Exercises
Practice problems sized for a coffee break. Each states the question a programme actually asked and expects an answer you can defend.
A cluster presentation ranks twelve communes on malnutrition, colours the worst three red, and truncates the axis so the gap looks decisive. Nothing on the slide is factually wrong. Find the four decisions that made it dishonest and produce the version that is not.
A report says 12.4% and the analyst has left. The notebook produces 11.8%, the spreadsheet says 12.9%, and the only thing everyone agrees on is that the report is the one the minister saw. You have an afternoon to work out which is right.
A board pack says the feeding programme raised literacy by 7 points and recommends doubling it. The number is right, the counterfactual is missing, and you have an hour before the meeting to work out what to say instead.
A note claims TSFP cures better than OTP — 75.3% against 72.6%. Adjust for admission MUAC and nothing changes. Then look at where the two programmes overlap, and find 106 children with no comparator at all.
The cluster reports 76% water coverage and the district reports 82% from the same survey. Both are defended, both are computed correctly, and the meeting is in an hour. Work out which is right, and write the paragraph that stops it happening again.
The epidemic curve is due at four o'clock. Forty-seven of 975 cases have no onset date and they are not spread evenly. Draw the curve three ways, say which one you would send, and what the missing dates could still be hiding.
A funding proposal reports a 9.7-point literacy gain at p < 0.001. The baseline paper had 40 items and the endline paper had 50, 156 of the weakest students never sat the second one, and the test treated the same children as two independent samples.
Four organisations compute the same indicator from the same register and publish four different numbers. Nobody made an error. Reproduce all four, then write the reference sheet that would have prevented the meeting.
A slide says penta3 coverage is 78%. Answer the six interrogation questions from the data, find the two that the extract cannot answer, and write the provenance block that should have been on the slide.
A donor wants to know why maize is up 113% since the baseline. The baseline was taken at the harvest and the follow-up at the lean peak. Separate the calendar from the deterioration, and write the reply.
A supervisor wants to know why one office loses contact with so many cases. The answer is in a column the monthly report never used, and getting to it means expanding 1,108 cases across the months they were open.
271 chlorine readings are the string "<0.1". Someone has coerced the column to numeric and reported 46.5% compliance. Reconstruct both figures, say which is right, and write the two sentences that go under the table.
A donor asks why the draft report said 73.6% and the final one says 82.3%. Nothing in the data changed. Reconstruct all three denominators, say which Sphere means, and write the answer.
A dashboard ranks twenty-four schools on attendance and two sit twenty points below the rest. A dropout response is being planned for them. Find out what actually happened before it is funded.
A colleague's analysis script produces four estimates from the household survey. Every one is computed correctly and every one is wrong, in four different ways. Find each, fix it, and quantify what it cost.
Five findings from a colleague's draft. Two are real, two are chance dressed as evidence, and one is a district failure written up as twenty-seven facility failures. Test each against the data and rewrite the survivors.
Four dplyr pipelines, each returning a plausible number that answers a different question from the one asked. Find the leftover grouping, say what each actually computed, and rewrite it.
One CSV, six defects, and a read that has to survive all of them. Write the read_csv call and the four checks that prove it worked, then say what each defect would have cost.
You inherit a cleaning script with six undocumented steps and a coverage figure that has already been reported. Reconstruct the log, measure what each step cost, and identify the two steps that are wrong.
A ranking sends a support team to the three worst-attending schools. All three are there because of a join decision, and the school that genuinely has the worst attendance is fourth.
Five reporting scenarios where the numerator is agreed and the denominator is not. Choose one, justify it, and state what your choice excludes.