Logo site TopDissertations
+
Order Now
Logo site TopDissertations
Order Now

Diversity in Dissertation Research: Inclusive Methods and Sampling

Home  /  Blog  /  Diversity in Dissertation Research: Inclusive Methods and Sampling

What Is Diversity in Dissertation Research?

Diversity in dissertation research isn’t just about making sure your sample looks different. It’s about designing your entire research process so that it actively includes perspectives, experiences, and populations that are typically excluded from academic studies.

Think of it this way: if you’ve studied organizational trust, and all your participants work in corporate offices, your findings can’t tell you anything about people working in retail, healthcare, education, or informal economies. That’s not diversity — that’s a narrow slice.

A truly diverse sample ensures your research is more representative of the broader population. It reduces bias, strengthens validity, and produces findings that are applicable beyond a single privileged group.

But diversity goes deeper than demographics. The ERC’s Scientific Council notes that considering diversity in research designs has far-reaching implications for both the effectiveness and the ethics of your findings:

Failing to include diverse participants in any type of research may lead to biased and stereotyped conclusions. Researchers might unintentionally generalise findings from a homogeneous sample to the entire population, perpetuating stereotypes and reinforcing biases.

This isn’t just a social concern. It’s a methodological one. Biased samples produce biased results, and biased results produce flawed policy, poor clinical interventions, and theoretical models that don’t actually reflect reality.

  • Diversity in research means actively recruiting participants from underrepresented and historically excluded groups, not just treating demographic diversity as an afterthought
  • Intersectionality — people belong to multiple social groups simultaneously — is the single most important framework for designing inclusive samples
  • Five proven inclusive sampling strategies: purposive sampling, community-based participatory research (CBPR), stratified/quota sampling, maximum variation sampling, and barrier-removal techniques
  • Funders now expect EDI best practice embedded throughout research design — proposals that don’t address diversity often fail to get funded
  • Common mistakes: assuming one-size-fits-all sampling, ignoring power dynamics, and treating “diversity” as a checkbox rather than a design principle

Why Diversity Matters — Beyond the Checkbox

For decades, researchers treated diversity as an administrative requirement. Today, the academic community understands it as a foundational research quality principle. Here’s why it matters in every phase of your dissertation.

1. Methodological Rigor

Homogeneous samples produce narrow findings. If your research question involves human behavior, social dynamics, health outcomes, or educational experiences, a single-group sample cannot capture the full picture. As Research Design Review’s qualitative methodology experts emphasize, the scope of your sample directly affects whether you can credibly interpret your results.

2. Ethical Imperative

Ethical research demands fairness. If your research produces interventions or policies that will be applied to a broad population, excluding segments of that population from the research that informs those policies is unethical. As the ERC article explicitly states, excluding certain populations from research can raise ethical concerns “especially if those populations are later expected to use the interventions prompted by the research’s results.”

3. Funder Requirements

Major research funders now require EDI (Equality, Diversity, and Inclusion) best practice embedded throughout proposals. As Oxford Medical Sciences notes, “A good idea may fail to get funded if EDI is not adequately addressed.” The UK’s Medical Research Council (MRC), the National Institute for Health and Care Research (NIHR), and the Wellcome Trust all require EDI integration in research design. This is no longer optional — it’s a baseline expectation.

4. Real-World Applicability

Diverse samples produce findings that actually apply to the people you’re studying. A clinical trial that only includes male participants may miss critical treatment differences. A study on workplace stress that only samples corporate employees may miss patterns that matter to gig workers, caregivers, or people with disabilities. Diverse samples = real-world impact.

Inclusive Sampling Strategies for Your Dissertation

This is the practical core of this guide. Here are five evidence-based inclusive sampling strategies you can apply to your research design, with concrete examples.

Strategy 1: Purposive Sampling for Diversity

Purposive sampling means selecting participants based on specific characteristics relevant to your research question, rather than relying on convenience or random selection. For diversity, the goal is to intentionally recruit from multiple demographic, cultural, or experiential groups.

