Identifying Critical Factors of Work-Life Imbalances among Employees in the Newspaper Industry

Kavitha L

Assistant Professor & Head, Department of Commerce, MMNSS College, Kottiyam, Kerala, 691571, India

Corresponding Author Email: kavithaleela@gmail.com

Abstract

This study attempts to identify the reasons that lead to problems in work and personal life using PCA and confirmatory factor analysis. The present study uses the scientific method to identify a few variables that influence the WLB of employees in the newspaper industry. The study focuses on the responses to 10 WLB statements from 385 respondents selected from various newspaper establishments in Kerala, a state in India. The survey method was used for data collection, using a structured questionnaire on a 5-point scale. The principal component analysis with varimax rotation and ordinal regression has been performed to identify various problems affecting the WLB and their impact on their personal lives. The study found all ten statements statistically valid based on Cronbach’s internal consistency alpha. Thus, all ten statements can be considered valid for assessing employees' problems. The present study will benefit newspaper organizations in designing an effective workplace with an effective work-life balance and minimum work interference in employees' personal lives.

Keywords: Family Work Interference; Factor Analysis; Principal Component Analysis; Work- Family Interference; Work-Life Balance

Introduction

Work-Life balance is an emerging phenomenon in today’s modern world organizations. Behaving in a different way can improve employees’ work life balances. Organizations must be aware of the issues they must address or eliminate to enable, employees to achieve proper work-life balance. It was found that organizations should provide better facilities and amenities to employees so that they can lead better-balanced lives. It also enables them to achieve their personal and professional goals within a short span. Moreover, it enables them to improve their behavior in the workplace. So, giving significance to life over work or work over professional life is not going to yield anticipated returns to both the employees and the organizations. Work-life balance is gaining significant importance in human resource management as it aims to assist employees in bringing an equilibrium between their professional and personal lives with a glut of programs and interventions (Cahill et al., 2015). Employees who are facing distractions due to work-life balance pressures end up costing companies too much by lesser productivity, increased absenteeism, and increased medical expenses, which has severe repercussions on business earnings (Allen et al., 2013). Imbalance means disparity between two corresponding factors, and in this study, it is the work and non-work domain of employees in the newspaper industry. The employees in the newspaper industry are constantly under pressure, and this study will help in suggesting ways to avoid imbalances in work and personal life by listing our needs and commitments, handling stress in positive ways, and occasionally adjusting work timings so that they will become more productive in a briefer time. Work-life inequities often decrease job performance because longer hours spent at work tend to be unexplored normally rather than used completely. Family dysfunction and stress-related illness can be eliminated to a certain extent by achieving a proper balance between work and free time. Preserving a vigorous balance between work and life is generally well-thought-out as essential to improving the eminence of life as a whole. The current study proposes to identify the principal components of problems faced by employees in the newspaper industry.

Literature Review

Several studies have been conducted on the various aspects of WLB and have secured significant relevance due to varying employee perceptions, assorted workforce, shortage of skilled employees, and shifting parts of male and female workers. Flexibility in work and family-supportive practices leads to increased commitment among employees, which leads to increased productivity and lesser employee turnover. Work environments with supportive management and supple work arrangements as work-life conflict moderators were suggested by Lowe & Schellenberg (2006). Lamovsek et al. (2022) found that the best work and life balance pattern is reached when they get sufficient societal support, proactivity, and less formalization. Mauno, Kinnunen, & Ruokolainen, (2007) conducted a study on employees working in different sectors of an economy- health care, manufacturing sectors, supermarkets, and banks- and showed that interference from work to family was more predominant than interference from family to work. It was also found that no gender difference between work and family and family-to-work conflict exists. The matter without any discrimination has a bearing on both genders. Women's involvement in today’s workforce is increasing, and families nowadays are shifting from combined to nuclear family structures. Wey Smola, & Sutton (2002) found that today’s youth emphasize the accomplishment of work-life balance more than their preceding generation. Sandhu, & Mehta (2006), in their study, project that gender role attitude and spillover roles have an integral part in disturbing the career of employed women. WLB initiatives are essential in managing stress, improving performance, and making employees more motivated and loyal to the organization. Employers have to give due importance to the benefits package as it acts as an essential tool for retaining employees. Kedia, Shukla, & Sinha (2023) highlighted that work-life balance is critical for reducing stress levels and avoiding burnout. The study also shows that a healthier and more productive workplace can be fostered by properly supporting employees to balance their work and private everyday tasks effectively. This can lead to improved mental health and overall job gratification. In their study, Borowiec and Drygas (2022) indicated the connection between a lower level of work-life balance and worse mental and physical health. The adoption of work-life balance policies has proven a relationship between the utilization of the policies and decreased work-family conflict (Shin & Enoh, 2020; Oludayo & Omonijo, 2020)

