Table of Contents
Type: Full Masters Dissertation | Subject: Business and Management | Level: Masters | Word Count: ~14,000 words
This model Full Masters Dissertation was produced by an Essays UK specialist as reference material for learning purposes only. For support in this field, see our business and management dissertation writers.
Overview: This model MSc Management dissertation presents a complete, structurally faithful example of a survey-based human resource management study examining how hybrid working arrangements relate to employee engagement in UK financial services firms. It demonstrates how a full quantitative dissertation is assembled — abstract, introduction, literature review, methodology, findings, discussion and conclusion — with an illustrative primary dataset, descriptive and inferential statistics, a branded results table and Harvard (Cite Them Right) referencing. All figures are illustrative and generated for teaching purposes; the document is intended solely as a learning reference for how a Masters-level business and management dissertation is organised and argued, not as a source of empirical fact.
Background: The accelerated adoption of hybrid working — a model that combines scheduled office attendance with regular remote work — has reshaped how knowledge-intensive organisations manage and motivate their people. In the UK financial services sector, where regulation, client confidentiality and collaborative deal-making have historically favoured office presence, the move to hybrid arrangements raises pressing questions about whether such flexibility strengthens or erodes employee engagement.
Aim: To examine the relationship between hybrid working arrangements and employee engagement among professionals in UK financial services firms, and to identify which attributes of the hybrid experience most strongly predict engagement.
Methods: A positivist, deductive and quantitative design was adopted. A cross-sectional self-completion questionnaire was administered online to employees of banking, insurance, asset management and fintech organisations, yielding 247 usable responses. Five predictor constructs — schedule flexibility and autonomy, perceived organisational support, communication and connectivity quality, work–life balance, and professional isolation — were measured alongside employee engagement using adapted, previously validated multi-item scales. Data were analysed in IBM SPSS using reliability analysis, correlation and multiple linear regression. All data are illustrative and were generated for this model dissertation.
Findings: In the illustrative dataset the five predictors jointly explained 50.4% of the variance in employee engagement (R² = .504, F(5, 241) = 49.02, p < .001). Perceived organisational support (β = .31) was the strongest positive predictor, followed by work–life balance (β = .24), schedule flexibility and autonomy (β = .18) and communication quality (β = .14); professional isolation was a significant negative predictor (β = −.12). All five hypotheses were supported. An exploratory analysis suggested engagement was highest among respondents working two to three days remotely rather than at the extremes of the distribution.
Conclusion: The model study indicates that hybrid working is not inherently beneficial or harmful to engagement; rather, its effect depends on the organisational conditions surrounding it, with felt support and protected work–life boundaries mattering most. Recommendations are offered for human resource practice in financial services, together with the study’s limitations and avenues for future research. As a cross-sectional, self-report design it cannot establish causation, and the findings are presented as an illustrative teaching example rather than as empirical evidence.
Keywords: hybrid working; employee engagement; perceived organisational support; work–life balance; UK financial services; quantitative survey.
| Abbreviation | Meaning |
|---|---|
| COMM | Communication and Connectivity Quality |
| ENG | Employee Engagement |
| FLEX | Schedule Flexibility and Autonomy |
| HRM | Human Resource Management |
| ISO | Professional and Social Isolation |
| JD–R | Job Demands–Resources (model) |
| POS | Perceived Organisational Support |
| SDT | Self-Determination Theory |
| UWES | Utrecht Work Engagement Scale |
| VIF | Variance Inflation Factor |
| WLB | Work–Life Balance |
The organisation of work in knowledge-intensive industries has changed more in the past five years than in the preceding two decades. What began as an emergency response to a public health crisis — the sudden dispersal of office populations to the home — has settled into a durable reconfiguration of where, when and how professional work is performed. The dominant expression of this reconfiguration is hybrid working: a model in which employees divide their contracted hours between a central workplace and one or more remote locations, typically the home, according to arrangements that may be fixed by the employer, negotiated with a line manager, or left largely to individual discretion. Hybrid working is now a mainstream feature of the UK labour market for office-based roles, and employers across professional sectors have moved from asking whether to offer it to asking how best to design it.
Nowhere is that design question more consequential than in financial services. The sector — encompassing retail and investment banking, insurance, asset and wealth management, and a fast-growing fintech segment — is one of the United Kingdom’s largest contributors to economic output and employment, and it is intensely dependent on the discretionary effort, judgement and retention of a highly skilled workforce. Historically, financial services firms have been among the most office-centric of employers. Trading floors, deal teams, underwriting functions and client-facing advisory roles were organised around physical co-presence, partly for operational reasons, partly for reasons of supervision, control and regulatory oversight, and partly because of an entrenched cultural belief that presence signals commitment. The migration of these firms to hybrid models therefore represents not merely an administrative adjustment but a challenge to a deeply held set of assumptions about how value is created and how performance is observed.
At the centre of those assumptions sits the construct of employee engagement. Engagement — broadly, the extent to which employees invest their physical, cognitive and emotional energies in their work roles — has become one of the most closely monitored human resource outcomes in large organisations because of its demonstrated association with performance, discretionary effort, customer satisfaction, retention and wellbeing. For financial services firms competing for scarce specialist talent, engagement is not a soft metric but a strategic asset. The question that confronts human resource leaders is whether the flexibility and autonomy of hybrid working strengthen engagement by giving employees control, trust and better balance, or whether the loss of informal contact, visible recognition and spontaneous collaboration erodes the very conditions on which engagement depends. The evidence to date is mixed and, importantly, much of it has been produced in general or cross-sector samples rather than in the distinctive regulatory and cultural setting of UK finance.
This dissertation addresses that question directly. It treats hybrid working not as a single, undifferentiated condition but as a bundle of experiences — flexibility and autonomy over one’s schedule, the support one perceives from the organisation, the quality of communication and connectivity, the balance one is able to strike between work and non-work life, and the degree of professional isolation one feels — and asks how each of these attributes relates to engagement. In doing so it aims to move the debate beyond the binary of “office good, home bad” towards a more useful account of which features of the hybrid experience managers can influence in order to sustain an engaged workforce.
The timing of such an inquiry is significant. The initial period of hybrid working was characterised by improvisation, as firms and employees adapted whatever arrangements they could to an unplanned situation. The present period is different in kind: hybrid working has become a matter of deliberate organisational design, negotiated in recruitment, formalised in policy and increasingly contested as some employers seek to reassert office attendance while employees, particularly specialists with bargaining power, resist. Decisions made now about attendance expectations, workspace investment, managerial training and technology will shape the engagement, retention and productivity of the financial services workforce for years to come. Studying engagement under settled, voluntary hybrid conditions — rather than under the duress of the earlier emergency — therefore offers a more reliable guide to those decisions than much of the evidence generated during the disruption itself, and it is this settled, designed form of hybrid working with which the present study is concerned.
Although hybrid working has been widely adopted, the managerial and academic understanding of its consequences for engagement remains incomplete and, in places, contradictory. Three specific problems motivate this study. First, much of the existing commentary conflates remote working with hybrid working, treating any time spent away from the office as equivalent; yet the hybrid model is distinctive precisely because it blends locations, and its effects may differ from those of fully remote or fully office-based arrangements. Second, a substantial share of the evidence base derives from the exceptional conditions of the pandemic, when remote working was involuntary, universal and accompanied by acute external stressors, making it unreliable as a guide to the steady-state, voluntary hybrid arrangements that now prevail. Third, and most importantly for this study, the financial services sector has characteristics — strict regulation, confidentiality requirements, dense collaborative networks and a strong presence-oriented culture — that may condition the engagement effects of hybrid working in ways that general samples cannot capture.
The practical consequence of this incomplete understanding is that human resource and operational leaders in financial services are making significant decisions about office mandates, attendance policies, technology investment and managerial training with limited sector-specific evidence about what actually sustains engagement under hybrid conditions. Some firms have reacted by reasserting office attendance requirements; others have embraced flexibility as a recruitment and retention advantage. Both strategies carry risks, and both are frequently justified by appeals to engagement without a clear, evidence-based account of the mechanisms involved. This study seeks to supply part of that account.
The rationale for the study is both theoretical and practical. Theoretically, examining hybrid working through the lens of established engagement theory — in particular the job demands–resources model, organisational support theory and self-determination theory — offers an opportunity to test whether these frameworks, largely developed in conventional workplace settings, retain their explanatory power in the hybrid context. By decomposing the hybrid experience into theoretically grounded resources (autonomy, support, balance, connectivity) and a demand (isolation), the study contributes to the refinement of engagement theory for a changed world of work.
