Where can I find data on the percent missing for each prescreen question? This will help me decide whether I should add the question to my survey or rely on the prescreen data. Thank you.
Welcome to the forums.
As far as I know the percentage is not displayed but the effect on the number of participants is shown with the addition of each prescreener.
Thus you can see effect of adding a prescreener on the total number of participants, or of a subset depending upon the order in which you apply the prescreeners, and then calculate the percentage yourself, I am afraid.
I am sorry I can’t be of more help.
Sorry, I also misunderstood. I interpretted “Where can I find data on the percent missing for each prescreen question?” to mean what percent of the participants would be excluded by the addition of each prescreener. But had I read your “Statistics on Missing Data on Presceeners” then I would have known you want to know the percentage of participants that have not responded to the prescreener or have not kept their information up to date. Sorry about that.
I am not aware of a way of obtaining this data directly but you can select all the items on a prescreener as acceptable and see how that affects the subject pool.
E.g. there is now a total subject pool of 128,225 participants
If I select employment and all its options(
- Due to start a new job within the next month
- Unemployed (and job seeking)
- Not in paid work (e.g. homemaker’, 'retired or disabled)
(using the “select all” button) and press “Apply.” there are now only 68,877 participants which suggests that approximately half the participants (46.3%) have not responded to, or not updated their response to, this question.
I have often been amazed by enormity of this percentage but I don’t know how to explain it in any other way.
If it is true, then I guess t hat Prolific might be well advised to make the fact that prescreening data is incomplete more prominent. I have put in a feature request in the past…
AND JUST THEN! from today, I think, prescreener questions have not been responded to or updated appear on the top “Studies” page of the participant GUI. I expect this will mean that participants keep there data up to date to a far higher degree and that there will be much less missing data.
By the way, selecting a prescreener and all its options allows you to obtain the answer given by the participants on your survey without asking them in your survey since responses to include prescreeners are included in the demographics download. I am not sure how this lies ethically, but it can be useful.
PS I wish Prolific would keep “community leaders” in the loop regarding changes to the GUI as requested in the past