Methodology

Work-life imbalance can arise from personal or social life responsibilities, such as caring for an ailing and older family member, and result in a physical and mental strain that can severely affect the workers’ workplace performance and efficiency (Thörel, Pauls, & Göritz 2022). Many studies indicate that work-life imbalance causes occupational stress; one of the significant contributions of the present study is that it provides valid scientific evidence that occupational stress significantly negatively influences work-life balance. The present study is predominantly grounded on primary data. The primary data were collected from employees of the top three circulated Malayalam newspaper dailies, who are leaders in the newspaper industry in Kerala in terms of circulation. For data collection, a structured survey questionnaire was used, and the employees in the newspaper industry were questioned about their demographic details and the various problems they faced in their work and life. Issues related to the main aspects were discussed in detail with experts and other eminent persons in the industry before the preparation of the questionnaire.

The problems associated with work-life imbalance were analyzed using a two-step process. First, an exploratory factor analysis was conducted using principal component analysis to identify the underlying factors. Then, a confirmatory factor analysis was performed to validate the identified factors. This rigorous approach ensures the reliability and validity of the findings.

Problems faced in Work-Life Imbalance of employees in newspaper industry in Kerala – EFA

A principal component analysis (PCA) was conducted on a 10-item questionnaire that measured the ‘Problems faced in Work-Life Imbalance’ of employees in the newspaper industry in Kerala. The questionnaire was rated on a 5-point scale, ranging from 'strongly disagree 'to 'agree strongly '. The sample set of 385 observations was randomly split into two sets, with 193 and 192 samples each. The first set was used for exploratory factor analysis, and the second set was used for confirmatory factor analysis. The profile of the 193 respondents used for the exploratory factor analysis is presented in Table 1, providing a snapshot of the study's findings.

Table 1: Snapshot Study Finding’s

Sources: Researcher’s Computed Data

The suitability of PCA was assessed prior to analysis, and the descriptive statistics of the variables subjected to dimension reduction are shown in Table 2.

Table 2: Descriptive Statistics

Sources: Researcher’s Computed Data

The most prominent measure of Problems faced in Work-Life Imbalance was found to be ‘non- availability of flexitime work arrangement is always causing difficulties’ (𝑋𝑋�= 2.7617 ± 1.218) while ‘Unable to concentrate either in work or in family matters’ (𝑋𝑋�= 2.0207 ± 0.963) was the least important measure of Problems faced in Work-Life Imbalance.

Assumptions

The applicability of the following assumptions is checked before proceeding with the factor analysis.

Multivariate normality

The 193 observation data series was examined to determine whether a multidimensional data sample could legitimately come from a multivariate normal distribution. The squared Mahalanobis distances were calculated for the data and shown against the quantiles of a Chi-square distribution to evaluate the assumption of multivariate normality (DeCarlo, 1997; Field, 2009). The solid line in the scatterplot symbolizes the theoretical quantiles of a normal distribution. A reasonably straight line connecting the spots indicates normality. Figure 1 displays the normality scatterplot.

Figure 1: Scatterplot Testing Multivariate Normality

Sources: Researcher’s Computed Data

In general, normality of the input data is not necessary for linear factor analysis. Distributions that are slightly skewed are acceptable. Although common factors and the input data are not assumed to be expected, the model's unique factors are acknowledged.

Factorability

To assess the data's factorability, Pearson correlations were calculated to determine the intercorrelations for each variable. According to Tabachnick, Fidell & Ullman (2019), correlation coefficients should exceed .30 to justify compressing the data into factors. Table 3 depicts the Pearson’s coefficient of correlation between the variables in the analysis.

