Skip to content
ConductExperience
Menu

Quota sampling

Quota sampling fills a target number of people for each group you name, such as 50 coffee drinkers and 50 tea drinkers. People take part as they arrive, until a group is full. Then the next person who belongs to it is turned away.

In ConductExperience a quota counts the answers to one question of a screener, across all of a study's links.

Try a quota

The example has a quota of 2 for Coffee and 2 for Tea, and none for Water. Choose an answer and press Continue, as the next person to reach the link. The product's own screener decides each person, and the tables show what the Links page and the download hold.

A screener with a quota of 2 for Coffee and 2 for Tea

Which do you drink most?

Each time you press Continue, the next person reaches the link and gives that answer. Nothing you do here is saved or sent.

The person who reaches the link next sees this, once they press Continue.

What the Links page shows

AnswerFinishedIn progressTarget
Coffee002
Tea002
Water00No limit

The people so far

The screener's columns of participants.csv, a row for each person, turned away or not.

participant_codestatusscreener
No one yet.

What quota sampling is

You choose the groups and a target for each. A person is counted in their group when they take part, and a group that has reached its target takes no one else. It is a non-probability method: it controls how many people each group has, not who they are.

Quota sampling and stratified random sampling

Quota sampling takes the people who arrive until each group is full. Stratified random sampling draws people at random within each group, from a list of everyone in the population.

CompareQuota samplingStratified random sampling
Needs a list of the populationNo.Yes: the list is the sampling frame.
What it controlsHow many people each group has.How many people each group has, and who in it is asked, by chance.
What you can say of a groupIt has the number you set. Its people are not a random draw.Each person in it had a known chance of being chosen, so the usual margin of error applies.

Quotas in ConductExperience

  • A quota sets a target for each answer of one required question with fixed answers, across all the study's links. When an answer's quota is full, people who give it are turned away: they read one sentence, and a panel participant goes back to the quota-full address.
  • Participants never see the screener's rules, so they cannot work out which answer gets them in. Two people who race for a quota's last place cannot both get it, and a person turned away stays turned away.
  • The Links page shows each answer's Finished, In progress and Target, and the Participants page marks who was turned away.
  • A quota can count a question of three kinds: a single choice, a yes or no, or a rating scale. An answer with no target has no limit.
Writing screener questions, with rules and quotas

When a quota is full

  • A person who gives an answer whose quota is full reads one sentence. Unless you write your own, it says: “This study already has enough people with your answers, so it ends here.”
  • A person from a panel goes to the link's Quota-full address. Without one, they go to the screen-out address.
  • A person turned away holds no place: the Links page does not count them.

What the download holds

  • participants.csv has a row for each person, turned away or not.
  • Its status column says turned_away for a person the screener turned away, and its screener column says fits, turned_away or quota_full.
  • The Links page shows each answer's Finished, In progress and Target, as in the example.

Build a quota

  1. Start a study, and put a questionnaire first. It holds the question the quota counts.
  2. Write that question with fixed answers, and make it required and shown to everyone: a single choice, a yes or no, or a rating scale.
  3. Choose Use as a screener on the part. Under it, pick the Quota question and set a Target for each answer.
  4. Preview the study as a participant, publish it and make a link. A quota counts across all of a study's links.
Links, panels and the screen-out and quota-full addresses

Limits

  • A target is up to 10,000 people.
  • A screener runs on links. In the lab and on tablets, it runs as a plain questionnaire.
  • The quota question must be required and shown to everyone: no rule hides it, no end rule comes before it, and it does not repeat, offer only some options or sit on a screen drawn at random.
  • A quota counts the answers of one question. It sets how many people each group has, and does not make a sample represent a population.

Related

Planning a study?

Start one now, or tell us what you want to measure.