Practically, the study is significant for several stakeholder groups. For human resource professionals in financial services, it offers evidence about which levers — many of them within managerial control — are most strongly associated with engagement, and therefore where limited investment and attention are best directed. For employees, a clearer understanding of the conditions that sustain engagement under hybrid working may inform the negotiation of working patterns that protect both performance and wellbeing. For the sector as a whole, which faces persistent competition for specialist talent and ongoing scrutiny of its conduct and culture, evidence about the engagement effects of hybrid working speaks directly to questions of retention, productivity and the sustainability of flexible-working policies. The study is presented, it should be stressed, as a model and a structural reference; its illustrative data are intended to demonstrate method and argument rather than to furnish definitive sector statistics.
The overarching aim of this study is to examine the relationship between hybrid working arrangements and employee engagement among professionals employed in UK financial services firms, and to determine which attributes of the hybrid experience most strongly predict engagement. To fulfil this aim, four objectives were set:
Consistent with the aim and objectives, the study was guided by one central and three subsidiary research questions:
From the theoretical framework developed in Chapter 2, five directional hypotheses were derived for formal testing:
The study is bounded in several respects. It concerns office-based professional and managerial roles within UK financial services firms for which hybrid working is feasible; it does not address branch, call-centre or operational roles that require continuous physical presence. It adopts a cross-sectional design and therefore captures relationships at a single point in time rather than causal sequences unfolding over time. It relies on employee self-report and does not incorporate objective performance data or employer records. These boundaries are revisited critically in Chapter 6.
The remainder of the dissertation is organised into five further chapters. Chapter 2 reviews the literature on hybrid working and employee engagement, establishes the theoretical framework and develops the conceptual model and hypotheses. Chapter 3 sets out the research methodology, justifying the philosophical stance, design, sampling, measurement, analysis and ethics. Chapter 4 presents the findings, including sample characteristics, descriptive statistics, reliability analysis, correlations and the regression model. Chapter 5 discusses these findings in relation to the literature and draws out theoretical and practical implications. Chapter 6 concludes, offering recommendations, acknowledging limitations and suggesting directions for future research.
This chapter critically reviews the scholarship relevant to the relationship between hybrid working and employee engagement, and uses that review to build the conceptual framework that the empirical study tests. It begins by conceptualising hybrid working and employee engagement as distinct constructs, tracing how each has been defined and measured. It then establishes a theoretical framework drawn from three complementary traditions — the job demands–resources model, organisational support theory and self-determination theory — that together explain how features of the hybrid experience might influence engagement. The chapter proceeds to synthesise the empirical evidence on hybrid and remote working and engagement, before locating that evidence within the specific context of UK financial services. It closes by developing the conceptual model and hypotheses, identifying the research gap, and summarising the argument.
Hybrid working is best understood as a particular point on a continuum of spatial and temporal flexibility rather than as a single, uniform arrangement. At one pole of this continuum lies fully co-located work, in which all contracted hours are performed at a central workplace; at the other lies fully distributed or remote work, in which the employee has no routine physical presence at an employer site. Hybrid working occupies the broad middle ground, blending scheduled office attendance with regular remote work. Within that middle ground, however, arrangements vary enormously along several dimensions: the proportion of time spent remotely; whether the split is fixed by the employer, agreed at team level, or chosen by the individual; whether office days are coordinated across a team or left to personal preference; and whether remote work is confined to the home or extended to other locations. This heterogeneity matters because the engagement consequences of hybrid working are unlikely to be uniform across such different configurations, and because studies that treat “hybrid” as a single category risk obscuring the mechanisms that actually drive outcomes (Wang et al., 2021).
A useful way to resolve this heterogeneity, and the approach taken in this dissertation, is to characterise the hybrid experience not by a label but by the qualities it affords the employee. Drawing on the broader flexible-working literature, four such qualities recur as salient: the degree of autonomy employees exercise over when and where they work; the extent to which they feel supported by their organisation while working at a distance; the quality of the communication and technological connectivity that links them to colleagues and managers; and the balance they are able to maintain between work and non-work domains. A fifth quality — professional and social isolation — appears as the principal risk that distributed work introduces. Conceptualising hybrid working through these five attributes rather than through a crude office-versus-home dichotomy allows the construct to be connected directly to established engagement theory, and it is this attribute-based conception that structures the remainder of the review.
It is also important to distinguish voluntary, steady-state hybrid working from the involuntary, universal remote working that characterised the pandemic period. The latter was accompanied by school closures, caring burdens, health anxiety and an absence of choice, all of which plausibly depressed wellbeing and complicate inference about the inherent effects of distributed work (Wang et al., 2021). Evidence generated under those conditions must therefore be interpreted cautiously when drawing conclusions about the designed, negotiated hybrid arrangements that now predominate. This dissertation is concerned with the latter.
A further conceptual distinction concerns the locus of control over the hybrid arrangement. Flexible-working scholarship has long differentiated between flexibility that is granted and governed by the employer and flexibility that is genuinely exercised by the employee, and the two can diverge sharply in practice. A nominal hybrid policy that in fact dictates fixed office days, monitors attendance and treats remote work as a concession is experientially very different from an arrangement in which employees exercise real discretion over their patterns. Because self-determination theory locates the motivational value of flexibility in experienced autonomy rather than in formal entitlement (Deci and Ryan, 2000), this distinction matters for engagement: it is the felt autonomy of the arrangement, not its label, that is theorised to drive outcomes. The present study therefore measures the autonomy employees perceive themselves to hold, rather than inferring it from the existence of a policy, and this choice is reflected in the operationalisation set out in Chapter 3.
Employee engagement has a longer intellectual history than hybrid working but an equally contested definition. The construct is commonly traced to Kahn (1990), who conceived of personal engagement as the harnessing of organisational members’ selves to their work roles, such that they employ and express themselves physically, cognitively and emotionally during role performance. Kahn argued that engagement arises when three psychological conditions are met — psychological meaningfulness, psychological safety and psychological availability — a formulation that already implies a dependence on the surrounding work environment and on relationships with colleagues and managers. This relational, condition-dependent view is important for the present study because it suggests that any change in how and where work is organised, such as a shift to hybrid working, may alter the conditions on which engagement rests.
A second, highly influential tradition defines engagement as a positive, fulfilling, work-related state of mind characterised by vigour, dedication and absorption, and operationalises it through the Utrecht Work Engagement Scale (Schaufeli and Bakker, 2004). Vigour refers to high levels of energy and mental resilience at work; dedication to a sense of significance, enthusiasm and pride; and absorption to being fully concentrated and happily engrossed in one’s work. This state-based conception is the one adopted in the present study, both because it is the most widely validated across sectors and national contexts and because its component dimensions map cleanly onto the kinds of experience — energy, meaning, concentration — that hybrid working might plausibly affect.
A third tradition, associated with Saks (2006), broadens the construct by distinguishing job engagement from organisation engagement and by situating engagement within a social-exchange framework, in which employees reciprocate the resources and support they receive from their organisation with greater engagement. This exchange-based account is particularly valuable for a study of hybrid working because it foregrounds the role of perceived support and fairness — precisely the organisational conditions that distributed work can strengthen or strain. Taken together, these three traditions converge on an understanding of engagement as an energising, meaningful and absorbing psychological state that is sustained by the resources and relationships an organisation provides and that employees reciprocate with discretionary effort. It is this understanding that the theoretical framework below seeks to connect to the hybrid experience.
Three complementary theories are used to explain how the attributes of hybrid working might influence engagement: the job demands–resources model, organisational support theory and self-determination theory. No single theory accounts fully for the phenomenon, but together they provide a coherent mechanism-based framework.
The job demands–resources (JD–R) model (Bakker and Demerouti, 2017) proposes that every working environment can be described in terms of job demands — physical, psychological, social or organisational aspects that require sustained effort and carry physiological or psychological costs — and job resources, which are aspects that help achieve goals, reduce demands or stimulate growth. The model holds that resources drive a motivational process that fosters engagement, while demands drive a health-impairment process that can lead to strain and disengagement. Applied to hybrid working, the framework suggests that attributes such as schedule flexibility and autonomy, perceived organisational support, communication quality and work–life balance function as resources that should raise engagement, whereas professional isolation functions as a demand that should lower it. The JD–R model thus supplies the overarching logic for the hypotheses and justifies treating the hybrid experience as a mix of resources and demands rather than a single condition.
