Validating Cyberloafing Antecedents through Confirmatory Factor Analysis: Evidence from Malaysian Employees

Validating Cyberloafing Antecedents through Confirmatory Factor Analysis: Evidence from Malaysian Employees

¹Shahrul Niza Samsudin*, ¹Nor Saidi Mohamed Nasir., ²Mohd Sufiean Hassan

¹Faculty of Business, Hospitality and Technology, Universiti Islam Melaka, Malaysia

²Faculty of Communication and Media Studies, UiTM, Malaysia

*Corresponding Author

DOI: https://dx.doi.org/10.47772/IJRISS.2025.914MG00105

Received: 04 June 2025; Accepted: 08 June 2025; Published: 10 July 2025

ABSTRACT

This study presents the results of a confirmatory factor analysis (CFA) conducted to validate the measurement model of workplace stressors associated with cyberloafing among Malaysian employees. Building on an earlier exploratory factor analysis, the study examines the structure and validity of four latent constructs; workplace ostracism, role ambiguity, role conflict, ad role overload. Data were collected from 242 employees in various public and private sector organizations. The measurement model was tested using Structural Equation Modelling (SEM) with AMOS 26.0. The CFA results indicated that the proposed four-factor model achieved acceptable good fit: χ²/df = 2.003, CFI = 0.932, TLI = 0.919, RMSEA = 0.065, and SRMR = 0.045. All factor loadings were significant and exceeded the 0.60 threshold. Composite reliability (CR) values for each construct ranged from 0.89 to 0.94, and average variance extracted (AVE) values exceeded the 0.50 benchmark, indicating strong convergent validity. Discriminant validity was also confirmed via the Fornell-Larcker criterion. These findings provide robust psychometric support for the use of this four-construct model in future cyberloafing research. The validated instrument enhances the theoretical understanding of workplace antecedents to cyberloafing and provides a reliable measurement tool for researchers and practitioners.

Keywords: Cyberloafing, Confirmatory Factor Analysis, Workplace Stressors, Structural Equation Modelling, Malaysia

INTRODUCTION

The rapid integration of internet technology into the workspace has introduced new challenges for organizations, particularly in the form of cyberloafing, which is the employee’s use of digital resources for non-work-related purposes during working hours (Lim, 2002). While some view this behaviour as a harmless coping mechanism, excessive or habitual cyberloafing may result in productivity loss, weakened organizational discipline, and information security risks (Hadlington & Parsons, 2017; Henle & Blanchard, 2008). As such. Understanding the antecedents that lead employees to engage in cyberloafing has become a critical area of inquiry in organizational behaviour and human resource management.

Existing literature has identified a range of antecedents to cyberloafing, particularly those related to job stress and negative workplace experiences (Koay et al., 2017; Soh et al., 2022). Stress-inducing factors such as role ambiguity, role conflict, role overload, and workplace ostracism are frequently cited as drivers of withdrawal behaviours, including cyberloafing (Arshad et al., 2016; Ferris et al., 2008; Rizzo et al., 1970). However, the accurate measurement of these constructs within the cyberloafing context remains a methodological challenge. While prior studies have employed adapted instruments, few have provided comprehensive validation of their factor structures in a cyberloafing-specific setting.

To address this gap, a previous exploratory factor analysis (EFA) conducted during the pilot phase of this research identified a four-factor structure comprising workplace ostracism, role ambiguity, role conflict, and role overload as key antecedents to cyberloafing. While the initial EFA results demonstrated acceptable psychometric properties, further validation is required to confirm the consistency and robustness of this measurement model across a larger and more representative sample.

The present study aims to validate this four-factor model using confirmatory factor analysis (CFA). Specifically, the objectives are to (1) test the overall model fit of the proposed measurement structure, and (2) evaluate the construct validity including convergent and discriminant validity of the latent constructs. By establishing the validity and reliability of this measurement model, this study contributes a psychometrically sound tool for future cyberloafing research and provides practical implications for diagnosing workplace stressors that may lead to cyberloafing.

