UJ DEVHONS: Research Methodology for Development Practitioners Study Guide

Research methodology is the backbone of credible development practice: it determines how practitioners understand problems, generate evidence, and justify interventions. For a University of Johannesburg (UJ) DEVHONS (Development Honours) learning context, the “Research Methodology for Development Practitioners” focus equips students to design ethically sound, rigorous, and politically aware studies that can survive scrutiny from supervisors, communities, and academic assessors. This study guide distils the course into exam-ready concepts, practical methods, and assessment-aligned skill sets—especially with attention to South Africa’s realities across universities, colleges, and TVETs where development work is both locally grounded and globally informed.

This guide is written for development practitioners and honours-level students at UJ. It emphasises how to move from a researchable question to a feasible design; how to choose methods (qualitative, quantitative, mixed methods) appropriately; how to handle sampling, data collection, validity, and analysis; and how to align research ethics with South African legal and institutional expectations. Throughout, it uses concrete examples that reflect common development topics: livelihoods, service delivery, youth employment, gender-based violence, informal settlements, community health, educational access, and local governance.

Section 1: Development Research as Practice—Paradigms, Positioning, and Problem Formulation

Why development methodology is different

Development practitioners often begin with lived experience: the community has a problem, a programme is failing, or an intended outcome has not materialised. Research methodology matters because development problems are rarely “pure” technical issues; they are shaped by power relations, institutional constraints, resource limitations, policy environments, and social norms. A methodology that ignores those realities may produce findings that are technically correct but practically unusable—or ethically harmful.

In South Africa, this is especially important. Development work intersects with:

Persistent inequality shaped by historical patterns of land dispossession and labour migration.
Service delivery challenges linked to municipal capacity and governance.
High levels of social vulnerability associated with unemployment, housing insecurity, and health burdens.
Community heterogeneity (language, class, gender identities, migration histories, informal work patterns).

A development honours student is expected to approach research not only as a technical procedure but as an interpretive and ethical process. The course’s methodology lens helps students link their research design to development outcomes—such as empowerment, inclusion, improved wellbeing, and institutional accountability.

Research paradigms: choosing a worldview without ignoring practical constraints

In methodology courses, “paradigm” refers to deeper assumptions about reality (ontology), knowledge (epistemology), and how inquiry should proceed (methodology). A practitioner-researcher must know these debates to avoid mixing incompatible assumptions.

The most common paradigms you will encounter in development research include:

Positivist / post-positivist approaches
- Assume there is a reality that can be measured.
- Knowledge is produced through observation, measurement, and analysis.
- Methods often include surveys, experiments, and statistical modelling.
- Useful when you need to estimate prevalence, quantify relationships, or track change over time.
Interpretivist / constructivist approaches
- Assume meanings are socially constructed and vary across contexts.
- Knowledge is generated through understanding lived experiences and interpreting narratives.
- Methods often include interviews, focus groups, ethnography, and document analysis.
- Useful when you need to understand how people make sense of programmes, policies, or institutions.
Critical approaches (critical realism, critical theory, participatory variants)
- Focus on power relations, ideology, and structural constraints.
- Knowledge is not neutral; research can challenge or reinforce inequality.
- Methods often include mixed methods, participatory research, and policy ethnography.
- Useful when you need to study why outcomes differ across groups and how structures shape those outcomes.
Pragmatism
- Treats methods as tools selected based on research questions and practical needs.
- Allows mixed methods and methodological pluralism.
- Useful for development practitioners who need actionable evidence.

Exam angle: You are usually not asked to “pick a paradigm forever.” Rather, you should explain how your research question and context justify your methodological choices. In many UJ honours dissertations, students defend a mixed approach using pragmatism: e.g., combine survey prevalence data with interviews explaining why certain households fail to access services.

Positionality: the practitioner’s role in the research relationship

Development methodology requires attention to positionality, meaning how the researcher’s identity, status, language, institutional affiliation, and prior engagement shape access to participants and interpretation of findings.

