Skip to main content

Get Survey Results

GET/v2/project/{project_key}/survey/{survey_id}/results

The survey's report: the funnel and every question's answers, filtered to a group or compared across groups.

Query parameters​

ParameterTypeRequiredDescription
filtersJSON arraynoOnly people who gave all of these answers, up to 5. Each { "question", "is" }: a question number and the answer: a choice answer's id, "yes" / "no", an NPS group ("promoter", "passive", "detractor") or a score. Open-text questions can't be filtered on
start_datestringnoYYYY-MM-DD, in the timezone of the tz header (hours from UTC, default 0)
end_datestringnoYYYY-MM-DD
segmentstringnoA quick segment, as on the experiment report: new, returning, desktop, mobile, direct, referral, social, or country-XX
compareintegernoA question number. Every other question's answers are split by that question's answers, for example students versus instructors. Works with choice, Yes / No, NPS and rating questions

Example​

What did instructors say? That is question 1, answer id 1:

curl -G "https://api-{region}.mida.so/v2/project/YOUR_PROJECT_KEY/survey/42/results" \
  -H "Authorization: Bearer YOUR_GENERATED_API_KEY" \
  --data-urlencode 'filters=[{"question":1,"is":1}]'

Success response​

{
  "success": true,
  "funnel": { "shown": 2, "answered": 2, "completed": 2 },
  "questions": [
    {
      "question": 1,
      "qid": 1,
      "type": "choice",
      "text": "Are you visiting as a student or instructor?",
      "n": 2,
      "multi": false,
      "options": [
        { "id": 0, "text": "Student", "n": 0 },
        { "id": 1, "text": "Instructor", "n": 2 }
      ]
    },
    { "question": 2, "qid": 2, "type": "yesno", "text": "Would you use a free Education Plan?", "n": 0, "yes": 0, "no": 0 },
    { "question": 3, "qid": 3, "type": "yesno", "text": "Would you teach your class with Mida?", "n": 2, "yes": 1, "no": 1 }
  ],
  "written_answers": [],
  "filters": [{ "question": 1, "is": 1 }]
}
  • funnel.shown: visitors the survey was shown to. answered: answered at least one question. completed: answered every question on their own path. In a survey that branches, that path depends on their answers. An optional open-text question can be skipped.
  • Each question carries n, the number of answers, plus its type's counts: options[].n for choice (a multiple-choice answer counts once per pick), yes / no, counts by score with mean for ratings, and nps with promoters, passives and detractors for NPS.
  • written_answers: the newest open-text answers, and what people wrote after a choice with ask.

With compare, the response also has:

"compare": {
  "question": 1,
  "groups": [
    { "key": "o0", "label": "Student", "is": 0, "n": 6 },
    { "key": "o1", "label": "Instructor", "is": 1, "n": 2 }
  ],
  "questions": {
    "2": { "o0": { "type": "yesno", "n": 3, "yes": 1, "no": 2 }, "o1": { "type": "yesno", "n": 0, "yes": 0, "no": 0 } }
  },
  "note": "Groups under 30 people are too small to compare reliably."
}

questions is keyed by question number, then by group key. A group under 30 people is too small to compare: report its counts ("1 of 3 students") rather than a percentage.

Error responses​

StatusMeaning
400A filter or compare that names no question, an answer that question doesn't have, an open-text filter, more than 5 filters, or an unknown segment
401Invalid or missing API key
403Surveys need a paid plan, or the key's user is a Viewer (for changes)
404Project or survey not found. A survey from another project is not found