Topic Selection and HR Problem Diagnosis

Introduction

A dissertation in human resource management begins long before data collection. It begins with the ability to recognise a significant workforce issue, understand its organisational context, and convert a broad concern into a focused and researchable problem. Topic selection and problem diagnosis are therefore not administrative preliminary tasks; they are foundational intellectual activities that determine the quality, relevance and feasibility of the entire project.

Organisations routinely experience concerns such as employee turnover, skills shortages, weak engagement, workplace conflict, inequitable promotion, ineffective performance management, absenteeism, burnout, poor adoption of digital systems and difficulties attracting diverse talent. These concerns may be visible, but their causes are rarely obvious. A high resignation rate, for example, may reflect pay dissatisfaction, ineffective supervision, limited career opportunities, workload, commuting difficulties, psychological insecurity or competition from other employers. The HR researcher must distinguish symptoms from causes and assumptions from evidence.

A well-selected dissertation topic should meet four tests. It should be important to an organisation or workforce; connected to HR theory and strategy; feasible within the available time, access and resources; and capable of being investigated ethically through appropriate evidence. This chapter explains how to apply these tests and how to conduct a preliminary organisational diagnosis that leads to a clear research problem, research questions and a defensible project scope.

Learning Objectives

By the end of this chapter, students should be able to:

1. Identify contemporary workforce issues relevant to HR practice and research.
2. Distinguish a symptom, an issue, a problem statement and a research question.
3. Consult stakeholders systematically and ethically.
4. Conduct a preliminary organisational diagnosis using qualitative and quantitative evidence.
5. Apply frameworks such as the 5W1H method, root-cause analysis and the socio-technical systems perspective.
6. Evaluate potential topics for significance, feasibility, originality and ethical acceptability.
7. Convert a practical concern into a focused, researchable HR problem aligned with organisational strategy.

1. From Workforce Concern to Research Problem

A workforce concern is an observed condition that attracts attention. It may be expressed informally: “Our new employees leave quickly” or “Managers are not using the performance system.” A problem statement is more precise. It identifies the undesirable condition, the population affected, the context and the evidence indicating that the condition exists. A research problem goes further by identifying a knowledge or decision gap that can be investigated through systematic inquiry.

The following distinctions are important:

The difference between a symptom and a cause is central to diagnosis. Turnover is not necessarily the underlying problem; it may be an outcome of several conditions. Similarly, low training attendance may not indicate low motivation. Sessions may be scheduled during operationally difficult periods, may lack relevance, or may be inaccessible to shift workers.

A useful problem statement normally contains five elements:

• Condition: What is happening?
• Population: Who is affected?
• Context: Where and under what circumstances?
• Evidence: What indicates that the issue is real?
• Consequence or gap: Why does it matter, and what is not yet known?

For example: “In the customer-support division, voluntary turnover among employees with less than two years of service increased during the previous 12 months. Exit interviews suggest concerns about workload and career progression, but the organisation has not systematically examined how job demands, development opportunities and supervisor support relate to employees’ intention to remain.”

This statement is stronger than “The company has a retention problem” because it is bounded, evidence-oriented and capable of guiding research design.

2. Selecting Contemporary HR Issues

Contemporary HR issues arise from changes in labour markets, technology, regulation, employee expectations and organisational strategy. Important areas include:

• hybrid and remote work, including fairness between remote and on-site employees;
• artificial intelligence in recruitment, workforce planning and performance management;
• employee wellbeing, stress, burnout and psychosocial risk;
• diversity, equity, inclusion and belonging;
• skills shortages, reskilling and lifelong learning;
• ageing workforces and intergenerational collaboration;
• employee voice, participation and social dialogue;
• precarious work, platform work and contractor management;
• pay equity, transparency and reward fairness;
• workforce analytics, privacy and algorithmic bias;
• sustainable HRM and the relationship between people practices and environmental goals;
• ethical leadership, trust and organisational justice.

A contemporary issue is not automatically a suitable dissertation topic. A topic should be selected because it is relevant to a particular organisation or workforce and can be investigated within the project’s boundaries. “The future of AI in HR” is too broad for most undergraduate dissertations. “How do employees perceive the transparency and fairness of AI-assisted shortlisting in one organisation?” is more manageable.