Organisational support theory (Eisenberger et al., 1986) deepens one strand of this logic. It holds that employees form a global belief — perceived organisational support — concerning the extent to which the organisation values their contribution and cares about their wellbeing, and that this belief, grounded in the norm of reciprocity, obliges them to help the organisation reach its goals, which they do in part through heightened engagement. The theory is especially pertinent to hybrid working because distance can either signal distrust and neglect, depressing perceived support, or, where the organisation actively invests in remote employees’ development, inclusion and wellbeing, signal genuine care that strengthens it. The theory therefore predicts that perceived organisational support will be a powerful route through which hybrid arrangements affect engagement.
Self-determination theory (Deci and Ryan, 2000) completes the framework by specifying why autonomy should matter. The theory posits three basic psychological needs — autonomy, competence and relatedness — whose satisfaction underpins intrinsic motivation and wellbeing. Hybrid working bears directly on all three: it can enhance autonomy by giving employees control over when and where they work; it can affect competence through the adequacy of remote tools and support; and it can threaten relatedness through isolation from colleagues. Self-determination theory thus explains both the motivational promise of autonomy (supporting H1) and the motivational cost of isolation (supporting H5), while reinforcing the relational emphasis of organisational support theory.
The three theories are complementary rather than competing, and their integration is deliberate. The JD–R model supplies the macro-architecture — the division of the work environment into resources and demands and the twin motivational and health-impairment processes — but it is relatively agnostic about why particular resources energise and particular demands deplete. Organisational support theory and self-determination theory supply that missing mechanism: the former through the norm of reciprocity, which explains why felt support obliges engagement, and the latter through basic-need satisfaction, which explains why autonomy motivates and isolation demotivates. Mapping the five attributes onto this integrated framework yields a clear set of expectations: support, autonomy, balance and connectivity function as resources that should raise engagement through need satisfaction and reciprocal exchange, while isolation functions as a demand that should lower it through the frustration of relatedness. The engagement outcome is itself defined, following Schaufeli and Bakker (2004), as the vigour, dedication and absorption that the motivational process is theorised to produce. This integrated, attribute-based framework is the conceptual spine of the study, and the conceptual model and hypotheses developed in Section 2.7 follow from it directly.
The empirical literature connecting flexible, remote and hybrid working to engagement is sizeable but uneven, and it yields a nuanced rather than a one-directional picture. Several threads can be distinguished. A first thread, consistent with self-determination theory, finds that the autonomy and flexibility afforded by distributed work are associated with higher motivation, satisfaction and engagement, because employees value control over their schedules and the ability to work when and where they are most productive (Wang et al., 2021). This thread supports the expectation that schedule flexibility and autonomy will relate positively to engagement. However, the same literature cautions that autonomy without structure can blur boundaries and generate its own demands, so the relationship is unlikely to be simple or unbounded.
A second thread emphasises the centrality of organisational support and leadership. Where managers maintain regular, high-quality contact with remote team members, communicate expectations clearly, and signal that distributed employees are valued and included, engagement is sustained or enhanced; where support is withdrawn or becomes purely transactional, engagement suffers (Saks, 2006; Eisenberger et al., 1986). This thread underpins the expectation that perceived organisational support and communication quality will both relate positively to engagement, and it suggests, consistent with social-exchange reasoning, that support may be the single most important condition.
A third thread concerns work–life balance. Hybrid working can improve balance by eliminating commuting, enabling better integration of work and personal responsibilities, and allowing employees to structure their days around periods of peak energy. Yet the literature also documents the opposite risk: that the collapse of the spatial boundary between work and home can extend working hours, foster an always-on culture and erode recovery, thereby undermining the very balance that flexibility promised (Wang et al., 2021). The net effect is likely to depend on whether boundaries are actively protected, but on balance the literature supports the expectation of a positive association between work–life balance and engagement.
A fourth thread addresses the principal hazard of distributed work: professional and social isolation. Reduced spontaneous interaction, fewer informal learning opportunities, weaker social ties and a diminished sense of belonging are repeatedly identified as costs of working at a distance, and they map directly onto the frustration of the relatedness need in self-determination theory and onto the health-impairment process in the JD–R model (Bakker and Demerouti, 2017; Deci and Ryan, 2000). This thread supports the expectation of a negative association between isolation and engagement. Finally, a smaller body of work suggests that the relationship between the amount of remote work and positive outcomes may be curvilinear rather than linear, with moderate levels of remote working outperforming both minimal and extensive levels — an intuition that this study probes exploratorily in relation to the number of remote days worked.
Reading these threads together, two observations stand out and together justify the design of the present study. The first is that the threads are not in genuine contradiction so much as in tension: the autonomy and balance that make hybrid working attractive are the same features that, if mismanaged, generate boundary erosion and isolation. The literature thus points towards a contingency account, in which the net effect of hybrid working on engagement depends on the configuration of resources and demands that a particular arrangement delivers, rather than towards any universal verdict. The second observation is methodological. Much of the evidence rests on comparisons between remote and office workers, or on before-and-after pandemic comparisons, both of which confound the hybrid arrangement with a host of other differences. A design that instead measures the specific attributes of the hybrid experience within a hybrid-working population, and relates them to engagement, is better placed to isolate the mechanisms at work. This is precisely the approach adopted here, and it follows directly from the shape of the existing evidence.
The financial services sector provides a distinctive setting in which these general relationships may be amplified, attenuated or reshaped. Three contextual features are particularly relevant. First, the sector is heavily regulated, and regulatory expectations around supervision, record-keeping, information security and the oversight of certified staff have historically been easier to satisfy through physical co-presence. Hybrid working in finance therefore proceeds under constraints that are less salient in many other professional sectors, and employees may experience more monitoring, more restrictions on where remote work can occur, and more friction around confidential tasks — all of which could condition how autonomy and support are perceived.
Second, financial services work is frequently collaborative and network-dependent. Deal teams, trading desks, underwriting functions and advisory groups rely on dense, fast-moving interaction, informal knowledge transfer and the rapid formation of trust, much of which has traditionally been accomplished through co-location. The isolation risk of hybrid working may consequently bear more heavily on engagement in finance than in sectors where work is more individual and modular. Third, the sector has a strong presence-oriented culture in which visible attendance has often been read as a proxy for commitment and ambition, particularly in front-office and client-facing roles. Where such a culture persists, employees who take up hybrid or remote options may fear reputational or career penalties, which could depress perceived support and engagement even where formal policies are generous.
At the same time, UK financial services firms compete intensely for scarce specialist talent, and flexible working has become a significant factor in recruitment and retention, especially among younger professionals and those with caring responsibilities. Industry bodies have reported a broad, durable shift towards hybrid arrangements for office-based finance roles since the pandemic (Chartered Institute of Personnel and Development, 2022; Office for National Statistics, 2023), even as some firms have periodically sought to increase mandated attendance. This tension — between a talent market that rewards flexibility and an operating and cultural environment that has historically favoured presence — makes the engagement consequences of hybrid working an especially live question for the sector, and it is the reason this study situates its general hypotheses within the financial services context specifically.
A further contextual nuance is the heterogeneity of the sector itself. The four sub-sectors examined in this study differ in ways likely to shape the hybrid–engagement relationship. Traditional banking and insurance organisations tend to carry the longest institutional histories and, with them, the most entrenched presence-oriented norms and the heaviest regulatory and legacy-technology burdens. Asset and wealth management sits between relationship-driven client service and portfolio work that can be performed effectively at a distance. Fintech firms, by contrast, are frequently younger, flatter and digitally native, and many were designed around distributed collaboration from their inception, which may make hybrid working feel less like a departure from the norm and more like the norm itself. These differences imply that the same hybrid policy may be experienced very differently across the sector, and although the present study treats financial services as a single population for the purpose of testing its general hypotheses, it records sub-sector so that this heterogeneity can be acknowledged and so that future work can examine it directly. The point reinforces the study’s central stance that context conditions outcome.
Synthesising the theoretical and empirical review, the study advances a conceptual framework in which employee engagement is the dependent variable and five attributes of the hybrid experience are the predictors. Four of these attributes are theorised, following the JD–R model, as job resources that drive the motivational process, while the fifth is theorised as a job demand that drives the health-impairment process. Perceived organisational support is expected, on the logic of organisational support theory and social exchange, to carry particular weight.