LITERATURE REVIEW AND THEORETICAL BACKGROUND

Cyberloafing is often framed not merely as a behavioural response to job stressors, but as a result of individual intention and rationalization processes. A growing body of literature has explored the antecedents of cyberloafing, with particular emphasis on workplace-related stressors and negative organizational experiences. To establish a theoretically grounded and empirically valid measurement model, this study focuses on four constructs previously identified in the exploratory phase; workplace ostracism, role ambiguity, role conflict, and role overload that have been linked to cyberloafing behaviour in previous research. The selection of these constructs is supported by the Theory of Planned Behaviour (TPB), which posits that attitudes, perceived norms, and behavioural control shape intentional behaviour (Ajzen, 1991). Under this framework, workplace stressors can shape employee’s cognitive evaluations and perceived control, thereby influencing their intention to engage in cyberloafing. The following present a review of the conceptual and theoretical foundations of each construct in relation to cyberloafing behaviour.

Cyberloafing and Its Antecedents

Cyberloafing refers to the use of the internet by employees for non-work-related activities during working hours, including browsing social media, online shopping, and streaming entertainment (Lim, 2002). While some research frames cyberloafing as a form of micro-rest or recovery behaviour, it is widely acknowledged as a counterproductive work behaviour when done excessively or in violation of organizational policies (Henle & Blanchard, 2008; Koay et al., 2017).

Research has increasingly focused on the antecedents of cyberloafing, particularly those related to job-related stress and negative workplace experiences (Mercado et al., 2017). Stressors such as role ambiguity, role conflict, role overload, and workplace ostracism have been consistently identified as predictors of withdrawal behaviours, including cyberloafing (Arshad et al., 2016; Soh et al., 2022). These stressors reduce employee’s psychological resources and engagement, making them more likely to disengage from their tasks through digital distractions.

Rizzo et al., (1970) posits that unclear or conflicting expectations (i.e. role ambiguity and role conflict) disrupt employee’s ability to perform effectively. When individuals experience uncertainty about their job responsibilities or receive contradictory instructions, stress level increase, potentially triggering deviant coping responses such as cyberloafing.

Role overload, meanwhile, represents a condition in which employees perceive their workload as excessive or unmanageable. High job demands when unbalanced by adequate resources can lead to strain and withdrawal behaviours (Bakker & Demerouti, 2007). Excessive job demands such as overload, without corresponding support or autonomy, are linked to disengagement and burnout, which may manifest as cyberloafing.

Workplace ostracism is another key antecedent of disengagement, defined as being ignored or excluded by others at work (Ferris et al., 2008). Ostracized employees may experience psychological distress and a sense of disconnection from organizational goals. This social exclusion reduces intrinsic motivation and may lead to compensatory behaviours such as cyberloafing (Soh et al., 2022).

These theoretical perspectives support the inclusion of the four constructs in the proposed measurement model. Prior exploratory factor analysis confirmed their empirical distinctiveness, but further validation is needed to confirm their factorial structure and measurement quality using confirmatory techniques.

Measurement Model Development

The development of the measurement model in this study is grounded in the Theory of Planned Behaviour (TPB) (Ajzen, 1991), which posits that behavioural intention is shaped by three key components; attitudes toward the behaviour, subjective norms, and perceived behavioural control. In the context of cyberloafing, TPB suggests that employees are more likely to engage in this behaviour when they perceive it as acceptable (attitude), believe others also engage in it or condone it (subjective norms), and feel they have control over their actions (perceived behavioural control). These components are shaped by employee’s experiences in the workplace, particularly the presence of stress-inducing conditions that influence their cognitive and emotional responses.

Drawing on this framework, the current study examines four workplace stressors, which are workplace ostracism, role ambiguity, role conflict, and role overload as antecedents that may influence employee’s intention to cyberloafing. These stressors can alter employee’s attitudes toward their work environment, reduce their sense of control, and indirectly contribute to withdrawal behaviours such as cyberloafing. Each construct has been widely discussed in prior research and was identified as a distinct factor during the exploratory phase of this study. The goal of the present research is to validate the factor structure of these constructs through confirmatory factor analysis (CFA), thereby establishing their psychometric properties in the context of cyberloafing research.

