A note on doing an MBA dissertation project with 1 primary research method: a discussion with generative AI

A note on doing an MBA dissertation project with 1 primary research method: a discussion with generative AI

 

My query posted to generative AI: Some of my part-time MBA students express difficulties to employ more than 1 primary research method for their dissertation projects, notably on the difficulties from obtaining informed consents from organizational gatekeepers. My problem is, to do a part-time 4-month dissertation project, which could be the research weakness of a research methodology design that is pragmatic and evidence-based to reply on a single primary research method, e.g. a questionnaire survey on the researcher friends with 40 respondents? I understand that the students have no problems to carry out 1 to 2 secondary research methods, e.g. Internet-based.

A single primary method is not automatically a weakness in a 4-month part-time MBA dissertation; the real issue is whether the method matches the research question, produces credible evidence, and is transparently justified. The main weakness of a small questionnaire survey of 40 friends is not that it is single-method, but that it may suffer from convenience sampling, limited representativeness, and weak generalizability.

What the real weakness is

A pragmatic research stance does not require mixed methods for its own sake; it asks what method best addresses the problem in the time and access constraints you face. So, if students can only manage one primary method, that is acceptable provided they are honest about scope, sampling, and limitations.

For a survey of 40 respondents drawn from the researcher’s friends, the likely weaknesses are:

·        Sampling bias, because friends are unlikely to represent the target population well.

·        Limited statistical power, because 40 cases is often too small for strong inference unless the study is very exploratory.

·        Response bias and social desirability, since acquaintances may answer in ways that please the researcher.

·        Weak external validity, meaning the findings should be treated as indicative rather than broadly generalizable.

How to frame it well

For a short dissertation, a single-method design can still be evidence-based if the research question is modest and the conclusions are carefully bounded. A survey of 40 can work for exploratory MBA research, especially if the aim is to identify patterns, perceptions, or preliminary relationships rather than to claim population-level truth.

A stronger framing would be:

·        Use the survey as a pilot or exploratory study.

·        Define the population narrowly.

·        State that the sample is non-probability and convenience-based.

·        Emphasize that findings are suggestive, not definitive.

·        Supplement with 1–2 secondary sources, such as industry reports, company websites, annual reports, or publicly available databases.

Managing gatekeeper problems

If organizational gatekeepers are difficult to access, that is a common fieldwork challenge and not a sign of poor research design by itself. In such cases, secondary research and accessible primary data sources are often a realistic compromise, especially for a time-limited dissertation.

A practical solution is to align the research design with what can actually be completed ethically and on time. For example, if access to employees is blocked, a student might study publicly available organizational data, customer perceptions, or professional respondents reached through personal networks, while clearly acknowledging the access constraint.

Best dissertation approach

For a 4-month part-time MBA dissertation, the safest pragmatic design is often:

1.    One primary method with a clearly defined and feasible sample.

2.    One or two secondary methods for triangulation or contextual support.

3.    A narrow research question that does not require deep access to organizations.

In your example, a questionnaire with 40 respondents is not a fatal weakness, but it becomes weak if the dissertation claims broad generalizability or robust theory testing. A better claim is that it provides exploratory evidence from an accessible sample, useful for identifying trends and informing future research.

A thesis-safe wording

You can write something like this in the methodology section:

·        “Given the time constraints of a 4-month part-time dissertation and the limited access to organizational gatekeepers, the study adopts a pragmatic design using one primary survey method supplemented by secondary documentary evidence.”

·        “The use of a convenience sample limits generalizability, but the design is appropriate for an exploratory MBA-level investigation.”

That wording is academically defensible because it shows methodological realism, ethical awareness, and transparent limitation handling.

 

My question posted to generative AI: Kindly offer 3 practice advices in this case, to make the dissertation project with 1 primary research method, to achieve an adequate level of intellectual sophistication up to the expected MBA academic level?

Here are three practice advices that help a single-method MBA dissertation reach an intellectually sophisticated, defensible academic level.

1. Build a tight “chain of reasoning” from philosophy → question → design → safeguards

Do not just list methods; explicitly justify each choice as a logical consequence of your research philosophy and question. For a pragmatic dissertation:

·        Start by stating why pragmatism fits your topic (e.g., you need practical, evidence-based answers to a business problem rather than pure theory testing).

·        Show how pragmatism leads to your chosen approach (often deductive or abductive) and then to a single primary method that is the most feasible and effective way to answer your question.

·        Explain how this shapes your design: sampling strategy, questionnaire structure, variables, and analysis plan.

