensive review of literature was carried out to have a thorough understanding of the measures and constructs of WLB, teaching satisfaction and job satisfaction. An initial draft of the questionnaire was developed. Stage 2: Modification in draft questionnaire on the basis of inputs/suggestions from academics/researchers During the second stage, the expertise of academics in India and Oman was sought through personal meetings and e-mails. Four researchers/academics were contacted; two were domain experts and two were RM experts. They were requested to critique the content and format of the questionnaire, enabling the researcher to finalize the constructs and its items and to modify the questionnaire leading to a stronger and more valid instrument. Stage 3: Translation of the questionnaire into Arabic for effective data collection As the data collection had to be carried out in Oman, it was considered vital to get the research instrument translated in Arabic, the national language of Oman. A large number of HEI teachers are more conversant in Arabic in comparison to English, hence, there was a need have a bilingual questionnaire. To avoid discrepancies in translation, back translation method is suggested by Green and White (1976). In a study conducted by Berkanovic (1980) on Hispanic individuals, some of the respondents were interviewed in Spanish whereas the rest of the group was interviewed in English. As Abu-Shanab (2011) pointed out, any research conducted using a different language but failed to utilize the backward translation would yield inaccurate results. Backward translation will definitely help decrease the influence of the language and increase the reliability of the research instrument (Brislin 1976; Su and Parham 2002). Therefore, the questionnaire was translated into Arabic by a professional translator. Two Omani researchers, experts in spoken/written English and Arabic, translated the Arabic questionnaire back to English. Some changes were made to the draft questionnaire on the basis of the

5

8

inputs of the experts. Here it is important to note that all questionnaires were bilingual, giving the respondent a choice to read in either or both languages.

Method of Analysis A measurement model is a graphical structure that describes the relationship between latent variables and observed variables (Silva and Scheines 2005) and assists in describing how effectively the observed indicators serve as a measurement instrument for the latent variables. According to Garver and Mentzer (1999), specifying the measurement model involves assigning indicators, which are actual items of the questionnaire to a construct or latent variable. An extensive measurement analysis gives confidence in terms of findings that reflect accuracy in the constructs. Furthermore, empirically reliable and valid scales can be used on different populations in studies to be conducted in the future. Measurement scales must exhibit unidimensionality, reliability and validity (Green et al. 2006). For this study, measurement analysis was performed on three scales viz. WIPL, PLIW and WPLE. In order to assess the unidimensionality of the study scales, both exploratory and confirmatory factor analyses were performed. Scale reliability and validity were also assessed.

Assessing Scale Unidimensionality Exploratory Factor Analysis (EFA) In most empirical studies “numerous variables are used to characterize objects” (Rietveld and Van Hout 1993, 251). Hence, Exploratory Factor Analysis (EFA) attempts to bring inter-correlated variables together under more general, underlying variables. EFA is easy to adopt, suitable for survey questions, used as a basis for other instruments such as regression analysis with factor scores, and easy to combine with other techniques like Confirmatory Factor Analysis (Ahire et al. 1996). The conventional approach to scale validation consists of performing an EFA to identify major factors according to item-factor loadings and refining the scales using Cronbach’s alpha (Ahire et al. 1996, Hurley, et al. 1997). EFA was performed on each scale separately to check and confirm that all items load on a single construct. To determine the likelihood of the data factoring well, Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy and Bartlett’s Tests of Sphericity were performed before


9

proceeding with EFA (Rietveld and Van Hout 1993). KMO is a measure that helps understand the degree of inter-correlations among the variables and if greater than 0.50, one can proceed with factor analysis (Malhotra 2004; Rietveld and Van Hout 1993). KMO values for all scales for this study were within acceptable range, informing that the data was suitable for factor analysis. An important measure is Bartlett’s Test of Sphericity to “test the null hypothesis that the original correlations matrix is an identity matrix” (Field 2000, 457). A significant Bartlett’s Test of Sphericity is required (Malhotra 2004). For this study p = 0.000 (associated probability is less than 0.05) for all scales, thus indicating that the data was suitable for factor analysis. The results of KMO and Bartlett Test of Sphericity for all scales are given in Table 5.1: Table Error! No text of specified style in document..1 Results of KMO and Bartlett Test of Sphericity KMO and Bartlett's Test of Sphericity Measures KMO Measure of Sampling Adequacy Bartlett's Test of Sphericity Approx. Chi-Square df Sig.