METHODOLOGY

To validate the proposed four-factor measurement model of workplace stressors associated with cyberloafing, this study employed a quantitative approach grounded in structural equation modelling. Confirmatory Factor Analysis (CFA) was used to test the factor structure derived from the exploratory phase and to assess the reliability and validity of the constructs. This section outlines the research design, sampling procedures, instrument development, and statistical techniques used to evaluate the measurement model.

Research Design and Participants

This study employed a quantitative, cross-sectional survey design to validate the measurement model of workplace stressors associated with cyberloafing. Data were collected from 242 full time employees working in various public and private sector organizations in Malaysia. A convenience sampling method was used to reach participants, with the inclusion criteria requiring a minimum of six months of full-time work experience to ensure familiarity with organizational roles and dynamics.

Informed consent was collected from all respondents. The survey was administered online. The final sample included respondents across a range of industries including education, healthcare, manufacturing, and services. Demographic variables such as age, gender, sector, and tenure were also collected to profile the sample.

The final sample consisted of 242 full-time employees across public and private sector organizations in Malaysia. Participants were drawn from diverse industries, including education (27%), healthcare (18%), manufacturing (16%), and government services (15%), with the remaining 24% representing retail, finance, and telecommunications. Of the respondents, 56% identified as female and 44% as male. Most participants (61%) were between the ages of 26 and 40. In terms of education, 73% held at least a bachelor’s degree, and 48% reported having over five years of work experience. This broad representation across industry, gender, age, and education enhances the generalizability of the findings.

Instrumentation

The measurement instrument included four constructs identified during the exploratory phase od the research, which is workplace ostracism, role ambiguity, role conflict, and role overload. All items were adapted from previously validated scales in the literature. Workplace Ostracism was measured using the 10-item scale developed by (Ferris et al., 2008), capturing experiences of being excluded or ignored in the workplace. Role Ambiguity, Role Conflict, and Role Overload were measured using adapted items from (Rizzo et al., 1970), a well-established scale in organizational psychology.

Respondents were asked to rate their level of agreement on a 10-point scale, ranging from 1 (strongly disagree) to 10 (strongly agree). Based on exploratory factor analysis results, the final measurement model included 4 items for workplace ostracism, 5 items for role ambiguity, 6 items for role conflict, and 8 items for role overload.

Data Analysis Procedure

Confirmatory Factor Analysis (CFA) was conducted using AMOS version 26.0 to test the validity and fit of the proposed four-factor measurement model. CFA was selected to verify the structure identified in the pilot phase and to assess the psychometric properties of the scale. Model fit was assessed using a range of commonly accepted indices, including the chi-square to degrees of freedom ratio (χ²/df), the Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI), the Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR). Acceptable model fit was determined based on the following thresholds: χ²/df < 3.0, CFI and TLI ≥ 0.90, RMSEA ≤ 0.08, and SRMR ≤ 0.08.

Construct validity was evaluated through convergent and discriminant validity tests. Convergent validity was assessed by examining standardized factor loadings (≥ 0.60), Composite Reliability (CR ≥ 0.70), and Average Variance Extracted (AVE ≥ 0.50). Discriminant validity was tested using the Fornell-Larcker criterion, where the square root of AVE for each construct should exceed its correlations with other constructs. Internal consistency was assessed using Cronbach’s Alpha and composite reliability values. All analyses followed established SEM guidelines, and data were screened for assumptions including missing values and multivariate normality prior to model estimation.

RESULTS

Before presenting the results of the Confirmatory Factor Analysis (CFA), the dataset was examined for missing values, outliers, and violations of normality. The data met all assumptions required for CFA, and no significant issues were detected. Following this, the hypothesized four-factor measurement model was tested to determine how well it fit the observed data. The fit indices, standardized factor loadings, and validity measures are reported in the following subsections.

