🤔 Ask Anything Thread (Newbie Friendly)

Hi @P.P,

That’s exactly what I’m after, thanks for the input!

Dan

1 Like

Thanks! One more question: If I launch a survey for, say, 200 respondents, and then want to add more respondents to that same survey, can I do that with the same exact link? Basically I just want to make sure its obvious to people who have already taken it that they shouldn’t take it again.

Hello @Karyn_Amira,

Yes you can increase places (both of active and complete studies) from the top right Action menu >> Increase places. Please have a look at the page that I linked below which explains all the details! Let me just point out two things now:

  • The number you enter in the Increase places box should be the total number of places you want and not the number you want to increase by

  • This option is not available for rep sample studies nor for studies with the ‘balance by sex’ feature.

Hope it helps!
Veronica

Is this the prescreener that asks: “How many units of alcohol do you drink on average per week? (1 unit of alcohol = 1 small glass of wine; half pint of beer; pub measure of spirits)?”

If so, participants are not required to update it :slight_smile:

That HC article has now been updated. Thanks for flagging!

1 Like

Hi, I am hoping to run a questionnaire using Prolific as I have had a terrible time trying to get more qualitative data. My questionnaire would consist of open-ended questions where I would be looking for details of their experiences while at university and looking for work. I only need around 30 recent graduates to complete this, do you think something like this would get completed? Thank you!

Hi guys,

to increase methodological rigor I want to ask how prolific asks for age and how it is verified.
I could not try myself because account creation for participating in surveys is restricted.

Furthermore I would like to ask if for gender prolific has a 'non-binary’ option.

strong text

Hello @twenzel and welcome to the Forum!

There is a variable “Gender (new)” that exactly provides the Non-binary option:

Gender identity has instead more options, as you can see below:
image

Sex is instead coded as a Male/Female variable but the question participants had to respond was cautiously worded as " What sex were you assigned at birth, such as on an original birth certificate?":

As for age, as all questions in Prolific, they rely on self-reported answers. In particular participants were asked " What is your date of birth?". By applying for registration to use Prolific, participants undertake the Terms and Conditions you can read here https://www.prolific.co/assets/docs/Participant_Terms.pdf (where is also stated that they must be at least 18 years old).

Hope it helps
-Veronica

Hi @Jennifer_Craik_Nicoll, welcome! :boom:

My 2 cents on the topic.

  • There are a few prescreeners that might be of your interest and that you can combine as wished in order to get your desired selected sample: Highest education level completed; Employment Status; Current education level; Student Status; Job Seeking; Employment Status; Age.
    You can try explore them and their answers (for example, do you want those graduated at a BSc or at a MSc level? Unemployed? Seeking for a job?). Remember that for multiple screeners an AND logic is appplied (there is no OR functionality) while within screeners an OR logic is applied.

  • If the questionnaire is very long, I would be thinking to add Attention Checks in order to make participants pay more attention over the whole survey. Here on the Forum we are starting collecting proposals for new valid attention checks: Community Attention Check Library - Add Your Own! - #4 by Josh . In general, you can refer to this page for more details on what a fair attention check should be like: Using attention checks as a measure of data quality – Prolific

Hope it helps!
-Veronica

Hi,

I started my first experiment about a week ago.

I increased the place for participants (by 2) a couple of days ago and I am still waiting for participants to do my experiment.
Is it likely to wait for 2 participants for so long (2-3 days?)
For the first batch, I waited a day or two for the same amount of places.

Thanks in advance,
Shay Grunwald

Hi,

I want to duplicate a completed study, like I used to, but now the option does not appear to work. This is with a study that includes equal number of males and females with the new option to do that so maybe that has something to do with it?

Hey @Shay_Grunwald,

I guess you used the Increase places tool available under the top right Action menu. If you did so, which number you entered in that space? The number you should enter there must be the total number of places you want to be available for your study after increasing places and not the number you want to increase by. I’m just wondering if this can explain your problem. If this is not the case, there might be two other options: i) it’s a bug and you should contact @Josh to have a look at your study (providing him with study title and details); ii) its’ related to your sample: are the prescreeners too restrictive? Maybe you selected a super selective sample with just few eligible participants?

Hope it helps!
Veronica

Welcome to the Forum @paualexander !
Unfortunately this option is not available for this type of studies. You can find this info in the Help Center both for the ‘balance sample by sex’ and copying a study features.

Cheers,
Veronica

Hi, Thanks :grinning:
like you said, I entered the total number.
I don’t think that the prescreened was too restrictive (the number of possible participants is pretty large).
I’ll try to contact Josh as you suggested.
Thanks again,
Shay

Dear Paula
Welcome to the forums!

I am sorry but yes, the use of gender balancing does prevent the use of the Duplicate Study option I am afraid.