WIPL

PLIW

WPLE

0.892

0.780

0.723

317.237 14 .000

62.421 2 .000

88.028 2 .000

Confirmatory Factor Analysis (CFA) Structural Equation Modelling (SEM) is a powerful statistical technique that combines measurement model or Confirmatory Factor Analysis (CFA) and structural model into a simultaneous statistical test. SEM is valuable to carry out inferential data analysis and testing of hypothesis (Hoe 2008). Chin (1998) adds that SEM has the flexibility to model relationships amongst predictors and variables and statically tests theoretical assumptions and empirical data through CFA. Researchers have opined that CFA is a rigorous test of assessing the measurement model when using SEM (Medsker et al. 1994). Although the results obtained in EFA confirmed unidimensionality of all the scales except JS scale, it was decided to perform CFA as a further test of scale unidimensionality, reliability and validity. High loading values obtained in CFA provide strength to the fact that the measurement model is acceptable. CFA values were further used to assess scale reliability and validity. A goodness of fit index (GFI) of a value very close to 0.90 or higher for

5

10

the model suggests that evidence for unidimensionality existed (Jöreskog and Sörbom 2002). The GFI indices for all scales were above 0.90, indicating that scales were unidimensional. A summary of items in each scale with their GFI values are given in Table 5.2 along with the loading value range. Table Error! No text of specified style in document..2 GFI values of WLB Constructs GFI Values for all Scales WIPL PLIW WPLE

Number of Items with GFI values 7 Items (GFI=0.90) 4 Items (GFI=0.95) 4 Items (GFI=0.93)

Loading Value Range 0.64-0.84 0.70-0.90 0.60-0.77

Assessment of Reliability After establishing the unidimensionality of scales, the researcher tested the statistical reliability of the scales before proceeding with the validation analysis (Anderson and Gerbing 1988; Mentzer et al. 2001; Steenkamp and Trijp 1991). For a scale to be valid, it should first be established as reliable (Peterson 1994). Reliability refers to the degree of dependability, stability and internal consistency of a scale. According to Garver and Mentzer (1999), reliability is defined as the extent to which a questionnaire, test, observation or any measurement procedure gives same results on repeated trials. Conceptually, reliability is defined as the degree to which measures are free from random error and therefore yield consistent results. Two types of reliability estimates were calculated: (1) indicator reliability and (2) scale reliability. Indicator Reliability Indicator reliability is a very crucial measure to find out reliability of individual indicators. They are calculated by finding the squared factor loadings and are called communalities. The squared values generally range from 0-1 in order to be acceptable (Jöreskog and Sörbom 2002). In SEM terms, reliability of an indicator is defined as variance in the indicator that is not accounted for by measurement error (Wu 2005). Some researchers have informed that indicator reliability should preferably be 0.5 or close (Long 1983; Schumacker and Lomax 2004). The indicator reliability of the three scales, viz. WIPL, PLIW, and WPLE were high or close to the desired values. Table 5.3 illustrates indicator reliability for indicators in each scale.


11

5

12

Table Error! No text of specified style in document..3 Indicator Reliability Indicator Reliability of the Scales Indicators 1 2 3 4 5 6 7