Topic selection should consider both strategic alignment and employee significance. Strategic alignment asks whether the issue affects organisational objectives such as service quality, productivity, innovation, risk management or workforce sustainability. Employee significance asks whether the issue affects dignity, fairness, safety, development, inclusion or quality of working life. Strong HR research usually connects both dimensions.

3. Stakeholder Consultation

Stakeholders are people or groups who are affected by the issue, possess relevant knowledge, make decisions about it or may be affected by the research findings. They may include employees, line managers, HR professionals, senior leaders, trade union or employee representatives, occupational health staff, customers and external partners.

Stakeholder consultation has three purposes. First, it helps the researcher understand the organisational context. Second, it tests whether an apparent issue is sufficiently important and feasible. Third, it prevents the researcher from imposing an inaccurate interpretation on the organisation.

Consultation should not become a search for approval of a predetermined conclusion. The researcher should ask open questions, listen for competing perspectives and record uncertainty. Typical questions include:

• What workforce outcome is causing concern?
• When and where does it occur?
• Which groups experience it most strongly?
• What evidence already exists?
• What explanations have been proposed?
• What actions have already been attempted?
• What would a useful research finding enable the organisation to decide or improve?
• What information cannot ethically or legally be accessed?

Different stakeholders may describe the same issue differently. Senior managers may emphasise productivity, managers may emphasise capability, and employees may emphasise workload or fairness. These perspectives are not simply competing opinions; together they reveal how the issue operates across the organisation.

A stakeholder map can classify participants according to influence and interest:

Consultation must respect voluntary participation, confidentiality and power relationships. An employee should not feel that refusal will affect employment. Managers should not be allowed to identify individual responses unless this is explicitly agreed and ethically justified.

4. Preliminary Organisational Diagnosis

Organisational diagnosis is a systematic attempt to understand how an organisation functions and why a particular condition may exist. It is preliminary because a dissertation usually cannot diagnose every aspect of an organisation. The purpose is to establish a credible problem definition and identify suitable evidence for the main study.

4.1 The diagnostic levels

A useful diagnosis examines three connected levels:

1. Individual level: perceptions, attitudes, skills, wellbeing, behaviour and employment experiences.
2. Team or managerial level: leadership, communication, workload allocation, team climate, supervision and local practices.
3. Organisational level: strategy, structure, policies, culture, technology, labour-market position and external regulation.

An issue located at one level may be produced by conditions at another. For example, employee performance may appear to be an individual capability problem while actually reflecting unclear goals, inadequate systems or contradictory managerial priorities.

4.2 The 5W1H diagnostic frame

The 5W1H framework provides a simple starting structure:

• What is happening?
• Who is affected?
• Where does it occur?
• When did it begin or change?
• Why might it matter?
• How might the condition be produced or experienced?

The framework does not prove causation. It organises preliminary information and prevents vague descriptions. It is particularly useful during an initial meeting with an organisational contact.

4.3 Evidence sources

A diagnosis should use evidence appropriate to the question. Possible sources include HR records, employee surveys, exit interviews, absence data, recruitment metrics, training records, policy documents, job descriptions, performance reports, focus groups and interviews. Each source has strengths and limitations.

Triangulation means comparing evidence from different sources. If exit records, employee interviews and absence data point toward workload pressure, confidence in the preliminary diagnosis increases. Triangulation does not mean that all sources must agree; disagreement may itself be analytically important.

4.4 Root-cause analysis

Root-cause analysis seeks to move beyond immediate symptoms. The “five whys” technique repeatedly asks why a condition occurred. It should be used carefully: organisational problems usually have multiple causes, and a linear chain can oversimplify reality.

For example:

• Why are training completion rates low? Employees do not attend.
• Why do they not attend? Sessions conflict with shift patterns.
• Why are sessions scheduled this way? The timetable was designed for standard office hours.
• Why was shift work not considered? Workforce segmentation was not included in the training planning process.

The emerging research problem may therefore concern the accessibility and design of learning provision rather than employee commitment.

A fishbone or cause-and-effect analysis can group possible causes under categories such as people, process, technology, leadership, work design and external environment. These categories are prompts, not evidence. Proposed causes must be tested rather than treated as established facts.