Table 3: Correlation Matrix

Sources: Researcher’s Computed Data

All variables appeared to be appropriate forfactor analysis because they all had at least one correlation value larger than 0.30.

Multicollinearity

Although variables should be intercorrelated with one another, variables that are too highly correlated can cause problems in EFA. To assess multicollinearity, the determinant of the correlation matrix was calculated. Multicollinearity in the data is indicated by a determinant with a value of less than 0.00001. (Field, 2009).

The correlation matrix's determinant value was 003 (0.00001), indicating that the data lack multicollinearity and that the model's predictions may be accurate.

The factorability tests indicate whether or not trying to extract factors from a set of variables is likely to be worthwhile. Table 4 displays the factorability test results.

Table 4: KMO and Bartlett's Test

Sources: Researcher’s Computed Data

The KMO statistic demonstrates the amount of shared variance. The KMO typically needs to be at least 0.6 for factorization to be beneficial. Since factorization is likely to produce important answers concerning any underlying factors, a KMO of 0.866 is considered commendable. The hypothesis that the correlation matrix among the variables is an identity matrix, indicating that they do not share a similar variance, is tested by Bartlett's test for sphericity. This hypothesis is disproved because the p- value is less than 0.05, indicating no identity matrix. Bartlett's test is susceptible and is typically disregarded unless there are less than five samples for each variable. In this instance, the average number of samples per variable is 19.3. The initial communality estimates have been set, assuming that standard variables cause all of the data's variability because the main components approach was used. The table displays the initial communalities and commonalities following extraction.

Table 5: Communalities

Sources: Researcher’s Computed Data

The eigenvalues for each of the factors are displayed on the screen plot in Figure 2. The eigenvalues are related to the proportion of the data's variability that may be attributed to the components. The horizontal line at 1.0 served as the threshold for choosing to extract two components.

Sources: Researcher’s Computed Data

Figure 2: Scree Plot

Only two factors have been extracted since only 2 of them had eigenvalues greater than or equal to

1.0. The eigenvalues and total variances explained before and after varimax rotation are shown in Table 6.

Table 6: Total Variances Explained

Sources: Researcher’s Computed Data

Together, the two elements are responsible for 66.797% of the variance in the original data before and after varimax rotation. In contrast, factor 2 explains more, with 30.368% after rotation, up from 13.888% before rotation, while factor 1 explains less, with 36.429% after rotation, down from 52.91% before rotation. Table 7 displays the factor loading matrix following varimax rotation.

Table 7: Rotated Component Matrix

Sources: Researcher’s Computed Data

The factors identified were named as:

1. Loss of Self-Confidence

   This factor, which accounted for 36.429 percent of total variances after rotation, consisted of ‘Getting angry over subtle issues spoils relations with co-workers,’ ‘Feeling unhappy and depressed always affects mental health,’ ‘Inability to take career break further induces mental pressure,’ ‘Often loses self-confidence and feels worthless,’ ‘Unable to concentrate either in work or in family matters,’ ‘The profession by itself is creating turbulence not only in mind but also in family’ and ‘Feeling of inability to take care of children and elders at home is always a concern’.

2. Excessive Workhours

The second factor, which accounted for 30.368 percent of total variances, consisted of ‘the non- availability of a flexitime work arrangement is always causing difficulties,’ ‘Excessive work hours that affect family commitments have always been a concern,’ and ‘the Decision to continue in a job or quit for the more relaxed family is always haunting life.’

Problems faced in Work-Life Imbalance – Confirmatory factor analysis

The second portion of randomly split data, consisting of 192 observations, was subjected to confirmatory factor analysis using structural equation modelling. The sample respondents' profile is shown in Table 5.26 under section 5.13 of this report.

The factors identified as principal components through exploratory factor analysis from split sample 1 were subjected to confirmatory factor analysis using split sample 2 data in the model shown in Figure C.

Sources: Researcher’s Computed Data

Figure C Problems faced in Work-Life Imbalance – Measurement Model

The first-order constructs of ‘Loss of Self-confidence’ and ‘Excessive work hours’ identified as principal components from dimension reduction are reflections of the second-order construct, namely ‘Problems faced in Work-Life Imbalance.’ Table 11 shows the details of acronyms used in the model diagram.