Confirmatory Factor Analysis

Figure 1: Confirmatory Factor Analysis for Cyberloafing

Figure 1 presents the Confirmatory Factor Analysis (CFA) model assessing the measurement structure of five latent constructs: Workplace Ostracism (WOx), Role Ambiguity (RAx), Role Conflict (RCx), Role Overload (ROx), and Cyberloafing (CLx). Each latent construct is represented by multiple observed indicators, all of which demonstrate satisfactory standardized factor loadings, exceeded the recommended threshold of 0.60 (Hair et. al., 2019). The CFA was conducted using SEM-AMOS, and the model shows an acceptable fit to the data, as indicated by fit indices in Table 1. These values fall within the commonly accepted cut-off points, suggesting adequate convergent validity of the measurement model. The model also depicts the correlation among the independent constructs and their relationship with dependent variable, Cyberloafing (CLx). Overall, the CFA results confirm that the measurement model is statistically sound and suitable for subsequent structural model analysis.

Measurement Model Fit

Confirmatory Factor Analysis was conducted to evaluate the fit of the proposed four-factor measurement model comprising workplace ostracism, role ambiguity, role conflict, and role overload. The initial model demonstrated an acceptable to good fit with the data based on the following indices summarized in Table I, indicating an overall good fit for the four-factor model.

Table 1: Model Fit Indices for the Four-Factor Measurement Model

Source: (Hair et al., 2010, 2018, 2019; Khan et al., 2019; Sarstedt et al., 2017; Sarstet et al., 2016; Zainuddin Awang., 2015)

These results meet the commonly accepted thresholds for model adequacy, suggesting that the hypothesized measurement model is a good representation of the observed data.

Factor Loadings and Convergent Validity

Table 2 presents the standardized factor loadings obtained from CFA. All items demonstrated strong loading values (≥ 0.60), confirming that the observed indicators adequately represent their respective latent constructs. These results support convergent validity of the measurement model.

Table 2: Standardized Factor Loadings from CFA

Source: Author’s compilation

All standardized factor loadings were statistically significant and exceeded the recommended minimum threshold of 0.60, indicating that the observed variables were good indicators of their respective latent constructs. Composite Reliability (CR) values ranged from 0.89 to 0.94, while Average Variance Extracted (AVE) values ranged from 0.68 to 0.83. These results indicate strong convergent validity for all four constructs.

Table 3: Standardized Factor Loadings, Composite Reliability, and Average Variance Extracted

Source: Author’s compilation

(All loadings significant at p<0.001. CR = Composite Reliability, AVE = Average Variance Extracted)

Discriminant Validity

Discriminant validity was assessed using the Fornell-Larcker criterion, which compares the square root of AVE values to inter-construct correlations. As shown in Table III, the square root of each construct’s AVE (shown on the diagonal) exceeded its correlations with the other constructs, confirming adequate discriminant validity.

Table 3: Discriminant Validity Index (Fornell-Larcker Criterion)

Source: Author’s compilation

(Diagonal values are square roots of AVE; off-diagonal values are latent correlations)

DISCUSSION

This study aimed to validate a four-factor measurement model of workplace stressors associated with cyberloafing through confirmatory factor analysis (CFA). The results provided strong evidence for the model’s validity and reliability, supporting the inclusion of workplace ostracism, role ambiguity, role conflict, and role overload as distinct but interrelated constructs. The model demonstrated acceptable to good fit across all indices (χ²/df = 2.003, CFI = 0.932, TLI = 0.919, RMSEA = 0.065, SRMR = 0.045), and all factor loadings exceeded the recommended threshold of 0.60. Furthermore, composite reliability (CR) and average variance extracted (AVE) values confirmed the model’s convergent validity, while the Fornell-Larcker criterion demonstrated discriminant validity between all constructs.