4.5 Socio-technical diagnosis

The socio-technical perspective views organisations as systems in which social elements—people, roles, leadership and culture—interact with technical elements such as workflows, information systems, equipment and job design. HR problems often arise from misalignment between these elements.

For example, a new HR information system may promise efficiency but create frustration if employees lack training, managers receive conflicting instructions and the system requires duplicate data entry. A narrow explanation blaming “resistance to change” would overlook design and implementation factors.

The socio-technical approach encourages questions such as:

• What new technology or process has changed work?
• How have roles and decision rights changed?
• Do employees have the skills and support required?
• Does the formal system match actual work practices?
• Which groups benefit and which groups carry additional burdens?

5. From Diagnosis to Research Design

Once an issue is diagnosed, the researcher must narrow it. A useful topic is neither so broad that it becomes unmanageable nor so narrow that it lacks significance. Scope can be controlled through population, location, period, variables and outcome.

Compare the following:

• Broad: “Employee wellbeing in modern organisations.”
• Improved: “Work-related stress among hospital administrative employees.”
• Focused: “How do workload and supervisor support influence perceived burnout among administrative employees in one hospital department?”

The final version identifies a population, explanatory concepts and an outcome. It does not claim that the causes have already been proven.

A research question may be descriptive, exploratory, explanatory or evaluative:

Research objectives should use clear verbs such as examine, explore, compare, assess, identify and evaluate. They should not promise more than the method can deliver. A cross-sectional survey can examine association, but it usually cannot establish that one variable causes another.

5.1 The FINER test

A practical topic-selection test is FINER:

• Feasible: Can the study be completed with available time, access, skills and resources?
• Interesting: Is it intellectually meaningful to the researcher and stakeholders?
• Novel: Does it add a useful context, population, comparison or interpretation?
• Ethical: Can it be conducted without unacceptable harm, coercion or privacy violations?
• Relevant: Does it contribute to HR knowledge, practice, policy or organisational strategy?

A topic can be scored using a simple rating scale from 1 to 5 for each criterion, but the total score should not replace judgement. A highly relevant topic with no realistic data access should be redesigned or rejected.

5.2 Alignment matrix

Before finalising a topic, prepare an alignment matrix:

Misalignment is a common weakness. For example, a project may claim to study retention but collect only data about job satisfaction, with no evidence about intention to remain or actual turnover. The title, questions, objectives, literature review, methods and conclusions must point toward the same problem.

6. Ethics, Governance and Data Protection

HR research involves sensitive information and unequal power relationships. Ethical diagnosis requires informed consent, voluntary participation, confidentiality, secure data handling and honest reporting. Researchers should collect only data necessary for the stated purpose and should avoid unnecessarily identifying small groups.

Particular risks include:

• employees believing participation is compulsory;
• managers requesting access to individual responses;
• disclosure of health, disciplinary or protected-characteristic data;
• re-identification from combinations of job title, location and demographic information;
• research findings being used to punish individuals or groups;
• algorithmic systems reproducing bias while appearing objective.

The researcher should obtain institutional ethical approval where required and comply with applicable data-protection law and organisational policy. Data should have defined retention and deletion arrangements. If an interview reveals serious risk of harm, safeguarding procedures may limit ordinary confidentiality; these limits should be explained before participation.

AI tools can assist with topic brainstorming, transcription, coding support or document organisation, but they do not replace researcher judgement. Sensitive organisational data should not be entered into an unapproved public AI system. AI-generated summaries may omit context, reproduce bias or invent interpretations. Researchers remain responsible for verifying sources, protecting confidentiality and disclosing significant AI assistance according to institutional rules.

7. Applied Case: Diagnosing Early-Career Turnover

This is an illustrative fictional case.

Background

Northbridge Services, a 600-employee business-process organisation, reported that voluntary turnover among employees recruited within the previous two years had increased. Senior management proposed improving salaries, while HR suspected that the onboarding programme and first-line supervision were more important.

Problem

The initial statement—“young employees are leaving because they are dissatisfied”—was inadequate. It used an unverified causal claim and treated age as the central explanation. The researcher reframed the issue around early-career employees’ work experiences and intention to remain.