Common positionality issues in South Africa include:

Researcher affiliation with government, NGOs, universities, or donor-funded programmes.
Power imbalances between “outsiders” (researchers) and “insiders” (community members).
Language and translation challenges.
Expectations created by past programme participation (“You ask questions; you must be connected to help.”).

Practical exam-ready response: In a methodology section, you typically include:

Your role in relation to the study context.
Potential influence on responses (social desirability bias, fear of repercussions, gratitude effects).
How you will mitigate those influences (confidentiality, careful consent processes, reflexive journals, triangulation).

Problem formulation: from a broad development concern to a researchable question

A frequent honours weakness is a fuzzy problem statement: “Youth unemployment is a problem; we will study it.” This is too broad. Methodology requires turning a concern into a precise, answerable question.

A strong problem formulation usually includes five elements:

Context and justification
- What is happening locally?
- Why does it matter for development outcomes?
- What gap exists in existing research or programme performance?
The development problem
- What exactly is not working? (access, participation, skills, barriers, governance, funding, coordination)
The conceptual focus
- Which theoretical lens guides interpretation? (capabilities, livelihoods, gender analysis, empowerment, social capital, institutional theory)
The unit of analysis
- Are you studying individuals, households, organisations, communities, projects, or policy processes?
The researchable question (or questions)
- Should specify population, geography (or institutional setting), and outcomes.

Example: refining a research question in a South African development context

Weak question: “How can we reduce youth unemployment in Tshwane?”

Refined possibilities (choose one angle):

“What barriers prevent youth in a specific informal settlement community near Tshwane from accessing employability programmes, and how do these barriers differ by gender?”
“To what extent do municipal youth employment interventions in Tshwane improve job readiness outcomes, as measured by changes in self-efficacy and placement within six months?”
“How do employers in Tshwane perceive youth employability credentials, and what organisational factors shape hiring decisions?”

Notice how each refined question:

Chooses a unit (youth, employers, intervention participants).
Specifies barriers/outcomes.
Suggests feasible data sources (youth interviews, employer surveys, programme records).

Operationalising variables and concepts (even in qualitative work)

Even qualitative studies need operational clarity. Key concepts like “empowerment,” “participation,” “agency,” or “resilience” must be translated into observable indicators or interview themes.

For quantitative work, operationalisation requires explicit variable definitions. For qualitative work, it means defining what counts as evidence for your concept. For example:

Empowerment might be evidenced by:
- decision-making autonomy,
- confidence expressed in interviews,
- perceived ability to negotiate with institutions,
- access to resources and networks.
Barriers to access might include:
- transport costs,
- documentation requirements,
- information deficits,
- stigma,
- childcare constraints,
- fear of fraud or programme misuse.

Exam advantage: When you demonstrate operationalisation, you show the examiner that you understand methodology as translation between concepts and data.

Practical activity: constructing a submission-ready problem statement

For your exam or draft proposal, practice writing in this structure:

Problem statement (150–250 words)
Main research question
Sub-questions (3–5)
Objectives (general + specific)
Justification (why now; why this population; what gap)
Expected contribution to practice (e.g., programme improvement, policy relevance, implementation strategy)

This structure is common across UJ-style honours submissions and helps you link methodology to development outcomes.

Section 2: Research Designs for Development Practitioners—Sampling, Data Collection, and Ethics in South Africa

Choosing a research design: aligning method to the purpose

A research design is the overall “blueprint” for how you will conduct your study, including:

approach (qualitative, quantitative, mixed),
strategy (case study, survey, ethnography, evaluation, action research),
time horizon (cross-sectional, longitudinal),
unit of analysis,
data sources and procedures.

Development methodology must balance:

rigour (credible evidence),
feasibility (time, budget, access),
ethics (consent, confidentiality, harm minimisation),
usefulness (actionable findings).

Below are common designs in development research.

Cross-sectional survey design

Best for measuring prevalence, associations, and descriptive patterns at one time point.

Example: survey of caregivers about clinic access barriers.
Strength: efficient, quantifiable, comparable.
Limitation: cannot easily infer causality.

Longitudinal design

Best for observing change over time.