These findings support the use of this four-construct model as a valid tool for assessing workplace stressors that may lead to cyberloafing. The inclusion of these specific antecedents is theoretically grounded in the Theory of Planned Behaviour (TPB), which posits that behavioural intention is shaped by individual attitudes, social norms, and perceived behavioural control (Ajzen, 1991). The stressors assessed in this study such as excessive workload or unclear roles can influence these psychological processes by shaping how employees perceive their environment and how they rationalize engaging in counterproductive behaviours like cyberloafing.

Compared to previous research, this study extends the literature by offering a rigorously validated measurement model specific to cyberloafing antecedents. While earlier studies have explored the influence of role stressors and ostracism to cyberloafing (Arshad et al., 2016; Soh et al., 2022), few have provided psychometric validation of the instruments used. By conducting both EFA and CFA in sequential phases, this study addresses that gap and provides a tool with strong psychometric properties. The particularly high composite reliability values observed (ranging from 0.89 to 0.98) highlight the internal consistency and robustness of the measurement model in a Malaysian context.

Practically, the validated instrument can serve as a diagnostic tool for organizational researchers and HR professionals to assess work-related stress conditions that may lead to disengaged or counterproductive behaviours. Understanding how different stressors influence behavioural intentions can help organizations design targeted interventions such as clarifying job roles, managing workload, or fostering inclusive workplace cultures to reduce the likelihood of cyberloafing.

CONCLUSION

This study validated a four-factor measurement model comprising workplace ostracism, role ambiguity, role conflict, and role overload as key antecedents of cyberloafing behaviour. Using confirmatory factor analysis (CFA) with data from 242 Malaysian employees, the model demonstrated good fit and strong psychometric properties. All constructs showed high internal consistency, convergent validity, and discriminant validity, confirming the robustness of the proposed measurement instrument.

The findings offer both theoretical and practical contribution. From a theoretical standpoint, the validated model reinforces the relevance of the Theory of Planned Behaviour (TPB) in understanding how workplace stressors may shape employee’s intention to cyberloafing. By operationalizing stress-related constructs through a reliable measurement model, this study lays the foundation for future structural model testing involving behavioural intentions and actual cyberloafing behaviour.

Practically, the instrument provides organizations and researchers with a diagnostic tool for identifying stress conditions that may contribute to disengagement and counterproductive work behaviour. Interventions that address role clarity, reduce overload, and foster inclusive work environments may help mitigate cyberloafing by altering employee’s perceptions and attitudes.

Future research could extend this work by validating the model across different cultural or industry contexts, testing the predictive power of the constructs on actual cyberloafing behaviour, or exploring potential moderators such as personality traits, digital literacy, or perceived organizational justice.