The reasoning for each hypothesis can be summarised as follows. Schedule flexibility and autonomy satisfy the need for autonomy identified in self-determination theory and function as a motivational resource in the JD–R model, leading to H1: schedule flexibility and autonomy are positively associated with engagement. Perceived organisational support, by signalling that the organisation values and cares for distributed employees, invokes the norm of reciprocity and should elicit greater engagement, leading to H2: perceived organisational support is positively associated with engagement. High-quality communication and connectivity maintain the informational and relational resources on which engaged performance depends while working at a distance, leading to H3: communication and connectivity quality is positively associated with engagement. Work–life balance, where genuinely achieved, conserves the energy and recovery that sustain vigour and absorption, leading to H4: work–life balance is positively associated with engagement. Finally, professional and social isolation frustrates the need for relatedness and acts as a demand that depletes engagement, leading to H5: professional and social isolation is negatively associated with engagement. The study also examines, as an exploratory question rather than a formal hypothesis, whether the number of days worked remotely relates to engagement in a non-linear pattern.
Three gaps emerge from the review and define the contribution of this study. The first is contextual: despite the sector’s economic importance and its distinctive regulatory and cultural characteristics, relatively little engagement research has focused specifically on UK financial services under steady-state hybrid working, with much of the evidence drawn from general or cross-sector samples or from the exceptional pandemic period. The second is conceptual: studies frequently treat hybrid or remote working as a single, undifferentiated condition, obscuring which attributes of the experience actually drive engagement; by decomposing the hybrid experience into five theoretically grounded attributes and testing their relative contribution, this study responds to that gap. The third is practical: human resource leaders in the sector need evidence about which levers most strongly relate to engagement so that limited attention and investment can be directed effectively, yet such comparative, attribute-level evidence is scarce. This study addresses all three gaps within the bounds of a model, illustrative design.
This chapter has conceptualised hybrid working as a bundle of five attributes and employee engagement as an energising, meaningful and absorbing psychological state sustained by organisational resources and relationships. It has established a theoretical framework combining the JD–R model, organisational support theory and self-determination theory, and it has synthesised the empirical evidence into a set of directional expectations. It has situated these expectations within the distinctive context of UK financial services, developed a conceptual model and five hypotheses, and identified the contextual, conceptual and practical gaps the study addresses. The next chapter sets out the methodology through which these hypotheses are tested.
This chapter justifies and describes the methodology used to test the hypotheses developed in Chapter 2. Following the logic of the research “onion” (Saunders, Lewis and Thornhill, 2019), it moves from the outer layers of philosophy and approach, through design, sampling and data collection, to the inner layers of analysis, before addressing validity, reliability and ethics. The guiding principle throughout is coherence: each methodological choice is shown to follow from the study’s positivist stance, its deductive and quantitative character, and its aim of testing relationships between defined variables in a specified population.
The study adopts a positivist philosophical position. Positivism holds that social phenomena can be studied as an objective reality through observable, measurable facts, and that knowledge is advanced by testing theory-derived hypotheses against empirical data in a value-neutral manner (Saunders, Lewis and Thornhill, 2019). This stance is appropriate here for three reasons. First, the constructs of interest — engagement and the five hybrid-working attributes — have been operationalised and measured in prior research using validated quantitative scales, demonstrating that they can be treated as measurable variables. Second, the study’s purpose is explanatory: it seeks to test whether, and how strongly, defined variables are related, which is the characteristic purpose of positivist inquiry. Third, the hypotheses are derived deductively from established theory, consistent with the positivist cycle of theory, hypothesis, measurement and test. An interpretivist alternative, which would treat engagement as a socially constructed meaning to be explored qualitatively, was considered and rejected as misaligned with the study’s explanatory, hypothesis-testing aim, although its value for exploring the lived experience of hybrid working is acknowledged in the recommendations for future research.
Consistent with its positivist philosophy, the study follows a deductive approach, moving from theory to the testing of hypotheses, rather than an inductive approach that would build theory from data. The chosen strategy is a survey, implemented through a cross-sectional, self-completion questionnaire. The survey strategy is well suited to the collection of standardised quantitative data from a relatively large sample at modest cost, and it permits the statistical examination of relationships between variables across many respondents (Bryman and Bell, 2015). A cross-sectional design — capturing data at a single point in time — was adopted in preference to a longitudinal design on grounds of feasibility within the constraints of a Masters dissertation; its inability to establish causal direction is acknowledged as a limitation in Chapter 6. The study is mono-method quantitative: a single quantitative data-collection technique and analysis procedure are used, which maintains internal consistency with the positivist, deductive design.
The target population comprised employees in office-based professional and managerial roles within UK financial services firms — spanning retail and investment banking, insurance, asset and wealth management, and fintech — whose roles are compatible with hybrid working. Because no sampling frame listing all such employees exists, and because access to named individuals within regulated firms is difficult, probability sampling was not feasible. A non-probability sampling strategy combining convenience and snowball sampling was therefore used: the questionnaire link was distributed through professional networks and relevant online professional communities, and initial respondents were invited to share it with eligible colleagues. Eligibility was confirmed by two screening questions at the start of the survey, which required respondents to be currently employed in a UK financial services firm and to work under a hybrid arrangement (at least one regular office day and at least one regular remote day per week); those who did not meet both criteria were routed out of the survey.
A minimum sample size was estimated with reference to the requirements of multiple linear regression. A common rule of thumb recommends at least 10 to 15 cases per predictor, and more formal guidance for detecting a medium effect with five predictors at conventional power indicates a requirement well below 150 (Field, 2018). A target of at least 200 usable responses was set to provide a comfortable margin and to support stable estimates. In the event, 263 responses were received and 247 were retained after screening and data cleaning (see Section 4.2), exceeding the target. The limitations of non-probability sampling for generalisation are acknowledged in Chapter 6; the approach is defended here as a pragmatic and widely used response to the access constraints characteristic of research in regulated sectors.
Two features of the sampling design deserve particular comment because they bear on the credibility of the findings. First, distributing the survey through professional networks risks a self-selection bias, whereby those with strong views about hybrid working, or particularly positive experiences of it, are more likely to respond. This risk cannot be eliminated in a non-probability design, but it was mitigated by framing the invitation neutrally, by avoiding emotive language about hybrid working in the recruitment text, and by securing a reasonably broad spread across sub-sectors, role levels and remote-working intensities, as the sample profile in Chapter 4 confirms. Second, snowball sampling can produce clustering within particular firms or networks, which would threaten the independence of observations. To limit this, the chain was initiated from several unconnected starting points across different sub-sectors, and respondents were not asked to identify their employer, which both protected anonymity and reduced the temptation to over-recruit from a single organisation. These measures do not confer the representativeness of a probability sample, but they strengthen the defensibility of the dataset for the illustrative, method-demonstrating purpose of this dissertation.
Data were collected through a structured online questionnaire administered on a secure survey platform. An online self-completion instrument was selected because it enables efficient distribution to a geographically dispersed population, allows respondents to participate at a convenient time, reduces social-desirability pressure relative to interviewer-administered formats, and permits automated routing and data capture that reduce error (Bryman and Bell, 2015). The questionnaire comprised four sections: an information and consent section; two screening questions; a demographic and work-pattern section capturing gender, age band, sub-sector, organisational tenure, role level and the number of days worked remotely per week; and the main measurement section containing the six multi-item scales.
All substantive items were measured on a five-point Likert scale ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). Established, previously validated scales were adapted to the hybrid-working context to maximise construct validity rather than new items being written from scratch. Employee engagement was measured with a nine-item adaptation of the Utrecht Work Engagement Scale, capturing vigour, dedication and absorption (Schaufeli and Bakker, 2004). Perceived organisational support was measured with items adapted from the established survey of perceived organisational support (Eisenberger et al., 1986). Schedule flexibility and autonomy, communication and connectivity quality, work–life balance, and professional and social isolation were each measured with four to six items drawn and adapted from the flexible- and remote-working literature. Wording was adjusted so that every item referred explicitly to the respondent’s current hybrid arrangement. The isolation items were positively scored for analysis so that a higher score denoted greater isolation.