WIPL

PLIW

WPLE

0.7 0.7 0.7 0.6 0.6 0.5 0.7

0.5 0.5 0.8 0.7 -

0.4 0.6 0.4 0.6 -

Scale Reliability Reliability is the ability of an instrument to measure consistently (Tavakol et al. 2008). It is closely related to validity and cannot be achieved without measuring the validity of the tool. However, reliability of an instrument is independent of its validity (Nunnally and Bernstein 1994). Calculating Cronbach’s alpha coefficient of internal consistency is the most common method used. Gliem and Gliem (2003) have explained that the test of reliability is measured by Cronbach’s alpha, which requires only a single test administration to provide a unique estimate of reliability for a given scale. Cronbach’s Alpha: Cronbach’s alpha is a pervasive objective measure of reliability. Alpha was developed by Lee Cronbach in 1951, to provide a measure of the internal consistency of a test or scale and is expressed as a number between 0 and 1 (Tavakol et al. 2008). Cronbach’s alpha is an index of reliability related to the variation accounted for by the true score of the underlying construct (Hatcher 1994). Alpha coefficient ranges from 0-1 and is used to explain the reliability of factors extracted from multi-point scales (i.e. rating scale: 1 to 5). A value of Cronbach’s alpha of 0.6 or more is used as a criterion for a reliable scale (Hair et al. 1998; Nunnally and Bernstein 1994; Werts et al. 1974). Low alpha values could be due to low number of questions, poor interrelatedness between items or heterogeneous constructs. Reliability assessments of the three scales obtained through Cronbach’s alpha values are presented in Table 5.4. It is clear that all values suggest high reliability.


13

Table Error! No text of specified style in document..4 Scale Reliability Values of Cronbach’s alpha Scale Reliability – Cronbach’s Alpha Scale WIPL PLIW WPLE

Cronbach’s alpha 0.9 0.9 0.8

SEM based Methods Internal consistency, measured through alpha, is crucial but not a sufficient condition for measuring unidimensionality or reliability in test items. Sometimes coefficient alpha values present underestimated or overestimated reliability (Cortina 1993, Garver and Mentzer 1999). Hence, for estimating reliability, the researcher calculated the SEM constructreliability and variance-extracted measures. Furthermore, Garver and Mentzer (1999) add that these are easy and reliable techniques and can be calculated for each construct. Fornell and Bookstein (1982), Fornell and Larcker (1981), and Garver and Mentzer (1999) describe the concepts of Construct Reliability (CR) and Variance Extracted (VE) as follows: • Construct Reliability (CR): In SEM, CR is a LISREL-generated estimate of internal consistency analogous to Cronbach’s alpha. • Variance Extracted (VE): VE is for finding construct reliability and is known as variance extraction measure. Figure Error! No text of specified style in document..1 Formulae for CR and VE

“sli” denotes the standardized loadings for the indicators of a variable; “ei” denotes the corresponding error terms for the chosen variable. Variance extracted estimates assess the amount of variance captured by a construct’s measure in relation to variance due to random measurement error.

5

14

Table Error! No text of specified style in document..5 Scale Reliability – CR and VE Scale Scale Reliability – CR and VE Scale WIPL PLIW WPLE

Construct Reliability (CR) 0.9 0.9 0.8

Variance Extracted (VE) 0.6 0.6 0.5

Assessment of Validity Validity of a scale is the extent to which differences in observed scale scores reflect true differences among objects of characteristic being measured (Malhotra and Birks 2007). Researchers typically establish construct validity by presenting correlations between a measure of a construct and a number of other measures that should, theoretically, be associated with it, which is known as convergent validity, or vary independently of it, which is called discriminant validity (Westen and Rosenthal 2003). A scale has validity if it is measuring the concept that it was intended to measure (Bagozzi 1981). Construct validity is a complex process and is used to assess what construct or characteristic the scale is in fact measuring (Cronbach and Meehl 1955). Construct validity can be found by using several types of validity tests such as convergent, discriminant, predictive and criterion validity tests. For this study convergent, discriminant, and nomological validity were measured with the help of the measurement model, while criterion validity was measured through the structural model. Convergent Validity Convergent validity is the extent to which items in a scale correlate positively with each other (Campbell and Fiske 1959) implying that all items (measured variables) correctly measure the designed construct or the latent variable (Garver and Mentzer 1999) and helps understand that constructs in a scale converge on one scale or are correlated to each other. CFA is frequently used to evaluate convergent validity (Bagozzi 1981, Nunnally and Bernstein 1994). Kaplan and Sacuzzo, (1993) explained that internal consistency is a type of convergent validity. Since unidimensionality and high internal consistency of three scales for this study had already been established, evidence of moderate convergent validity already existed. Moreover,


