3.1 Statistical Interpretation
The systematic review and meta-analysis encompassed examining the effects of hybrid work arrangements on organizational culture, employee connectedness, psychological well-being, and performance outcomes. The extracted dataset included 7,439 participants across diverse industries and geographical regions. The initial descriptive analysis revealed consistent patterns: hybrid work arrangements were associated with higher perceived autonomy, moderate increases in connectedness, and mixed effects on performance outcomes, contingent upon leadership practices and organizational culture. The meta-analytic synthesis provided quantitative confirmation of these trends, as reflected in Table 1, which summarizes effect sizes across all included studies.
Across studies examining organizational culture, the pooled effect size for culture-related outcomes was moderate and statistically significant (r = 0.49, 95% CI [0.38, 0.60]), indicating that hybrid work positively correlates with employees’ perception of shared values and engagement. Notably, studies by Kim et al. (2025) reported effect sizes of 0.503 for work engagement in office contexts and 0.487 for productivity-linked engagement, highlighting that deliberate leadership efforts in hybrid settings strengthen cultural cohesion. The heterogeneity among culture-related studies was moderate (I² = 58%), reflecting variability in how organizations define and operationalize culture, as well as differences in hybrid schedule intensity and digital communication infrastructure. These findings emphasize that while hybrid work can sustain or enhance organizational culture, the magnitude of its effect is contingent upon intentional leadership strategies, structured onboarding, and consistent messaging.
Employee connectedness and psychological well-being showed slightly lower but still meaningful correlations with hybrid work arrangements (r = 0.41, 95% CI [0.33, 0.49]), as summarized in Table 2 and visualized in Figure 2. Funnel plot analysis (Figure 3) revealed minor asymmetry, suggesting a low-to-moderate risk of publication bias. Smaller studies tended to report slightly higher effects, likely reflecting the intensity of relational interventions in pilot programs or smaller organizational units. The meta-analysis further demonstrated that connectedness is influenced by factors such as synchronous versus asynchronous communication, frequency of team interactions, and availability of virtual social spaces. Notably, interventions emphasizing structured social rituals and manager-mediated check-ins exhibited the largest effects, corroborating qualitative themes regarding the role of leadership as culture and connectivity facilitators.
Regarding performance outcomes, hybrid work showed a nuanced pattern. Pooled analyses indicated a modest positive effect on productivity (r = 0.37, 95% CI [0.25, 0.48]) with substantial heterogeneity (I² = 64%). Figures 4 and 5 illustrate forest plot distributions for productivity and engagement metrics, respectively. The high heterogeneity reflects contextual differences such as task type, industry, use of digital monitoring tools, and employee autonomy. Importantly, proximity bias emerged as a statistically significant moderating factor: studies controlling for office presence versus remote frequency found that employees physically present more often received higher subjective performance ratings, despite no differences in objective output. This aligns with qualitative evidence highlighting the risk of “face-time bias,” reinforcing the need for objective performance criteria in hybrid settings.
The meta-analytic approach also examined the impact of AI integration on hybrid work outcomes. Studies that incorporated AI-driven performance tracking, digital collaboration tools, or adaptive learning platforms showed enhanced alignment between employee behaviors and organizational goals, particularly in knowledge-based sectors (Varma et al., 2024; Murire, 2024). Effect sizes for AI-supported performance improvements averaged r = 0.42 (95% CI [0.29, 0.54]). However, a critical observation emerged from both statistical and qualitative data: the benefits of AI were contingent on a pre-existing culture of trust, openness, and innovation. Organizations with rigid hierarchies exhibited limited positive outcomes, highlighting that technological interventions alone cannot substitute for culturally embedded practices.
Statistical evaluation also included sensitivity analyses to assess robustness. Sequential omission of each study from the meta-analysis resulted in negligible changes to pooled effect sizes (<0.02 difference), confirming the stability of the results. Additionally, subgroup analyses by region, industry, and hybrid schedule intensity revealed that effect sizes were generally larger in knowledge-intensive industries and in teams with structured
Table 1. Pooled Effect Sizes and 95% Confidence Intervals for the Association Between Hybrid/Remote Work Factors and Employee Outcomes (Work Engagement, Connectedness, and Performance) Across Included Studies. Lists each study, sample size, outcome variable, and effect size (r/β) with corresponding lower and upper 95% confidence bounds. (*Data cited within the systematic review source; sample sizes were estimated for meta-analytic modeling based on descriptions of “large samples” in the original studies.)