The study comprised one dependent variable, five independent variables and a set of control variables. Table 3.1 summarises how each construct was operationalised. Each multi-item scale was computed as the mean of its constituent items, producing a continuous score between 1 and 5 for each respondent on each construct. Control variables — age band, organisational tenure and number of remote days — were included to account for alternative explanations of engagement and to isolate the association attributable to the hybrid-working attributes.
| Variable | Type | Items | Example Item (Adapted) |
|---|---|---|---|
| Employee engagement (ENG) | Dependent | 9 | “At my work I feel bursting with energy.” |
| Schedule flexibility and autonomy (FLEX) | Independent | 5 | “I have genuine control over when and where I carry out my work.” |
| Perceived organisational support (POS) | Independent | 6 | “My organisation really cares about my wellbeing when I work remotely.” |
| Communication and connectivity quality (COMM) | Independent | 5 | “The tools and channels we use keep me well connected to my team.” |
| Work–life balance (WLB) | Independent | 5 | “My current working pattern lets me balance work with my personal life.” |
| Professional and social isolation (ISO) | Independent | 5 | “I often feel cut off from my colleagues.” |
| Controls | Control | 3 | Age band; organisational tenure; remote days per week. |
Table 3.1 Operationalisation of study variables. All scale items measured on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Items and figures are illustrative.
Before full distribution, the questionnaire was piloted with a small convenience group of twelve respondents who met the eligibility criteria. The pilot served three purposes: to check that item wording was clear and unambiguous in the hybrid-working context; to confirm that the survey logic, routing and estimated completion time were satisfactory; and to obtain a preliminary indication of scale reliability. Pilot feedback led to minor rewording of three items judged ambiguous, the addition of a “prefer not to say” option on the gender question, and a reduction in the estimated completion time from fourteen to approximately ten minutes after two redundant items were removed. Pilot responses were not included in the main dataset. The pilot increased confidence that the final instrument would function as intended and would yield reliable, interpretable data.
Data were exported from the survey platform and analysed in IBM SPSS Statistics. Analysis proceeded in four stages. First, the data were screened and cleaned: incomplete responses, respondents failing the screening criteria, straight-line response patterns and implausible completion times were removed, and negatively worded items were reverse-scored as required. Second, scale scores were computed and the dataset was described using frequencies for categorical variables and means and standard deviations for continuous variables, and the reliability of each multi-item scale was assessed using Cronbach’s alpha, with a value of .70 taken as the conventional threshold for acceptable internal consistency (Field, 2018). Third, the bivariate relationships between the predictors and engagement were examined using Pearson product-moment correlation coefficients. Fourth, multiple linear regression was used to test the hypotheses simultaneously, with engagement as the dependent variable and the five attributes entered as predictors alongside the controls, yielding standardised beta coefficients, their significance, the proportion of variance explained, and collinearity diagnostics. Before interpreting the regression, the assumptions of linearity, independence of residuals, homoscedasticity, normality of residuals and the absence of problematic multicollinearity were checked (Field, 2018). An exploratory one-way analysis of variance compared engagement across categories of remote days worked to probe the possibility of a non-linear pattern.
Several steps were taken to protect validity and reliability. Construct validity was supported by the use of established, previously validated scales adapted only minimally to the hybrid context, and by the pilot’s confirmation of item clarity. Content validity was addressed by deriving every construct and item directly from the theoretical framework in Chapter 2. Internal reliability was assessed empirically through Cronbach’s alpha (reported in Section 4.4), all values exceeding the .70 threshold. The risk of common-method variance, inherent in single-source self-report designs, was mitigated procedurally by assuring respondents of anonymity to reduce evaluation apprehension, by varying item wording and scale anchors, and by separating the predictor and outcome items within the instrument. External validity is necessarily limited by the non-probability sample and is discussed in Chapter 6. These measures do not eliminate the limitations of a cross-sectional self-report design, but they strengthen confidence that the instrument measured the intended constructs consistently.
The study was designed to comply with recognised principles of research ethics and would, in a live project, be subject to formal institutional ethical approval before any data collection. Participation was voluntary and based on informed consent: the opening page explained the study’s purpose, what participation involved, the voluntary and anonymous nature of the survey and the right to withdraw before submission, and respondents confirmed consent before proceeding. No personally identifying information was collected, and responses were anonymous, protecting confidentiality and reducing the risk of harm to respondents whose candid views about their employers might otherwise expose them. Data were stored securely on an encrypted, access-controlled platform and used solely for the purposes of the study. Because the topic is low-risk and participants were working adults, the principal ethical issues concerned consent, anonymity and secure data handling, all of which were addressed. It is reiterated that the data presented in this model dissertation are illustrative and were generated for teaching purposes; no real respondents were surveyed.
This chapter has justified a positivist, deductive, quantitative methodology, implemented through a cross-sectional online survey of UK financial services employees working under hybrid arrangements. It has described the non-probability sampling strategy and the resulting sample, the design of the questionnaire and the adaptation of validated scales, the operationalisation of the variables, the pilot, and the four-stage analytical procedure. It has set out the measures taken to protect validity and reliability and the ethical safeguards applied. The next chapter reports the findings produced by this methodology.
This chapter presents the findings of the study in the sequence set out in the analysis plan. It reports the response rate and data-screening outcomes, describes the demographic and work-pattern profile of the sample, presents descriptive statistics and reliability coefficients for the six scales, reports the bivariate correlations between the predictors and engagement, and then presents the multiple regression model used to test the five hypotheses. An exploratory comparison of engagement across levels of remote working is also reported. All figures are illustrative and were generated for this model dissertation; they are presented to demonstrate how quantitative results are organised, reported and interpreted, not as empirical evidence about the sector.
The questionnaire link was accessed 311 times. Of these, 263 respondents submitted a response, and 16 were removed during screening and cleaning: 7 failed the eligibility criteria (not currently working under a hybrid arrangement), 5 provided substantially incomplete responses, 3 displayed straight-line response patterns across the scales, and 1 completed the survey in an implausibly short time. This left a final analytical sample of 247 usable responses, an effective usable response rate of 79.4% of submitted responses. Missing data among retained cases were minimal (below 2% on any item) and were handled by pairwise deletion in the correlation analysis and listwise deletion in the regression. The sample size of 247 comfortably exceeds the minimum required for stable estimation of a five-predictor regression model (Field, 2018).
Table 4.1 summarises the demographic and work-pattern characteristics of the 247 respondents. The sample was reasonably balanced by gender and skewed, as expected for an office-based professional population, towards the 30–49 age range. All four target sub-sectors were represented, with banking the largest. Respondents were distributed across junior, managerial and senior levels, and the modal hybrid pattern was two to three days per week worked remotely, consistent with sector norms for the period.
| Characteristic | Category | n | % |
|---|---|---|---|
| Gender | Female | 133 | 53.8 |
| Male | 111 | 44.9 | |
| Prefer not to say | 3 | 1.2 | |
| Age band | 18–29 | 77 | 31.2 |
| 30–39 | 94 | 38.1 | |
| 40–49 | 51 | 20.6 | |
| 50 and over | 25 | 10.1 | |
| Sub-sector | Banking | 89 | 36.0 |
| Insurance | 59 | 23.9 | |
| Asset and wealth management | 54 | 21.9 | |
| Fintech | 45 | 18.2 | |
| Role level | Junior / analyst | 101 | 40.9 |
| Manager | 94 | 38.1 | |
| Senior / executive | 52 | 21.0 | |
| Remote days per week | 1 day | 35 | 14.2 |
| 2 days | 82 | 33.2 | |
| 3 days | 83 | 33.6 | |
| 4 or more days | 47 | 19.0 |
Table 4.1 Demographic and work-pattern profile of respondents (n = 247). Percentages may not sum to 100 owing to rounding. Illustrative data.
Table 4.2 reports the mean, standard deviation and Cronbach’s alpha for each of the six scales. All scales demonstrated acceptable to excellent internal consistency, with alpha coefficients ranging from .79 to .91, comfortably exceeding the conventional .70 threshold (Field, 2018). The descriptive pattern is itself informative: respondents reported relatively high schedule flexibility and autonomy (M = 3.92) and work–life balance (M = 3.78), moderately high perceived organisational support (M = 3.64) and engagement (M = 3.71), somewhat lower communication and connectivity quality (M = 3.41), and a below-midpoint level of isolation (M = 2.87). In other words, the sample on average experienced its hybrid arrangements positively, with connectivity emerging as the comparatively weaker resource and isolation present but not dominant.
