10/27 - Fix 'er up Friday!

That video caused this to come up as recommended

 

Yes!!

On top of that, I'm fairly certain that "50% of marriages fail" statistic is a false one we've all just come to accept as true.

 

Harvard LDS [A3P3S1LVRNHGOH] Study about people - $1.00 | PANDA

Fair

Unrated

Unrated

$8.35 / hour
00:07:11 / completion time

Pros:
Cons: Watch multiple videos.
Make sure on your own from a list of IDs that you have not previously participated.

To read Harvard LDS's full profile check out TurkerView!​

 

Oh man, I would love that.

When i was little, after some standardized testing, they pulled my dad in for a real serious meeting, and I was petrified because I thought it meant I was dumb or something.

Turns out I have a very strongly eidectic memory. I can often reference facts I've seen a couple times at most, a nd if Im feeling amped, I can probably tell you what page of what book I've seen it on. Or, alternately, I can quote something exactly and then get upset because I can't remember what study it's from, but I can visualize the entire page its on.

And this is why I'm incredibly annoying to fight with, because I can and will reference something you did 5 years ago with total accuracy.

Yes! It's 4 or 5 year old Birdie! I was a cat, and I was 100% commited to my character of cat.

 

I think the presenter is a sex offender.

 

Russian mail order bride!

 

That led me to this....


just... wtf.

 

Title: Making Judgments Part 1 of 2 | PANDA Worker: Preview | Accept | Requester Requester: IMC Lab [A1T434NH13YMQU] (Contact) TO 1: [Pay: 4.28] [Fast: 4.70] [Comm: 4.83] [Fair: 4.82] [Reviews: 40] [ToS: 0] TO 2: [Rate: $11.72/hr] [Pen: 0.53 days] [Res: -- of 0] [Rec: 78% of 9] [Rej: 0] [ToS: 0] [Brk: 0] Reward: $3.50 Duration: 2 hours Available: 1 Description: Read different principles, remember events from your past, and make ratings about them. Requirements: MoralViolations DoesNotExist ; MoralExceptionsAll DoesNotExist ; MoralExceptionEtiquette DoesNotExist ; HIT approval rate (%) GreaterThan 85; Location EqualTo US; Adult Content Qualification EqualTo 1;

HIT exported from Mturk Suite v1.25.8​

 

I spent a good portion of last night watching all of these multiple times

 

Title: Digital Camera Study Oct27 2017 | PANDA Worker: Preview | Accept | Requester Requester: Joydeep Srivastava [AO3N31NV0HT9E] (Contact) TO 1: [Pay: 3.00] [Fast: 3.00] [Comm: 5.00] [Fair: 5.00] [Reviews: 4] [ToS: 0] TO 2: [Rate: $6.00/hr] [Pen: 1.00 days] [Res: -- of 0] [Rec: -- of 0] [Rej: 0] [ToS: 0] [Brk: 0] Reward: $0.40 Duration: 3600 Available: 1 Description: 3-4 min academic research study about the decision making process people use to evaluate and choose products ($0.40). Requirements: Total approved HITs GreaterThan 500; HIT approval rate (%) GreaterThan 98; Location EqualTo US

HIT exported from Mturk Suite v1.25.8​

 

Title: STIR | PANDA Requester: STIR 2017 [A2TFQGAIVC3JKF] (Contact) (TO): [Pay: N/A] [Fair: N/A] [Comm: N/A] [Fast: N/A] Description: STIR is a personalized waking service offering morning “gifts” for people around the world. Our service is their first encounter with a new day. Time: 10 minutes HITs Available: 1 Reward: $2.00 Qualifications: None

 

I was thinking of setting up a DB of completion times like this, but I ultimately concluded that there was too much noise in the data and that I would need to go to college for statistics in order to extract something meaningful from the data.

Speaking of statistics, the guy that runs the Dasanfall website provided such a great statistical engine that it ended up adding a meta for experienced players in Planetside 2.

The specificity of this website is what I would personally try to emulate:

http://stats.dasanfall.com/ps2/player/Daddy

Stats are tough, because you have to account for so many things if the stats are meant to estimate the potential worth of a batch:

- How good is the person "on average" at completing work? A low $/hr on a hit means less for the aggregate if that same person earns less $/hr than average on every task.

- Was the person distracted/drunk/exhausted while completing the work?

- Was the work completed across multiple tabs/windows (meaning the person wasn't actually "working on" the HIT the moment it was opened)?

- Esp. for surveys: Did the person complete the survey beforehand (i.e., from an email) and then just insta-submit after accepting the HIT?

- etc...

 

Well I googled aye aye captain but this was the first result, and you're a girl (as far as one can be a girl on the Internet).. so

Haha! I think a lot of these scenarios (not all, but the old "malice vs stupidity" ratio) comes down to people getting different versions of a survey. I remember a few months ago there was a MASSIVE argument here about how fast one person completed a survey.. I did it in the same time, and it was painfully obvious I simply got screened out but they paid for it the same (kinda like the $2 HIT the other day). But similar things happen a lot with Qualtrics just RNG'ing you to a different version of a Requester's survey, some folks get writing some don't, etc etc.

I think incentivizing people into submitting more data will probably be more reliable than hoping someone readjusts their time based on info they don't have.. can only report how fast it took you to do it. I've seen a few reviews that mention "my time was slow because I wrote feedback" or whatever as well. Can't really fault folks for reporting what was their accurate time, even though its obvious that may not be perfectly accurate to the majority of folks. But all of those minor errors would be fixed with.. more (high quality) dataaaa.

I'm kinda waiting on @Totally Not , to release his uber broken OP tracking script, then I can hijack it for my own nefarious purposes

Also, always feel free to report reviews you don't think are accurate. There's absolutely no harm in doing so (presuming someone isn't just flagging every review they don't like, ala whatshisfacefromTO) until the flag is confirmed in the back end.. and the mod panel is turning out pretty decent if I can toot my own horn lol.

 

This report will, of course, be the anniversary present for his wife.

Make sure it's precise and poetic.

 

Jesus Christ is my .... not even gonna say it.

 

Title: Decision making survey. Less than 5 mins. | PANDA Worker: Preview | Accept | Requester Requester: Altsuss [A1ATES40ZPE477] (Contact) TO 1: [Pay: 3.38] [Fast: 4.79] [Comm: 4.73] [Fair: 4.93] [Reviews: 132] [ToS: 0] TO 2: [Rate: $9.90/hr] [Pen: 0.02 days] [Res: 100% of 1] [Rec: 67% of 3] [Rej: 0] [ToS: 0] [Brk: 1] Reward: $0.30 Duration: 1 hour Available: 23 Description: Show us how you would make these decisions! Requirements: HIT approval rate (%) GreaterThan 95; Location EqualTo US;

HIT exported from Mturk Suite v1.25.8​

 

Title: Experience of work on Amazon Mechanical Turk(~ 10 minutes) | PANDA Worker: Preview | Accept | Requester Requester: virksomhetskommunikasjon [AETZTYKP2X9UJ] (Contact) TO 1: [Pay: 4.44] [Fast: 5.00] [Comm: 5.00] [Fair: 5.00] [Reviews: 26] [ToS: 0] TO 2: Not Available Description: Give us your opinion about your experience of work on Amazon Mechanical Turk. Time: 4 hours HITs Available: 1 Reward: $2.50 Qualifications: Total approved HITs GreaterThanOrEqualTo 100; Inc: [831508-61042] Exists ; HIT approval rate (%) GreaterThanOrEqualTo 95; Location In US;