Proposal Development and Project Management

Introduction

An accounting and finance dissertation is not merely an extended essay. It is a managed research project that must demonstrate a worthwhile problem, a defensible method, credible evidence, ethical conduct, and delivery within agreed constraints. A proposal is the document that converts an initial interest into an executable plan. It allows a student, supervisor, and assessment panel to judge whether the proposed study is significant, feasible, methodologically appropriate, and capable of producing a contribution.

Proposal development and project management are therefore closely connected. Proposal development addresses what will be studied and how; project management addresses how the work will be organised, monitored, controlled, and completed. A strong proposal makes assumptions visible, identifies risks early, allocates resources realistically, and establishes milestones that support timely completion.

In accounting and finance, projects may involve archival financial statements, market data, surveys of accountants, interviews with managers, case studies, experiments, or mixed methods. Each design creates different requirements for access, data quality, ethics, software, analytical skill, and time. A proposal should demonstrate awareness of these requirements rather than treating methodology as a generic paragraph.

Learning Objectives

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

1. Explain the purpose and structure of an accounting or finance research proposal.
2. Convert a broad topic into a focused research problem, question, and objectives.
3. Select and justify an appropriate research philosophy, design, method, and analysis plan.
4. Prepare a feasible schedule, resource plan, risk register, and supervision arrangement.
5. Apply project-management tools to monitor scope, time, quality, and deliverables.
6. Address ethical, governance, data-protection, and research-integrity requirements.
7. Evaluate the strengths and limitations of a proposed project and defend its feasibility.

1. The Proposal as a Decision and Control Document

A proposal serves at least four purposes. First, it communicates the intellectual rationale for the study. Secondly, it provides a methodological argument: the proposed evidence and analysis must be capable of answering the research question. Thirdly, it functions as a feasibility test. A topic may be interesting but impossible if data cannot be obtained, the sample is too small, or the project exceeds the available time. Finally, the proposal becomes a control document against which progress can be assessed.

A proposal is not a promise that every detail will remain unchanged. Research often requires justified adjustments after preliminary reading or data inspection. However, changes should be documented and agreed with the supervisor. Uncontrolled changes create scope creep, weaken coherence, and make it difficult to determine whether the project has achieved its objectives.

A defensible proposal normally contains:

• a working title;
• background and rationale;
• a clearly defined research problem;
• research aim, questions, and objectives;
• a focused literature and theoretical framework;
• methodology and methods;
• sampling or data-selection procedures;
• analysis plan;
• ethical and governance considerations;
• resources and feasibility assessment;
• milestones and deliverables;
• risks and mitigation actions;
• supervision and communication arrangements; and
• references and, where appropriate, appendices.

The components must form a chain of logic. The problem should lead to the questions; the questions should determine the data; the data should determine the analysis; and the analysis should support the proposed conclusions. A proposal is weak when these links are asserted rather than demonstrated.

2. From Topic to Researchable Problem

A topic is a broad area such as sustainability reporting, audit quality, budgeting, cryptocurrency taxation, or working-capital management. A research problem identifies a specific uncertainty, inconsistency, practical difficulty, or unexplored relationship within that area. The problem should be relevant to accounting or finance and sufficiently narrow for the available time and evidence.

A useful problem statement answers four questions:

1. What is known?
2. What remains uncertain or problematic?
3. Why does the uncertainty matter?
4. How will the proposed project address it?

For example, “sustainability reporting is important” is a background claim, not a research problem. A stronger formulation might be: “Although listed firms increasingly publish climate-related information, it is uncertain whether the extent of disclosure is associated with cost-of-debt differences among mid-cap firms in a specified market and period.” The second statement identifies a population, variables, relationship, and boundary.

Aim, questions, and objectives

The aim is a concise statement of the overall purpose. Research questions express the precise issues to be answered. Objectives state the actions required to answer those questions. Objectives should begin with observable verbs such as identify, examine, compare, evaluate, estimate, explore, or assess.

A coherent example is:

• Aim: To examine whether climate-disclosure quality is associated with the cost of debt of mid-cap listed firms.
• Question: What relationship exists between climate-disclosure quality and cost of debt after considering firm size, leverage, and profitability?
• Objectives: develop a disclosure measure; collect firm and debt information; test the association using an appropriate model; interpret the results in relation to theory and practice.

Objectives should be neither a list of administrative tasks nor a set of conclusions in disguise. “To prove that disclosure reduces borrowing costs” is inappropriate because it presupposes the result. “To collect data” is necessary but not, by itself, an intellectual objective.