The standard deviations add a second layer of meaning. The comparatively large dispersion on communication and connectivity quality (SD = 0.92) and isolation (SD = 0.95) indicates that respondents’ experiences on these dimensions varied widely — some felt well connected and others distinctly cut off — whereas the tighter dispersion on engagement itself (SD = 0.74) suggests a more consistent central tendency. This variability is analytically useful rather than problematic: it is precisely the spread on the predictors that allows their relationship with engagement to be estimated. The relatively modest mean for connectivity, combined with its wide spread, also foreshadows a practical theme developed in Chapter 5, namely that the quality of distributed communication is the attribute on which firms in the sample appear least consistent, and therefore the one on which there may be most room for improvement, even if its marginal contribution to engagement proves smaller than that of support or balance.
| Scale | Items | Mean | SD | Cronbach’s α |
|---|---|---|---|---|
| Employee engagement (ENG) | 9 | 3.71 | 0.74 | .91 |
| Schedule flexibility and autonomy (FLEX) | 5 | 3.92 | 0.78 | .84 |
| Perceived organisational support (POS) | 6 | 3.64 | 0.86 | .88 |
| Communication and connectivity quality (COMM) | 5 | 3.41 | 0.92 | .81 |
| Work–life balance (WLB) | 5 | 3.78 | 0.81 | .86 |
| Professional and social isolation (ISO) | 5 | 2.87 | 0.95 | .79 |
Table 4.2 Descriptive statistics and scale reliability (n = 247). Scores range from 1 to 5. Illustrative data.
Pearson correlations were computed to examine the bivariate relationships among the study variables (Table 4.3). All four resource variables correlated positively and significantly with engagement, and isolation correlated negatively and significantly, consistent with the direction of the hypotheses. The strongest zero-order correlation with engagement was for perceived organisational support (r = .57, p < .01), followed by work–life balance (r = .52), schedule flexibility and autonomy (r = .48), communication quality (r = .44) and isolation (r = −.39). The inter-correlations among the predictors were moderate (the highest being .51 between POS and WLB), suggesting related but distinct constructs and giving an early indication that multicollinearity was unlikely to be problematic in the regression.
| Variable | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| 1. ENG | — | |||||
| 2. FLEX | .48** | — | ||||
| 3. POS | .57** | .42** | — | |||
| 4. COMM | .44** | .35** | .47** | — | ||
| 5. WLB | .52** | .45** | .51** | .33** | — | |
| 6. ISO | −.39** | −.28** | −.41** | −.46** | −.34** | — |
Table 4.3 Pearson correlation matrix (n = 247). ** p < .01 (two-tailed). Illustrative data.
To test the five hypotheses simultaneously and to establish the relative predictive contribution of each attribute, a standard multiple linear regression was conducted with engagement as the dependent variable and the five hybrid-working attributes as predictors, controlling for age band, organisational tenure and the number of remote days worked. Before interpretation, the regression assumptions were examined and found satisfactory: inspection of residual plots supported linearity and homoscedasticity; the Durbin–Watson statistic (2.03) indicated independent residuals; the normal probability plot of residuals was approximately linear; and collinearity diagnostics were sound, with all variance inflation factors below 1.9 and all tolerance values above 0.5, confirming that multicollinearity was not a concern.
The model was statistically significant and explained just over half of the variance in engagement: R = .71, R² = .504, adjusted R² = .494, F(5, 241) = 49.02, p < .001. The control variables did not add significant explanatory power once the attributes were entered and are omitted from the summary for clarity. The branded results table below (Table 4.4) reports the standardised coefficients, their significance and the collinearity diagnostics for the five predictors.
| Predictor | B | SE B | β | t | p | VIF |
|---|---|---|---|---|---|---|
| (Constant) | 1.02 | 0.23 | — | 4.43 | <.001 | — |
| Perceived organisational support (POS) | 0.27 | 0.05 | .31 | 5.12 | <.001 | 1.63 |
| Work–life balance (WLB) | 0.22 | 0.06 | .24 | 3.98 | <.001 | 1.54 |
| Schedule flexibility and autonomy (FLEX) | 0.17 | 0.06 | .18 | 3.05 | .003 | 1.41 |
| Communication and connectivity quality (COMM) | 0.12 | 0.05 | .14 | 2.41 | .017 | 1.47 |
| Professional and social isolation (ISO) | −0.10 | 0.05 | −.12 | −2.18 | .030 | 1.39 |
| Dependent variable: employee engagement. R = .71, R² = .504, adjusted R² = .494, F(5, 241) = 49.02, p < .001. Controls (age, tenure, remote days) entered but non-significant. All figures illustrative. | ||||||
The model confirms all five hypotheses. Perceived organisational support was the strongest predictor of engagement (β = .31, p < .001), followed by work–life balance (β = .24, p < .001), schedule flexibility and autonomy (β = .18, p = .003) and communication and connectivity quality (β = .14, p = .017). Professional and social isolation was a significant negative predictor (β = −.12, p = .030). The pattern of coefficients indicates that, while all five attributes matter, the relational and boundary resources — feeling supported and achieving balance — carry more weight than either raw flexibility or technical connectivity, and that isolation, though a genuine cost, is the weakest of the five effects once the resources are accounted for.
The exploratory one-way analysis of variance comparing engagement across the four categories of remote days worked was significant, F(3, 243) = 4.11, p = .007. Post-hoc comparisons indicated that respondents working two days (M = 3.82) or three days (M = 3.86) remotely reported significantly higher engagement than those working four or more days (M = 3.49); the single-day group (M = 3.63) fell between these extremes. This pattern is consistent with the inverted-U intuition drawn from the literature: a moderate dose of remote working appears most favourable for engagement, while the most extensive remote working is associated with somewhat lower engagement, plausibly through heightened isolation and weaker connectivity. This finding is exploratory and should be treated with caution.
Table 4.5 summarises the outcome of the five hypothesis tests. All five hypotheses were supported by the illustrative data.
| Hypothesis | Statement | Result |
|---|---|---|
| H1 | Flexibility/autonomy → engagement (+) | Supported (β = .18, p = .003) |
| H2 | Perceived support → engagement (+) | Supported (β = .31, p < .001) |
| H3 | Communication quality → engagement (+) | Supported (β = .14, p = .017) |
| H4 | Work–life balance → engagement (+) | Supported (β = .24, p < .001) |
| H5 | Isolation → engagement (−) | Supported (β = −.12, p = .030) |
Table 4.5 Summary of hypothesis testing. Illustrative data.
This chapter has reported the study’s findings. After screening, 247 usable responses were analysed. All six scales were reliable, and the sample on average experienced its hybrid arrangements positively. All four resource variables correlated positively, and isolation negatively, with engagement. The regression model explained 50.4% of the variance in engagement and supported all five hypotheses, with perceived organisational support and work–life balance the strongest predictors. An exploratory analysis suggested an inverted-U relationship between the amount of remote working and engagement. The next chapter interprets these findings in the light of the literature and draws out their implications.
This chapter interprets the findings presented in Chapter 4 in relation to the theoretical framework and empirical literature reviewed in Chapter 2. It addresses the research questions in turn, explains why the observed pattern of results arose, and draws out the theoretical and practical implications. The discussion is organised around the overall relationship between hybrid working and engagement, the relative importance of the five attributes, the implications for engagement theory, and the implications for human resource management in UK financial services firms. Throughout, it is reiterated that the findings derive from an illustrative dataset and are offered to demonstrate interpretive reasoning rather than to assert fact.
The central research question asked what relationship exists between hybrid working arrangements and employee engagement. The findings indicate that the relationship is real, substantial and — crucially — conditional. The five attributes of the hybrid experience together explained just over half of the variance in engagement, a large effect by the conventions of organisational research, which demonstrates that how hybrid working is experienced is strongly bound up with how engaged employees feel. Yet the result that matters most for interpretation is not the size of the overall effect but its decomposition: engagement was not predicted by hybrid working as such, but by the quality of the resources that the arrangement delivered and the demand it imposed. This supports the study’s central argument, developed from the JD–R model, that hybrid working is neither inherently good nor inherently bad for engagement; its consequences depend on its design and management (Bakker and Demerouti, 2017).
This conditional interpretation helps to reconcile the apparently contradictory findings in the wider literature. Studies that report positive engagement effects of flexible working and studies that report negative effects may both be correct, because each is likely to have sampled organisations in which the underlying resources and demands differed. Where hybrid arrangements are accompanied by genuine support, protected boundaries, real autonomy and good connectivity, the present findings suggest they should enhance engagement; where they are accompanied by neglect, boundary erosion, token flexibility and poor connectivity, the same nominal arrangement should depress it. The policy implication — that firms should attend less to the fact of hybrid working and more to its quality — follows directly and is developed below.
