Research Questions, Objectives, Hypotheses and Proposal Development

Introduction

A dissertation proposal is a reasoned plan for producing knowledge, not merely an administrative form. It explains what will be studied, why the issue matters, how evidence will be collected and analysed, and what contribution the study is expected to make. In human resource management (HRM), a well-designed proposal connects an organisational problem—such as turnover, skills shortages, employee wellbeing or performance management—with concepts, evidence and an appropriate research design.

The quality of a proposal depends especially on alignment. The research problem should lead to research questions; questions should be translated into objectives; objectives should determine the methods; and the methods should generate evidence capable of answering the questions. Hypotheses are appropriate when the study tests predicted relationships, but they are not required for every dissertation. A qualitative study may instead use exploratory questions and sensitising concepts.

Learning Objectives

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

1. Distinguish a research problem, purpose, question, objective and hypothesis.
2. Formulate focused, feasible and ethically appropriate HRM research questions.
3. Write objectives that are observable and aligned with the questions.
4. Develop testable hypotheses and identify variables and expected relationships.
5. Establish the scope, significance and anticipated contribution of a study.
6. Prepare a defensible proposal containing rationale, literature, methodology, timetable, resources and limitations.
7. Critically evaluate alignment, feasibility, ethics and methodological quality.

From Organisational Issue to Research Problem

An organisational issue is a situation that attracts attention. A research problem is a clearly evidenced gap, uncertainty, contradiction or practical difficulty that can be investigated systematically. For example, “employee turnover is high” is an issue. A research problem might be: “Although the organisation has introduced flexible working, voluntary turnover among early-career employees remains high, and the relative influence of perceived career opportunities and supervisor support is unclear.”

A useful problem statement normally establishes four elements:

• Context: the organisation, sector, population or setting.
• Evidence of the problem: relevant reports, statistics, prior studies or stakeholder accounts.
• Knowledge gap: what is not adequately known or explained.
• Research response: what the proposed study will investigate.

A problem should not be manufactured merely to justify research. It should be supported by credible sources and be sufficiently bounded for a bachelor’s dissertation. Broad topics such as “motivation in organisations” require narrowing by population, location, time period, variables and purpose.

Purpose and Significance

The purpose statement gives the study’s central direction. It commonly begins: “This study aims to examine…” or “This study seeks to explore…”. Significance explains why the study is worth conducting. Contributions may be:

• Theoretical: clarifying, extending or challenging an established explanation.
• Empirical: providing evidence about a population or setting under-researched in the literature.
• Methodological: demonstrating the usefulness of a design or measurement approach.
• Practical: informing HR policy, managerial decisions or employee support.
• Societal: addressing fairness, inclusion, decent work or employee wellbeing.

Claims must be proportionate. A small cross-sectional survey cannot normally establish universal causal laws, but it may provide useful evidence about associations in a defined organisation.

Research Questions

A research question is a precise, answerable question that directs evidence collection and interpretation. Strong questions are significant, focused, researchable, ethically acceptable and feasible within available time and resources.

Common forms include:

A question should identify, where relevant, the population, setting, phenomenon and time frame. “How does leadership affect performance?” is too broad. “What is the relationship between perceived transformational leadership and self-reported task performance among full-time employees in three medium-sized technology firms?” is more focused, although the researcher must define and measure both constructs.

Question Quality and Scope

Questions can be narrowed using the SPICE logic: Setting, Perspective, Intervention or phenomenon, Comparison, and Evaluation. Not every study requires all five elements. A qualitative question might ask: “How do newly promoted supervisors in public hospitals experience training during their first year?”

Avoid questions that:

• contain an assumed conclusion, such as “Why is the unfair appraisal system harmful?”;
• ask several unrelated questions at once;
• require unrestricted access to confidential information;
• use vague terms such as “good,” “effective” or “impact” without definition;
• imply causality when the design can only show association.

A main question can be supported by two to four subquestions. For example:

Main question: How does hybrid working influence perceived work–life balance among administrative employees?

