Research Methods, Data Analysis and Report Writing Training Course

Participants can translate a management problem into focused research questions and objectives.

• Problem statements separate observed conditions from assumed causes
• Stakeholder analysis identifies evidence users, interests and decision authority
• Research questions framed using population, phenomenon, context and timeframe
• SMART research objectives specify measurable outputs and intended use
• Conceptual frameworks identify plausible relationships between key variables
• Scope boundaries define geographic coverage, participant groups and exclusions

Participants can build a testable theory of change and aligned results framework.

• Theory of Change maps causal pathways from activities to intended outcomes
• Assumptions and risks documented at each causal link
• Logical Framework Approach aligns objectives, indicators, verification and assumptions
• Outcome statements describe measurable changes in behaviour, practice or condition
• External factors distinguished from programme-controlled implementation factors
• Results chains tested for plausibility, sequence and measurable contribution

Participants can specify valid indicators and prepare an indicator reference sheet.

• Indicator definitions specify numerator, denominator, unit and disaggregation
• Output, outcome and impact indicators matched to results statements
• SMART criteria used to test indicator utility and feasibility
• Baseline, target and milestone values established using available evidence
• Indicator reference sheets record source, frequency, responsibility and limitations
• Disaggregation plans address sex, age, disability, location and vulnerability

Participants can select a research design appropriate to the question, resources and ethical constraints.

• Cross-sectional designs estimate conditions at a single point
• Longitudinal designs track change through repeated measurements over time
• Experimental and quasi-experimental designs assess plausible programme effects
• Case study designs investigate processes, context and implementation mechanisms
• Mixed-methods designs combine numerical patterns with explanatory accounts
• Design choices balance validity, feasibility, cost, timing and ethical risk

Participants can prepare a practical sampling plan and justify its limitations.

• Target populations, sampling frames and study units clearly defined
• Probability sampling uses random, systematic, stratified and cluster methods
• Non-probability sampling applies purposive, quota, snowball and convenience approaches
• Sample size considers precision, confidence level, variability and design effect
• Stratification improves representation of priority groups and locations
• Sampling plans document selection procedures, replacements and non-response handling

Participants can construct a structured questionnaire that produces analysable data.

• Questionnaire flow moves from screening questions to sensitive topics
• Closed questions use mutually exclusive and collectively exhaustive response categories
• Likert scales measure attitudes with balanced response options
• Recall periods reduce memory error in household and beneficiary surveys
• Skip logic prevents inapplicable questions and inconsistent respondent pathways
• Pilot testing identifies ambiguity, burden, missing options and sequencing problems

Participants can design interviews, focus groups and observations for credible qualitative evidence.

• Semi-structured interview guides use open questions and focused probes
• Focus group discussions explore shared norms, differences and group dynamics
• Key informant selection identifies informed perspectives and institutional knowledge
• Observation checklists record behaviour, conditions, interactions and service standards
• Field notes distinguish direct observations from researcher interpretations
• Question sequencing builds rapport before discussing sensitive or contested issues

Participants can apply ethical safeguards throughout data collection, analysis and reporting.

• Informed consent explains purpose, procedures, risks, benefits and voluntary participation
• Confidentiality procedures protect identities in records, datasets and reports
• Do-no-harm assessment identifies foreseeable physical, social and emotional risks
• Safeguarding protocols define referral actions for disclosures and protection concerns
• Data minimisation limits collection to information necessary for stated purposes
• Ethical review requirements assessed before fieldwork begins with vulnerable populations

Participants can plan fieldwork controls that improve completeness, accuracy and timeliness.

• Fieldwork plans assign roles, schedules, locations, transport and supervision
• Enumerator training covers tool intent, probing, neutrality and consent procedures
• Back-checks verify interviews through selected respondent follow-up contacts
• Spot checks observe data collection against approved procedures
• Data-quality assessments examine validity, reliability, precision, integrity and timeliness
• Issue logs record errors, corrective actions, ownership and closure dates

Participants can prepare a documented, analysable dataset from raw research data.

