The COVID-19 pandemic accelerated the shift to remote work, highlighting the need to understand factors influencing work-from-home productivity. This study focuses on the association between family-work conflict (FWC) and work-from-home productivity (WFHP) among Taiwanese employees during the pandemic's first wave. While Taiwan had a relatively successful pandemic response, the mandatory work-from-home measures implemented during the outbreak provide a valuable context for this research. Previous research on work-from-home (WFH) has explored its impact on various aspects of employee experience, including job satisfaction, work engagement, and work-life balance. However, several gaps remain in the literature, particularly concerning the mediating role of self-efficacy (SE) in the relationship between FWC and WFHP, and the moderating effect of working hours. This research aims to address these gaps using role conflict and resource drain theories. Specifically, it investigates (1) how FWC affects WFHP and the mechanism behind this effect; and (2) how working hours moderate the association between FWC and WFHP. The study uniquely examines the WFHP of employees in Taiwanese small and medium-sized enterprises (SMEs) during the pandemic, a demographic often overlooked in previous studies. This context is critical because the abrupt shift to WFH imposed different challenges on SMEs and their employees compared to larger corporations.

The literature review examines existing research on WFH, focusing on its advantages and challenges, particularly during the COVID-19 pandemic. Studies investigating the impact of WFH on productivity, employee performance, and gender disparities are reviewed. The theoretical foundation of this research is based on Role Conflict Theory and Resource Drain Theory. Role conflict theory suggests that individuals experience stress when faced with competing demands from multiple roles (e.g., work and family), leading to reduced productivity. Resource drain theory posits that individuals have limited resources (time, energy) that need to be allocated across different roles. The review further discusses the relationship between FWC and WFHP, exploring the mediating role of work engagement (WE) and self-efficacy (SE). WE is defined as an employee's emotional, cognitive, and physical investment in their work. SE refers to an employee's confidence in their ability to successfully perform their work. The literature suggests a negative relationship between FWC and WFHP and examines the potential mediating role of WE and SE in this relationship. The moderating role of working hours is also considered, with studies indicating a potential negative association between long working hours and productivity.

This study employed a quantitative approach using a questionnaire survey to collect data from employees in Taiwanese SMEs who worked from home during the COVID-19 level 3 alert. A convenience sampling technique was used to recruit participants, with questionnaires distributed to 1307 employees. After data cleaning, 785 valid responses were included in the analysis. The questionnaire included sections measuring demographic information, FWC, WE, SE, and WFHP using a five-point Likert scale. Data analysis involved descriptive statistics, principal component analysis (PCA), confirmatory factor analysis (CFA), and mediation analysis (Process Macro, Model 5). PCA and CFA were used to assess construct validity and reliability. The Process macro was used to test the hypothesized relationships, including the mediating roles of WE and SE and the moderating role of WFHDs. Harman's single-factor test addressed potential common method variance.

The findings support all proposed hypotheses. FWC was found to significantly and negatively affect WFHP (β = -0.26, p < 0.001). FWC also negatively affected WE (β = -0.30, p < 0.001) and SE (β = -0.25, p < 0.001). WE showed a positive effect on WFHP (β = 0.11, p < 0.01), and SE also demonstrated a positive impact on WFHP (β = 0.37, p < 0.001). Importantly, both WE (β = -0.03, LLCI = -0.0634, ULCI = -0.0063) and SE (β = -0.09, LLCI = -0.1242, ULCI = -0.0621) partially mediated the negative relationship between FWC and WFHP. Furthermore, the negative relationship between FWC and WFHP was significantly stronger for employees with more WFHDs (β = -0.07, p < 0.01). Finally, a significant gender difference in WFHP was observed, with male employees reporting higher WFHP than female employees (F(1, 783) = 20.478, p < 0.001, η² = 0.03). All Cronbach's alphas were above 0.8, and CFA model fit indices showed good model fit (CMIN/df = 2.830, RMSEA = 0.048, CFI = 0.952, IFI = 0.952).

The findings confirm the negative association between FWC and WFHP, highlighting the impact of family responsibilities on work performance during WFH. The mediating effects of WE and SE demonstrate the importance of employee well-being and confidence in mitigating the negative effects of FWC. The study's findings contribute to the understanding of the interplay between family, work, and productivity in the context of WFH, particularly during a pandemic. The significant gender disparity in WFHP underscores the need for targeted interventions to address the specific challenges faced by women. The discovery of SE as a mediator fills a gap in existing literature and suggests that promoting self-efficacy among employees may be crucial in boosting WFHP. The moderating effect of working hours indicates that implementing regulations on WFH duration may be beneficial in enhancing overall productivity and work-life balance.

This study makes several important contributions. It provides a predictive model for WFHP among SME employees in Taiwan during the COVID-19 pandemic, highlighting the mediating roles of work engagement and self-efficacy and the moderating role of work-from-home days. The findings emphasize the importance of family-friendly policies, employee well-being initiatives, and regulations on WFH duration for enhancing productivity and reducing work-family conflict. Future research could utilize a larger, more representative sample, longitudinal studies to establish causality, and explore the moderating effects of additional demographic factors across diverse cultural contexts.

The study's reliance on a convenience sample limits the generalizability of the findings. The cross-sectional design prevents the establishment of causal relationships between variables. Future research should address these limitations by employing probability sampling and longitudinal studies. Additional research should also consider exploring the potential moderating effects of other demographic variables and the impact of specific WFH policies implemented by different organizations.