·        End with how you safeguard quality and ethics: validity, reliability, bias control, informed consent, data protection.

This chain of reasoning shows methodological sophistication: you are not “using a survey because it’s easy,” but “using this survey design because it is the most appropriate pragmatic solution to my question under these constraints”.

2. Narrow the research question and aim for exploratory depth, not broad generalization

A small sample (e.g., 40 respondents) is academically acceptable if the question is narrow and the claims are modest. Instead of trying to “prove” general truths, aim to:

·        Identify patterns, perceptions, or preliminary relationships in a specific context.

·        Explore how a concept operates in a particular setting (e.g., a specific industry, company type, or region in Hong Kong or Singapore).

·        Generate insights that are useful for practice and that can inform future research.

Do this by:

·        Defining a clear, bounded population (e.g., “mid-level managers in Hong Kong retail firms” rather than “all managers”).

·        Stating that the study is exploratory or pilot in nature, and that findings are indicative rather than definitive.

·        Using your single method to go deep on a few well-chosen variables or constructs, with strong theoretical grounding from the literature review.

This approach shows intellectual sophistication: you understand the limits of your design and are not overclaiming, which is a key marker of good research.

3. Strengthen rigor within the single method through careful design, analysis, and triangulation with secondary sources

Even with one primary method, you can demonstrate methodological rigor by:

·        Design rigor:

·        Use validated scales or constructs from existing literature rather than inventing items from scratch.

·        Ensure your questionnaire has clear instructions, logical flow, and avoids bias (e.g., leading questions, double negatives).

·        Explicitly describe how you will collect data (online platform, time window, recruitment channel) and how you ensure informed consent and anonymity.

·        Analysis rigor:

·        Use appropriate statistical techniques for your sample size (e.g., descriptive statistics, correlations, simple regression, or group comparisons if justified).

·        Check basic assumptions (e.g., normality, linearity) and report limitations transparently.

·        Present results clearly with tables and graphs, and interpret them in relation to your research questions and theory.

·        Triangulation with secondary methods:

·        Use 1–2 secondary methods (e.g., analysis of company reports, industry data, publicly available statistics, or policy documents) to contextualize and support your survey findings.

·        Compare and contrast survey results with existing evidence: do they align, contradict, or refine what the literature or industry reports suggest?.

This combination shows that while you rely on one primary method, your overall design is robust, evidence-based, and intellectually mature.

---

If you embed these three practices—(1) a clear chain of reasoning, (2) a narrow, exploratory question with bounded claims, and (3) high internal rigor and secondary triangulation—a single-method dissertation can fully meet MBA-level academic expectations, even with a small sample and limited gatekeeper access.

 

My question posted to Generative AI: Highlight 4 main characteristics of an exploratory research for part-time MBA students with regard to our discussion here.

Based on our discussion about single-method, time-constrained part-time MBA dissertations, here are 4 main characteristics of exploratory research:

1. Purpose is to clarify and map, not to prove or generalize

The goal is to understand "what is happening," "why it might be happening," or "how it works" in a new or under-studied area, rather than to confirm hypotheses or claim population-level truths.

For your students, this means framing their research as identifying patterns, perceptions, or preliminary relationships in a specific context (e.g., Hong Kong retail managers), not as producing definitive generalizable findings.

2. Flexible and open-ended design that evolves with insights

Exploratory research is not rigid; researchers adjust their approach based on what they discover as data collection progresses.

In practice, this allows part-time MBA students to refine their questionnaire, focus on key variables, or narrow their scope mid-project if early results show what is most relevant.

3. Foundational: builds stronger questions for future research

Exploratory studies are designed to lay the groundwork for more structured, descriptive, or experimental research later.

This is academically appropriate for a 4-month dissertation: students can honestly state that their findings are indicative and suggest directions for future research, which demonstrates intellectual maturity rather than weakness.

4. Cost-effective and feasible with limited sample sizes

Compared to large-scale studies, exploratory research is typically less expensive and more budget-friendly, making it suitable for students with limited access, time, and resources.

A questionnaire with 40 respondents from accessible networks is acceptable if the study is framed as exploratory, where small samples are standard and the focus is on depth of insight rather than statistical power.

---

These characteristics show that exploratory research is not a "weak" design but a legitimate, academically rigorous approach for MBA students facing real-world constraints like gatekeeper access and tight timelines.

 

*** MBA students need to discuss with their dissertation project supervisors to ensure that they are reasonably satisfied with their research methodology design. This note is an academic discussion paper with generative AI, not an official policy statement from the university.

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