• Data dictionaries define variables, labels, codes, formats and allowable values
• Unique identifiers link respondent records while protecting personal information
• Range checks identify values outside plausible or permitted limits
• Missing-data codes distinguish refusals, unknown responses and inapplicable items
• Duplicate records detected through identifiers, timestamps and matching fields
• Cleaning logs document edits, rationale, source files and version control

Participants can summarise quantitative data and select clear charts for intended users.

• Frequency distributions summarise categorical responses and response completeness
• Means, medians and modes describe central tendency appropriately
• Ranges, interquartile ranges and standard deviations describe variation
• Percentages calculated using valid denominators and clearly stated bases
• Bar charts, line charts and histograms match data types
• Tables present totals, disaggregation and rounding without misleading readers

Participants can compare groups and interpret basic statistical tests without overstating findings.

• Cross-tabulations compare outcomes by sex, location, age and participation status
• Chi-square tests assess associations between categorical variables
• T-tests compare mean values between two defined groups
• Confidence intervals express precision around survey estimates
• P-values interpreted alongside effect size, sample size and context
• Statistical significance distinguished from operational and policy significance

Participants can code qualitative material and develop defensible themes from participant accounts.

• Transcripts and field notes prepared for systematic coding
• Deductive codes derived from research questions and conceptual frameworks
• Inductive codes capture unanticipated issues emerging from participant accounts
• Codebooks define labels, inclusion criteria, exclusions and illustrative examples
• Thematic matrices compare patterns across participant groups and locations
• Negative cases examined before finalising explanatory themes and conclusions

Participants can integrate quantitative and qualitative evidence into coherent findings.

• Convergent designs compare quantitative and qualitative findings collected concurrently
• Explanatory designs use interviews to interpret statistical patterns
• Joint displays combine numerical results, themes and interpretation
• Triangulation checks convergence, complementarity, contradiction and evidence gaps
• Discrepant findings investigated through sampling, measurement and context review
• Integrated conclusions distinguish observed evidence from plausible interpretation

Participants can prepare an evaluation plan, terms of reference and review process.

• Evaluation purposes distinguish accountability, learning, improvement and strategic decisions
• Terms of reference specify scope, questions, methods, deliverables and governance
• Evaluation matrices align questions, indicators, sources, methods and analysis
• OECD DAC criteria operationalised through context-specific evaluation questions
• Independence, conflicts of interest and evaluator competence assessed early
• Inception reports confirm methodology, workplan, limitations and stakeholder engagement

Participants can develop evidence-based conclusions and recommendations that management can use.

• Findings present verified evidence without premature judgement or speculation
• Conclusions answer research questions through reasoned synthesis of findings
• Recommendations specify action, owner, timeframe, resources and expected result
• Prioritisation matrices rank recommendations by importance, feasibility and urgency
• Evidence strength assessed through source quality, consistency and limitations
• Management responses record acceptance, actions, accountability and follow-up dates

Participants can structure and write a report that meets management and donor requirements.

• Executive summaries state purpose, methods, key findings and priority actions
• Methods sections document design, sampling, tools, analysis and limitations
• Results chapters organise evidence around questions, criteria or themes
• Charts and tables support claims with titles, sources and interpretation
• Limitations explained honestly without weakening valid conclusions unnecessarily
• Annexes include tools, detailed tables, terms and reference materials

Participants can present findings to technical and non-technical audiences with appropriate evidence.

• Audience analysis identifies decision authority, technical knowledge and information needs
• Briefing notes condense findings, implications and requested management decisions
• Slide narratives lead with conclusions before supporting methodological detail
• Data visualisation choices prevent misleading comparisons and unsupported claims
• Facilitated validation meetings test interpretation with stakeholders and implementers
• Feedback incorporated while preserving methodological independence and evidence integrity

Participants can build, review and present a complete research or evaluation package for a defined programme decision.

• Decision brief defines the management question, users and intended action
• Theory of Change and indicator framework establish the evidence model
• Sampling plan and data-collection tools address defined research questions
• Cleaned dataset or coded evidence base supports documented findings
• Analysis tables, themes and triangulation produce defensible conclusions
• Report draft presents methods, findings, recommendations and stated limitations
• Peer review checks methodological coherence, evidence quality and report usability
• Management presentation communicates priority actions and implementation implications