15

Bagozzi (1981) suggests that all items should load on their hypothesized dimensions and the estimates should be greater than 0.50. In this study, all scales had loadings of more than 0.5 indicating convergent validity. The convergent validity of all five scales was checked with BentlerBonett Normed Fit Index (NFI) using LISREL for this study. This index measures the extent to which different approaches to measuring a construct produce the same results (Ahire et al. 1996). A value of 0.90 and above demonstrates strong convergent validity (Hartwick and Barki 1994). The Bentler-Bonett coefficient for all constructs was greater than 0.90, indicating high convergent validity for this study. The values of NNFI also support the convergent validity and have been found to be acceptable. The values are presented in Table 5.6: Table Error! No text of specified style in document..6 Convergent Validity NFI and NNFI Values of Scales WIPL PLIW WPLE

NFI 0.94 0.96 0.91

NNFI 0.91 0.90 0.80

Discriminant Validity Discriminant validity proves that none of the items in a scale measure other constructs or other scales. Bagozzi and Yi (1988) inform that discriminant validity measures the degree to which a construct and its indicators differ from another construct. Discriminant validity can be assessed in several ways (Mentzer et al. 2001) one being comparing Cronbach’s alpha of a construct to its correlations with other model variables (Sila and Ebrahimpour 2005). If the value of alpha is sufficiently larger than the average of its correlations with other variables, this is evidence of discriminant validity (Ghiselli et al. 2001). For this study, the difference between the alpha value for each of three constructs and the average correlation of each construct with other constructs was fairly large. This provides evidence of discriminant validity for three scales as shown in Table 5.7. Table Error! No text of specified style in document..7 Discriminant Validity Scale WIPL

Correlations PLIW WPLE

Average of Correlations

Alpha Values

Existence of DV

5

16 WIPL PLIW WPLE

1.00 -

0.57 1.00 -

0.21 0.12 1.00

0.4 0.4 0.2

0.9 0.9 0.8

Nomological Validity Ahire et al. (1996) and Garver and Mentzer (1999) recommend assessing nomological validity by examining correlations between constructs in the measurement theory. The covariance matrix Phi (Ф) of construct correlations is useful in this assessment. SEM is considered ideal for testing correlations and determining nomological validity of the constructs of interest. SEM takes into account measurement error by estimating measurement error variances from the data, whereas traditional correlation techniques do not. The latter would underestimate true correlations due to inherent measurement errors (Ahire et al. 1996). Nomological validity was assessed for three study scales viz. WIPL, PLIW and WPLE, and it was found that all correlation values were positive and significant thus giving evidence of nomological validity. Since the above constructs are theoretically related it is assumed that constructs will also be statistical correlated. As presented in Figure 5.2, correlation values are positive giving sufficient proof of nomological validity. Figure Error! No text of specified style in document..2 Nomological Validity


17

Discussion, Conclusions and Directions for Future Research Discussion and Conclusions This study was initiated with an understanding that the scenario of HRM is changing and the concept of WLB is taking shape among professionals. The development of HRM has happened to an extent that employees are reaping the benefits of the changed job conditions, flexibility within organizations and supportive work environment. There is conceptual clarity amongst employers on the role and responsibilities of employees. Taking care of employees and providing them with requisite support is one of the most important HRM functions. Globalization, knowledge, technology, and innovation are prominent reasons for the changing nature of work and the workforce on a global scale. In order to have a competitive edge, organizations need to consider the employees as the prime asset and need to support and develop them to enhance their satisfaction levels. This study conducted in Oman attempts to develop and empirically validate the WLB scale. The sector chosen for research was the Higher