How to apply it:

  • Identify the demographic or experiential dimensions most relevant to your research question (e.g., age, socioeconomic status, cultural background, geographic location)
  • Define subgroups within your target population
  • Set minimum recruitment targets for each subgroup
  • Document your sampling frame so your committee can evaluate whether diversity was achieved

Dissertation example:

“To capture variation in technology adoption, we purposively sampled participants from three distinct sectors: rural agricultural communities, suburban small businesses, and urban retail enterprises. Each sector represented a different socioeconomic profile, technological infrastructure level, and decision-making structure.”

Strategy 2: Community-Based Participatory Research (CBPR)

CBPR is considered the gold standard for inclusive sampling. It involves partnering with communities themselves in the research process — not as passive subjects, but as active co-designers of the research question, sampling approach, and data interpretation.

Why it works: Community partners can identify barriers to participation that researchers might miss (transportation, language, cultural sensitivity, trust issues). They provide access to populations that are otherwise hard to reach.

Dissertation example:

“Rather than recruiting from clinical clinics alone, we partnered with two community health organizations serving predominantly immigrant populations. These partners helped design culturally appropriate recruitment materials and facilitated introductions to community members who might otherwise have been excluded from the study.”

Source: Carter et al., 2023, cited 26 times for the CBPR evidence base.

Strategy 3: Stratified and Quota Sampling

Stratified sampling divides your population into subgroups (strata) based on relevant characteristics, then samples from each stratum. Quota sampling sets minimum recruitment targets for each subgroup. Both increase accountability by defining who needs to be recruited and ensuring underrepresented groups aren’t overlooked.

Source: Roscoe, 2021 (NSF, cited 22 times).

How to apply it:

  • Define your strata based on your research question
  • Set quotas for each stratum (e.g., “at least 20% of participants must be from group X”)
  • Use stratified random selection within each stratum
  • Track recruitment progress against your quotas

Practical tip: If your research question involves health outcomes across age groups, set quotas that match the population distribution rather than recruiting who is most convenient.

Strategy 4: Maximum Variation Sampling

Maximum variation sampling is a qualitative technique that deliberately selects participants with widely differing characteristics. The goal isn’t to find patterns of similarity — it’s to uncover patterns of difference.

How it works: Research Design Review notes that this approach “maximizes the diversity relevant to the research question.” If your study examines workplace stress, maximum variation sampling would include employees from different industries, career stages, work arrangements (remote, hybrid, on-site), and organizational sizes.

When to use it: This approach is ideal when you suspect that diversity will have a significant impact on your results. For instance, if you’re studying how people cope with academic pressure, the experience will differ dramatically between graduate students, first-generation college students, parents balancing work and study, and international students.

Strategy 5: Barrier Removal Techniques

Roscoe’s fifth strategy focuses on removing the practical barriers that exclude underrepresented participants. This includes addressing cost, distance, communication needs, scheduling conflicts, and cultural mistrust.

Practical barrier-removal techniques:

  • Offer transportation assistance or travel stipends
  • Provide materials in multiple languages
  • Schedule data collection at times and locations convenient for participants
  • Use community-based recruitment (go to where people are, rather than expecting them to come to you)
  • Offer alternative participation formats (phone interviews, virtual meetings, home visits)
  • Partner with trusted community organizations as gatekeepers

The Five-Level Framework: Applying Inclusion Across the Research Cycle

Diversity isn’t just a sampling concern. It runs through every phase of your dissertation. Here’s a framework for integrating inclusive methods across the entire research lifecycle.

Level 1: Research Question Design

Your research question should account for diversity from the start. Consider whether your question implicitly assumes a homogeneous population. If your research question is about “employee retention,” that could mean very different things for corporate workers, healthcare workers, teachers, and gig economy workers.

Ask yourself:

  • Does my research question exclude certain populations by default?
  • What populations are already overrepresented in existing literature on my topic?
  • How might the experience I’m studying differ across demographic or cultural groups?