Example: follow youth participants in an employability programme across 6 months or 12 months.
Strength: stronger inference about change.
Limitation: higher cost, risk of attrition.

Case study design

Best for in-depth understanding of a bounded system (a project, community, NGO programme).

Example: case study of a community-based waste management initiative and its governance dynamics.
Strength: rich context, multiple sources of evidence.
Limitation: must justify boundaries and generalisability claims.

Evaluation design (formative and summative)

Best for assessing programme performance.

Example: formative evaluation of a skills training programme to improve design before scale-up.
Strength: directly supports decision-making.
Limitation: must manage politics, access, and stakeholder expectations.

Participatory action research (PAR)

Best for research that aims to create change with participants, not only about them.

Example: co-designed livelihood experimentation with women entrepreneurs.
Strength: empowerment and relevance.
Limitation: complexity, power dynamics, ethical and governance requirements.

Exam tip: Examiners look for coherence: the chosen design must follow from the research question and objectives. If your question is “why and how,” a purely cross-sectional survey may be inadequate unless paired with qualitative explanation.

Sampling strategies: selecting participants without losing credibility

Sampling is a major methodological grading area. You need to show:

who you will include,
why that group is relevant,
how you will recruit them,
how sampling affects representativeness and bias.

Common sampling strategies include:

Probability sampling
- Ensures known chance of selection.
- Examples: simple random sampling, systematic sampling, stratified sampling.
- Useful for generalisable quantitative findings.
Non-probability sampling
- Selection not based on chance.
- Examples: purposive sampling, convenience sampling, snowball sampling.
- Useful for qualitative depth and hard-to-reach populations.
Stratified purposive sampling
- A hybrid strategy often used in mixed methods.
- Example: purposively recruit participants across strata (gender, age group, employment status).

Example: purposive sampling for a study on youth employability

Suppose your research question is:

“How do employability programme participants in Tshwane experience barriers to internships, and how do these differ by gender?”

You might use purposive sampling to include:

participants who completed the programme,
participants who dropped out,
participants currently in internship placement,
and ensure diversity across gender and educational levels.

You would justify why each group is needed to answer “how” and “why,” not just “what proportion.”

Sample size: defending “enough” for your design

In qualitative work, sample size is typically guided by data saturation—the point when additional interviews no longer produce substantially new themes. In quantitative work, sample size is justified using statistical power, expected effect sizes, or minimum thresholds.

Because exam questions often require a defensible justification rather than a single “magic number,” adopt a structure:

Qualitative interviews: e.g., 15–30 in-depth interviews depending on diversity and saturation.
Focus groups: e.g., 2–6 groups depending on language and composition.
Surveys: e.g., 150–500 depending on population size, desired precision, and available resources.

If you choose to state numbers in your assignment, ensure they align with time and logistics. For example:

25 interviews, each 60 minutes, requires about 25 hours plus transcription time.
If transcribing verbatim, allocate time and budget.

Consistency principle: Once you pick sample sizes in your draft plan, keep them consistent across objectives, instruments, analysis plans, and ethics processes.

Data collection methods: selecting instruments and managing quality

Quantitative instruments

Structured questionnaires
Likert-scale items
Knowledge/attitude scales
Usage metrics (service access frequency)
Programme outcome indicators

Quality considerations:

Pilot testing (item clarity, comprehension)
Translation and back-translation (if multilingual)
Reliability checks (e.g., internal consistency for scales)
Minimising measurement bias (neutral wording, consistent administration)

Qualitative instruments

Semi-structured interview guides
Focus group discussion prompts
Observation checklists
Document analysis protocols

Quality considerations:

Interview training and standardisation
Managing group dynamics in focus groups
Recording, transcription, and anonymisation
Reflexivity and field notes

Mixed methods

Mixed methods designs require careful integration. The main ways studies integrate methods include:

Convergent design: collect quantitative and qualitative data concurrently, then compare/merge findings.
Explanatory sequential design: collect quantitative results first, then qualitative interviews to explain patterns.
Exploratory sequential design: qualitative first to generate variables/themes, then quantitative to test patterns.

Exam-ready statement: Integration is not just “using two methods.” You must explain how the data inform each other and how you will interpret contradictions.