18

5

Education Sector and the academic staff members were considered as subjects for the present study. The teaching faculty employed in private higher education institutions; Government University; Colleges of Applied Sciences and Colleges of Technology were contacted for the study. Males and females as well as Omani nationals and expatriates were the respondents. They held Bachelors, Masters, Doctorates or Post Doctorate degrees and they were teaching either bachelors, masters or at both the levels. They had teaching experience ranging from 1 to >10 years and had been employed for a minimum of 2 years. Finally, 621 responses were considered for evaluation and analysis. In order to establish unidimensionality of the scale used, scale validation, reliability and validity tests were performed. The scores of the reliability and validity tests were acceptable thus establishing unidimensionality of the scale. The scale adapted in this study has been validated by previous researchers; still a need was felt to further validate in the context of Oman (Hayman 2005; Ho and Au 2006). This study attempts to establish the relationship of three constructs, viz. WIPL, PLIW and WPLE. Interestingly, studies related to WLB have been carried out in the western world and hence follow a western model. The scales have been tested in the context of western countries and definitely not in the Middle East context. Hence, it becomes very interesting to note that even though the scales of WIPL, PLIW and WPLE have been validated in the western world in different studies, they have not been consolidated at any time. This study not only consolidates and contextualizes the scales but also validates them in the context of Oman. In Oman, where this conceptual model was tested, it was found that a good spread of Omani and expatriate teachers were present in HEIs. Additionally, a judicious mix of males and females can also be seen from the data collected by the researcher to carry out the study. This study has hence, laid the foundation for further research in Oman on WLB. It implies that work-life balance is a global phenomenon. Even though they are western models, when validated in the Omani context they reveal similar results, proving that they are similar for human beings around the world. Even if the context is different, the behaviour of human beings (i.e. employees) remains consistent. This study validated the work-life balance scale by finding the validity and reliability of the three constructs of the scale namely WIPL, PLIW and WPLE. It is indicative in the study that all the three constructs are reliable and valid and that they can be used by researchers throughout the world for research purposes.


19

Directions for Future Research Now that the WLB scale has been validated by the study, an opportunity for using the scale has been provided for the researchers. This tool can be combined with job satisfaction and other issues to study the relationship and impact on the employees. Future researchers can test the scale with moderating variables like gender, differing level of education and different levels of management. Furthermore, the researchers can consider the limitations of this study to enhance the scale that would suit the culture and other factors of the working conditions.

References Abu-Shanab, Emad. "Quality Assurance Initiatives: The Influence of Language." In The International Arab Conference on Quality Assurance in Higher Education, Zarqa, Jordan. 2011. Agha, Kakul. “The Coping-up Strategies Used by Staff to Integrate Family and Work Lives: A Case from Middle East College, Oman.” In Proceedings of Applied Information and Communications Technology, Muscat, Sultanate of Oman. Elsevier Publications, 2014. 785-790. ISBN No. 978-89-3510-72850. Agha, Kakul, and Feza, T. Azmi. “ICT Interventions and Workforce Flexitime: An Exploratory Study at MECIT, Oman.” In Proceedings of International Conference on Applied Information and Communications Technology, Muscat, Sultanate of Oman. 2011. ISBN No. 978-818424-738-1. Ahire, Sanjay L., Damodar Y. Golhar, and Matthew A. Waller. "Development and Validation of TQM Implementation Constructs." Decision Sciences 27, no. 1 (1996): 23-56. Anderson, James C., and David W. Gerbing. "Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach." Psychological Bulletin 103, no. 3 (1988): 411-423. Arthur, Jeffrey B., and Trish Boyles. "Validating the Human Resource System Structure: A Levels-Based Strategic HRM Approach." Human Resource Management Review 17, no. 1 (2007): 77-92. Bagozzi, Richard P. "Evaluating Structural Equation Models With Unobservable Variables and Measurement Error: A Comment." Journal of Marketing Research 18, (1981): 375-381. Bagozzi, Richard P., and Yi, Y. “On the Evaluation of Structural Equation Models.” Journal of Academic Marketing Science 16, (1988):74–94. Baltes, Boris B., Thomas E. Briggs, Joseph W. Huff, Julie A. Wright, and