REFERENCES

1. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
2. Ajzen, I. (2020). The theory of planned behavior: Frequently asked questions. Human Behavior and Emerging Technologies, 2(4), 314–324. https://doi.org/10.1002/hbe2.195
3. Alias, N. E., Samsudin, Z., & Ismail, A. (2021). Job demands, resources, and counterproductive work behaviour: The role of work engagement. Asia-Pacific Journal of Business Administration, 13(3), 235–251. https://doi.org/10.1108/APJBA-08-2020-0309
4. Arshad, R., Ab Malek, A., Mahmood, R., & Alavi, K. (2016). Role stressors and cyberloafing: A case of Malaysian academics. Asian Social Science, 12(2), 252–260. https://doi.org/10.5539/ass.v12n2p252
5. Askew, K. L., Buckner, J. E., Taing, M. U., Ilie, A., & Bauer, J. A. (2014). Explaining cyberloafing: The role of the theory of planned behavior. Computers in Human Behavior, 36, 510–519. https://doi.org/10.1016/j.chb.2014.04.006
6. Bakker, A. B., & Demerouti, E. (2007). The Job Demands–Resources model: State of the art. Journal of Managerial Psychology, 22(3), 309–328. https://doi.org/10.1108/02683940710733115
7. Ferris, D. L., Brown, D. J., Berry, J. W., & Lian, H. (2008). The development and validation of the Workplace Ostracism Scale. Journal of Applied Psychology, 93(6), 1348–1366. https://doi.org/10.1037/a0012743
8. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
9. Gkorezis, P., & Bellou, V. (2022). Workplace cyberloafing: A review of the literature. Personnel Review, 51(7), 1551–1569. https://doi.org/10.1108/PR-11-2020-0892
10. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate Data Analysis (7th ed.). Pearson Education.
11. Hair, J. F., Risher, Jeffrey J; Marko, S., & Ringle, C. M. (2018). When to use and how to report the results of PLS-SEM. European Business Review, 31(1), 2–24. https://doi.org/10.1108/EBR-11-2018-0203
12. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate data analysis (8th ed.). Cengage Learning.
13. Hadlington, L., & Parsons, K. (2017). Can education and training be used to reduce cybersecurity negligence? Work & Stress, 31(4), 396–411. https://doi.org/10.1080/02678373.2017.1367735
14. Henle, C. A., & Blanchard, A. L. (2008). The interaction of work stressors and organizational sanctions on cyberloafing. Journal of Managerial Issues, 20(3), 383–400. https://www.jstor.org/stable/40604536
15. Khan, G. F., Sarstedt, M., Shiau, W. L., Hair, J. F., Ringle, C. M., & Fritze, M. P. (2019). Methodological research on partial least squares structural equation modeling (PLS-SEM): An analysis based on social network approaches. Internet Research, 29(3), 407–429. https://doi.org/10.1108/IntR-12-2017-0509
16. Koay, K. Y., Soh, P. C. H., & Chew, K. W. (2017). Do employees’ private demands spill over into workplace deviance? Employee Relations, 39(3), 347–362. https://doi.org/10.1108/ER-07-2016-0130
17. Lim, V. K. G. (2002). The IT way of loafing on the job: Cyberloafing, neutralizing, and organizational justice. Journal of Organizational Behavior, 23(5), 675–694. https://doi.org/10.1002/job.161
18. Mercado, B. K., Giordano, C., & Dilchert, S. (2017). Counterproductive work behaviors in response to job stressors: A review and meta-analytic investigation. Journal of Research in Personality, 66, 98–111. https://doi.org/10.1016/j.jrp.2016.11.008
19. Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.
20. Podsakoff, N. P., LePine, J. A., & LePine, M. A. (2007). Differential challenge stressor–hindrance stressor relationships with job attitudes, turnover intentions, and withdrawal behavior: A meta-analysis. Journal of Applied Psychology, 92(2), 438–454. https://doi.org/10.1037/0021-9010.92.2.438
21. Rizzo, J. R., House, R. J., & Lirtzman, S. I. (1970). Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 15(2), 150–163. https://doi.org/10.2307/2391486
22. Sarstedt, M., Ringle, C. M., & Hair, J. F. (2017). Partial Least Squares Structural Equation Modeling. Springer International Publishing. https://doi.org/https://doi.org/10.1007/978-3-319-05542-8_15-1
23. Sarstedt, M., Ringle, C. M., & Hair, J. F. (2016). Partial Least Squares Structural Equation Modeling. In Practical Assessment, Research and Evaluation (Vol. 21, Issue 1). https://doi.org/10.1108/ebr-10-2013-0128
24. Soh, P. C. H., Tan, K. H., & Soh, K. L. (2022). Workplace ostracism and cyberloafing: The mediating role of moral disengagement. Management Research Review, 46(7), 1312–1327. https://doi.org/10.1108/MRR-07-2022-0490
25. Yücel, İ., & Uslu, Ö. (2020). Job stress and cyberloafing: The moderating role of perceived organizational support. Kybernetes, 49(2), 457–474. https://doi.org/10.1108/K-12-2018-0666
26. Zainuddin Awang. (2015). SEM Made Simple: A Gentle Approach to Learning Structural Equation Modeling (Z. Awang, Ed.). MPWS Rich Publication.