Piloting and fieldwork logistics

Piloting tests the research tools and procedures. It helps identify:

confusing questions,
unsafe or sensitive wording,
inadequate response options,
misunderstandings due to language,
unrealistic time burdens for participants.

For fieldwork in South Africa:

consider transport realities,
account for community schedules (church days, market days),
build buffer time for interruptions,
ensure safe storage of devices and recordings.

Ethics and legal/institutional compliance in South Africa

Ethics is not a formality; it is part of methodology. In development research, ethical risks include:

harm caused by disclosure of sensitive information,
coercion in power-laden contexts,
privacy violations when communities identify participants,
emotional distress when discussing trauma or violence,
“benefit expectation” where participants believe the researcher can provide assistance.

Key ethical principles:

Informed consent
- Understandable, voluntary, and documented appropriately.
- For participants with limited literacy, use verbal explanation plus consent procedures consistent with institutional guidelines.
Confidentiality and anonymity
- Remove identifying details in transcripts.
- Store recordings securely with password protection.
- Use participant codes rather than names.
Non-maleficence
- Avoid questions that can cause harm without appropriate support pathways.
- If studying sensitive topics (e.g., GBV, mental health), plan referral options.
Beneficence and fairness
- Ensure participants are not exploited.
- Consider whether communities benefit from results in accessible ways.
Respect for persons and community processes
- When required, obtain permission from community gatekeepers in addition to individual consent.
- Ensure participatory elements are genuine, not symbolic.
Data management
- Maintain a data protection plan: storage, access control, retention period, and disposal.

Exam angle: If your study involves vulnerable groups, anticipate exam questions on how you reduce risk, ensure support, and avoid extracting “data without care.”

Building trust and reducing bias in data collection

Bias can be introduced through:

leading questions,
interviewer effects,
social desirability bias,
selective recruitment,
non-response bias.

Strategies to mitigate:

train interviewers to maintain neutrality,
use probing questions that encourage elaboration without steering,
anonymise and emphasise confidentiality,
triangulate findings using multiple data sources.

Illustrative scenario: If interviewing households about corruption in service delivery, participants may fear reprisal. A methodology must include secure interview locations, anonymisation, and careful wording to avoid actionable accusations.

Section 3: Data Analysis and Evidence Building—Validity, Trustworthiness, and Development Relevance

The logic of analysis: turning data into claims

Analysis is the bridge between raw data and development decisions. It should demonstrate:

how patterns were identified,
how interpretation was grounded in evidence,
how conclusions connect to research questions and objectives.

In honours-level development studies, you are expected to articulate:

what your analytic framework is,
what software or procedures you used (if applicable),
how you ensure rigour (quantitative validity, qualitative trustworthiness),
how you report findings transparently.

Quantitative analysis: core procedures and common pitfalls

Descriptive statistics

Used to summarise data:

frequencies and percentages,
means and standard deviations,
cross-tabulations.

Example exam-friendly approach:

“Table 1 presents demographic characteristics by gender.”
“Table 2 shows service access frequency distribution.”

Inferential statistics

Used to test hypotheses or examine associations:

chi-square tests for categorical variables,
t-tests or ANOVA for group differences,
correlations for linear relationships,
regression models for predicting outcomes while controlling variables.

Pitfall: applying tests without meeting assumptions. A rigorous methodology requires acknowledging assumptions:

normality, independence, multicollinearity,
adequate cell sizes in contingency tables.

Interpreting results: linking numbers to meaning

Quantitative analysis must avoid “number dumping.” Instead:

explain effect direction and magnitude,
describe practical significance, not only statistical significance,
connect findings to development mechanisms (why might the relationship exist?).

Example:

A regression shows that having transport access predicts programme attendance.
Analysis should then interpret plausible mechanisms: travel costs, mobility constraints, and information flow.

Qualitative analysis: coding, themes, and interpretive depth

Coding strategies

Common options:

Deductive coding: start with a framework derived from theory or research questions.
Inductive coding: allow codes to emerge from the data.
Hybrid coding: combine both.