Scope and delimitations

Scope defines what the study covers: population, geography, period, variables, sector, and unit of analysis. Delimitations are deliberate boundaries set by the researcher, such as excluding private firms because comparable public data are unavailable. A limitation is a constraint that may affect the study, such as non-response or measurement error. The distinction matters: delimitations are design choices, whereas limitations are weaknesses or uncertainties that must be acknowledged.

3. Building the Intellectual Framework

Literature review

A literature review is an analytical synthesis, not a sequence of article summaries. It should classify previous research, compare findings, assess methods, identify theoretical explanations, and establish the precise gap addressed by the project.

An effective review often moves through four levels:

1. Conceptual level: definitions and boundaries of key constructs.
2. Theoretical level: explanations of why variables or behaviours may be related.
3. Empirical level: evidence from prior studies and its consistency.
4. Methodological level: strengths and weaknesses of previous designs, samples, measures, and analyses.

Sources should be selected for relevance and authority. Peer-reviewed research, professional standards, regulatory publications, official statistics, company reports, and reputable institutional evidence may all be appropriate, but their purposes differ. A corporate report can provide a disclosure example; it is not automatically independent evidence of a causal relationship.

Theory and conceptual models

Theory provides a reasoned explanation for expected patterns. For example, agency theory may explain why managers provide information to reduce information asymmetry; signalling theory may explain voluntary disclosure intended to communicate superior performance; institutional theory may explain convergence in reporting practices due to regulation, professional norms, or imitation.

A conceptual model translates theory into constructs and proposed relationships. In a quantitative study, the model may identify an outcome variable, explanatory variable, and control variables. In a qualitative study, it may sensitise the researcher to themes, actors, processes, and institutional conditions without imposing rigid categories.

A conceptual model must not be confused with proof. A proposed relationship is an expectation to be investigated, not a finding. Definitions and measurement decisions should be explicit. For instance, “audit quality” might be represented by auditor size, restatement frequency, discretionary accrual measures, or a composite index. Each operationalisation captures only part of the construct.

4. Methodology and Method Selection

Methodology is the reasoning that connects assumptions about knowledge to the research design. Methods are the practical techniques used to collect and analyse evidence.

Research philosophies

A positivist or post-positivist orientation commonly emphasises observable evidence, structured measurement, and testing relationships. It is often compatible with archival data, surveys, and statistical analysis, while recognising that measures are imperfect.

An interpretivist orientation emphasises how participants construct meaning and how accounting practices operate within social contexts. Interviews, observations, and document interpretation are common methods.

A pragmatic orientation selects methods according to the research problem and may support mixed methods. It does not mean that methodological standards can be ignored; rather, it requires the researcher to explain why each method is useful for the question.

Common designs in accounting and finance

The correct method is not the most sophisticated method. It is the method that answers the question with acceptable validity, reliability, credibility, and feasibility.

Sampling and selection

A population is the complete set of relevant units, such as all firms listed on a particular exchange during a defined period. A sample is the subset studied. Probability sampling gives units a known chance of selection and supports statistical generalisation under suitable assumptions. Non-probability sampling, including purposive or convenience sampling, may be appropriate for qualitative or exploratory work but requires careful claims about transferability.

For archival studies, inclusion and exclusion rules should be stated before data collection where possible. For example, the researcher may include non-financial firms continuously listed between 2019 and 2023 and exclude firms lacking annual reports or required variables. These rules reduce selection ambiguity and improve reproducibility.

Quantitative analysis planning

The analysis plan should identify variables, transformations, tests, and assumptions before results are known. A simple linear regression may be written as:

Here, y_i is the outcome for observation i, x_i is the principal explanatory variable, z_i is a control variable or vector of controls, β₀ is the intercept, β₁ and β₂ are coefficients, and ε_i is the unexplained error. The model assumes, among other conditions, an appropriate functional form, independent or suitably modelled errors, and no perfect multicollinearity. Regression association does not by itself establish causation. A proposal should explain how confounding, reverse causality, missing data, outliers, and clustered or repeated observations will be considered.

For descriptive work, the arithmetic mean is:

where x_i is an observation and n is the number of observations. The mean is expressed in the same units as x_i and can be distorted by extreme values; the median may therefore be reported as a robustness measure. Statistical tests should be selected because of the research question and data properties, not simply because they are available in software.

Qualitative analysis planning

A qualitative proposal should specify recruitment, interview or document procedures, recording and transcription, coding strategy, and credibility practices. Thematic analysis may involve familiarisation, initial coding, development of candidate themes, review, definition, and reporting. The process is iterative, but an audit trail should show how interpretations were reached. Credibility may be strengthened through triangulation, participant clarification where appropriate, peer discussion, reflexive notes, and attention to disconfirming evidence.