The magnitude of the overall effect also warrants comment. An adjusted R² of .494 means that the five attributes accounted for almost half of the variation in engagement across respondents, leaving roughly the same proportion unexplained. This residual variance is not a weakness of the account so much as a reminder of the many other determinants of engagement — the intrinsic character of the work, the quality of individual relationships, personality, career stage, pay and reward, and firm-specific culture — that lie outside the hybrid-working lens adopted here. The contribution of the model is to show that the attributes of the hybrid experience form a coherent and substantial block of predictors, not to claim that they are the only ones that matter. For the sector, this is a hopeful message: a large share of the engagement variance attributable to hybrid working flows through attributes — support, balance, autonomy, connectivity — that are, to a considerable degree, within managerial control, rather than through fixed features of the work or the workforce.
The second research question asked which attributes most strongly predict engagement. The regression answered this clearly: perceived organisational support was the strongest predictor, followed by work–life balance, then flexibility and autonomy, then communication quality, with isolation the weakest (though still significant) effect once the resources were accounted for. The third research question, concerning the amount of remote working, was answered by the exploratory analysis of variance, which suggested an inverted-U pattern in which moderate remote working (two to three days) was associated with the highest engagement. Each of these results is interpreted in turn in the following section.
The most theoretically significant finding is the primacy of perceived organisational support. That support out-predicted raw flexibility is consistent with organisational support theory and with a social-exchange account of engagement (Eisenberger et al., 1986; Saks, 2006): what engages employees under hybrid working is not simply the freedom to work from home but the belief that their organisation values them and cares about their wellbeing while they do so. This interpretation has an important corollary for the presence-oriented culture of financial services. Where distance is read by employees as a signal of organisational indifference or distrust — as it may be in firms that grant hybrid working grudgingly or that penalise it reputationally — perceived support, and hence engagement, is likely to suffer regardless of the formal generosity of the policy. Conversely, firms that actively invest in the inclusion, development and wellbeing of distributed employees should be able to convert hybrid working into an engagement advantage.
The second-strongest predictor, work–life balance, reinforces this relational reading while adding a resource-conservation dimension. The finding that balance strongly predicts engagement is consistent with the JD–R logic that conserved energy and adequate recovery sustain the vigour and absorption components of engagement (Schaufeli and Bakker, 2004; Bakker and Demerouti, 2017). It is notable, however, that balance was reported as relatively high in the sample yet remained a powerful differentiator, which suggests that the firms best able to protect boundaries — discouraging an always-on culture, respecting non-working hours, and designing workloads realistically — reap a disproportionate engagement return. The risk identified in the literature, that hybrid working collapses the work–home boundary and erodes the very balance it promises, implies that this resource cannot be taken for granted and must be actively managed.
Flexibility and autonomy, though a significant positive predictor, carried less weight than support or balance. This is a theoretically interesting result. It suggests that autonomy, while valued and consistent with the motivational premise of self-determination theory (Deci and Ryan, 2000), is not sufficient on its own to drive engagement; employees who are given control but who feel unsupported or unable to maintain balance do not appear to convert that control into engagement as effectively. In the language of the JD–R model, autonomy is a resource, but its motivational potential is realised most fully in combination with other resources. Communication and connectivity quality, the weakest of the positive predictors, retained a significant effect, confirming that the informational and relational infrastructure of distributed work matters, but suggesting that once a baseline of adequate connectivity is reached, further gains in tooling yield smaller engagement returns than gains in support or balance.
Finally, the negative effect of isolation, though significant, was the smallest of the five. Two readings are plausible. The first is substantive: in a sample working predominantly two to three days remotely, isolation may simply have been contained by regular office contact, limiting its depressive effect on engagement. The second is statistical: because isolation correlated with the resource variables (notably support and connectivity), part of its bivariate association with engagement was absorbed by those resources in the multivariate model. Both readings are compatible with the exploratory finding that the most extensive remote working was associated with the lowest engagement, which is most readily explained by the heightened isolation and weaker connectivity that accompany near-total remote work. Together these results suggest that isolation is best addressed not in isolation but by strengthening the resources — support and connectivity — that buffer it.
The study carries several implications for engagement theory. First, it supports the extension of the JD–R model to the hybrid context: the attributes of hybrid working behave, empirically, as the model predicts, with resources driving engagement and a demand depressing it (Bakker and Demerouti, 2017). This is a useful confirmation, because much of the model’s original evidence derives from conventional workplaces, and it was not obvious a priori that the resource–demand logic would transfer cleanly to arrangements in which the workplace itself is partly dissolved. Second, the primacy of perceived support strengthens the case for integrating organisational support theory and social-exchange reasoning into accounts of engagement under flexible working: the relational quality of the employment relationship appears to matter more than the spatial facts of where work is done (Eisenberger et al., 1986; Saks, 2006). Third, the comparatively modest effect of raw autonomy qualifies a simple self-determination reading: autonomy is necessary but, in this setting, not the dominant driver, which suggests that the satisfaction of the relatedness need (through support and connectivity) and the conservation of resources (through balance) may matter at least as much as the satisfaction of the autonomy need (Deci and Ryan, 2000). Collectively, these implications argue for a multi-theory, attribute-based approach to studying flexible working rather than reliance on any single framework.
For human resource and operational leaders in financial services, the findings translate into a clear order of priorities. Because perceived organisational support was the strongest predictor, the highest-return intervention is likely to be the deliberate cultivation of felt support for distributed employees: equipping and holding managers accountable for regular, high-quality contact with remote team members; ensuring that development, recognition and progression are not quietly rationed to those who are most physically present; and signalling, credibly and consistently, that the organisation values employees’ contribution and wellbeing wherever they work. In a sector whose culture has historically equated presence with commitment, this requires active effort to prevent “proximity bias” from undermining the engagement of hybrid workers.
Because work–life balance was the second-strongest predictor, the protection of boundaries is the second priority: discouraging an always-on expectation, respecting non-working hours in a sector prone to long hours, and designing workloads so that the flexibility of hybrid working does not silently convert into work intensification. Because flexibility and autonomy matter but are not sufficient alone, firms should offer genuine rather than token choice over working patterns while recognising that autonomy delivers its engagement return only when support and balance are also present. Because connectivity retains a significant effect, investment in reliable collaboration tools and in team practices that keep remote members informed and included remains worthwhile, though with diminishing returns beyond an adequate baseline. Finally, the exploratory inverted-U finding offers tentative guidance on dosage: a moderate hybrid split — of the order of two to three remote days — appears most favourable for engagement, suggesting that both heavy-handed full-office mandates and near-total remote arrangements may be sub-optimal, and that coordinated, moderate hybrid patterns deserve consideration. These recommendations are offered as reasoned inferences from illustrative data and should be validated before being acted upon.
A cross-cutting implication concerns the pivotal role of the line manager. Each of the top priorities — cultivating felt support, protecting boundaries and making autonomy genuine — is enacted, day to day, by managers rather than by policy documents. Yet many managers in financial services were themselves promoted in, and for, a co-located world, and may lack the specific capabilities that sustaining engagement at a distance requires: structuring regular one-to-one contact that is developmental rather than merely supervisory, reading the engagement and wellbeing of team members whom they see less often, allocating stretch opportunities and recognition equitably between office-present and remote staff, and modelling healthy boundaries themselves. The finding that perceived support dominates the engagement model therefore points not to a new policy but to an investment in managerial capability: equipping managers to lead hybrid teams, and holding them accountable for doing so, is likely to be the single most consequential lever available to the sector. Treating hybrid working as a leadership challenge rather than an estates or technology question reframes it in a way the evidence here supports.
This chapter has interpreted the findings as showing that hybrid working relates to engagement conditionally, through the resources it delivers and the demand it imposes, rather than as an undifferentiated condition. Perceived organisational support and work–life balance emerged as the dominant positive predictors, flexibility and connectivity as significant but lesser ones, and isolation as a modest negative one, with a moderate dose of remote working associated with the highest engagement. The chapter drew out implications for engagement theory — supporting a multi-theory, attribute-based approach — and for human resource practice in financial services, where cultivating felt support and protecting balance are the highest priorities. The final chapter concludes the study and offers recommendations.