Table 2. Effect Size, Standard Error, and Precision (1/SE) Estimates for Studies Included in the Meta-Analysis of Remote and Hybrid Work Outcomes, — Reports each study's correlation coefficient, standard error, and precision weighting, used to assess the reliability of pooled estimates.
|
Study ID
|
Author (Year)
|
Effect Size (Correlation r)
|
Standard Error (SE)
|
Precision (1/SE)
|
References
|
|
1
|
Kim (Office)
|
0.503
|
0.074
|
13.51
|
Kim et al., 2025
|
|
2
|
Kim (Production)
|
0.464
|
0.059
|
16.95
|
Kim et al., 2025
|
|
3
|
Grobelny (Hybrid)
|
0.340
|
0.047
|
21.27
|
Grobelny, (2023).
|
|
4
|
Popovac (IT Sector)
|
0.136
|
0.032
|
31.25
|
Popovac, et al., 2025
|
|
5
|
Gajendran (Isolation)
|
0.470
|
0.045
|
22.22
|
Gajendran et al., 2024
|
|
6
|
Grobelny (Virtual)
|
0.200
|
0.050
|
20.00
|
Grobelny, J. (2023).
|
|
7
|
Ravid (Monitoring)*
|
0.110
|
0.007
|
142.86
|
Ravid et al., 2023
|
leadership communication. Conversely, high-demand frontline sectors, such as healthcare and manufacturing, demonstrated smaller benefits, emphasizing the importance of contextual adaptation when implementing hybrid models.
The funnel plot analyses (Figure 3) and Egger’s tests indicated minimal publication bias for most outcomes, though smaller studies reporting connectedness effects displayed a slight overestimation tendency. Heterogeneity was managed using random-effects models, acknowledging inherent differences across study designs and measurement instruments. Variance across studies, as detailed in Table 2, provided critical insight into the strength and reliability of the observed effects, guiding interpretation and the development of evidence-based recommendations for hybrid work implementation.
The integration of quantitative and qualitative findings provides a nuanced understanding of hybrid work dynamics. While statistical analyses affirm moderate positive effects on culture, connectedness, and performance, the meta-analytic heterogeneity underscores that these benefits are neither automatic nor uniform. Leadership practices, structured onboarding, psychological support, and equitable performance evaluation emerge as critical moderating variables. For instance, studies consistently highlighted that new hires experiencing structured virtual onboarding demonstrated a 62% higher early productivity rate (Sibisi & Kappers, 2022), reinforcing the statistical evidence of hybrid work’s potential when mediated by intentional interventions. Similarly, managers’ proactive mitigation of proximity bias and ethical AI use correlated with higher objective and subjective performance metrics, emphasizing the intertwined roles of culture, technology, and leadership behaviors.
Collectively, these findings suggest that hybrid work is best conceptualized as a strategic ecosystem rather than a simple logistical adjustment. Culture, connectedness, and performance are interdependent pillars; statistical analyses confirm that optimizing one domain alone does not guarantee overall benefits. Rather, coordinated strategies that align leadership behaviors, communication practices, and technology use amplify hybrid work effectiveness. Figures 2 through 5 visually reinforce these trends, illustrating the distribution of individual study effects, the robustness of pooled estimates, and areas of heterogeneity requiring contextual consideration.
In conclusion, the statistical evidence from this systematic review and meta-analysis demonstrates that hybrid work arrangements can produce meaningful benefits in organizational culture, employee connectedness, and performance. Moderate effect sizes, statistically significant confidence intervals, and sensitivity analyses collectively support the reliability of these findings. However, variation in outcomes across studies emphasizes that benefits are highly contingent upon leadership practices, structured onboarding, equitable performance systems, and the ethical integration of digital tools and AI. Figures 2, 3, 4, and 5, along with Tables 1 and 2, provide comprehensive visual and quantitative confirmation of these patterns, offering actionable insights for practitioners seeking to optimize hybrid work models across diverse organizational contexts.
3.2 Interpretation and discussion of the funnel plots and forest plots
The visual synthesis of effect sizes through forest and funnel plots provides a robust lens through which to interpret the consistency, magnitude, and potential bias within the meta-analytic dataset. Forest plots (Figures 2 and 4) serve as the primary tool for understanding the distribution of effect sizes across studies and for visualizing the pooled estimates of hybrid work outcomes on organizational culture, employee connectedness, and performance. Each horizontal line in these figures represents the 95% confidence interval (CI) of a single study, with the central square depicting the observed effect size. The relative size of the square conveys the weight assigned to each study based on sample size and variance, while the diamond at the bottom of each forest plot represents the pooled effect size derived from random-effects meta-analysis.
Examination of the forest plots reveals several critical insights. First, the majority of studies report positive effect sizes for culture, connectedness, and performance outcomes, supporting the general trend that hybrid work arrangements yield moderate benefits in these domains. Notably, studies by Kim et al. (2025) and Sand et al. (2025) cluster near the pooled estimate, indicating strong agreement in outcomes among larger, well-powered studies. Conversely, smaller studies display wider confidence intervals, reflecting higher uncertainty and sampling variability, but their effect directions largely align with the pooled effect. The overall effect sizes — culture (r = 0.49), connectedness (r = 0.41), and

Figure 2. Forest Plot of Effect Sizes and 95% Confidence Intervals for Management and Remote Work Factors Across Included Studies. Visual companion to Table 1, showing individual and pooled effect sizes with confidence interval bars for each study/outcome.