20

5

George A. Neuman. "Flexible and Compressed Workweek Schedules: A Meta-Analysis of Their Effects on Work-Related Criteria." Journal of Applied Psychology 84, no. 4 (1999): 496-513. Berkanovic, Emil. "The Effect of Inadequate Language Translation on Hispanics' Responses to Health Surveys." American Journal of Public Health 70, no. 12 (1980): 1273-1276. Bonney, Norman. "Overworked Britons? Part-Time Work and Work-Life Balance." Work, Employment & Society 19, no. 2 (2005): 391-401. Bordia, Prashant, and Gary Blau. "Pay Referent Comparison and Pay Level Satisfaction in Private Versus Public Sector Organizations in India." International Journal of Human Resource Management 9, no. 1 (1998): 155-167. Boyar, Scott L., Carl P. Maertz Jr, Allison W. Pearson, and Shawn Keough. "Work-Family Conflict: A Model of Linkages between Work and Family Domain Variables and Turnover Intentions." Journal of Managerial Issues 15, no. 2 (2003): 175-190. Brislin, Richard W. "Comparative Research Methodology: Cross‐Cultural Studies." International Journal of Psychology 11, no. 3 (1976): 215229. Byrne, Una. "Work-Life Balance Why Are We Talking About It At All?" Business Information Review 22, no. 1 (2005): 53-59. Becker, Franklin, Kristen L. Quinn, and Livit U. Callentine. The Ecology of the Mobile Worker. Workscape 21: The Ecology of New Ways of Working. State University of New York: Ithaca College of Human Ecology at Cornell University, 1995. Campbell, Donald T., and Donald W. Fiske. "Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix." Psychological Bulletin 56, no. 2 (1959): 81-105. Chin, Wynne W. "Commentary: Issues and Opinion on Structural Equation Modeling." MIS Quarterly, 22 (1998): vii-xvi. Clark, Sue Campbell. "Work/Family Border Theory: A New Theory of Work/Family Balance." Human Relations 53, no. 6 (2000): 747-770. Cortina, Jose M. "What is Coefficient Alpha? An Examination of Theory and Applications." Journal of Applied Psychology 78, no. 1 (1993): 98104. Cronbach, Lee J., and Paul E. Meehl. "Construct Validity in Psychological Tests." Psychological Bulletin 52, no. 4 (1955): 281-302. Doyle, Judith, and Richard Reeves. Time Out: The Case for Time Sovereignty. London: The Work Foundation, 2001. EOC. Working Outside the Box: Changing Work to Meet the Future. Survey Report. EOC, Manchester, 2007. Retrieved July 6, 2013, from


21

http://www.equalityhumanrights.com/uploaded_files/Employers/workin g_outside_box_summary.pdf Fairbrother, Kerry, and James Warn. "Workplace Dimensions, Stress and Job Satisfaction." Journal of Managerial Psychology 18, no. 1 (2003): 8-21. Field, Andy. Discovering Statistics using SPSS for Windows. CA: Thousand Oaks, Sage Publications, 2000. Fisher-McAuley, Gwenith, J. Stanton, J. Jolton, and James Gavin. "Modelling the Relationship between Work-Life Balance and Organizational Outcomes." In Annual Conference of the Society for Industrial-Organizational Psychology. Orlando, pp. 1-26. 2003. Fornell, Claes, and David F. Larcker. "Evaluating Structural Equation Models with Unobservable Variables and Measurement Error." Journal of Marketing Research 18, no. 1 (1981): 39-50. Fornell, Claes, and Fred L. Bookstein. "Two Structural Equation Models: LISREL and PLS Applied to Consumer Exit-Voice Theory." Journal of Marketing Research 19, no. 4 (1982): 440-452. Forsyth, Stewart, and Andrea Polzer‐Debruyne. "The Organisational Pay‐ Offs for Perceived Work-Life Balance Support." Asia Pacific Journal of Human Resources 45, no. 1 (2007): 113-123. Frankenhaeuser, Marianne, Ulf Lundberg, Mats Fredrikson, Bo Melin, Martti Tuomisto, Anna‐Lisa Myrsten, Monica Hedman, Bodil Bergman‐Losman, and Leif Wallin. "Stress On and Off the Job as Related to Sex and Occupational Status in White‐Collar Workers." Journal of Organizational Behavior 10, no. 4 (1989): 321346. Friedman, Stewart D., and Jeffrey H. Greenhaus. Work and Family-Allies or Enemies?: What Happens When Business Professionals Confront Life Choices. New York: Oxford University Press, 2000. Garver, Michael S., and John T. Mentzer. "Logistics Research Methods: Employing Structural Equation Modeling to Test for Construct Validity." Journal of Business Logistics 20, no. 1 (1999): 33-57. Gerson, Kathleen. The Unfinished Revolution: How a New Generation is Reshaping Family, Work, and Gender in America. New York: Oxford University Press, 2010. Ghiselli, Richard F., Joseph M. La Lopa, and Billy Bai. "Job Satisfaction, Life Satisfaction, and Turnover Intent: Among Food-Service Managers." Cornell Hospitality Quarterly 42, no. 2 (2001): 28-37. Gliem, Rosemary R., and Joseph A. Gliem. "Calculating, Interpreting, and Reporting Cronbach’s Alpha Reliability Coefficient for Likert-Type Scales." Proceedings of the 2003 Midwest Research-to-Practice

