10/11 - Wicked Wednesday!

Are you saying her HITs are no longer of the hightest quality?

 

Angela Listy [A1M2H0HAEXONBY] Holiday Drinking Survey - $0.26

Good

Unrated

Unrated

$15.34 / hour
00:01:01 / completion time

Pros: nice and quick
Cons:

To read Angela Listy's full profile check out TurkerView!​

 

Title: Answer a short survey about resilience and creativity | PANDA Worker: Preview | Accept | Requester Requester: Yale Creativity Research [A2YA4469TDOZH8] (Contact) TO 1: [Pay: 5.00] [Fast: 0.00] [Comm: 0.00] [Fair: 0.00] [Reviews: 2] [ToS: 0] TO 2: [Rate: $42.86/hr] [Pen: 5.00 days] [Res: -- of 0] [Rec: 100% of 1] [Rej: 0] [ToS: 0] [Brk: 0] Reward: $5.00 Duration: 45 minutes Available: 1 Description: Participation in this study will involve completing surveys about your behavior in everyday life. Requirements: Masters has been granted; Total approved HITs is greater than 500; HIT approval rate (%) is greater than 95; US Bachelor's Degree is 1; Location is US

HIT exported from Mturk Suite v1.25.7​

 

While I think it's true that you can't repeat the Connect 4 HIT, that message looks like it's saying all the active HITs have been accepted.

 

Ready the pandas.

Galileo Play Connect 4 (1 hour - $4 base + up to $10 bonus)

3D343PJM16Y8J700HVDLA9DQWPHJ45

 

Title: Answer a few short questions | PANDA Worker: Preview | Accept | Requester Requester: Joshua Knobe [A3INOEQSBNJES5] (Contact) TO 1: [Pay: 3.86] [Fast: 4.76] [Comm: 3.96] [Fair: 4.72] [Reviews: 155] [ToS: 0] TO 2: [Rate: $8.20/hr] [Pen: -- days] [Res: -- of 0] [Rec: 100% of 3] [Rej: 0] [ToS: 0] [Brk: 0] Reward: $0.30 Duration: 15 minutes Available: 1 Description: Fun easy 2-3 minute study for money, contribute to philosophy of intuitions Requirements: HIT approval rate (%) is not less than 95; Location is US;

HIT exported from Mturk Suite v1.25.7​

like a minute.

 

To expand (and maybe set your mind at ease a bit):

Amazon recommends Requesters use blocks for retake prevention in their documentation (last I reviewed it) - so as incompetent as everyone likes to make them out to be I highly doubt a simple X number of blocks would ever trigger an account ban automatically.

There is a "reason" field that can be filled out when blocking a worker (its optional, FWIW) that workers cannot see, presumably that field would be utterly useless if someone at mTurk wasn't going to look at it if workers are precluded from ever seeing it, so logic would dictate there is some level of human interaction between workers receiving blocks and an account suspension - though it is Amazon so there may well be some ML/AI in there (highly doubt it, personally, but who knows).

Anecdotally, I have heard of someone triggering an account block by all of Google's Requester accounts at one time and being suspended shortly thereafter. Barcodes (Amazon Visual Search) also seem to have a high rate of block->speedy suspension. I've yet to hear of either case where it didn't turn out that the suspended worker didn't deserve the block/suspension, so its really not something I'd worry about too much.

Most people who do surveys probably have a bunch of blocks they'll never even know about lol.

My favorite block fun fact:
You can block up to 100,000 Workers.

Something tells me if blocks triggered an auto-ban we'd have seen some rascally dickhead grab some of those survey worker id exclude lists and just go ham since creating a new Requester account takes all of 5 seconds and requires nothing in the way of identity verification.

 

Joshua Knobe [A3INOEQSBNJES5] Answer a few short questions - $0.30 | PANDA

Good

Unrated

Unrated

$13.17 / hour
00:01:22 / completion time

Pros: quick surey
Cons: have to create your own completion code. I hate that

To read Joshua Knobe's full profile check out TurkerView!​

 

Joshua Knobe [A3INOEQSBNJES5] Answer a few short questions - $0.30 | PANDA

Good

Unrated

Unrated

$12.00 / hour
00:01:30 / completion time

Pros: Quick survey.
Cons: A couple paragraphs of reading.

To read Joshua Knobe's full profile check out TurkerView!​

 

A feature to mark strong disagreement with a requester review on turkerview could be useful.
(this has nothing to do with knobe bad timing LOL)

 

Anybody else getting "Sorry, this survey is not currently active."?

 

Title: Determine a Company's ZIP Code | PANDA Worker: Preview | Accept | Requester Requester: AWS HubSpot Turk Account [A2QNZJNNADT0V3] (Contact) TO 1: [Pay: 3.76] [Fast: 5.00] [Comm: 4.60] [Fair: 4.98] [Reviews: 46] [ToS: 0] TO 2: Not Available Reward: $0.05 Duration: 30 minutes Available: 1 Description: Help us find a ZIP code for this website. Requirements: None;

HIT exported from Mturk Suite v1.25.7​

 

thank you i figure im probably safe since ive turked for a gazillion years and never got a nastygram email, but with my block plus potential one coming was a little worried. Not that theres been a ton of money out there this week, but would rather not have to get a real job!

 

Percy Liang [A29YQT3EEC0ZEC] Negotiate with another user to split some shared items - $0.15 | PANDA

Underpaid

Unrated

Unrated

$ / hour
00:00:00 / completion time

Pros: If you like being poor, then just do these HITs all day.
Cons: Extremely underpaid and buggy HITs.

To read Percy Liang's full profile check out TurkerView!​

 

Thats why you put your own review up to counter the review you perceive as being bad

 

Irene [A2005EP47NDO85] Answer a 4 ( /-2) minute survey - $0.35

Fair

Unrated

Unrated

$7.88 / hour
00:02:40 / completion time

Pros:
Cons: Pay isn't great (was hoping it'd be less than 2 minutes).

To read Irene's full profile check out TurkerView!​

 

shame. he used to be a great requester. i may just return mine on your word.

 

One thing I've noticed is that you can 100% see some people work slower/faster than others. I was mulling over an "adjustment filter" of sorts the other day that you could use client-side to adjust the scores/completion times if you tend to work faster than the average or vice-versa.

In any case, flagging reviews and upvoting reviews is super helpful. They'll be weighted differently in the aggregate data in the future.

Ultimately, though, leaving your own review is the biggest help. Probably more than anything else I could possibly do manually.