Table 11: Variables in the Model – Legends

Sources: Researcher’s Computed Data

Internal Reliability Test

For each of the latent constructs shown in the model, the internal reliability of the model, which indicates how well the measuring items are holding together in assessing the relevant construct, was calculated using Cronbach's Alpha and is provided in Table 12.

Table 12 Reliability Statistics

Sources: Researcher’s Computed Data

As a measure of scale reliability Cronbach’s alpha is a measure of internal consistency, which shows how closely related a set of items are as a group. A reliability coefficient of 0.70 or higher is considered “acceptable” in most social science research situations. Hence, the model shown above can be said to be possessing internal reliability since the Cronbach’s Alpha for the different sets of measured variables for each latent construct was found to be acceptable since it is greater than 0.70 in both the cases.

Composite Reliability Test and Convergent Validity Test

Convergent validity and composite reliability, which measure the latent variables' internal consistency, were calculated and presented.

The following formula was used to determine the latent variable's internal consistency or composite reliability.

CR = (ΣҚ) 2 / [(ΣҚ)2 + (Σ1- Қ2)]

Where Қ = factor loadings on each item and 1-Қ = error

This convergent validity could be confirmed in addition to eliminating low factor loading items in a model that could lead to the construct failing convergent validity by determining the Average Variance Extracted (AVE) for the construct as follows.

AVE= Σ Қ2 / n

where n is the number of items in the measurement model and the factorloadings on each item. Table 13 displays the composite model's average variance explained (AVE) and composite reliability (CR).

Table 13: Average Variance Explained (AVE) and Composite Reliability

Sources: Researcher’s Computed Data

The CR value of 0.9451 is significantly higher than the threshold value of 0.70, indicating that the model underexamination has achieved composite reliability, reflecting the latent structures' reliability and internal consistency.

Convergent validity is stated to be attained if AVE exceeds the specified threshold value of 0.5. Since the AVE for the model underexamination is 0.5972, convergent validity is said to have been attained.

Discriminant Validity Test

The model has ensured discriminant validity, which shows that the composite model is free from redundant items, by identifying the redundant items in the model using a discrepancy measure called Modification Indices (MI) and constraining the redundant pair as a "free parameter estimate" for the redundant items that have a high value of Modification Indices.

Another prerequisite for discriminant validity is that endogenous construct correlation should not be more than 0.85. When the correlation is more than 0.85, there is substantial multicollinearity, or the two variables are redundant. Table 14 displays the correlations between the model's variables.

Table 14: Implied Correlations Between Variables

Sources: Researcher’s Computed Data

None of the model's inter-variable correlations were more than 0.85, indicating that multicollinearit y is not a concern.

Construct Validity Test

When the Fitness Indexes for a concept reach the necessary level, construct validity is attained. The fitness indices show how well-suited the items are to measuring each latent construct. Table 15 displays the fitness indices for the "Problems confronted in Work-Life Imbalance" model.

Table 15: Fitness Indices – ‘Problems Faced in Work Life Imbalance’ Model

Sources: Researcher’s Computed Data

All of the indices were within acceptable bounds, meaning that the threshold value requirements were satisfied. Consequently, the model is regarded as having an excellent fit. Table 16 displays the standard regression weights together with their probability values.

Table 16: Standardised Regression Weights –Problems Faced in Work Life Imbalance Model

Sources: Researcher’s Computed Data

All measured variables have a high likelihood of being significant standardised regression weights, which suggests that they are accurate predictors of the latent variables.

The lower than 0.001 "p" values of the measured variables' standardised regression weights make it clear that the regression weights for the measured variables are substantially different from zero at the 0.001 level (two-tailed). Table 17 displays the test findings and conclusions for the following null hypotheses.

