Path
Research Officer
For the officer running surveys rather than reading them — sampling, estimation with the design accounted for, and the plausibility checks that decide whether a result can be published at all.
Competencies
Sampling and design
AdvancedChoose a design that can answer the question asked, and state the precision it will deliver before fieldwork starts.
Design-aware estimation
AdvancedCompute an estimate and an interval that account for clustering, and know what the design effect is doing to your conclusion.
Data quality assurance in the field
AdvancedRun the plausibility checks that detect a mis-set scale, a rounding habit or a team measuring differently from the others.
Publication judgement
AdvancedDecide whether a result can be published, must be qualified, or requires the fieldwork repeated.
Before you start
- A grounding in survey methods — sampling, questionnaire design, non-response — which this path applies rather than teaches.
- No programming background. The language stage starts at the install.
The route through
Stage 1
Ground the numbers
Separate what a survey estimates from what it measures, and state the disaggregation before the fieldwork rather than after it.
Stage 2Choose one
Pick the language your team uses
Go deep in one of Python or R. R is the more common choice in survey work and the one the WHO growth-standard packages are written for; Python is the better answer if your team already has one.
These are alternatives, not a sequence. Take the one your team already uses — the second is far cheaper to add once you have the first.
Stage 3
Make the data trustworthy
Turn a raw export into a table you would defend line by line, and hand over a cleaning log that answers the auditor's question before it is asked.
Stage 4
Put the files together
Assemble several exports into one analysis table you would defend column by column, with every join proved, every grain stated and every denominator sourced.
Stage 5
Check it like an auditor
Assess your own data before a donor does — five dimensions with measures, a recount against the source, and a report where every finding has an owner and a date.
Stage 6
Define what you report
Write indicator definitions two analysts compute the same way, defend the denominator, and set a baseline and target that survive a mid-term review.
Stage 7
Put an interval on it
Analyse a cluster survey the way its design requires — weights from the frame, a measured design effect, and an interval that says what the sample can and cannot settle.
Stage 8
Know where it came from
Read a routine reporting system as the database it is, pull an extract you can point at months later, and answer what was counted for any figure it produces.
Stage 9
Apply it to nutrition
Take the methods into one sector — WHO growth standards, SMART plausibility, the IPC phases and the Sphere performance thresholds, on a survey and a treatment register.
Stage 10
Apply it to public health
Rates with person-time denominators, treatment cascades, coverage three ways, and an outbreak line list turned into a curve, attack rates and a case fatality you can defend.
Stage 11
Apply it to WASH
The JMP service ladders and the Sphere minimums on a household survey, then a monitoring register with repeat visits that turns one functionality rate into three and bounds each against the rounds nobody drove.
Stage 12
Apply it to food security
FCS, HHS, rCSI and the livelihood coping module built from raw components and found to disagree by a factor of seven, then a price series whose seasonality dwarfs the programme effect and the evidence table that reports both.
Stage 13
Apply it to protection
The analyses you must decline to publish, alongside the ones that matter — a consent-gated referral pathway, a nineteen-point equity gap located at a specific gate, and a caseload that explains two other tables.
Stage 14
Apply it to education
Gross against net enrolment on a projected denominator, two attendance numbers twenty-six points apart, a cohort through promotion and repetition, and two assessment rounds whose instruments differ.
Stage 15
Say how sure you are
An interval on every proportion, the right test for a comparison, an effect size beside every p-value, and the count of comparisons that turns two striking schools back into noise.
Stage 16
Model more than one thing at once
A coefficient is a comparison — which one, between which units, adjusted for what. An odds ratio your reader will misread, a covariate that removes 42% of the effect, and a model that explains three per cent and settles a targeting decision.
Stage 17
Say what caused it
A seven-point gain that is entirely the school year, a comparison group imbalanced on every characteristic measured, and the minimum detectable effect that decided the answer before any data existed.
Stage 18
Put it in front of them
The mark the comparison implies, an interval that stops a ranking, a palette that already means something to this audience, and a figure generated from the dataset so the chart and the sentence cannot drift.
Stage 19
Make it rerunnable
Raw data read-only and code the only thing edited, an environment pinned so a colleague's laptop gives your numbers, checks that stop the pipeline rather than producing a plausible wrong one, and a handover a successor can act on.
Stage 20
Put it in front of the decision
A page that answers one question rather than twenty, the definition panel that stops the monthly argument, and the report whose findings, limitations and recommendation survive being read separately.
Recommended projects
How the skill is assessed
Can compute a prevalence with a confidence interval that accounts for the cluster design.
EvidenceAn estimate using a between-cluster variance, with the design effect reported.
Can run a plausibility assessment and reach a stated verdict rather than a list of statistics.
EvidenceA plausibility report ending in accept, qualify or repeat, with the reasoning.
Can detect measurement bias between enumeration teams and distinguish it from a real difference.
EvidenceA between-team comparison that considers and rules out assignment as an explanation.
The estimate is the easy half. A prevalence figure computed carelessly and one computed carefully often differ by less than the interval around either, and the decision that matters is usually not “what is the number” but “can this number be published”.
The SMART project in this path is built on that. Its headline estimate is 14.9% against an emergency threshold of 15%, with an interval that crosses the threshold — so the survey cannot answer the question it was commissioned for until the measurement question is settled. Two plausibility checks fail, in two different teams, and only one of the two faults moves the estimate.
Treating the design as an afterthought is the commonest way this work goes wrong. The same survey analysed as a simple random sample reports an interval about 1.5 times too narrow, which would place the whole interval below the threshold and produce a confident answer to the wrong question.
What this path still needs
Three courses are published: Data Analysis Foundations, Python for Programme Data and R and the Tidyverse for Programme Data. Sampling and estimation sit in Survey Analysis and Sampling under Indicators and Measurement, and the inference content in Statistics and Modelling — both on the programme roadmap and neither written yet. See the programme.