Have Turkers take a fairly comprehensive demographic questionnaire once and consent to letting all future researchers see it. That way we don't need to ask them for their gender/age/income every time we survey them. Asking these things can activate psychological processes that can affect results (see: research on priming) so it would be great if we could ask them less frequently.

Currently HITs are in the order in which they were created. In order to facilitate easier navigation perhaps allow users to sort (1) alphabetically, (2) date created, (3) date completed, (4) latest activity (such as timed out, worker submission, etc.)

Currently, it is possible to exclude all workers who took a particular HIT by excluding that entire HIT ID, as long as the HIT was launched via TurkPrime. As many researchers flocking to TurkPrime have been using MTurk for years, it would be a huge benefit to be able to exclude HIT IDs retroactively, i.e., to be able to exclude an entire HIT that was originally launched via MTurk. Otherwise researchers are basically losing the capacity to capitalize on their cumulative use of Mturk when transferring to Turkprime.

Notifying workers of studies and then sending reminders can help reduce attrition in longitudinal studies - so glad you have this feature available! Adding a feature that records our messages (e.g., a sent folder) can help researchers better manage this process (e.g., send enough messages without harassing participants; determine how many contacts are necessary to get good attrition rates). We can of course keep track of our participant contacts separately from TurkPrime (or include ourselves as workers to receive and save the emails), but including this feature in the system would be efficient. Thanks!

Enable reposting HITs when workers are rejected, just like MTurk has, i.e. posting only one more HIT per rejection. This should probably by default exclude all Worker IDs that have already come in for that HIT.

As of now Turk Prime only excludes participants who complete the study but not those who did not complete the study but attempted bits of it. Technically, these participants have seen the study or survey protocol and if they attempt a similar study again, their responses would not be original and might be influenced by their participation before. Hence, it is important to exclude these participants.

I’d like to approve/reject HITs, provide feedback in case of rejection, and pay bonuses using the csv output file - similarly to what we do in mturk, where we upload the csv output file to approve/reject and provide feedback, but here I would also add the option of paying bonuses.

I really dislike the idea that some users are completing my HITS at 3 am in the morning. It's not representative of any actual behavior, and may lead to distorted data. Would it be possible to allow users to set restrictions on the time of day their HITs would be available for user?

Allow downloading of .csv files including Worker (or Turk Prime) ID number, and secret key. It doesn't have to be uploadable to accept or reject like on MTurk, but to be able to tell the difference between the set of Worker IDs you got data from vs. the set of Worker IDs you got HITs submitted for is important.

When you login to TurkPrime it would be more convenient for the default home page to be the Dashboard.

I would like to be able to send a worker an invitation to Part 2 of a study immediately upon their completion of Part 1. I have been waiting until I am finished with collection and then sending a mass email, but I lose a lot of people this way.

you should add a batch calculator to estimate the time to run the entire study and to model how much time in between it could take.

When emailing included workers, it would be nice if there is an option to only email workers that have not completed the HIT (or have, both options would be nice). I often want to send "reminder" emails to workers I've included in the HIT but have not completed it. Right now included worker emails go to all workers and it confuses the ones that already completed the survey.

I am manually approving payment and noticed that I cannot sort by worker ID on the "review pending" screen. This function would be extremely useful in identifying a number of participants that I need to reject from my study.