Table 17: Hypothesis Test Results – ‘Problems faced in Work Life Imbalance’ Model

Sources: Researcher’s Computed Data

Discussion

Das & Kushwaha (2013) in their study found four factors covering 16 items having Eigen values of more than 1 which accounted for about 52.9% of variation in data. According to his study thirteen statements were found to be statistically valid on the basis of Cronbach’s alpha of internal consistency. In the present study, the null hypothesis gets rejected in all the cases, indicating that the identified variables are significant predictors of the principal components recognised from exploratory factor analysis and these principal components are significant reflections of ‘Problems faced in WLB’ as the calculated value is greater than the table value at 0.01 significance level. Faldu, JVIMS, & Udaykumar (2022) found that healthy working conditions and colleagues support significantly influence work life balance of employees. Present study found loss of self-confidence, excessive work hours, inability to take care of children and elders at home, turbulence at work and family are all items that cause problems in work life imbalances among employees in newspaper industry. Moreover, the factorinability to concentrate in work and family, feeling worthless, inability to take career breaks also induces mental pressure. From the study we can infer that feeling depressed, getting angry over subtle issues, non- availability of flexi time (Flexible working arrangements are recommended for balancing work and family commitments Gregory & Milner 2009) excessive work, decision to quit for more relaxed family are all factors that accelerate problems in work life. Bakar, (2024) found an encouraging correlation between work life balance and various aspects of wellbeing, job satisfaction, mental health and overall life satisfaction. The study results also show that work life balance is steadily related with enhanced wellbeing products. Hence these factors can be considered as critical factors in identifying work life imbalances among employees in newspaper industry.

Results

On the basis of analysis, it was found that mahalanobis distance plot shows relatively straight line and no multicollinearity found. Determination of the new correlation matrix shows 0.003 which is greater than 0.00001. KMO measure of sampling adequacy of 0.866 is adequate. Chi square of 1107.99 at 45 degree of freedom, p<0.05, hence no identity matrix. On the basis of above discussion 2 principal components extracted which together explained 66.797%. The factors identified were Loss of Self Confidence and Excessive work hours. The identified variables are significant predictors of the principal components recognised from exploratory factor analysis and these principal components are significant reflections of problems faced in work life imbalance. These factors were then subject to confirmatory factor analysis. The internal reliability based on Cronbach’s alpha of 10 items is found greater than 0.70. Composite reliability is 0.9451 which is also above 0.70. AVE is greater than 0.5972 and hence the convergent validity of the construct is adequate. Since maximum coefficient of correlation is 0.646 multicollinearity is absent. It was observed that all the indices were in the acceptable range and the threshold value conditions being met. Hence the model is considered to be a good fit.

Conclusion

The key objective of the study was to identify the critical factors of work life imbalance on the working life of employees in newspaper industry. Principal component analysis was conducted with varimax rotation for identifying the factors. The study found 2 factors having Eigen value of more than 1. These factors accounted for about 66.797% of variation in data. The various factors out of 10 statements are found useful as a scale for measuring the work life imbalances of employees in newspaper industry. The study helped in identifying a suitable scale for work life balance in newspaper establishments and these establishments can emphasis the issues identified in this study for recreating the working environment and amenities so that a better balance can be achieved in the employees’ personal and professional lives.

Scope for Further Research

More variables that influence the work life imbalances can be incorporated and similar studies covering more sample size can be undertaken. The same study can be undertaken in other sectors or organisations. The study identified a useful scale for work life imbalances of employees in newspaper industry and other industries can also concentrate on the issues identified in this study.

Limitations of the Study

The study covers employees of a regional newspaper in the state of Kerala. The response from the employees were collected based on their opinions and perceptions. These could change according to their personal feelings and contexts or time. Most of the time the employees working in this industry were reluctant to reveal the real situations/plight they are facing. Developing a rapport with the employees helped to overcome this limitation.

Conflict of Interest

The authors declare that they have no competing interests.

Acknowledgment

The author sincerely express gratitude to Dr. Dileep A S and Dr. Suresh Kumar S in extending their time and support and insightful suggestions to strengthen the content of the article.

