Teaching · For Teachers

How many students
do you need to ask?

A school survey is only as good as the way people were chosen to answer it. Enter the population, the method and the precision you want. The guide works out the responses needed for a proportion, corrects for a small population and for clustering, turns that into invitations at your expected response rate and says what each method needs to be done properly.

Plan a sampleWhat you get
$12one-time · every device you sign in on
Buy · $12

Included with Caveat Classroom and Whole-school licenses. Or get it in the For Teachers toolkit, $79 for 79 items.

Study plans

The worked example is made up: a stratified survey of 480 ninth graders, a staff survey where inviting all 140 is simpler, and a cluster sample of 2,600 families by homeroom. Saved on this device only.

Loading the study form. If nothing appears, turn on JavaScript.

What you get

Teaching · For Teachers$12 · included with Classroom and Whole-school licenses

What you finish with
A record per study: question, population and size, method, margin of error, confidence level, expected response rate, design effect and five planning checks. Works out the responses needed for a proportion (with the finite population correction and design effect) and the invitations to send; flags when a census is simpler, when results cannot be generalized and what each method needs.
Who it is for
Teacher-researchers, school leaders, instructional coaches and graduate students in education running surveys or studies.
Format
Browser guide with a saved record per study. Prints one plan with its numbers, every plan in a table or a blank form; downloads a spreadsheet (.csv); exports and imports a file.
Access
$12 one-time, yours on every device you sign in on; the tool opens once you are signed in with it. Included with Caveat Classroom and Whole-school licenses. Also in the For Teachers toolkit ($79) and the Teaching Suite pass ($199), and 1 more bundle. Everything you type stays in this browser; nothing is sent to Caveat.
Where it lives
The For Teachers space in Caveat Teaching.
Support
Questions about access, an order or a correction: team@caveatacademic.com.

How to use it

  1. Write the question and population.
  2. Choose a method from the table.
  3. Set the margin and confidence.
  4. Estimate the response rate honestly.
  5. Invite the number shown, then follow up.

Plan a sample

Choosing a method

Six ways to choose who answersMethodBest whenSimple randomYou have a full list and no subgroupsSystematicA full list with no repeating patternStratifiedSubgroups must each be representedClusterPeople come in groups, such as classesConveniencePiloting; results describe only takersPurposiveInterviews with chosen informantsSix ways to choose who answersSimple randomBest when: You have a full list and no subgroupsSystematicBest when: A full list with no repeating patternStratifiedBest when: Subgroups must each be representedClusterBest when: People come in groups, such as classesConvenienceBest when: Piloting; results describe only takersPurposiveBest when: Interviews with chosen informants
Only the first four let you attach a margin of error to the result.

The formula

Responses needed for a proportion: n₀ = z² × 0.25 ÷ e², using p = 0.5 (the most cautious case) and z of 1.645, 1.96 or 2.576 for 90, 95 or 99 percent confidence. For a population of N, n = n₀ ÷ (1 + (n₀ − 1) ÷ N), then multiplied by the design effect for a cluster sample and rounded up.

Response rate is the weak point

Inviting more people does not fix a low response rate: if the people who do not answer differ from those who do, the result is biased however many answer. Reminders, a short survey and time in class do more than a larger invitation list.

Sources

  1. Cochran, W. G. (1977). Sampling Techniques, 3rd ed. Wiley: sample size for proportions and the finite population correction.
  2. Kish, L. (1965). Survey Sampling. Wiley: design effects for cluster samples.
  3. Lohr, S. L. (2022). Sampling: Design and Analysis, 3rd ed. CRC Press: stratified, systematic and cluster sampling.
  4. American Association for Public Opinion Research. Margin of sampling error and nonprobability samples (aapor.org).

The worked example is made up. The numbers assume a simple estimate of a proportion; other analyses need their own power calculations. Nothing you type is sent to Caveat. Support: team@caveatacademic.com.