5. Project Management Architecture

Project management controls five related dimensions: scope, time, resources, quality, and risk. These dimensions are interdependent. Adding a new research question expands scope and may increase data, analysis, and time requirements. Reducing time may require fewer cases, simpler analysis, or additional resources.

Work breakdown structure

A work breakdown structure (WBS) decomposes the project into manageable work packages. A typical dissertation WBS is:

1. approve topic and proposal;
2. conduct structured literature review;
3. refine framework and instruments;
4. obtain ethics and access approvals;
5. pilot the instrument or data process;
6. collect and document data;
7. clean, store, and analyse evidence;
8. draft chapters;
9. undertake supervisor review and revisions;
10. edit, reference, quality-check, and submit.

Each work package should have an owner, output, start and finish dates, dependencies, and acceptance criteria. “Complete data collection” is less useful than “obtain the planned dataset, record exclusions, and produce a data dictionary.”

Milestones and critical path

A milestone is a significant checkpoint, such as ethics approval or completion of a clean dataset. A dependency means that one task cannot reasonably begin or finish until another task has progressed. The critical path is the sequence of dependent activities that determines the earliest possible completion date. Delays on this path require immediate attention.

A simple schedule might include the following:

The timings are illustrative, not universal. A project involving external participants should allow for recruitment delays; an archival study should allow for data cleaning and missing records.

Resources and feasibility

Resources include time, databases, software, recording equipment, transcription support, travel, participant incentives where permitted, secure storage, and researcher skills. A resource plan should distinguish resources that are already available from those requiring approval or purchase.

A feasibility assessment can be organised around five questions:

• Access: Can the researcher obtain the participants, records, or database?
• Capability: Does the researcher possess or have time to acquire required technical skills?
• Time: Can the work be completed within the academic calendar?
• Cost: Are direct and indirect costs affordable?
• Governance: Can the project meet ethics, confidentiality, licensing, and data-protection requirements?

A small, well-executed project is academically stronger than an ambitious project that cannot be completed.

6. Risk, Quality, and Change Control

Risk is an uncertain event that may affect objectives. A risk register should record the event, cause, consequence, probability, impact, owner, early-warning indicator, and response. A simple priority score can be calculated as:

where R is the priority score, P is a rating of probability, and I is a rating of impact. If both ratings use a 1–5 ordinal scale, the score ranges from 1 to 25. This is a prioritisation aid, not a probability model; ordinal scores should not be interpreted as precise statistical probabilities.

Quality assurance should be planned rather than added at the end. Examples include duplicate data checks, a data dictionary, version control, citation verification, transcription checks, code review, documented exclusion rules, and a final consistency check between objectives, methods, results, and conclusions.

Change control requires the researcher to record the proposed change, reason, effect on objectives, effect on time and resources, ethical implications, decision, and date. Minor editorial changes need not create bureaucracy; substantive changes to sample, method, research question, or data handling should be discussed with the supervisor.

7. Supervision and Governance

Supervision is a structured professional relationship, not a substitute for independent research. At the outset, the student and supervisor should agree meeting frequency, preparation requirements, feedback turnaround expectations, communication channels, and escalation procedures. The student should circulate a short agenda and action log. Meetings should result in specific decisions, owners, and deadlines.

A supervision plan may assign different forms of support: the academic supervisor may guide conceptual and methodological coherence; a librarian may support systematic searching and referencing; a data specialist may advise on software or database use; and an ethics officer may clarify participant or data requirements. Responsibility for the final work remains with the student.

Ethical and governance issues include informed consent, voluntary participation, right to withdraw, confidentiality, secure storage, minimisation of personal data, lawful and transparent processing, avoidance of deception unless explicitly justified and approved, conflicts of interest, and responsible reporting. Public availability of information does not automatically remove ethical duties. Combining public data can create new privacy risks, and commercially licensed databases may restrict redistribution.

Research integrity requires accurate representation of sources and findings. Fabrication, falsification, plagiarism, undisclosed manipulation, selective reporting, and inappropriate authorship are unacceptable. Artificial-intelligence tools may assist with brainstorming, language editing, coding support, or organisation where institutional rules permit, but generated material must be checked, disclosed when required, and never treated as an authoritative source. Sensitive participant data should not be entered into unapproved systems. The student remains accountable for accuracy, confidentiality, citations, and intellectual decisions.