Level 2: Literature Review

A diverse literature review isn’t just about citing diverse authors (though that matters). It’s about showing you understand how different populations have been studied — and where the gaps are.

Key moves:

  • Identify studies that have explicitly addressed diversity in your field
  • Note where previous research has ignored underrepresented groups
  • Highlight how your study fills a diversity gap

Level 3: Data Collection

Your data collection methods need to be accessible and culturally appropriate for all participants. This means:

  • Using instruments validated across multiple demographic groups
  • Training interviewers in cultural competency
  • Offering multiple participation modes (written surveys, interviews, focus groups)
  • Ensuring informed consent materials are accessible and culturally sensitive

Level 4: Data Analysis

Here’s where many researchers fail. If you’ve recruited a diverse sample but analyze the data as if it’s a single group, you’ve wasted the diversity.

Research Design Review’s critical insight: When your sample is diverse, you need to think about how you’ll analyze by demographic segment. “At the conclusion of data collection, the analytic process will ultimately result in themes across the entire sample but importantly the researcher has the ability to look closely at the data associated with each of the participant groups.”

Practical strategies:

  • Disaggregate data by relevant demographic characteristics
  • Report subgroup findings alongside overall findings
  • Acknowledge when subgroup results differ significantly

Level 5: Dissemination

How you share your findings matters. Diverse research should be shared with diverse audiences. Consider:

  • Plain-language summaries for community participants
  • Multiple publication formats (journal article, conference presentation, policy brief)
  • Multi-language dissemination where relevant
  • Archiving in open-access repositories so underrepresented groups can access the findings

Discipline-Specific Considerations

Different fields face different diversity challenges. Here’s how inclusive methods look across major disciplines.

Health and Medical Research

Health research has the longest history of diversity scrutiny, particularly around sex and gender inclusion. The NIH’s policy on sex and gender in research design, the MRC’s EDI requirements, and the INCLUDE framework all set clear expectations for sex-inclusive clinical research. Key considerations:

  • Recruit participants across age, sex, gender identity, race, and socioeconomic status
  • Account for sex differences in study design (not just analysis)
  • Include pregnant women and people with disabilities (not just exclude them for “safety”)

Social Sciences

Sociology, psychology, education, and political science need diversity across demographic, cultural, and experiential dimensions:

  • Intersectionality is the dominant framework — consider how race, gender, class, and other identities interact
  • Include diverse cultural contexts, not just Western populations
  • Consider geographic diversity (rural, suburban, urban)

STEM (Science, Technology, Engineering, Mathematics)

STEM research traditionally has the least diversity, which produces notable biases:

  • Algorithmic bias in AI/machine learning research stems from non-diverse training data
  • Engineering studies often sample only male participants, missing critical design variables
  • Medical device research that only tests on certain body types produces dangerous outcomes

What to do: Actively recruit across gender, race, socioeconomic status, and geographic location. Design studies that account for biological and cultural variation.

Humanities

Humanities research often focuses on cultural interpretation, making diversity especially critical:

  • Avoid over-sampling Western European or Anglo-American sources
  • Include perspectives from multiple cultural, linguistic, and geographic contexts
  • Acknowledge positionality (your own position in the research landscape)

Common Mistakes (and How to Fix Them)

Mistake 1: Treating Diversity as a Post-Hoc Add-On

What it looks like: You recruit a sample, realize it’s homogenous, then try to find “more diverse” participants to bolt on.

Why it fails: Diversity has to be baked into your research design, not added after the fact. If you don’t specify sampling diversity in your proposal, funders will question your methodology.

Fix: Define your diversity framework in the proposal stage. Document which demographic or experiential dimensions are relevant to your research question and specify your recruitment strategy for each.

Mistake 2: Assuming “Representativeness” Means Random Sampling

What it looks like: You use simple random sampling and assume that’s enough for diversity.

Why it fails: Random sampling from a non-divisible population won’t capture underrepresented groups. If your population frame is drawn from urban areas, rural participants won’t appear in the random sample regardless of probability.

Fix: Use stratified or quota sampling to ensure all relevant groups are represented proportionally.