Subquestions:

1. How do employees describe the benefits and difficulties of hybrid working?
2. Is the number of home-working days associated with perceived work–life balance?
3. What role do supervisor expectations and workload play in this relationship?

Research Objectives

Objectives state what the researcher will do to answer the questions. They should be specific, observable and realistic. The main objective expresses the overall purpose; specific objectives divide it into manageable actions.

Useful action verbs include identify, describe, compare, examine, analyse, explore, assess, evaluate, interpret and determine. Avoid verbs that are difficult to demonstrate, such as understand or learn, unless accompanied by an observable analytical action.

A well-written objective includes an action, object, population or context and, when appropriate, a method or condition. For example: “To examine the association between supervisor support and turnover intention among customer-service employees using an anonymous questionnaire.”

Questions and objectives must correspond. The following alignment is appropriate:

The verb determine should be used cautiously. Determining a causal effect generally requires a design capable of supporting causal inference, such as a strong longitudinal or experimental design. A single-time-point survey usually supports terms such as examine, estimate or assess association.

Hypotheses and Conceptual Models

A hypothesis is a logically derived, empirically testable statement predicting a relationship or difference. It is not a guess; it follows from theory, prior findings or a justified conceptual argument.

A typical directional hypothesis has the form:

Employees who perceive higher supervisor support will report lower turnover intention than employees who perceive lower supervisor support.

The variables are:

• Independent or predictor variable: perceived supervisor support.
• Dependent or outcome variable: turnover intention.
• Direction: negative.
• Population: employees in the specified setting.

A null hypothesis states that no relationship or difference exists in the population, whereas an alternative hypothesis states that a relationship or difference exists. In statistical testing, researchers commonly evaluate evidence against the null hypothesis. Failure to reject a null hypothesis does not prove that no relationship exists; it indicates that the evidence is insufficient under the specified assumptions.

Examples include:

• Directional: H1: Perceived supervisor support is negatively associated with turnover intention.
• Non-directional: H1: Perceived supervisor support is associated with turnover intention.
• Difference: H1: Employees receiving structured induction report higher organisational commitment than employees receiving unstructured induction.
• Mediation: H1: The relationship between supervisor support and turnover intention is partly mediated by affective organisational commitment.

Mediation and moderation require careful conceptual and statistical justification. A mediator explains a possible process; a moderator changes the strength or direction of a relationship. For example, supervisor support may be related to wellbeing through reduced work–family conflict, while the relationship may differ according to employment status.

A conceptual model makes proposed relationships explicit. It may show supervisor support leading to commitment, which relates to turnover intention, with workload as a possible moderator. The model is not proof; it is a structured statement of what the study will examine.

Quantitative Notation and Caution

For a simple linear relationship, a researcher may represent the statistical model as:

Here, y_i is the outcome for person i, x_i is the predictor, β₀ is the intercept, β₁ is the expected change in the outcome for a one-unit change in the predictor, and ε_i is the unexplained error. This model is applicable when a linear approximation is defensible and assumptions concerning independence, measurement, residual behaviour and model specification are considered. A statistically estimated association does not by itself establish causation.

Researchers should specify how constructs will be operationalised. “Job satisfaction” might be measured with a validated multi-item scale, while “turnover intention” might be measured through agreement with intention-to-leave items. Reliability, validity, translation, response bias and common-method bias must be considered.

Proposal Architecture

A proposal is a persuasive argument whose sections perform distinct functions. Institutions vary in format, but a defensible proposal commonly includes the following.

Title

The title should be concise and informative. It may identify the key phenomenon, population and setting: Supervisor Support and Turnover Intention among Early-Career Employees in Regional Retail Organisations. Avoid titles that promise more than the design can deliver, such as The Definitive Causes of Employee Turnover.

Rationale and Background

The rationale moves from the broad HRM context to the specific problem. It should show why the topic matters now, identify the affected stakeholders and establish the evidence gap. Assertions about prevalence, discrimination, wellbeing or organisational performance require credible sources rather than anecdote.

