03/01 - Singing Saturday

Enjoy y'all!

 

@Janya

 

@Janya

 

what do I do with a

This survey has been previously completed and approved once by the same WorkerID.

rejectioN? dispute it because it technically should be their fault for not putting in filters/requirements?

 

You can try disputing it, by telling them that you have no way of knowing if you've done the same survey for the same person before, and that we all see the same tasks OVER AND OVER again... but I doubt you'll get anywhere.
from their perspective, they think you'll recognize if you've done a task before even if it's been over a year, they don't know what it's like on our side, and because of that you probably won't get them to budge.

If you do message them, be cordial, don't get angry because that doesn't help your case


but honestly, sometimes you just gotta take the hit and move on.

25 of my 29 rejections are because I thought something was a batch HIT (it had previously been a batch where we were allowed to do as many as we could grab) but I hadn't realized they changed. that was a sickening day for me, but looking back, I still have over 99.99% approved so it didn't kill me


sometimes you just gotta move on no matter how flustered you are

 

Often when I see language in a HIT description warning about submitting a HIT more than once, I'll send the requester a short message letting them know they should be using exclusionary quals to prevent people from taking the HIT again. I explain that there is no way for us to keep track of which HITs we have done and exclusion quals makes that simple for both parties. On more than one occasion, they've written back to thank me for the suggestion as they had no idea they could do that.
I realize you're past that point here, but saying something similar might work if they are not robots.

 

but wouldn't they have to pay to create qualifications? sometimes that's probably a bit of a barrier. I mean, look at how many 15-20 minute surveys for 2¢ get posted. I'm not sure that it's always flat out being unaware so much as simply not having the money.

 

As @Brone said, I would message the requester and say you didn't remember doing the previous hit and ask for a reversal.

Every once in a while, if I feel like the requester clearly screwed up and they won't reverse the rejection, I will message the IRB at the University they are affiliated with. This doesn't really apply for rejections for retakes, as that seems to be more on us, even if there wasn't a warning/ auto screen out, but I was able to get Dr. McBride to reverse a rejection on the $5 Complete a short survey daily commutes to work hit that was up Feb. 4th. She had insisted we would have been screened out in 10 seconds, then later admitted they had screwed up the screener, as it was her first time using mturk. Hoping if anyone else was affected by this one, they had their rejections reversed too.

Can I ask what hit you were rejected for? I got two rejections for repeats in the last month or so.

One was Nicola Redl's 'News Survey' for $1.30. That one really annoyed me cause I knew I had done it before, but I checked that it didn't say no repeats (even screenshot the consent form), and the article that I read at the beginning was different than the 1st time I did it, so I thought they may have relaunched the survey or made changes, and we could repeat it... I messaged her, and she said I wasn't supposed to be allowed to retake it, so even though it was an error on her end, I don't expect to get a reversal. I'm considering messaging colorado.edu's IRB, since I have the info from the screenshot on this one, but probably won't tbh.

The other one was Zhuowen Chen's Welcome to Our Smartphone Purchase Survey for $1. I messaged the requester, but didn't hear back, so I am just going to have to take this rejection I guess. On this one, I think I had done it 1.5-2 months earlier and didn't realize I was repeating..


ETA: This is a good reason to screenshot consent forms, etc, especially on new requesters on mturk and prolific, as having that info can be handy down the road.