A typical honours process:

familiarisation with transcripts (reading, re-reading),
initial coding (open coding),
code refinement (merging, splitting),
theme development (grouping codes into themes),
interpretation with theoretical and contextual linkage.

Trustworthiness: ensuring interpretive rigour

Instead of “validity” (a term often emphasised in quantitative work), qualitative research uses trustworthiness criteria such as:

Credibility: member checking, triangulation, prolonged engagement.
Transferability: thick description so others can judge similarity to their contexts.
Dependability: documenting research process and decisions.
Confirmability: reflexivity, audit trail, and showing evidence for interpretations.

Exam-ready phrasing: You can state that your study enhances credibility through triangulation and that your audit trail records coding decisions.

Mixed methods integration: how to avoid superficial merging

Mixed methods analysis must show integration logic:

Are qualitative findings used to explain quantitative outliers?
Do both strands confirm each other?
Do they diverge, and if so, what does divergence mean?

Common integration approaches include:

joint displays (tables/figures aligning themes with statistics),
narrative integration (explaining how each strand informs the other),
meta-inferences (final integrated conclusions).

Practical exam response: If quantitative data show low programme uptake but qualitative interviews reveal fear of debt or documentation barriers, then integration yields a mechanism-based interpretation.

Validity, reliability, and evidence strength (exam-critical concepts)

Quantitative reliability

Consistency of measurement, often assessed by scale reliability (e.g., Cronbach’s alpha in many contexts).
Ensures the instrument measures consistently.

Qualitative reliability/trustworthiness

Consistency of interpretation and transparency in analytic procedure.
Use a codebook and document changes.

Internal validity and causality caution

Even if you use regression, remember that correlation is not causation unless your design supports causal inference. Development studies often face:

confounding factors,
selection bias,
measurement errors.

Your analysis plan should acknowledge limitations and explain how you control for known confounders or how design reduces bias.

Building development-relevant evidence: beyond academic outputs

The core promise of development practitioner research is that findings improve practice. Evidence building should demonstrate:

implications for programme design,
actionable recommendations,
feasibility considerations (resources, governance, political constraints).

A development-focused analysis section usually includes:

Findings linked directly to objectives.
Interpretation using theory and contextual understanding.
A section on implications and trade-offs.
Recommendations for stakeholders (programme managers, local government, NGOs, community groups).

Example: analysis workflow for a hypothetical programme study

Consider a hypothetical case: a youth mentorship and internship readiness programme in Tshwane.

Quantitative strand

Survey youth participants at programme start and after 6 months.
Measure outcomes such as:
- job readiness confidence (scale),
- internship placement status (binary),
- perceived access to networks (index).

Qualitative strand

Interviews with participants and employers.
Explore themes:
- credibility of credentials,
- employer trust and safety concerns,
- mentorship quality,
- barriers to internship attendance.

Integration

If the survey shows no significant improvement in confidence, interviews may reveal mentorship is inconsistent or language barriers reduce benefit.
If internship placement improves only for one gender group, employers’ hiring norms and perceived mobility constraints might be explaining mechanisms.

Your integrated findings are not just “two sets of results”—they are a single development explanation.

Reporting findings: structure that earns marks

Examiners respond well to:

clear linkage between research questions and reported results,
tables/figures with appropriate titles and explanations,
quotations that illustrate themes (with participant codes),
a consistent narrative: method → analysis → findings → implications.

Common reporting components:

brief reminder of objectives,
thematic headings aligned to sub-questions,
evidence presented systematically,
interpretive statements supported by data.

Section 4: Proposal Writing, Chapter Structure, and Methodology for UJ DEVHONS Assessments

How honours methodology is assessed

In UJ honours development studies settings, the assessment emphasis often includes:

coherence (question → design → methods → analysis),
ethical rigour,
feasibility and credibility,
theoretical alignment,
methodological justification (why this method, why this population),
scholarly academic writing quality.

Many students lose marks not because their idea is wrong, but because:

the sampling plan is underdeveloped,
ethics risk is not addressed,
methods do not match the research question,
analysis plans are vague,
concepts are not operationalised.