References

Allen, T. D., Johnson, R. C., Kiburz, K. M., & Shockley, K. M. (2013). Work–family conflict and flexible work arrangements: Deconstructing flexibility. Personnel Psychology, 66(2), 345-376. https://doi.org/10.1111/peps.12012

Bakar, A. Y. A. (2024). The Application of ‘Work-Life Balance’in Relation with Employees’well- Being: An Exploratory Study. Journal of Law and Sustainable Development, 12(2), e3243-e3243. https://doi.org/10.55908/sdgs.v12i2.3243

Borowiec, A. A., & Drygas, W. (2022). Work–life balance and mental and physical health among Warsaw specialists, managers and entrepreneurs. International Journal of Environmental Research and Public Health, 20(1), 492. https://doi.org/10.3390/ijerph20010492

Cahill, K. E., McNamara, T. K., Pitt-Catsouphes, M., & Valcour, M. (2015). Linking shifts in the national economy with changes in job satisfaction, employee engagement and work–life balance. Journal of Behavioral and Experimental Economics, 56, 40-54.

https://doi.org/10.1016/j.socec.2015.03.002

Das, S. C., & Kushwaha, M. S. (2013). Identifying critical factors of work–life balance and its impact on insurance employees in India–an exploratory factor analysis. Time’s Journey, 2(1), 1-14. https://doi.org/10.17632/W6FKTM8JWX.1

DeCarlo, L. T. (1997). On the meaning and use of kurtosis. Psychological Methods, 2(3), 292. https://psycnet.apa.org/doi/10.1037/1082-989X.2.3.292

Faldu, R., JVIMS, J., & Udaykumar, T. K. (2022). "Genderdifferences: Work Life balance of banking employees in Gujarat State in India". https://doi.org/10.13140/RG.2.2.23827.04640

Field, A. (2009). Discovering statistics using SPSS. London (UK): Sage Publications Ltd. https://www.scirp.org/reference/ReferencesPapers?ReferenceID=1866193. Accessed on 10th June, 2023.

Gregory, A., & Milner, S. E. (2009). Work-life balance: a matter of choice? Gender, Work and Organization, 16(1). 1-13. https://doi.org/10.1111/j.1468-0432.2008.00429

Kedia, S., Shukla, A., & Sinha, S. (2023). Balancing Act: A Comprehensive study of Work Life Balance and its Impact on Employee Performance & Job Satisfaction. European Chemical Bulletin, 12 (Special Issue12), 668-678. https://doi.org/10.48047/ecb/2023.12.si12.057

Lamovšek, A., Černe, M., Radević, I., & Božič, K. (2023). The key to work–life balance is (enriched) job design? Three-way interaction effectswith formalization and adaptive personality characteristics. Applied Research in Quality of Life, 18(2), 647-676. https://doi.org/10.1007/s11482-10100-9

Lowe & Schellenberg., Use of Family Friendly Workplace Practices- IRPP Working Paper, number 2, September 2006. http://www.oecd.org/els/family/database.htm. Accessed on 15th May, 2023.

Mauno, S., Kinnunen, U., & Ruokolainen, M. (2007). Job demands and resources as antecedents of work engagement: A longitudinal study. Journal of Vocational Behavior, 70(1), 149-171.

https://doi.org/10.1016/j.jvb.2006.09.002

Oludayo, A. O., & Omonijo, D. O. (2020). Work-life balance: The relevance of social support. Academy of Strategic Management Journal, 19(3), 1-10.

https://www.abacademies.org/articles/Work-life-balance-the-relevance-of-social-support-1939- 6104-19-3-557.pdf

Sandhu, H. S., & Mehta, R. (2006). Work‐family conflict among women executives in service sector: an empirical study. Journal of Advances in Management Research, 3(2), 68-80. https://doi.org/10.1108/97279810680001246

Shin, D., & Enoh, J. (2020). Availability and use of work–life balance programs: Relationship with organizational profitability. Sustainability, 12(7), 2965.https://doi.org/10.3390/su12072965

Wey Smola, K., & Sutton, C. D. (2002). Generational differences: Revisiting generational work values for the new millennium. Journal of Organizational Behavior: The International Journal of Industrial, Occupational and Organizational Psychology and Behavior, 23(4), 363-382. https://doi.org/10.1002/job.147

Tabachnick, B. G., Fidell, L. S., & Ullman, J. B. (2019). Using Multivariate Statistics. 2nd Edn, Vol.5. NY, NY: Pearson, UK. https://www.pearsonhighered.com/assets/preface/0/1/3/4/0134790545.pdf
Thörel, E., Pauls, N., & Göritz, A. S. (2022). The association of work-related extended availability with recuperation, well-being, life domain balance and work: A meta-analysis. Organizational Psychology Review, 12(4), 387-427. https://doi.org/10.1177/20413866221116309