Analysis

Preliminary records showed that turnover was concentrated in two customer-support teams rather than across the whole organisation. Exit interviews mentioned workload, limited feedback and uncertainty about progression. However, exit interviews represented only employees who had already left and did not show whether current employees held similar views.

The researcher consulted HR, team leaders and current employees. Interviews indicated that onboarding was formally consistent but that supervisors varied considerably in the quality of follow-up support. A review of schedules also found that new employees in one team handled complex cases earlier than planned.

The diagnosis suggested interacting factors: work demands, supervisor support, role clarity and perceived development opportunities. Pay remained relevant but was not the only plausible explanation.

Research approach

A focused study was proposed: “How do role clarity, supervisor support and perceived development opportunities relate to intention to remain among employees with less than two years of service in Northbridge’s customer-support division?”

A mixed evidence strategy was considered. A questionnaire could measure perceptions across the division, while semi-structured interviews could explore how employees experienced onboarding and supervision. The design would not claim to prove causation. It would identify patterns and provide contextual explanations.

Outcome

The diagnosis led management to revise supervisor check-ins, clarify progression information and review the timing of complex-case assignments. The dissertation itself would evaluate employee perceptions and relationships among the selected concepts rather than promise a complete solution to turnover.

Learning points

1. A visible outcome may have several causes.
2. Existing organisational data are useful but rarely sufficient on their own.
3. Stakeholder perspectives can reveal differences between formal policy and actual practice.
4. A focused question is more valuable than a broad claim about “young workers.”
5. Research findings should inform decisions without overstating what the evidence proves.

8. Common Errors and Best Practices

Students commonly select topics because they are fashionable, personally familiar or easy to describe. Popularity does not establish research value. Other errors include choosing a topic before checking access, using a problem statement that contains an unsupported cause, attempting to study an entire industry, confusing employee opinions with organisational performance, and promising recommendations before collecting evidence.

Best practice involves:

• beginning with a genuine workforce concern;
• reviewing relevant academic and professional literature early;
• consulting more than one stakeholder group;
• separating evidence from interpretation;
• defining the population and setting precisely;
• selecting concepts that can be investigated with available data;
• recognising alternative explanations;
• checking ethics and access before committing;
• aligning title, problem, questions, objectives and method;
• stating limitations honestly.

Chapter Summary

Topic selection and HR problem diagnosis form the foundation of a credible dissertation. A workforce concern must be distinguished from a symptom, a practical problem and a research problem. Contemporary HR issues provide important starting points, but they must be connected to a specific organisational context and a feasible evidence strategy.

Stakeholder consultation helps reveal different experiences and prevents premature conclusions. Preliminary diagnosis should examine individual, team and organisational levels and draw on suitable evidence, including records, surveys, interviews, documents and observation. The 5W1H framework, root-cause analysis and the socio-technical perspective help organise diagnosis, while triangulation strengthens interpretation.

A focused research problem identifies the affected population, context, important concepts, evidence gap and organisational significance. The FINER test and an alignment matrix help evaluate feasibility, relevance, ethics and coherence. Because HR research involves sensitive personal and employment data, confidentiality, voluntary participation, data protection and responsible use of technology are essential. A strong dissertation does not merely describe an organisational concern; it converts that concern into a bounded, ethical and researchable investigation aligned with HR strategy and industry needs.

Key Terms

Workforce issue: A condition affecting employees, work or organisational capability.

Symptom: An observable indicator of a deeper or broader problem.

Problem statement: A precise description of an undesirable condition, its context and significance.

Research problem: A practical or knowledge gap that can be investigated systematically.

Research question: A focused question that a study is designed to answer.

Organisational diagnosis: Systematic examination of how and why an organisational condition exists.

Stakeholder: A person or group affected by, knowledgeable about or influential in an issue.

Triangulation: Comparison of evidence from multiple sources or methods.

Root cause: An underlying condition contributing to an observed outcome.

5W1H: A diagnostic framework based on what, who, where, when, why and how.

Socio-technical system: An organisational system involving interaction between social and technical elements.

Strategic alignment: Consistency between a research topic and organisational or HR objectives.

Feasibility: The practical possibility of completing a study with available resources and access.

Novelty: The useful contribution a study makes beyond existing knowledge.

Research scope: The defined boundaries of a study.