22

5

Conference in Adult, Continuing, and Community Education, 2003: 8288. Retrieved July 6, 2013, from http://www.ssnpstudents.com/wp/wpcontent/uploads/2015/02/Gliem-Gliem.pdf Green, Amie S., Eshkol Rafaeli, Niall Bolger, Patrick E. Shrout, and Harry T. Reis. "Paper or Plastic? Data Equivalence in Paper and Electronic Diaries." Psychological Methods 11, no. 1 (2006): 87-105. Green, Robert T., and Phillip D. White. "Methodological Considerations in Cross-National Consumer Research." Journal of International Business Studies 7, no. 2 (1976): 81-87. Grover, Steven L., and Karen J. Crooker. "Who Appreciates Family‐ Responsive Human Resource Policies: The Impact of Family‐Friendly Policies on the Organizational Attachment of Parents and Non‐ Parents." Personnel Psychology 48, no. 2 (1995): 271-288. Hair, J. F., R. E. Anderson, R. L. Tatham, and W. C. Black. Multivariate Data Analysis. 5th (ed). Englewood Cliffs, NJ: Prentice-Hall Inc. USA, 1998. Hartwick, Jon, and Henri Barki. "Explaining the Role of User Participation in Information System Use." Management Science 40, no. 4 (1994): 440-465. Hatcher, Larry, and Edward J. Stepanski. A Step-By-Step Approach to Using the SAS System for Univariate and Multivariate Statistics. NC: SAS Institute, 1994. Hayman, Jeremy. "Psychometric Assessment of an Instrument Designed to Measure Work- Life Balance." Research and Practice in Human Resource Management 13, no. 1 (2005): 85-91. Heraty, Noreen, Michael J. Morley, and Jeanette N. Cleveland. "Complexities and Challenges in the Work-Family Interface." Journal of Managerial Psychology 23, no. 3 (2008): 209-214. Hoe, Siu Loon. "Issues and Procedures in Adopting Structural Equation Modeling Technique." Journal of Applied Quantitative Methods 3, no. 1 (2008): 76-83. Hurley, Amy E., Terri A. Scandura, Chester A. Schriesheim, Michael T. Brannick, Anson Seers, Robert J. Vandenberg, and Larry J. Williams. "Exploratory and Confirmatory Factor Analysis: Guidelines, Issues, and Alternatives." Journal of Organizational Behavior 18, no. 6 (1997): 667-683. Jöreskog, Karl G., and Dag Sörbom. LISREL 8: Structural Equation Modeling With the SIMPLIS Command Language. Scientific Software International, Lincolnwood 2002. Kaplan, Robert M., and Dennis P. Saccuzzo. Psychological Testing: Principles - Applications and Issues. 3rd ed., Brooks Cole, CA: Pacific