This dissertation examined the relationship between hybrid working and employee engagement among professionals in UK financial services firms, and sought to identify which attributes of the hybrid experience most strongly predict engagement. Motivated by the sector’s economic importance, its distinctive regulatory and cultural characteristics and the mixed and often pandemic-era evidence base, the study adopted a positivist, deductive, quantitative design, surveying 247 financial services employees working under hybrid arrangements and analysing the data through reliability analysis, correlation and multiple regression. Drawing on the job demands–resources model, organisational support theory and self-determination theory, it decomposed hybrid working into five attributes and tested their relationships with engagement.
The four objectives set in Chapter 1 were met. The first objective — to review the literature and identify the mechanisms linking hybrid working and engagement — was fulfilled in Chapter 2, which established an attribute-based, multi-theory framework in which four resources and one demand connect hybrid working to engagement. The second objective — to measure engagement and the five attributes — was achieved through the reliable survey instrument reported in Chapters 3 and 4. The third objective — to analyse the strength, direction and relative contribution of the relationships — was met by the correlation and regression analyses, which showed that the five attributes jointly explained just over half of the variance in engagement, that all five hypotheses were supported, and that perceived organisational support and work–life balance were the strongest predictors. The fourth objective — to develop recommendations for practice — is addressed below.
The overarching conclusion is that hybrid working is best understood not as a uniform condition with a fixed effect on engagement but as a design space whose engagement consequences depend on the resources it delivers and the demands it imposes. In the financial services setting examined, felt organisational support and protected work–life balance mattered most, autonomy and connectivity mattered but less, isolation imposed a modest cost, and a moderate dose of remote working was most favourable. For a sector historically oriented towards presence, the practical message is that the route to an engaged hybrid workforce runs less through attendance rules than through the deliberate management of support, balance, autonomy and connection.
On the basis of the findings, the following recommendations are offered to human resource and operational leaders in UK financial services firms, in descending order of expected impact:
Within the bounds of a model, illustrative design, the study makes three contributions corresponding to the gaps identified in Chapter 2. Contextually, it focuses on steady-state hybrid working in UK financial services specifically, rather than on general samples or the exceptional pandemic period. Conceptually, it decomposes hybrid working into five theoretically grounded attributes and establishes their relative predictive contribution, moving beyond the undifferentiated treatment common in the literature. Practically, it furnishes human resource leaders with an ordered set of levers, grounded in the relative strength of the predictors, for sustaining engagement under hybrid working.
The study’s conclusions must be read against several limitations. First, the cross-sectional design captures associations at a single point in time and cannot establish causal direction; it is possible, for example, that more engaged employees perceive greater support rather than the reverse, or that the relationships are reciprocal. Second, the reliance on single-source self-report data raises the possibility of common-method variance, which was mitigated procedurally but not eliminated, and means that no objective performance or employer data corroborate the self-reported engagement. Third, the non-probability convenience and snowball sample limits the generalisability of the findings to the wider population of UK financial services employees and may over-represent those with positive hybrid experiences or strong professional-network connections. Fourth, the study measured engagement and its predictors but did not model potential mediators (such as trust) or moderators (such as managerial quality or role type) that may shape the relationships. Fifth, and most fundamentally, the data are illustrative and were generated for this model dissertation; the results therefore demonstrate analytical method and interpretive reasoning and should not be cited as empirical evidence about the sector.
These limitations point to several avenues for future research. A longitudinal design would allow the causal direction of the relationships to be examined and would capture how engagement evolves as hybrid arrangements mature. A multi-source design, combining employee self-report with manager ratings or objective performance and retention data, would address common-method concerns and test whether the engagement effects observed here translate into performance outcomes. Probability or large-scale panel sampling would strengthen generalisability. Mediation and moderation analyses — examining, for instance, the role of trust, managerial capability, role type or sub-sector — would deepen the explanatory account. Finally, qualitative and mixed-methods work would illuminate the lived experience behind the statistical relationships, particularly the mechanisms by which presence-oriented cultures shape perceived support, and would complement the quantitative approach taken here. Research specifically testing the inverted-U dosage relationship, which was only exploratory in this study, would also be valuable.
Hybrid working has become a defining feature of professional employment, and nowhere is its management more consequential, or more contested, than in a financial services sector poised between a presence-oriented heritage and a flexibility-rewarding talent market. This model dissertation has argued, through an illustrative quantitative study, that the sector’s engagement question is not whether to permit hybrid working but how to design and manage it — and that the answer lies above all in the felt support firms extend to their distributed employees and the boundaries they help them protect. As a teaching reference, the document is intended to show how such an argument is constructed, measured and defended at Masters level; as a substantive claim, its conclusions await testing against real-world data.
Bakker, A.B. and Demerouti, E. (2017) ‘Job demands–resources theory: taking stock and looking forward’, Journal of Occupational Health Psychology, 22(3), pp. 273–285.
Blau, P.M. (1964) Exchange and Power in Social Life. New York: Wiley.
Bryman, A. and Bell, E. (2015) Business Research Methods. 4th edn. Oxford: Oxford University Press.
Chartered Institute of Personnel and Development (2022) Flexible and Hybrid Working Practices. London: CIPD.
Deci, E.L. and Ryan, R.M. (2000) ‘The “what” and “why” of goal pursuits: human needs and the self-determination of behaviour’, Psychological Inquiry, 11(4), pp. 227–268.
Eisenberger, R., Huntington, R., Hutchison, S. and Sowa, D. (1986) ‘Perceived organizational support’, Journal of Applied Psychology, 71(3), pp. 500–507.
Field, A. (2018) Discovering Statistics Using IBM SPSS Statistics. 5th edn. London: Sage.
Kahn, W.A. (1990) ‘Psychological conditions of personal engagement and disengagement at work’, Academy of Management Journal, 33(4), pp. 692–724.
Office for National Statistics (2023) Characteristics of Homeworkers, Great Britain. Newport: Office for National Statistics.
Saks, A.M. (2006) ‘Antecedents and consequences of employee engagement’, Journal of Managerial Psychology, 21(7), pp. 600–619.
Saunders, M., Lewis, P. and Thornhill, A. (2019) Research Methods for Business Students. 8th edn. Harlow: Pearson.
Schaufeli, W.B. and Bakker, A.B. (2004) ‘Job demands, job resources, and their relationship with burnout and engagement: a multi-sample study’, Journal of Organizational Behavior, 25(3), pp. 293–315.
Shuck, B. and Wollard, K. (2010) ‘Employee engagement and HRD: a seminal review of the foundations’, Human Resource Development Review, 9(1), pp. 89–110.
Wang, B., Liu, Y., Qian, J. and Parker, S.K. (2021) ‘Achieving effective remote working during the COVID-19 pandemic: a work design perspective’, Applied Psychology, 70(1), pp. 16–59.
The following extract illustrates the structure and item style of the questionnaire. All substantive items used a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). The full instrument contained the six scales summarised in Table 3.1 together with the demographic and work-pattern questions reported in Table 4.1. Items marked (R) were reverse-scored where appropriate.
Section A – Screening. A1. Are you currently employed by a UK financial services firm? (Yes / No). A2. Does your current role involve a hybrid pattern of at least one regular office day and one regular remote day each week? (Yes / No). Respondents answering “No” to either question were thanked and routed out.
Section B – Work pattern and demographics. B1. Which sub-sector best describes your employer? B2. What is your role level? B3. On average, how many days per week do you work remotely? B4. How long have you worked for your current employer? B5. Age band. B6. Gender.
Section C – Employee engagement (adapted UWES-9). C1. “At my work I feel bursting with energy.” C2. “I am enthusiastic about my job.” C3. “I am immersed in my work.” (and six further items across vigour, dedication and absorption).
Section D – Hybrid-working attributes. D1 (FLEX). “I have genuine control over when and where I carry out my work.” D2 (POS). “My organisation really cares about my wellbeing when I work remotely.” D3 (COMM). “The tools and channels we use keep me well connected to my team.” D4 (WLB). “My current working pattern lets me balance work with my personal life.” D5 (ISO, R). “I often feel cut off from my colleagues.” (with additional items for each construct as summarised in Table 3.1).
Note: this questionnaire extract is illustrative and was created for this model dissertation; no respondents were surveyed.
You May Also Like
All work is written by human writers. 100% AI free, guaranteed.
100% money back guarantee if you find plagiarism in our work.
COMPANY DETAILS
