If you need to recruit a custom sample, and cannot obtain the specific population you require using our existing prescreening filters, it is not permitted to screen participants within your survey or study description. You can read more about why this is not allowed here.
Instead, you can run two separate studies on Prolific to recruit your sample.
Your first study would be a short study designed to filter your participants of interest. Your second study would be a longer one to actually collect the data. Please see below for a more detailed breakdown:
The first study
This should be a short screening survey, where participants answer your intended 'screening' questions and provide their Prolific ID. We recommend that your screening question is kept vague so as to not lead participants to answer in a certain way. For example, it would be better to ask 'Do you drink any of the following drinks?' than 'Do you drink [specific brand]?'
If you are located in the UK, EEA, Canada, New Zealand, Japan, Israel, Argentina, South Korea, or the USA, we recommend using our in-built Survey Builder to create your first study. Using this tool, you can run a short study of up to 5 questions directly from Prolific, making it easy to set up your initial screening study.
You can already prescreen the sample to some extent using our existing options - e.g. for any demographic requirements you may have.
You may need to survey quite a large number of participants in order to achieve your desired sample size for the main study. However, because this first study is very short, you can reward a small amount and therefore test more participants.
Importantly, you should approve every complete submission to this study, as it wouldn't be appropriate to reject people who do not meet your criteria.
After the first survey, you will have a list of participant IDs of those who meet your criteria. As always, we recommend that you automatically record participant IDs in your study. If you’re using the Survey Builder, participant IDs will be recorded for you.
When it comes to recruiting a custom sample, we recommend keeping your screening questions vague so that participants are not influenced to answer a certain way. One method is to ask a variety of questions so that the participants don't know which one you're actually measuring. You could, for example, ask the following questions with 'yes/no' answer options:
- "I live in the UK"
- "I'm Catholic"
- "I use an electric toothbrush"
- "I run regularly"
- "I have a pet dog"
- "I have at least 2 uncles"
You can then adjust your questions and format the answers to what works best for you.
The second study
This study should direct participants to your main survey / experiment.
This will ensure that you only pay for full submissions from participants who meet your screening requirements.