Research Design, SurveyCTO Data Collection and Data Analysis Training Course

Participants can construct research questions and hypotheses aligned with a programme theory of change.

• Theory of change pathways linking activities, outputs, outcomes and impacts
• Research questions structured using population, variables, location and timeframe
• Causal assumptions, external factors and rival explanations documented explicitly
• Null and alternative hypotheses written for measurable programme relationships
• Conceptual frameworks used to identify independent, dependent and confounding variables
• Contribution analysis questions distinguished from experimental attribution claims

Participants can specify measurable indicators and select appropriate measurement approaches.

• Indicator reference sheets defining numerator, denominator, source and frequency
• Output, outcome and impact indicators matched to theory-of-change levels
• Disaggregation plans covering sex, age, disability, location and vulnerability
• Validity, reliability, sensitivity and feasibility assessed before indicator selection
• Likert scales, indices and composite measures designed for respondent surveys
• Standard indicator definitions reviewed against donor and sector guidance

Participants can design ethical field procedures that protect respondents and sensitive data.

• Informed consent scripts covering purpose, voluntariness, risks and withdrawal
• Do-no-harm assessment applied to sensitive household and protection questions
• Safeguarding referral protocols prepared for disclosures during interviews
• Data minimisation principles applied to personally identifiable information collection
• Role-based access controls specified for forms, servers and exported datasets
• Anonymisation, pseudonymisation and secure deletion procedures compared for survey records

Participants can select and document an appropriate sampling strategy for a field study.

• Sampling frame assessed for coverage errors, duplicates and missing groups
• Simple random, systematic and stratified sampling procedures compared
• Cluster and multistage sampling used for dispersed household populations
• Purposive, quota and convenience sampling limitations stated transparently
• Strata defined using programme geography, beneficiary groups and key characteristics
• Replacement rules documented before fieldwork begins to prevent enumerator bias

Participants can estimate sample sizes and explain the precision and limitations of survey results.

• Confidence level, margin of error and expected proportion explained
• Sample-size calculations performed for proportions, means and population comparisons
• Design effect applied when cluster sampling replaces simple random sampling
• Finite population correction considered for small beneficiary populations
• Non-response inflation added using realistic contact and refusal assumptions
• Power, effect size and minimum detectable difference introduced for evaluations

Participants can draft respondent-centred questionnaires that produce analysable data.

• Questionnaire flow sequenced from introductions through sensitive closing questions
• Open, closed, ranked and matrix questions selected by measurement need
• Double-barrelled, leading and ambiguous wording identified and rewritten
• Recall periods matched to respondent memory and programme exposure
• Response categories made mutually exclusive, exhaustive and analytically useful
• Translation and back-translation procedures planned for multilingual surveys

Participants can create a structured XLSForm workbook for SurveyCTO deployment.

• XLSForm survey, choices and settings worksheets configured correctly
• Question types selected for text, integers, decimals, dates and selections
• Choice lists coded with stable values and human-readable labels
• Variable names follow consistent conventions for exports and analysis
• Form metadata captures submission time, device identifiers and enumerator names
• SurveyCTO form upload process and version control procedures applied

Participants can build SurveyCTO logic that guides interviews and prevents invalid entries.

• Relevance expressions apply skip patterns from respondent eligibility answers
• Required fields configured without forcing inappropriate sensitive disclosures
• Constraint expressions prevent impossible ages, dates and numeric values
• Calculation fields derive ages, totals, scores and eligibility flags
• Groups and field-list appearance improve interview flow on mobile devices
• Hints, guidance messages and constraint messages support enumerator corrections

Participants can configure advanced form elements for complex monitoring and assessment instruments.

• Repeat groups capture household members, facilities or distributed items
• Roster designs link repeat records to household-level identification variables
• Cascading selects filter administrative units, sites and service options
• Media fields collect photographs, audio files and scanned documents
• Barcode and QR code scanning support asset and distribution verification
• Randomisation functions reduce ordering effects in experimental questionnaire modules

Participants can test, document and approve mobile forms before field deployment.