Stakeholder mapping: Classification of stakeholders according to interest and influence.

Ethical approval: Formal review confirming that proposed research meets ethical standards.

Data minimisation: Collection of only the personal data necessary for the research purpose.

Workforce analytics: Use of workforce data to support HR understanding and decisions.

Intention to remain: An employee’s stated likelihood of continuing employment with an organisation.

Review Questions

Short Answer Questions

1. Distinguish between a symptom, a practical issue and a research problem.
2. What are the five elements of a strong HR problem statement?
3. Why is stakeholder consultation important during topic selection?
4. What is triangulation, and how can it strengthen organisational diagnosis?
5. State the five criteria in the FINER test.

Long Answer Questions

1. Explain how a broad workforce concern can be converted into a focused research question. Use an HR example.
2. Discuss the contribution and limitations of the 5W1H framework and root-cause analysis in HR diagnosis.
3. Explain how individual, team and organisational factors may interact in producing an HR problem.
4. Discuss the ethical risks involved in researching employees and propose appropriate safeguards.
5. Evaluate the importance of strategic alignment and feasibility in selecting a dissertation topic.

Analytical / Application Questions

1. An organisation reports rising absence among warehouse employees. Design a preliminary diagnostic plan using at least three evidence sources and identify possible alternative explanations.
2. A manager claims that remote employees are less committed because performance has declined. Reframe this claim as a neutral, researchable HR problem and propose suitable research questions.
3. Use the FINER test and an alignment matrix to evaluate a proposed study on AI bias in recruitment for a medium-sized organisation with limited access to recruitment records.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes a research problem from a symptom?

A. A research problem is always expressed as a numerical measure.

B. A research problem identifies a knowledge or decision gap that can be investigated.

C. A research problem is simply the most visible organisational outcome.

D. A research problem is a personal opinion about management.

Correct Answer: B. A research problem identifies a knowledge or decision gap that can be investigated.

Question 2

Which activity is an example of triangulation?

A. Using only one interview with a senior manager

B. Repeating the same survey question three times

C. Comparing HR records, employee interviews and survey findings

D. Selecting the largest available sample

Correct Answer: C. Comparing HR records, employee interviews and survey findings

Question 3

In the 5W1H framework, the question “Who is affected?” primarily identifies the:

A. Population

B. Technology

C. Research instrument

D. Statistical test

Correct Answer: A. Population

Question 4

Which topic is most appropriately focused for an undergraduate dissertation?

A. The future of work

B. All causes of global employee turnover

C. Human resource management in the twenty-first century

D. The relationship between supervisor support and retention intentions among new employees in one department

Correct Answer: D. The relationship between supervisor support and retention intentions among new employees in one department

Question 5

Which is an ethical concern specific to HR research?

A. Using headings in a report

B. Employees believing participation is compulsory

C. Selecting a narrow research question

D. Comparing two evidence sources

Correct Answer: B. Employees believing participation is compulsory

Question 6

The socio-technical perspective emphasises the interaction between:

A. Finance and marketing

B. Employees and customers only

C. Social arrangements and technical systems

D. Research questions and references

Correct Answer: C. Social arrangements and technical systems

Question 7

Which statement about a cross-sectional study is most appropriate?

A. It always proves causation.

B. It can examine relationships but usually cannot establish temporal causality by itself.

C. It cannot collect employee perceptions.

D. It eliminates response bias.

Correct Answer: B. It can examine relationships but usually cannot establish temporal causality by itself.

Question 8

Which criterion is included in the FINER test?

A. Feasibility

B. Popularity

C. Profitability only

D. Automation

Correct Answer: A. Feasibility

Question 9

Why should a researcher avoid entering sensitive employee data into an unapproved public AI tool?

A. AI tools cannot produce text.

B. The data may be exposed or processed without adequate confidentiality controls.

C. AI tools are never useful in research.

D. Organisations cannot use digital systems.

Correct Answer: B. The data may be exposed or processed without adequate confidentiality controls.

Question 10

A high resignation rate is best treated initially as:

A. Proof that pay is the cause

B. A symptom requiring further diagnosis

C. Evidence that interviews are unnecessary

D. A complete research conclusion

Correct Answer: B. A symptom requiring further diagnosis