This section provides an exam-ready structure for proposals and methodology chapters.

Suggested chapter structure (methodology-focused)

While specific UJ templates may vary by supervisor, a typical development honours structure includes:

Introduction and background
Literature review
Theoretical framework
Methodology
- research approach and design,
- study setting and rationale,
- sampling strategy and sample size,
- data collection methods,
- instruments and procedures,
- data analysis plan,
- ethical considerations,
- limitations and delimitations.
Findings
Discussion and recommendations
Conclusion
References and appendices (consent forms, guides, instruments)

Your study guide focus here is methodology chapter craft.

Writing a methodology chapter: elements that must appear

A methodology chapter usually needs the following explicit subsections:

Research paradigm and approach
- justify qualitative/quantitative/mixed choice.
Research design
- explain strategy (case study, survey, evaluation, PAR),
- specify time horizon.
Study site and context
- describe location and relevant institutional or community features.
Population and sample
- define who you study,
- justify sampling technique,
- provide sample size and recruitment logic.
Data collection procedures
- list instruments (questionnaire/interview guide),
- describe steps (pilot, recruitment, scheduling, administration),
- include language considerations.
Data analysis procedures
- describe coding steps for qualitative,
- describe statistical tests and models for quantitative,
- describe integration steps for mixed methods.
Ethics
- consent,
- confidentiality,
- risk mitigation,
- data management.
Reliability/trustworthiness and validity
- describe how you ensure rigour.
Limitations
- be honest but analytical.
Delimitations

clarify boundaries of the study.

Linking theory to methodology without forcing it

A common issue is writing a literature review and theory chapter, then writing methodology as if theory never influenced choices. High-scoring methodology connects theory to:

coding frameworks,
interview guide themes,
survey variable construction,
interpretation logic.

Example:

If your theoretical lens is capabilities, then your methodology should measure or ask about practical freedoms: education, health, mobility, participation, and agency. If you only measure “income,” you weaken the theoretical alignment.

Instruments and operationalisation in proposal writing

For exams and proposals, you often need to show that your instruments are designed to answer your questions.

Designing an interview guide

A strong interview guide:

starts with broad context questions,
moves to programme experience and barriers,
includes probes for mechanisms (“what made it possible?”),
asks about perceptions and meanings,
ends with recommendations.

Example themes for a youth employability study:

pathway into the programme,
experiences with mentorship,
internship application barriers,
employer interaction and perceived fairness,
gendered experiences and support structures,
suggestions for improvement.

Designing a questionnaire

A strong questionnaire:

uses clear language,
measures constructs with multiple items,
includes demographic and control variables,
uses response categories that reflect participants’ realities.

Case example: a methodology proposal outline (exam model)

Imagine a DEVHONS student proposes a mixed methods study on barriers to internship access for youth in Tshwane.

Main research question

“What barriers shape youth internship access in Tshwane, and how do those barriers vary by gender and educational background?”

Objectives

quantify prevalence of different barriers,
examine associations between barriers and internship access outcomes,
explain how participants interpret institutional processes.

Design

explanatory sequential mixed methods:
- quantitative survey to identify patterns,
- follow-up interviews with selected youth and employers to explain mechanisms.

Sampling

quantitative: stratified sampling across gender and education categories;
qualitative: purposive selection based on survey results (e.g., those with high barriers and low access).

Ethics

informed consent,
confidentiality,
safe handling of potentially sensitive allegations.

Analysis

quantitative: logistic regression predicting internship access,
qualitative: thematic coding with an empowerment and institutional lens,
integration: compare qualitative themes with statistically significant barrier predictors.

This model demonstrates coherence and feasibility.

Handling limitations and delimitations like an honours candidate

Honours-level methodology should not claim impossibilities. Instead, it should:

identify what you cannot infer (e.g., causality),
explain how you handle uncertainties,
justify why your boundaries do not destroy usefulness.

Example limitations:

limited geographic scope (only Tshwane wards/communities),
non-response bias in surveys,
language barriers affecting interpretation despite translation.