Preliminary Literature Review

The literature review is an analytical synthesis, not a sequence of summaries. It should:

1. define central constructs;
2. explain relevant theories, such as social exchange theory, job demands–resources theory or psychological contract theory;
3. compare findings and methods across studies;
4. identify inconsistencies, limitations or contextual gaps;
5. justify the proposed conceptual framework and research questions.

A useful paragraph often follows this pattern: claim, evidence from several sources, comparison or qualification, and implication for the proposed study. Literature should be current but not exclusively recent; foundational theory remains important. Sources should be evaluated for authority, methodological quality, relevance and possible bias.

Methodology

The methodology explains how the research questions will be answered and why the choices are defensible.

Research philosophy and approach. Positivist approaches commonly emphasise measurement, hypothesis testing and systematic comparison. Interpretivist approaches emphasise meaning, context and participants’ lived experiences. Pragmatism permits methods to be selected according to the research problem. Deductive studies typically move from theory to hypotheses, while inductive studies develop concepts from evidence. These are tendencies, not rigid categories.

Design. A cross-sectional survey captures data at one point; it is efficient but weak for temporal and causal claims. A longitudinal design observes change over time. A case study examines a bounded organisation or process in depth. Interviews and focus groups can explore meaning, while documentary analysis can examine policies and records. Mixed methods may combine breadth and depth, but it requires additional planning and genuine integration rather than two disconnected studies.

Population and sampling. The population is the broader group to which the study relates; the sample is the group actually studied. Probability sampling supports statistical generalisation when a suitable sampling frame exists. Convenience, purposive and snowball sampling may be appropriate for access or qualitative depth, but their limitations must be acknowledged. Inclusion and exclusion criteria should be explicit.

Data collection. Instruments should be piloted for clarity, timing and technical problems. Interview protocols should use open, neutral prompts. Surveys should avoid double-barrelled questions, leading language and unnecessary personal data. Existing organisational records require permission, data-quality checks and attention to missing or inconsistent entries.

Analysis. Quantitative analysis may include frequencies, measures of central tendency, cross-tabulations, correlation or regression, selected according to the question and measurement level. Qualitative analysis may involve familiarisation, coding, category development, theme review and interpretation. The proposal should state how findings will be made credible—for example, through an audit trail, reflexive memoing, peer discussion, triangulation or participant clarification where appropriate.

Ethics and Governance

HRM research can involve power relationships and sensitive employment information. Participants should receive clear information about purpose, procedures, risks, benefits, confidentiality, voluntary participation and withdrawal. Consent must be informed and freely given; a manager should not pressure subordinates to participate. Researchers should avoid promising absolute anonymity when small samples or identifiable quotations make identification possible.

Governance includes institutional ethics approval, organisational permission, secure storage, access control, retention and deletion procedures. Data protection principles generally require purpose limitation, data minimisation, accuracy, security and lawful processing. Researchers must follow applicable national law and institutional policy. Findings should not be used to disadvantage individual employees, and conflicts of interest must be disclosed.

Timetable and Resources

A timetable converts intention into deliverables. A simple dissertation schedule might include:

Resources may include database access, recording or survey software, transcription support, travel, secure storage and statistical or qualitative-analysis software. A Gantt chart is useful, but the proposal should also identify dependencies: data collection cannot begin before approval and access.

Anticipated Limitations

Limitations are conditions that may constrain interpretation. They may include a small or non-probability sample, self-report bias, non-response, cross-sectional design, restricted organisational access, limited measures or researcher positionality. A strong proposal does not conceal limitations; it explains their likely effect and mitigation. For example, anonymity may reduce social desirability bias, while triangulation may strengthen interpretation. Delimitations—choices such as focusing on one sector or excluding contractors—should be distinguished from limitations.

Worked Proposal Example

Consider an illustrative undergraduate project in a logistics company. Exit records show increasing voluntary turnover among employees with less than two years’ service. Managers believe pay is the main cause, but informal employee feedback points to unpredictable schedules and weak supervisor communication.

Problem: The organisation lacks evidence about which work experiences are associated with early-career employees’ intention to leave.

Purpose: To examine how schedule predictability and supervisor communication relate to turnover intention among early-career warehouse employees.

