A note on formulating a dissertation proposal with less risk to research participants: for MBA students

A note on formulating a dissertation proposal with less risk to research participants: for MBA students

 

I am a part-time MBA student in Hong Kong, currently attempting to come up with a dissertation proposal. My concern is that, I have difficulties to employ primary research methods that require gatekeeper informed consent form. For research methods, e.g. research interview, that might gather sensitive information are also quite likely to be infeasible. In this regard, could you suggest a few research themes and research methods that are more feasible to consider under the present environment atmosphere with many organizations and people being more cautious in participating in dissertation project works as respondents?

Yes — in your situation, the most feasible dissertation designs are those that rely on publicly available data, document analysis, or lightly structured non-sensitive data, rather than requiring access to internal organizational information or sensitive personal disclosures. In Hong Kong, this is especially practical because many useful business and governance datasets are already available through public/open sources and listed-company disclosures.

Feasible themes

·        Corporate governance and disclosure quality of Hong Kong listed companies, using annual reports, ESG reports, and HKEX disclosure materials as data.

·        Dividend policy, board diversity, risk management, or ESG reporting changes among Hong Kong listed firms, again using only published documents and market data.

·        Sectoral resilience or adaptation in Hong Kong industries such as retail, hotels, logistics, or SMEs, using newspaper reports, company websites, public filings, and market statistics rather than internal interviews.

·        Comparative analysis of public policy or market trends in Hong Kong using official statistics and secondary datasets from government/open-data sources.

·                                                                 Content analysis of business communications, for example how companies frame sustainability, crisis response, talent retention, or innovation in public statements.

Methods that fit

·             Secondary data analysis: re-analyse existing datasets, government statistics, or published market data without collecting new personal or organizationally sensitive data.

·             Document analysis: examine annual reports, ESG reports, policy documents, news releases, and corporate websites.

·             Content analysis: code public texts for themes, frequency, tone, or disclosure patterns; this works well when interview access is difficult.

·             Comparative case study using public cases: compare firms, sectors, or policy responses based on publicly available information only.

·             Desk-based literature review with conceptual development: useful if access is very limited, especially when you want to build a strong theoretical framework first.

Topic ideas

Theme	Why it is feasible	Suitable method
ESG disclosure in Hong Kong listed firms	Public reports are accessible	Content analysis + secondary data
Board diversity and governance quality	Board-related disclosures are published	Document analysis + comparative analysis
Dividend policy communication	Dividend and governance disclosures are public	Content analysis
Retail or hotel resilience in Hong Kong	Public market reports and news coverage are available	Secondary data + document analysis
SMEs’ digital transformation in Hong Kong	Can use public policies, industry reports, and website materials	Case comparison + literature review

How to avoid gatekeeper issues

A useful design rule is to frame the study around industry-level or public-firm-level questions, not internal company operations or employee-specific experiences. For example, instead of studying “how employees in Company X perceive internal restructuring,” you might study “how Hong Kong retail firms communicate restructuring and resilience in public disclosures.” This keeps the project researchable without asking a manager to authorize internal access. It also reduces the chance that you will need sensitive informed consent arrangements for participants.

Practical recommendation

For a part-time MBA dissertation, the safest route is usually: one clear industry or governance theme + one public-data method + one manageable time window. A strong combination would be:

1.    pick a Hong Kong sector you already understand,

2.    define a question that can be answered from annual reports, public statistics, and company websites,

3.    use content analysis or secondary analysis as the core method.

A very workable example is: “How have Hong Kong listed retail companies changed their ESG and resilience disclosure practices from 2021 to 2025?” using annual reports, ESG reports, and publicly available market statistics.

 

What if the university still favor students to employ 1 to 2 primary research methods in the students' dissertation projects? What would you suggest I can consider coming up with primary research method design that expose to respondents with lower risks (e.g. reputation risk, sensitive corporate information leakage risk)?

You can satisfy a university requirement for 1–2 primary methods while keeping respondent risk very low by choosing designs that collect non-sensitive, non-identifiable data and avoid probing internal or reputational matters. Below are concrete, practical options (each linked to low risk, feasibility, and short note on ethics/gatekeepers).

Direct answer

·        Use low-risk primary methods such as structured anonymous surveys, brief public-facing intercept surveys, short expert-validation questionnaires with non-sensitive prompts, or lightweight vignette experiments; pair one primary method with a desk-based method (document/secondary data or content analysis) to meet the “1–2 primary” requirement while keeping risk minimal.