8. Writing a Persuasive Proposal

A proposal should be written for a critical reader who asks: Is the problem important? Is the question clear? Is the method appropriate? Can the study actually be completed? Are the claims proportionate to the evidence?

A useful writing sequence is:

1. draft the problem and provisional question;
2. search and organise the literature;
3. define constructs and boundaries;
4. select the design and justify alternatives considered;
5. specify data, sampling, analysis, and ethics;
6. build the schedule and risk register;
7. check alignment and revise.

Alignment can be tested in a matrix:

The proposal should distinguish evidence from assertion. “The method will establish causality” is usually too strong for a basic observational dissertation. “The method will estimate an adjusted association and discuss threats to causal interpretation” is more defensible.

9. Illustrative Case: A Proposal on Budgetary Participation

This fictional case illustrates proposal development rather than reporting actual research.

Background and problem

A medium-sized manufacturing company has introduced participative budgeting, but managers disagree about whether participation improves budget usefulness or merely increases negotiation time. The organisation wants evidence before extending the practice.

Proposed aim and questions

The aim is to evaluate how budgetary participation affects perceived budget usefulness among departmental managers. Questions ask how participation is experienced, whether perceived participation is associated with budget usefulness, and what organisational conditions influence the relationship.

Approach

A pragmatic mixed-method design is proposed. First, an anonymous survey measures perceived participation, budget usefulness, role ambiguity, and departmental characteristics using clearly described scales. Secondly, purposively selected managers are interviewed to explain survey patterns. The survey provides breadth; interviews provide context and help identify mechanisms that a numerical association cannot reveal.

The researcher must define the population, obtain organisational permission, explain voluntary participation, and prevent supervisors from seeing individual responses. A pilot tests wording and completion time. Quantitative analysis will begin with descriptive statistics and reliability checks, followed by an appropriately specified association model. Interview transcripts will be coded thematically, with a record of coding decisions and attention to negative cases.

Project plan and risks

The critical early dependency is organisational access. If permission is delayed, the researcher should not wait passively; an approved alternative case or a documentary study should be discussed. Other risks include low survey response, perceived pressure to participate, and common-method bias. Mitigation includes independent distribution, confidential collection, neutral wording, and triangulation through interviews and budget documents.

Expected deliverables

Deliverables include an approved proposal, ethics documentation, pilot report, anonymised dataset, analysis record, interview coding framework, dissertation chapters, and a concise management briefing. The outcome is not guaranteed to support participation. A credible project reports positive, negative, mixed, or inconclusive findings transparently.

Learning points

The case demonstrates that a strong proposal integrates organisational relevance with academic discipline. It limits the claim, combines methods for a stated reason, treats access as a project risk, and defines outputs beyond the final dissertation.

10. Common Weaknesses and Best Practices

Frequent weaknesses include an overly broad title, a literature review with no identified gap, research questions that cannot be answered by the chosen data, unexplained sample selection, unrealistic access assumptions, no analysis plan, ignored ethics, and a schedule that allocates most time to writing while treating data preparation as instantaneous.

Best practice is to:

• narrow the question before collecting data;
• justify every major design decision;
• pilot instruments and extraction procedures;
• maintain a decision and change log;
• create a data dictionary and secure versioned files;
• build contingency time into the schedule;
• discuss limitations before results are known;
• use supervisor meetings for decisions and feedback;
• match claims to design and evidence; and
• define completion criteria for each deliverable.

Project management should support research quality, not turn scholarship into administration. Plans are valuable because they make assumptions testable and problems visible early.

Chapter Summary

A dissertation proposal is both an academic argument and a project-control document. It should move logically from a significant and focused problem to research questions, objectives, theory, evidence, method, analysis, ethics, resources, milestones, risks, supervision, and deliverables. Methodological choices must be justified by the research question and by the nature of the evidence available. Project-management tools such as work breakdown structures, milestone schedules, risk registers, quality checks, and change logs improve feasibility and accountability. Ethical conduct, data protection, research integrity, and responsible use of technology are integral to the project rather than optional additions. The strongest proposals are focused, transparent, realistic, and capable of being defended before critical academic and professional audiences.

Key Terms

Research proposal: A structured plan explaining what a study will investigate, why it matters, and how it will be conducted.

Research problem: A specific uncertainty, difficulty, or gap that warrants investigation.

Research aim: A concise statement of the overall purpose of a study.

Research question: A focused question that the project is designed to answer.

Research objective: A specific action required to achieve the research aim.

Scope: The defined boundaries of a project.

Delimitation: A deliberate boundary imposed by the researcher.