Questions:

1. What levels of schedule predictability, supervisor communication and turnover intention are reported?
2. What is the relationship between schedule predictability and turnover intention?
3. How do employees describe the ways scheduling and communication influence their decisions?

Objectives:

1. To describe reported schedule predictability, supervisor communication and turnover intention.
2. To examine associations between the two predictors and turnover intention.
3. To explore employees’ explanations of these relationships.
4. To develop evidence-based recommendations for early-career retention.

Hypotheses for the quantitative component:

• H1: Greater schedule predictability is associated with lower turnover intention.
• H2: More effective supervisor communication is associated with lower turnover intention.

A sequential mixed-methods design could begin with an anonymous survey and follow with purposively selected interviews. The survey would not establish that scheduling causes turnover; interviews could help interpret plausible mechanisms. The researcher should avoid giving managers individual responses, report results in aggregate, and explain that participation will not affect employment.

The anticipated contribution is practical and contextual rather than universal: the organisation may learn which interventions deserve testing, while the dissertation may add evidence from a population underrepresented in local HRM research. Recommendations should follow the evidence—for example, clearer roster-notice standards or supervisor communication training—and should not claim that one study has solved retention.

Critical Evaluation and Contemporary Practice

Digital HR systems and artificial intelligence can assist literature discovery, transcription, coding or survey administration, but they do not replace scholarly judgement. Researchers should verify AI-generated references, protect confidential data, record substantial AI assistance where required, and examine outputs for bias and fabricated claims. Automated employee data may reproduce historical discrimination or infer sensitive characteristics without valid consent. Human oversight, explainability and a clear purpose are essential.

Contemporary HRM research increasingly addresses hybrid work, algorithmic management, employee monitoring, sustainability, inclusion, mental health and the quality of work. These topics require attention to unequal effects: an intervention beneficial to professional employees may disadvantage workers who cannot work remotely. Researchers should consider intersectionality, accessibility and whose voices are absent from the sample.

Best practice can be summarised as follows:

• begin with a real and evidenced problem;
• keep questions answerable within the degree and timetable;
• align every objective with a question and every method with an objective;
• use theory to explain, not decorate, the proposal;
• distinguish association from causation;
• pilot instruments and plan for missing data or non-response;
• obtain ethics and access approval before collecting data;
• protect participants from employment-related harm;
• state limitations honestly and moderate claims;
• design outputs for both academic and practitioner audiences.

Chapter Summary

A dissertation proposal establishes a logical and ethical plan for research. It begins by converting an organisational issue into an evidenced research problem and a justified purpose. Research questions should be focused, answerable and appropriate to the chosen design. Objectives translate questions into observable research actions. Hypotheses are theory-informed, testable predictions and should identify variables, direction and population where relevant. Conceptual models clarify expected relationships but do not prove them.

A defensible proposal integrates rationale, analytical literature review, philosophy, design, sampling, data collection, analysis, ethics, timetable, resources and limitations. Alignment is the central quality criterion: the problem, questions, objectives, hypotheses, methods and claims must fit together. In HRM, methodological quality must be accompanied by sensitivity to power, privacy, fairness and the possible organisational consequences of research.

Key Terms

Research problem: An evidenced issue, gap or uncertainty suitable for systematic investigation.

Purpose statement: A concise declaration of the study’s overall aim.

Research question: A focused question that directs investigation and evidence collection.

Subquestion: A narrower question supporting the main research question.

Research objective: A specific action the researcher will undertake.

Hypothesis: A testable prediction about a relationship or difference.

Null hypothesis: A statement that a specified relationship or difference is absent in the population.

Independent variable: A predictor or explanatory variable in a proposed relationship.

Dependent variable: The outcome variable being explained or predicted.

Operationalisation: Defining how an abstract construct will be observed or measured.

Conceptual framework: An organised representation of relevant concepts and proposed relationships.

Mediator: A variable that may explain the process linking a predictor to an outcome.

Moderator: A variable that may alter the strength or direction of a relationship.

Population: The broader group to which a study relates.

Sample: The units actually included in the study.

Cross-sectional design: A design collecting data at one point in time.

