Turn census, household-survey, ward-level or programme data into a finding that can survive a policy or development review. This skill gives a B.Plan graduate the reasoning layer behind defensible analysis: decide what the data can support, classify variables, clean and summarise them, compare groups, chart findings, account for sampling and uncertainty, and interpret association or regression without claiming causation. It is tool-agnostic and complements Statistical tool (R), which teaches how to execute an analysis in R.
Needed across
What you will be able to answer
A development-programme team gives you a household survey from two pilot districts and asks whether a sanitation intervention improved school attendance. What can you produce before the review meeting?
A short analytical note that first identifies the unit of observation, sampling frame, missing groups and variable types, then cleans and documents the table before calculating anything. It reports attendance and programme exposure with suitable summaries, cross-tabs and an honest chart; places intervals around sample estimates; and distinguishes a statistically detectable difference from a practically important one. Any correlation or regression result is stated in units and as an association, with plausible confounders such as household income and district selection named explicitly. Unless the study design supports causal inference, the note does not claim that the programme raised attendance.
One payment
₹99
The videos are free
Course outline
Concept 1
Concept 1 · What a dataset can and cannot tell you
Concept 2
After: what-the-data-can-answer
Nominal, ordinal and interval/ratio, and how the type rules out summaries and charts before you pick one.
Video title and channel appear once unlocked.
Concept 3
After: variables-and-measurement
Messy phone numbers standardised with regex replacement, and the data-type problem underneath them.
Video title and channel appear once unlocked.
Concept 4
After: variables-and-measurement
Standard deviation and the 68-95-99.7 rule worked on real adult height data.
Video title and channel appear once unlocked.
Concept 5
After: describing-a-distribution
Row percentages and column percentages off the same table, answering two different questions.
Video title and channel appear once unlocked.
Concept 6
After: comparing-groups
A line chart used on categorical data, and the false trend it invents.
Video title and channel appear once unlocked.
Concept 7
After: describing-a-distribution
Simple random against stratified, and how non-response quietly biases the result.
Video title and channel appear once unlocked.
Concept 8
After: sampling
A 95% confidence interval interpreted correctly, with the common misreading stated and refuted.
Video title and channel appear once unlocked.
Concept 9
After: uncertainty-and-intervals
A two-tailed test worked through on recovery times, from null hypothesis to p-value.
Video title and channel appear once unlocked.
Concept 10
After: charting-a-finding
A confounder worked in full — age driving both shoe size and reading ability.
Video title and channel appear once unlocked.
Concept 11
After: correlation, hypothesis-testing
A multiple regression equation read out loud, coefficient by coefficient.
Video title and channel appear once unlocked.
11 candidates did not meet the course criteria.
Planner-economist roles at multilaterals and development organisations (UN-Habitat, World Bank, ADB, GIZ, WRI, C40) working on urban policy, climate and infrastructure programs, in India or abroad. Rare for a fresh B.Plan directly — usually a master's + experience; freshers enter via internships and research-associate roles at India offices.
8 mapped employers
Explore path →