Please note you are unable to modify published studies that use the ‘balance by sex’ feature. These studies can’t be duplicated or have their study places increased after they have been published.

It is a shame but I guess there are insurmountable technical reasons.

Tim

USD v GBP – I want to run a study on a U.S. sample but I can only see payment and invoice information in GBP. Is there any way to change this setting?

Hey @Daniela_Donno, welcome to the Forum Community!

Yes, you can use Prolific in USD. If you have a GBP account and want to switch, you just need to contact support@prolific.ac or send a request via the form: Submit a request – Prolific. The switch is not automatic therefore you need to send a request (mail or form).

Hope it helps!- Veronica

Hi everyone !

I have a Qualtrics survey where participants are randomly assigned to one out of two different conditions. My problem is that one is twice as long as the other. I made a slightly high estimate of 20 and 40 mins.

I just ran a small pilot with 8 participants and I got times from 8 min to 1 hour and 7 minutes, which makes it very difficult to pay participants fairly. I thought about two potential solutions so far :

-Pay participants on the basis of 20 minutes, and manually double the reward for those who went to the long condition. But I am worried it will be quite labor-intensive for a few hundred participants. Plus, will participants be discouraged to take the study if they see they are paid for 20 minutes but that the survey can take up to 40 ?

-Run two separate study and forbid participants to take both, as I have seen it was possible. But then the different conditions won’t be randomized, as participants will be able to chose if they want to pass a 20 min or a 40 min survey.

Is there a better solution to this problem ? Like maybe an option to pay participants for exactly the time they spent on the survey (within reasonable boundaries) ? If not, do you have an opinion as to which of the two options is better ?

Thanks for your help !
Sandy

Dear Sandy

Welcome to the forums :slight_smile:

I think the standard procedure would be to run the two studies separately and sequentially, preventing those who took the first from taking the second.

There is the chance that those who are busy will pass on the longer study. This is less likely on the shorter since it is not so short as to be not worth logging in for. But those who pass on the longer of the two studies would probably have done the same – except perhaps timed out or returned their submission – had they been assigned to it randomly. Since the two studies are of a different length and participation is voluntary, I think that a truly random spread of participants is probably impossible.

You could include a question in your study to ask participants if they would have passed on the other one to see how common it is.

You might also consider prescreening for subjects would not pass on either study and then assign those that who would pass on neither study randomly but this would mean paying about 20 cents (approximate minimum payment for a single question prescreening study) X total number of participants X about 1.5 to 2 since not all will in fact come back.

If there is variation in time spent on the studies even within the same condition, you could mention a time linked bonus in the description. But mentioning the bonus in the description may encourage some participants to take a long time over the study. There is often such as rush on study participation that subjects do not get as far as reading average participation time. I don’t think I would mention it since participants know that Prolific ensures a minimum wage.

Should you decide to combine the studies and pay bonuses, the payment of bonuses need not be time consuming since you can reorder the submissions by the time taken, and then using the Firefox add on Check ThemAll (a misnomer since it can check only the selected)

you can select those participants who took longer, or within a time frame, check their checkboxes, and send bonus payments to the checked participants in bulk.

I am still not sure which of the two options is better (separate studies or combined studies and bonuses) but I hope this helps.

However, on second thoughts, there is also the issue that Prolific requires a minimum wage be paid to all participants based upon the median time taken. If the median were to end up in the longer of the two studies you might be required to pay extra to all your participants. You could probably get around this by recruiting more in the shorter study but it adds a complication which further suggests separating the two studies.

Tim

Hi @Sandy_Manolios (& @timtak )
Just thought I’d throw a few cents in this discussion. Trying to balance the time it takes for a study between two conditions that are so different (2x the time of the other) would be tricky, because 1) as Tim mentioned, Prolific has the minimum payment per hour based on the whole sample and 2) you probably want participants to be informed about the possibility of being assigned to each condition. I would hesitate to do the bonus procedure, because you don’t know how that might affect participants responses. For example, you might have people drop out when they realize they’re in the “lower paying” condition. Running two studies would violate the randomization procedure, which would also be problematic.

As Tim mentioned, one (probably the best) way to do this would be to use a pre-screening study in which you pay participants to self-select into the study, knowing that they will be assigned to one of the two conditions. Then once you have the Prolific IDs, randomly assign them to one or the other of two separate studies. Of course, because response rates are not 100% you’ll need more than your sample size and there might be some ~random~ error.

Another idea would be to estimate the median time across the two studies (probably slightly higher than the pilot to be on the safe side), post a single study that has the fact that there are two conditions with different times in the consent form, randomize as usual, and pay out the average across both (you could alternatively pay out the maximum for everyone which would probably be best if you have the funds).

Hope this helps a bit, and please post if you have other thoughts to work through! This is a tricky situation.

1 Like