23

Grove, 1993. Khan, Sami A., and Kakul Agha. "Dynamics of the Work-Life Balance at the Firm Level: Issues and Challenges." Journal of Management Policy and Practice 14, no. 4 (2013): 103-114. Long, J. S. Covariance Structure Models: An Introduction to LISREL. Beverly Hills, CA: Sage, 1983. Lundberg, Ulf, Bertil Mårdberg, and Marianne Frankenhaeuser. "The Total Workload of Male and Female White Collar Workers as Related to Age, Occupational Level, and Number of Children." Scandinavian Journal of Psychology 35, no. 4 (1994): 315-327. Malhotra, Naresh K. Review of Marketing Research. Howard House: Emerald Group Publishing Limited, 2004. Malhotra, Naresh K., and David F. Birks. Marketing Research: An Applied Approach. New York: Pearson Education, 2003. Medsker, Gina J., Larry J. Williams, and Patricia J. Holahan. "A Review of Current Practices for Evaluating Causal Models in Organizational Behavior and Human Resources Management Research." Journal of Management 20, no. 2 (1994): 439-464. Mentzer, John T., William DeWitt, James S. Keebler, Soonhong Min, Nancy W. Nix, Carlo D. Smith, and Zach G. Zacharia. "Defining Supply Chain Management." Journal of Business Logistics 22, no. 2 (2001): 1-25. Nunnally, J. C. and Bernstein, I. H. Psychometric Theory (3rd ed.). McGraw Hill, New York, 1994. Pollitt, D. “Faber Maunsell’s “Family” Feel Puts a New Slant on Work-Life Balance.” Human Resource Management International Digest 12, no. 4 (2004): 4-6. Rietveld, Toni, and Roeland Van Hout. Statistical Techniques for the Study of Language and Language Behaviour. Mishawaka, IN: Walter de Gruyter, 1993. Schumacker, Randall E., and Richard G. Lomax. A Beginner's Guide to Structural Equation Modeling. New York: Routledge, Psychology Press, 2004. Sila, Ismail, and Maling Ebrahimpour. "Critical Linkages among TQM Factors and Business Results." International Journal of Operations & Production Management 25, no. 11 (2005): 1123-1155. Silva, Ricardo, and Richard Scheines. "Generalized Measurement Models." (2005). Retrieved June 10, 2013, from http://www.cs.cmu.edu/~rbas/cald04-101.pdf Simpson, Ruth. "Presenteeism, Power and Organizational Change: Long Hours as a Career Barrier and the Impact on the Working Lives of

24

5

Women Managers." British Journal of Management 9, no.1 (1998): 3750. Smithson, Janet, and Elizabeth H. Stokoe. "Discourses of Work-Life Balance: Negotiating ‘Genderblind’ Terms in Organizations." Gender, Work & Organization 12, no. 2 (2005): 147-168. Sparks, Kate, Cary Cooper, Yitzhak Fried, and Arie Shirom. "The Effects of Hours of Work on Health: A Meta‐Analytic Review." Journal of Occupational and Organizational Psychology 70, no. 4 (1997): 391408. Standen, Peter, Kevin Daniels, and David Lamond. "The Home as a Workplace: Work-Family Interaction and Psychological Well-Being in Telework." Journal of Occupational Health Psychology 4, no. 4 (1999): 368-381. Steenkamp, Jan-Benedict EM, and Hans CM Van Trijp. "The Use of LISREL in Validating Marketing Constructs." International Journal of Research in Marketing 8, no. 4 (1991): 283-299. Su, Chia-Ting, and L. Diane Parham. "Generating a Valid Questionnaire Translation for Cross-Cultural Use." American Journal of Occupational Therapy 56, no. 5 (2002): 581-585. Tavakol, Mohsen, Mohammad Ali Mohagheghi, and Reg Dennick. "Assessing the Skills of Surgical Residents Using Simulation." Journal of Surgical Education 65, no. 2 (2008): 77-83. Werts, Charles E., Robert L. Linn, and Karl G. Jöreskog. "Intraclass Reliability Estimates: Testing Structural Assumptions." Educational and Psychological Measurement 34, no. 1 (1974): 25-33. Westen, Drew, and Robert Rosenthal. "Quantifying Construct Validity: Two Simple Measures." Journal of Personality and Social Psychology 84, no. 3 (2003): 608-618. Wu, C. H. "Factor Analysis of the Adult Attachment Scale-Taiwan Version." In 44th Annual Conference of the Taiwanese Psychological Association, Tau-Yuan, Taiwan. 2005. Yin, R. Case Study Research: Design and Methods (2nd ed.). Beverly Hills, CA: Sage Publishing, 1994.