Delimitations:

focus on internship access rather than long-term employment outcomes,
focus on one programme rather than all municipal interventions.

Preparing for methodology viva / exam questions

Typical examiner questions include:

Why did you choose this design and not another?
How did you determine sample size?
How will you ensure validity/trustworthiness?
What ethical risks exist in your study?
How will you handle translation and cultural interpretation?
How will you integrate quantitative and qualitative findings (if mixed methods)?
What is your plan for data security and storage?

Your study guide strategy:

always link answers to the research question and objectives,
mention mitigation steps,
use concise but specific methodological language.

Section 5: Advanced Topics and Practitioner-Focused Mastery—Participatory Methods, Programme Evaluation, and Ethical Evidence Use

Participatory research for development practitioners (beyond “consultation”)

Participatory approaches aim to reduce the extractive nature of research and improve relevance and legitimacy. However, participation can be superficial if it only collects opinions without shaping decisions.

In practitioner-focused methodology, participatory research should clarify:

who participates in defining the research problem,
what decisions participants influence,
how power imbalances are managed,
how feedback is shared with communities,
what happens if findings contradict programme assumptions.

Two forms commonly discussed:

Participatory action research (PAR)
- participants and researchers co-create knowledge and act on findings.
- cycles: plan → act → observe → reflect.
Participatory rural appraisal (PRA) and related methods
- includes mapping, ranking, seasonal calendars, problem trees, and group discussions.
- typically used for exploratory work and programme design.

Exam nuance: Examiners often ask how participation is genuine. You must describe governance processes, not just list activities.

Participatory techniques: practical examples

A practical portfolio of participatory methods includes:

Community mapping
- mapping service access points (clinics, schools, transport nodes),
- identifying unsafe areas or barriers.
Problem ranking
- communities rank issues (e.g., unemployment, safety, sanitation) based on criteria they define.
Seasonal calendars
- examine how livelihood activities change across months and how that affects programme participation.
Venn diagrams
- explore relationships among organisations (NGOs, shelters, municipal offices) and perceived importance.
Focus group deliberation
- structured discussions where participants evaluate policy options.

Methodological caution:

participants may replicate dominant views,
resource dynamics can shape who speaks,
facilitation biases can occur if researchers “steer” outcomes.

Mitigation:

use trained facilitators,
ensure inclusive group composition,
maintain facilitator reflexivity logs.

Programme evaluation methodology: using evidence responsibly

For development practitioners, programme evaluation is method-driven learning. Evaluation may be formative (improve design) or summative (assess outcomes and impacts).

Common evaluation frameworks:

Logic models
- connect inputs → activities → outputs → outcomes → impacts.
- helpful for structuring evidence collection.
Theory of Change (ToC)
- explains how and why a programme produces change, including assumptions.
Outcome harvesting
- identifies change and works backward to evidence.
Contribution analysis
- assesses whether a programme contributed to observed outcomes under complex conditions.

Exam-ready: If asked to justify design, show how evaluation type matches information needs:

If stakeholders need improvement: formative evaluation and process monitoring.
If stakeholders need accountability: summative evaluation with outcome measurement.

Ethical use of evaluation evidence: risks and responsibilities

Evaluation can create ethical tensions:

if results show underperformance, staff or communities may be blamed;
participants might feel exploited if their responses are used to justify budget cuts without support;
revealing sensitive failures can harm political legitimacy and community trust.

Ethical evidence use strategies:

report responsibly, avoid naming individuals,
engage stakeholders early about reporting boundaries,
provide communities with accessible feedback,
focus on learning and improvement, not punishment.

Counter-arguments: when certain methods may fail

A high-level methodology study guide includes not only “best practices” but also critiques.

Counter-argument 1: “Qualitative findings are not generalisable”

True in statistical terms. But qualitative research can offer:

transferability through thick description,
mechanism insights,
context-specific evidence.

Exam defence:

Explain generalisation strategy: analytic generalisation (theories and mechanisms), not population-level inferential claims.

Counter-argument 2: “Participatory methods are always empowering”

Participation can reproduce inequality:

those with confidence and social capital speak more,
facilitators can impose agendas,
participation can become tokenistic.