Low-risk primary method options (what they are, why low-risk, feasibility)

·        Anonymous online questionnaire (closed questions only). Collects aggregate, non-identifiable responses about opinions or perceptions on public topics (e.g., industry trends, attitudes to ESG disclosure). Low reputational risk because answers are anonymous and you avoid asking for firm-internal facts; easy to recruit via alumni lists, LinkedIn, or panels. Ethical forms are usually straightforward if anonymity is explicit.

·        Short structured surveys of consumers / customers (intercept or online panel). Ask about consumer behaviour, brand perceptions or willingness to pay—no internal corporate data, low sensitivity, fast to run, and no gatekeeper needed for firms if you recruit customers in public channels.

·        Vignette-based online experiment. Present short, fictional scenarios (vignettes) about public-firm behaviour (fictional company names or anonymised descriptions) and ask respondents for choices or ratings. This elicits causal or normative insight without exposing real companies or employees to reputational risk.

·        Expert-validation questionnaires with constrained topics. Send a short, structured questionnaire (3–8 items) to industry experts asking for general judgements (e.g., rank importance of drivers of retail recovery). Keep questions high-level and non-identifiable; ask for voluntary participation and offer anonymized reporting. This usually requires only minimal ethics paperwork because no sensitive firm data are requested.

·        Observation of public behaviour or content (structured naturalistic observation). Observe footfall patterns in public retail areas or code in-store window displays (no interaction, public setting). This is unobtrusive and typically low-risk for participants.

How to frame primary methods to reduce risk and gatekeeper needs

·        Ask about public-facing topics, not internal operations. For example: “How do consumers perceive retailers’ post-pandemic communications?” rather than “How did Manager X decide to restructure store Y?”.

·        Avoid collecting identifiers or linking responses to specific firms or individuals (use forced anonymity and aggregate reporting). This reduces privacy and reputational risk and simplifies ethics approval.

·        Use hypothetical or fictional stimuli (vignettes) if you need to study reactions to sensitive actions; respondents react to the scenario rather than identifying real people or companies.

·        Limit interview depth and scope: if you must do interviews, use very short (15–20 minute) semi-structured interviews focused on high-level industry perspectives rather than past actions or confidential decisions; obtain informed consent and offer anonymized quotes only.

·        Use opt-in recruitment from public channels or professional networks rather than asking a manager to introduce employees (avoids gatekeeper consent for accessing internal staff).

Practical pairings (one primary + one desk method) — low risk and defensible

·        Anonymous online survey of consumers + content analysis of firm annual reports (answers consumer perceptions vs. published disclosure).

·        Short vignette experiment with business students or alumni + secondary market/financial data analysis (tests responses to disclosure scenarios vs. actual market trends).

·        Expert short questionnaire (industry association members) + comparative document analysis of policy texts or corporate reports (validates expert judgments against documents).

·        Naturalistic observation of retail footfall patterns (public spaces) + desk-based analysis of industry sales statistics (links observed behaviour to published data).

Ethics and practical tips to improve chances of approval

·        Make anonymity explicit, collect no names/IDs, and report only aggregated results; document storage and deletion plans in your ethics form.

·        Use brief information sheets and one-click consent for online surveys; keep language simple and highlight voluntary nature.

·        If recruiting experts or alumni, use personal networks but emphasize independence and anonymized reporting; many gatekeepers accept this when no firm-confidential questions are asked.

·        Use existing datasets (HK government, HKEX, industry reports) to reduce primary data needs and justify limited, low-risk primary methods as complementary rather than central.

·        Discuss your low-risk design with your supervisor and attach a short rationale to ethics application showing how you avoid sensitive topics and reduce risk to participants (this often eases committee concerns).

Example low-risk dissertation design (concrete)

·        Topic: “How do Hong Kong consumers perceive retail firms’ ESG communications after COVID-19?”

·        Primary method: anonymous online consumer survey (10–12 closed items + 1 optional comment), recruited via social media and alumni list; consent via checkbox; no firm-specific questions.

·        Secondary/desk method: content analysis of ESG sections of annual reports for a matched sample of listed retail firms (2021–2025).

·        Risk management: anonymized responses, no employee/manager interviews, fictional vignettes if testing reactions to sensitive corporate actions.

** the discussion is supported by perplexity.ai

 

A collection of blog notes on using chatgpt for research purpose.