Longitudinal design: A design examining observations across time.

Triangulation: Using multiple sources, methods or perspectives to strengthen interpretation.

Delimitation: A boundary deliberately set by the researcher.

Research ethics: Principles and procedures protecting participants, data integrity and public trust.

Review Questions

Short Answer Questions

1. What distinguishes an organisational issue from a research problem?
2. State two characteristics of a strong research question.
3. How does an objective differ from a research question?
4. Distinguish a mediator from a moderator.
5. Why should a cross-sectional survey be cautious about causal claims?

Long Answer Questions

1. Explain how a research problem, purpose statement, questions and objectives should be aligned in an HRM dissertation.
2. Compare descriptive, exploratory, explanatory and evaluative research questions, giving an HRM example of each.
3. Explain how theory and prior research contribute to the development of hypotheses.
4. Discuss the major methodological components that should appear in a dissertation proposal.
5. Analyse the ethical issues that may arise when researching employees within their own organisation.

Analytical / Application Questions

1. An organisation reports rising absence but has no evidence about its causes. Develop a problem statement, one main research question, three objectives and one feasible design.
2. A student proposes to test whether flexible working “causes” higher wellbeing using one online survey. Critique the proposal and recommend more defensible wording and design choices.
3. Design a short proposal outline investigating whether supervisor support is related to early-career employee retention. Include scope, variables, sampling, ethics, timetable and two anticipated limitations.

Multiple Choice Questions with Answers

Question 1

Which statement best describes a research objective?

A. A broad organisational concern

B. A specific action the researcher will undertake

C. A prediction about a population parameter

D. A list of sources used in the literature review

Correct Answer: B. A specific action the researcher will undertake

Question 2

Which question is most appropriately focused for a bachelor’s HRM dissertation?

A. Why do all employees leave organisations?

B. What is the meaning of work in society?

C. How do newly promoted supervisors in public hospitals experience first-year training?

D. Is management good or bad?

Correct Answer: C. How do newly promoted supervisors in public hospitals experience first-year training?

Question 3

What does a directional hypothesis specify in addition to a relationship?

A. The research budget

B. The expected direction of the relationship

C. The name of the ethics committee

D. The number of literature sources

Correct Answer: B. The expected direction of the relationship

Question 4

Which design is generally weakest for establishing temporal order and causation?

A. A single-time-point cross-sectional survey

B. A longitudinal panel study

C. A randomised field experiment

D. A repeated-measures study

Correct Answer: A. A single-time-point cross-sectional survey

Question 5

In the statement ‘supervisor support is associated with turnover intention through organisational commitment,’ organisational commitment is proposed as a:

A. Moderator

B. Sampling frame

C. Mediator

D. Delimitation

Correct Answer: C. Mediator

Question 6

Which is an appropriate ethical practice in employee research?

A. Allowing managers to see individual survey answers

B. Making participation a condition of promotion

C. Obtaining informed and voluntary consent

D. Collecting every available personal detail

Correct Answer: C. Obtaining informed and voluntary consent

Question 7

What is the main purpose of a conceptual framework?

A. To replace data collection

B. To represent relevant concepts and proposed relationships

C. To guarantee statistically significant findings

D. To provide the dissertation timetable only

Correct Answer: B. To represent relevant concepts and proposed relationships

Question 8

Which sampling approach is especially suitable when participants are selected for relevant experience in a qualitative study?

A. Purposive sampling

B. Census of every person worldwide

C. Random assignment

D. Systematic deletion

Correct Answer: A. Purposive sampling

Question 9

Which claim is most defensible from a small, single-organisation cross-sectional survey?

A. The survey proves the universal cause of turnover

B. The findings may indicate associations in the defined setting

C. The findings automatically generalise to every industry

D. No limitations need to be reported

Correct Answer: B. The findings may indicate associations in the defined setting

Question 10

What should a proposal timetable show in addition to activities?

A. Only the final submission date

B. Dependencies and indicative time periods

C. The preferred statistical result

D. Participants’ individual responses

Correct Answer: B. Dependencies and indicative time periods