• Desk testing checks labels, codes, skips, constraints and calculations
• Scenario-based testing follows eligible, ineligible and exceptional respondent paths
• Pilot testing identifies timing, comprehension and translation problems
• Test cases record expected results, observed results and corrective actions
• User acceptance criteria agreed with programme, M&E and field teams
• Form change logs preserve version history and approval decisions

Participants can prepare operational plans for controlled and efficient mobile data collection.

• Fieldwork plans assign locations, caseloads, routes and supervision arrangements
• Enumerator training agendas cover consent, probing, neutrality and device use
• Daily briefing and debriefing routines address emerging fieldwork issues
• Device inventory procedures record assignment, charging, security and return
• Contact attempts and call-back rules protect sample integrity
• Incident escalation pathways cover safety, safeguarding and technical failures

Participants can manage form deployment, user access and incoming data submissions.

• Server folders organise forms, datasets, media files and team access
• User roles restrict form publishing, data viewing and export permissions
• Form publishing procedures distinguish draft, pilot and production versions
• Submission monitoring tracks completion rates, timestamps and enumerator productivity
• Data pull settings retrieve new submissions without duplicate records
• Audit trails support review of edits, corrections and submission history

Participants can detect, investigate and correct common survey-data quality problems.

• Data-quality dimensions assessed: completeness, accuracy, consistency, timeliness and uniqueness
• Range checks, logic checks and duplicate checks applied to exports
• Enumerator outlier analysis identifies unusual durations, patterns and response distributions
• Missing-data codes distinguished from blanks, refusals and not-applicable values
• Cleaning logs document corrections, reasons, source records and approval
• Raw, cleaned and analysis datasets stored as separate controlled versions

Participants can collect location data and create basic maps that support survey interpretation.

• GPS points, lines and polygons selected for appropriate geographic features
• Coordinate systems and latitude-longitude formats checked before mapping
• Location accuracy, privacy and security risks assessed before GPS capture
• Spatial joins connect survey records with administrative boundary datasets
• Choropleth maps display rates, proportions and index values correctly
• Point maps reveal service access, coverage gaps and geographic clustering

Participants can prepare documented analysis datasets from SurveyCTO exports.

• SurveyCTO exports generated for Excel, CSV, Stata and statistical workflows
• Codebooks document variable labels, value labels, formats and derivations
• Wide and long data structures distinguished for repeat-group analysis
• Categorical variables recoded into analysis-ready groups using documented rules
• Derived variables calculate scores, thresholds, rates and programme exposure
• Data dictionaries reconcile questionnaire variables with indicator reporting requirements

Participants can produce and interpret descriptive analyses for programme monitoring questions.

• Frequency tables summarise categorical responses, missingness and respondent profiles
• Means, medians, modes and standard deviations describe continuous variables
• Cross-tabulations compare outcomes by sex, age, location and beneficiary group
• Percentages use correct denominators and state excluded response categories
• Confidence intervals communicate uncertainty around estimated survey proportions
• Data visualisations select appropriate bar charts, histograms and line charts

Participants can select basic statistical tests and interpret results without overclaiming.

• Chi-square tests assess associations between two categorical variables
• t-tests compare mean outcomes between two independent respondent groups
• Correlation coefficients describe direction and strength of linear relationships
• P-values, confidence intervals and practical significance interpreted together
• Confounding variables considered before asserting programme-related outcome differences
• Analysis limitations stated for sampling error, bias and missing data

Participants can communicate survey findings in formats suited to programme and donor decisions.

• Analysis plans map research questions to variables, methods and outputs
• Findings statements distinguish observed results from interpretation and recommendation
• Results tables present denominators, disaggregation and uncertainty consistently
• Executive summaries prioritise decision implications over technical detail
• Data visualisations follow accessibility principles for colour, labels and annotation
• Evidence-use plans assign decisions, owners, deadlines and follow-up actions

Participants can build, review and present a complete research design, SurveyCTO data-collection package and analysis plan.

• Decision question, research objectives and theory of change presented coherently
• Indicator matrix, sampling strategy and sample-size assumptions justified clearly
• SurveyCTO form demonstrates logic, constraints, repeats and metadata capture
• Fieldwork plan addresses consent, supervision, devices and quality assurance
• GPS and mapping approach explains geographic data use and protections
• Analysis plan specifies cleaning, tables, tests, charts and disaggregation
• Peer review identifies methodological risks, form errors and reporting gaps