11/29 - Whew it's Only Wednesday!

Even though I'm pretty sure it's not true, I'm going to keep the myth alive that the unqualled Hotel Photos are their secret masters HIT.

 

Data never "says" anything. It's always open to interpretation, and it is usually interpreted ineffectively.
That's one of my pet peeves, is people who ignore the interpretation phase of coming to a conclusion based on data. I'm a skeptic and don't trust people's conclusions, including my own much of the time. I think it's arrogant to think you "know" something, especially since something that's true in one moment may become false in the next.



All I can say is that every single indication I've seen on Masters suggests to me that variety is extremely important, and this goes hand-in-hand with the actual purpose of Masters.

Like @ChrisTurk said when we were talking about Google's SEO black-box -- the best way to rank for Google these days is just to write good content that people like. Similarly, the best way to get Masters is simply to "demonstrate a high level of accuracy for a variety of requesters". Don't worry about padding stats, just literally do that one thing.

 

Th

There's Upwork qualifications??? Cool, maybe I've got a Masters there since I'm not blocked from anything
Anyway, I'd give you a link to the job that I was invited to but it's a chicks only job. PM me if you want it anyway

 

Surveys [A12QGMRLRQVKBY] How ads make us feel (B) - $1.00

Generous

Unrated

Unrated

$26.28 / hour
00:02:17 / completion time

Pros: Quick and easy
Pays good
Cons:

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

 

I've always been under the impression that there are specific HITs that need to be done in order to get Masters.

A few years back someone exported a HIT to whatever forum I was on at the time with the note "Do these if you want Masters ". I did them. It was a small batch, some kind of product categorization. I wasn't really expecting anything from it but the pay was pretty good so I did them regardless.

A few weeks later I had Masters.

Might have just been a coincidence but as someone who does like 99% surveys, I'm just gonna stick with my belief that those HITs led me to Masters.

 

J N [AL9XTUAY2PZGZ] Interpersonal relations - $0.50

Generous

Unrated

Unrated

$15.52 / hour
00:01:56 / completion time

Pros: Quick and easy
Cons:

To read J N's full profile check out TurkerView!​

 

Yale Behavioral Lab [A2IFZMK47ZIZ8X] Short study about a donation choice. - $0.40

Generous

Unrated

Unrated

$16.74 / hour
00:01:26 / completion time

Pros: Quick and easy
Cons:

To read Yale Behavioral Lab's full profile check out TurkerView!​

 

Unanimous A.I. [A1OVB3ATBAQ1KM] Qualification survey on economics, finance and general knowledge - $0.10

Unrated

Unrated

Unrated

$3.27 / hour
00:01:50 / completion time

Pros: Qual HIT
Cons:

To read Unanimous A.I.'s full profile check out TurkerView!​

 

That doesn't mean anything. What variety? It doesn't say a 'large' variety or a 'wide' variety. And yes, I do believe that every single policy Amazon releases is approved by a committee of Attorneys.

The 'variety' could be what you said and what others are saying... Maybe it's A9. Maybe it's a specific set of Requesters. Maybe it's specific batches or HITs from some requesters.

I think reading this as 'wide variety' is just not warranted. This doesn't prove anything.

 

Surveys [A12QGMRLRQVKBY] How ads make us feel (B) - $1.00

Generous

Unrated

Unrated

$19.46 / hour
00:03:05 / completion time

Pros: Quick, minor attention checks
Cons:

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

 

It's very possible, and around 10,000x more likely than the long-debunked bullshit "# of HITs and approval rate" hypothesis that people have stuck in their heads just because that's how the requesters ranked them when they got started on mTurk.

I think it's hilarious that people think "well, if it's not done by # of HITs and approval %, then it must be random". Like damn, that's a HUGE logical leap. The conclusion does not follow the evidence.

 

Yeah, there's different worked-based qualifications it looks like. I don't think these were around whenever I did stuff before.

 

Yeah, at the time I only had like 14k HITs done, maybe even less than that. I'm still barely over 20k now

 

Maybe it's any of those, but what's your point?
My point is that there is criteria beyond simply # of HITs and approval rate, and that it's not random, so you're playing right into what I'm saying. I guess they call this "agreement".

 

I am back!