Abstract
The widespread adoption of hybrid work arrangements, in which employees divide their working time between remote and office-based settings, has fundamentally altered how organisations manage, measure, and sustain workforce engagement and performance. While hybrid models offer tangible benefits such as increased autonomy, reduced commuting burden, and greater scheduling flexibility, they simultaneously introduce a set of organisational challenges that conventional management frameworks were not designed to address. Chief among these challenges is the difficulty of maintaining consistent levels of employee engagement across distributed work settings, where informal communication, social connection, and direct managerial oversight are structurally constrained. The absence of robust, data-driven understanding of which organisational factors most powerfully shape engagement and performance in hybrid environments represents a significant gap in both academic research and organisational practice. This thesis investigates the factors that influence employee engagement and performance in hybrid work environments through a quantitative, data-driven approach grounded in two complementary theoretical frameworks. The Job Demands–Resources (JD-R) model provides the conceptual basis for understanding how the balance between job demands, including digital overload, boundary ambiguity, and coordination complexity, and job resources, such as leadership support, communication quality, and technological infrastructure, determines employee engagement outcomes. Social Exchange Theory complements this framework by framing engagement as a product of perceived organisational investment, whereby employees who experience high levels of institutional support respond with elevated commitment and discretionary performance. Together, these frameworks guide the selection and operationalisation of the study’s core variables: communication effectiveness, leadership support, work-life balance, digital tool usage, job satisfaction, employee engagement, and employee performance. The study adopts the CRISP-DM methodology to structure its analytical pipeline across six stages: business understanding, data collection, data preprocessing, modelling, evaluation, and interpretation of findings. Primary data are collected through a structured survey distributed to employees working in hybrid arrangements, with Likert-scale responses coded numerically to enable rigorous statistical analysis. Data preprocessing is conducted in R, encompassing missing value treatment, outlier detection, normalisation, and variable encoding. Exploratory analysis includes descriptive statistics, distributional visualisations, correlation matrices, and geographic mapping of response patterns to establish a comprehensive understanding of the dataset prior to modelling. Three machine learning techniques are applied and compared: Random Forest, Logistic Regression, and Support Vector Machine. These methods are evaluated against one another using a consistent set of performance metrics, including F1-score, accuracy, precision, recall, and R-squared, to identify the algorithm that most accurately and reliably predicts employee engagement and performance outcomes from the survey data. Feature importance analysis drawn from the best-performing model provides the empirical basis for identifying which organisational factors exert the greatest influence on engagement and performance in hybrid settings, translating predictive model outputs into actionable organisational insights. The anticipated findings are expected to offer empirical evidence that advances understanding of hybrid work beyond descriptive accounts, providing a theoretically grounded and analytically rigorous account of the organisational conditions that sustain workforce engagement and drive performance in distributed environments. From a practical standpoint, the study aims to equip organisations, particularly those operating within the UAE and broader Gulf Cooperation Council context, with evidence-based guidance for designing hybrid work policies, calibrating managerial practices, and investing in the digital and relational infrastructure most likely to yield positive workforce outcomes. In doing so, the research contributes to the growing field of HR analytics by demonstrating how machine learning methods can be integrated with established organisational theory to generate insights of both scholarly and practical value.
Publication Date
8-2026
Document Type
Thesis
Student Type
Graduate
Degree Name
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research
Advisor
Ioannis Karamitsos
Recommended Citation
Alblooshi, Saif, "Factors Influencing Employee Engagement and Performance in HybridWork Environments" (2026). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/12710
Campus
RIT Dubai
