07/10 - Milksteak Monday

[loki3404]

I think for as long as I've been around mturk, they've been two separate things.

My memory is fading in my old age.

 

[aveline]

I did not read it that way, and they also say the forums benefit requesters.

I choose not to believe any of it. It's the fee increase that made all of the requesters leave, not the workers.

 

[Kerek]

I choose not to believe any of it. It's the fee increase that made all of the requesters leave, not the workers.

The workers have always been bad, the fee increase was what changed.

 

[loki3404]

>In particular, we find that from 24% to 83% of those passing the screener questions are impostors, and that deceit occurs in 49–89% of those who are “eligible” to misrepresent

This is incredibly depressing. I've returned so many fucking HITs because I knew I did not qualify at all. Makes me feel like this is a hopeless endeavor if the requester/worker relationship is going to be this antagonistic.

(Though they also did not mention, at all, fraudulent requesters, and seemed to solely focus on how awful MTurk workers are for daring to talk about their MTurk experiences in a public area...)

They more or less compare communities to trade unions.

Overall, the MTurk online forums help workers transform a difficult job of responding to studies into one that is more predictable, pleasant, and economically justifiable. In that way, forums benefit requesters by increasing the willingness of people to participate in research studies. Forums also encourage requesters to act in ways that support the joint system. In particular, the forums penalize requesters who pay a low hourly wage (Gleibs 2016), those who underreport the expected length of the study, those who annoy workers with unexpected or boring tasks, and those who block workers unjustifiably (Brawley and Pury 2016).

 

[aveline]

The workers have always been bad, the fee increase was what changed.

Maybe other workers, but I always follow every instruction and do quality work.

 

[Kerek]

Maybe other workers, but I always follow every instruction and do quality work.

You also submit one HIT every 45 days.

 

[kryss]

 

[aveline]

/s

 

[Kerek]

 

[Vanyanka]

Instead of working, I'm watching a million Prime Day deals. I feel it's a good use of my time.

 

[Kerek]

Instead of working, I'm watching a million Prime Day deals. I feel it's a good use of my time.

Really need a bocce ball set?

 

[Pleco]

" We should make clear in our research that no matter how badly you talk about Mturk workers, you will never surpass their ability for self-hatred, so just go to town"

 

[Shockwave]

I did not read it that way, and they also say the forums benefit requesters.

I understand I was being a bit hyperbolic, but reading further into the study they made a lot of mention of how forums are used to manipulate the outcomes of data and seemed suspicious about forums in general and recommended more requesters monitor them (which imo would be really great) In any case this study seems focused only on research-based HITs and not other stuff like transcription tasks or other verifications.

>While the forums restore some of the balance of power between requesters and Turkers, they may also discourage requesters from appropriately rejecting or blocking truly offending workers from their studies.

I guess this bugged me a bit too.

 

[Squatty]

Title: Questionnaire about personal preferences and emotional recognition | PANDA Worker: Preview | Accept | Requester Requester: Olivia Heck [A3M6EFBUYLZKS4] (Contact) TO 1: Not Available TO 2: Not Available Description: This is a psychology study that analyzes the effects of smiling on facial preferences. Time: 45 minutes HITs Available: 2 Reward: $0.25 Qualifications: US High School Graduate is 1;

HIT exported from Mturk Suite v1.21.11​

 

[Vanyanka]

Really need a bocce ball set?

Well, now I do. You know I'm a highly suggestible, mindless consumer.

 

[mboone73]

" We should make clear in our research that no matter how badly you talk about Mturk workers, you will never surpass their ability for self-hatred, so just go to town"

 

[JenniLeigh]

Instead of working, I'm watching a million Prime Day deals. I feel it's a good use of my time.

I'm trying to do both.

Need back to school stuff, kids clothes as cheap as possible (gonna end up getting second hand, but I can hope...) it's sock & undies stock up time....

 

[Kerek]

I understand I was being a bit hyperbolic, but reading further into the study they made a lot of mention of how forums are used to manipulate the outcomes of data and seemed suspicious about forums in general and recommended more requesters monitor them (which imo would be really great) In any case this study seems focused only on research-based HITs and not other stuff like transcription tasks or other verifications.

>While the forums restore some of the balance of power between requesters and Turkers, they may also discourage requesters from appropriately rejecting or blocking truly offending workers from their studies.

I guess this bugged me a bit too.

But all of that stuff about workers on forums is true. Posts are edited on a (near?) daily basis here, let alone across the totality of the forums. Workers aren't saints and are responsible for a lot of the problems with mturk (as shown by the study itself).

And requesters absolutely should reject and block more. If you cheat your way into a study, you are a thief and a liar, and harm other workers and the requester.

 

[Squatty]

Title: Answer a survey about your opinions on pro-social products | PANDA Worker: Preview | Accept | Requester Requester: Katy Johnson [A1AL9PLY47959O] (Contact) TO 1: Not Available TO 2: Not Available Description: We are conducting an academic survey about products that are pro-social in nature (meaning, they are associated with some kind of social good or social cause). Time: 30 minutes HITs Available: 1 Reward: $0.50 Qualifications: Masters has been granted; Location is US;

HIT exported from Mturk Suite v1.21.11​

 

[JenniLeigh]

I could cash out on Prolific, if I wasn't waiting on a survey from APRIL to be approved -____-

 

[aveline]

Posts are edited on a (near?) daily basis here

*Were

I don't know, things have been pretty nice around here over the last couple days.