Mistake 3: Ignoring Power Dynamics

What it looks like: You recruit from settings where you already have access (your university, your workplace) and assume that’s diverse.

Why it fails: Institutional settings systematically exclude people outside those institutions — working-class individuals, non-traditional students, people with caregiving responsibilities, people without stable housing.

Fix: Identify who is systematically excluded from your sampling frame. Design complementary recruitment channels (community organizations, social media outreach, community-based recruitment) to reach those outside institutional settings.

Mistake 4: Not Accounting for Intersectionality

What it looks like: You recruit for diversity on one dimension (e.g., race) but ignore how that intersects with gender, class, or other identities.

Why it fails: A Black woman’s experience is not the same as a Black man’s experience or a white woman’s. Single-axis diversity doesn’t capture lived reality.

Fix: Map the intersectional dimensions most relevant to your research question. Design your sampling to include people with different combinations of these identities.

Mistake 5: Reporting Only Aggregate Results

What it looks like: You find a significant effect in your overall sample but don’t check whether the effect differs across demographic groups.

Why it fails: A finding that applies to only one subgroup may not generalize — or it may reveal that the intervention works differently across populations.

Fix: Run subgroup analyses. Report findings by relevant demographic groups. If subgroup results diverge significantly, discuss what that means.

A Practical Checklist: Evaluating Your Diversity Framework

Before submission, verify each item below.

Research Design:

  • [ ] My research question accounts for demographic and experiential diversity
  • [ ] I’ve identified which underrepresented populations are relevant to my study
  • [ ] I’ve specified my sampling strategy for diversity (stratified, purposive, CBPR, maximum variation, or quota)
  • [ ] I’ve addressed intersectionality in my sampling design

Data Collection:

  • [ ] My recruitment materials are culturally sensitive and accessible
  • [ ] I’ve offered multiple participation modes (in-person, virtual, phone)
  • [ ] I’ve removed practical barriers (transportation, scheduling, language)
  • [ ] My informed consent process is accessible across literacy levels

Data Analysis:

  • [ ] I’m disaggregating data by relevant demographic characteristics
  • [ ] I’m reporting subgroup findings alongside overall results
  • [ ] I’m acknowledging limitations related to sample diversity

Dissemination:

  • [ ] I’ve planned how to share findings with the communities I studied
  • [ ] I’ve considered multi-language or plain-language dissemination options
  • [ ] My publication strategy reaches diverse audiences

Summary: Diversity as a Design Principle, Not a Checkbox

Here’s the most important thing to understand: diversity in dissertation research isn’t about ticking a box or meeting a quota. It’s about designing your research so that it actually captures the complexity of the world you’re studying.

The framework is simple:

  1. Define which demographic, cultural, and experiential dimensions matter for your research question
  2. Design your sampling strategy around those dimensions (purposive, stratified, CBPR, maximum variation, or barrier removal)
  3. Collect using accessible, culturally sensitive methods
  4. Analyze by subgroup — don’t let diversity disappear into aggregate results
  5. Disseminate to diverse audiences, including the communities you studied

If you follow these steps, your dissertation won’t just be methodologically sound — it will be ethically robust, practically applicable, and positioned to meet the diversity expectations of funders and reviewers alike.

As the ERC notes, “Diversity is a powerful force that fuels innovation and progress. When people from diverse backgrounds come together, they bring a myriad of ideas, skills, and talents to the table. It is within this dynamic exchange of thoughts and experiences that ground-breaking solutions are often discovered.”

Related Guides

Need Help Designing Your Research Methodology?

Writing a dissertation proposal that meets funder EDI requirements and demonstrates robust, inclusive methodology can be overwhelming. If you need expert guidance, our academic consultants can help you develop a methodology section that addresses diversity, equity, and inclusion across your entire research design.

Order Custom Dissertation Methodology Support or get a free consultation to discuss how we can help you build a rigorous, inclusive research plan.

External References:

Ready to publish? Explore our dissertation services or order your custom dissertation support today.