Exam defence:

Specify participatory governance and inclusion measures.
Describe how facilitation reduces steering.

Counter-argument 3: “Mixed methods solves all methodological problems”

Mixed methods combine strengths, but integration challenges remain:

contradictory findings can confuse,
different epistemic assumptions can complicate interpretation.

Exam defence:

Provide an integration plan (e.g., explanatory sequential design).
Describe how contradictions will be analysed rather than ignored.

Development relevance: turning findings into recommendations

A methodology that produces uninterpreted data is incomplete. Recommendations should be:

grounded in findings,
linked to mechanisms and stakeholders,
feasible within resource and governance constraints,
sensitive to unintended consequences.

A practical recommendation template:

Finding (what you observed)
Mechanism (why it occurs)
Recommendation (what to do)
Stakeholders (who should act)
Implementation considerations (time, costs, partnerships)
Monitoring indicators (how to know it worked)

Example: translating evidence into an actionable plan

Suppose findings from Tshwane show:

youth struggle to access internships due to transport costs and unclear documentation requirements,
employers express concerns about credential credibility,
mentorship quality varies by facilitator training.

Recommendations could include:

micro-grants or transport stipends tied to internship attendance,
a standard “documentation checklist” communicated before application,
employer briefing sessions and verification pathways,
training and quality assurance for mentors.

Monitoring indicators:

internship attendance rates,
number of applications completed with required documentation,
employer satisfaction surveys,
mentorship fidelity scores.

This turns methodology into development practice.

Data management, security, and long-term evidence stewardship

In modern research, data management is part of ethical methodology. Examiners may ask:

how data will be stored,
how long it will be retained,
who can access it,
how it will be anonymised,
how recordings will be protected.

A defensible plan includes:

encrypted storage where possible,
restricted access through role-based permissions,
anonymised transcripts and removed identifiers,
secure storage of consent forms separately from data.

Research integrity: avoiding common academic and ethical failures

Research methodology includes integrity:

accurate citation,
transparent methodology,
honest reporting of limitations,
avoiding fabrication or selective omission,
maintaining consistent coding and reporting practices.

A common methodological integrity practice is keeping:

audit trails for qualitative coding,
data cleaning logs for quantitative analysis,
version control for instruments and datasets.

Practitioner mindset: communicating findings to non-academic stakeholders

Development practitioners often need to present evidence to:

municipal officials,
NGOs and programme partners,
community structures (e.g., ward committees),
funders and donors.

Good communication practices:

present findings in accessible language,
use visuals (charts, maps, summary infographics),
provide an executive summary,
include limitations so decisions remain responsible.

If presenting sensitive findings, consider:

confidentiality protections,
careful phrasing,
focusing on system-level improvements rather than individual blame.

Final consolidation: an exam checklist for UJ DEVHONS methodology strength

Use this quick checklist when preparing for exams or drafting your methodology chapter:

Research question clarity: specific population, context, and outcome/mechanism focus.
Design fit: method matches purpose (descriptive vs explanatory vs evaluative).
Sampling justification: technique aligns with question and analytic needs.
Operationalisation: concepts defined in measurable/analysable terms.
Data collection rigour: pilot testing, training, language and translation plan.
Ethics: consent, confidentiality, risk mitigation, data management plan.
Analysis plan: clear qualitative coding strategy and/or quantitative statistical approach.
Validity/trustworthiness: credibility strategies or measurement reliability.
Integration (if mixed methods): explicit logic for combining strands.
Development relevance: findings lead to actionable recommendations and monitoring indicators.
Limitations and delimitations: honest, bounded, and analytical.

Closing note on mastery for development practitioners

Strong research methodology for development practitioners is not simply technical competence; it is disciplined, ethical, and context-aware evidence building. For UJ DEVHONS students, the goal is to demonstrate that your research design can produce credible findings and that your analysis can translate evidence into improvements in development outcomes. When you can show coherence between your question, design, sampling, ethics, analysis, and application, you have built the methodological foundation expected at honours level—ready for both academic evaluation and real-world development decision-making.