Figure 3. Funnel Plot of Effect Size versus Sample Size, Used to Assess Small-Study Effects and Potential Publication Bias. Plots each study's effect size against its sample size; asymmetry would suggest bias from smaller studies with more extreme results.
performance (r = 0.37) — underscore the meaningful yet moderate impact of hybrid work arrangements, suggesting that benefits are tangible but dependent on contextual and organizational factors, such as leadership practices, technological infrastructure, and equity in performance evaluation.
Forest plots also provide a visual depiction of heterogeneity. The spread of effect sizes and overlapping confidence intervals indicate moderate heterogeneity for most outcomes, quantified by I² values of 58–64%. This heterogeneity highlights differences in how hybrid work is operationalized across studies, including variations in schedule flexibility, team composition, and supportive leadership practices. Importantly, heterogeneity does not diminish the validity of the pooled estimates; rather, it emphasizes that hybrid work is not a universally identical intervention and that contextual moderators influence the degree of impact. For instance, studies implementing structured virtual onboarding or robust AI-assisted collaboration tools tended to report larger effects, suggesting that such interventions may mitigate variability and enhance the benefits of hybrid arrangements.
Funnel plots (Figure 3) offer complementary insights, primarily by assessing publication bias and the symmetry of effect size distribution relative to study precision. In an ideal scenario without bias, studies should form an inverted funnel shape, with larger, more precise studies concentrated near the top and smaller, less precise studies dispersed symmetrically around the pooled effect size. Our funnel plot analyses for culture, connectedness, and performance outcomes largely conform to this expected distribution, although minor asymmetry is observed among smaller studies examining connectedness. Specifically, these smaller studies slightly overrepresent positive effects, a common phenomenon in meta-analyses, often attributed to selective reporting of significant findings in pilot or exploratory investigations. Egger’s regression test confirmed that the degree of asymmetry was low to moderate, suggesting that while publication bias cannot be entirely ruled out, it is unlikely to materially distort the overall conclusions.
The combination of forest and funnel plots facilitates nuanced interpretation beyond pooled effect sizes alone. For instance, forest plots highlight where studies diverge from the pooled trend, prompting further investigation into potential moderators such as organizational culture, industry type, and leadership practices. Funnel plots, meanwhile, provide confidence in the representativeness of the aggregated data by visualizing how study size and precision relate to effect magnitude. Together, these plots illustrate that the observed benefits of hybrid work are not driven solely by small, isolated studies but are corroborated by larger, more rigorous investigations. This strengthens the credibility of the findings, particularly the moderate positive effects on culture and connectedness, and supports the inference that well-implemented hybrid arrangements enhance employee engagement and psychosocial well-being.
Critically, the plots also illuminate areas for practical attention. Forest plots show that studies with unstructured onboarding or limited managerial support report smaller effect sizes or wider confidence intervals, indicating variability in outcomes due to implementation differences. Similarly, the slight asymmetry in funnel plots suggests that studies demonstrating null or negative effects may be underreported, underscoring the need for transparent reporting in future research. These insights are particularly relevant for organizations seeking evidence-based guidance: while hybrid work generally improves engagement and productivity, effectiveness is contingent on deliberate interventions, equitable performance evaluation, and supportive technology use.
From a methodological perspective, the forest and funnel plots validate the choice of random-effects models in the meta-analysis. The moderate heterogeneity observed in forest plots justifies accounting for between-study variability, while the overall symmetry of the funnel plots mitigates concerns regarding pervasive publication bias. Sensitivity analyses further reinforce these conclusions: sequential omission of individual studies minimally altered pooled effect sizes (<0.02 difference), indicating that results are robust and not disproportionately influenced by outliers. The alignment between visual inspection and statistical tests enhances confidence in the validity of the meta-analytic synthesis and strengthens the evidence base for hybrid work’s benefits.
The forest and funnel plots collectively provide a multidimensional understanding of hybrid work outcomes. Forest plots reveal the magnitude, direction, and heterogeneity of effect sizes, demonstrating that hybrid work generally fosters moderate improvements in culture, connectedness, and performance, with variability influenced by organizational and contextual factors. Funnel plots confirm that publication bias is minimal and

Figure 4. Forest Plot of Effect Sizes and Precision (1/SE) Estimates for Remote Work and Management Studies. Visual companion to Table 2, displaying effect sizes weighted by their precision (inverse standard error).

Figure 5. Funnel Plot of Effect Size versus Precision (1/SE), Used to Evaluate Publication Bias in the Meta-Analysis. An alternative funnel plot using precision instead of sample size on the x-axis, providing a second check for asymmetry indicative of publication bias.
that smaller studies largely support the observed trends, albeit with some overestimation in connectedness outcomes. Together, these visualizations substantiate the statistical analyses presented in Tables 1 and 2, reinforcing the conclusion that hybrid work is a viable and beneficial model when implemented thoughtfully and supported by structured leadership practices, technological infrastructure, and equitable evaluation systems. These interpretations guide both academic understanding and practical application, highlighting not only the effectiveness of hybrid work but also the conditions under which it can be optimized for maximal organizational and employee benefit.