Limitation: A constraint or weakness that may affect interpretation.

Methodology: The reasoning linking assumptions about knowledge to research design.

Method: A practical technique for collecting or analysing evidence.

Conceptual framework: A structured representation of key concepts and expected relationships.

Operationalisation: The process of translating an abstract construct into observable measures.

Population: The complete set of units relevant to a study.

Sample: The units selected from the population.

Milestone: A significant project checkpoint or achievement.

Work breakdown structure: A decomposition of project work into manageable packages.

Risk register: A record of risks, consequences, ratings, owners, and responses.

Change control: A documented process for evaluating and approving project changes.

Research integrity: Honest, accurate, transparent, and responsible conduct of research.

Deliverable: A specified output produced by a project.

Review Questions

5 Short Answer Questions

1. What is the difference between a research topic and a research problem?
2. Why should research objectives use observable action verbs?
3. Distinguish methodology from method.
4. What information should a risk register contain?
5. Why is a data dictionary useful in an accounting or finance project?

5 Long Answer Questions

1. Explain how a proposal functions as both an academic argument and a project-management document.
2. Compare archival quantitative, survey, interview, and case-study designs, including their principal strengths and limitations.
3. Discuss how theory, conceptual frameworks, and operationalisation contribute to a defensible research design.
4. Explain how milestones, dependencies, resources, and critical-path thinking can improve dissertation feasibility.
5. Evaluate the ethical and research-integrity issues that may arise when using organisational records, interviews, or commercial databases.

3 Analytical / Application Questions

1. A student proposes to study “the effect of financial technology on banks worldwide” using interviews with five local employees. Diagnose the proposal’s weaknesses and redesign its aim, scope, questions, and method.
2. A proposed archival study has 30% missing observations for its main variable and only eight weeks remaining. Prepare a reasoned risk response covering feasibility, alternative designs, data-quality checks, and communication with the supervisor.
3. Using the illustrative budgetary-participation case, construct an alignment argument linking one research question to evidence, sampling, analysis, ethics, and a deliverable. Explain what claim the design can and cannot support.

Multiple Choice Questions with Answers

Question 1

Which statement best describes the purpose of a research proposal?

A. To guarantee that the expected findings will be obtained

B. To justify and plan a feasible investigation

C. To replace the final dissertation

D. To provide only a list of references

Correct Answer: B. To justify and plan a feasible investigation

Question 2

Which item is a deliberate boundary imposed by the researcher?

A. Limitation

B. Sampling error

C. Delimitation

D. Research finding

Correct Answer: C. Delimitation

Question 3

What should determine the selection of a research method?

A. The popularity of the method

B. The research question and available evidence

C. The amount of software available

D. The expected direction of results

Correct Answer: B. The research question and available evidence

Question 4

In the model y_i = β₀ + β₁x_i + ε_i, what does ε_i represent?

A. The intercept

B. The explanatory variable

C. The unexplained error term

D. The sample size

Correct Answer: C. The unexplained error term

Question 5

A work breakdown structure primarily helps a researcher to:

A. Guarantee statistical significance

B. Decompose project work into manageable packages

C. Remove the need for supervision

D. Determine the final research findings

Correct Answer: B. Decompose project work into manageable packages

Question 6

If probability is rated 4 and impact is rated 5 on a simple risk scale, the priority score R = P × I is:

A. 9

B. 15

C. 20

D. 25

Correct Answer: C. 20

Question 7

Which is the strongest example of a research objective?

A. To prove that the hypothesis is correct

B. To collect interesting information

C. To examine the association between disclosure quality and borrowing costs

D. To write a long dissertation

Correct Answer: C. To examine the association between disclosure quality and borrowing costs

Question 8

Why is a pilot study useful?

A. It guarantees representative results

B. It tests procedures, wording, and feasibility before the main study

C. It eliminates all research limitations

D. It replaces ethical approval

Correct Answer: B. It tests procedures, wording, and feasibility before the main study

Question 9

Which practice most directly supports research integrity?

A. Reporting only results that support the hypothesis

B. Changing exclusion rules after seeing results without disclosure

C. Maintaining accurate records and reporting methods transparently

D. Using unverifiable sources to increase the reference count

Correct Answer: C. Maintaining accurate records and reporting methods transparently

Question 10

What is the most defensible interpretation of an observational regression association?

A. It automatically proves causation

B. It establishes an adjusted association subject to model assumptions

C. It proves that omitted variables are irrelevant

D. It means the sample is unbiased by definition

Correct Answer: B. It establishes an adjusted association subject to model assumptions