Find missing voices in a sample
Define the population and the list or process you use to reach it. Report who was excluded, who did not respond and how the missing groups could change your decision.
Worked example
A phone survey reaches 100 shop owners and 70 express interest. If the list omits evening-only kiosks, 70% describes respondents. It cannot estimate all local merchants without evidence about the missing group.
Change the assumption
A follow-up survey reaches 20 previously omitted kiosks and only four express interest. Should you merge both percentages directly?
Check the reasoning
The groups differ in coverage and may have different population shares. Report their results separately and use a justified population weight if you need a combined estimate.
Check your reasoning
Would calling more listed owners solve the problem?
Reveal the explanation
It can improve precision within the list. Add a storefront validation sample to cover merchants the original list misses.
Try the same framework in a different case
Explore worked stories in the reading room · Practise the next analysis