thats not true. Approval rate and quality test and not hit count, if anything raise course hits from 500 to 1000. 1000 is a good number by than people know how to use mturk and do good work
You can only do 6 of these and you have to check if you have done any of these before, https://www.mturk.com/mturk/preview...id+Mease&hitId=2R8YQ4J3NAB604MAV6PYPT1TZHODQ8
So Rubin approved jobs at 6am this morning and it is stilling Approved Pending Payment, anyone else have this issue
Because I stated my opinion and made a suggestion I somehow must think that I'm superior? It was a suggestion, not a command. My opinion is just as good as anybody else's and it can be ignored or refuted civilly by those who don't like it. I don't appreciate your little pop-shot. I can agree with this. Except that a qualification test would be hard to make up because the HITs are reliant on other people's websites. A review of a persons work in small batches with no qual could be considered the test. Maybe it's not necessary to have a cap on the number of workers. It's just my opinion that if I were the requester, that's the way I would do it to ensure quality. Like you said, HITs approved doesn't necessarily ensure quality.
Some of you people urging for a rubin qualification are going to find yourself without a qualification.
LOL Oh there are people that know how to use and abuse that have 1000 approved. Just because you hit 1000 doesn't guarantee anything. That's still a Newb.
=-( Sorry. I didn't try it... CF makes me all rageface.... but I thought I'd pass it along since they don't reject.
Title: Create an original Truth or Dare question (US) Requester: Tyler Schnaidt Description: Make up your own truth or dare. The content must be original. Keep it brief: 30 words or less. Reward: $0.10 Qualifications: HIT approval rate (%) is not less than 95, Location is US Link: https://www.mturk.com/mturk/preview?groupId=2KYZBAT2D5NQE2GOYJRJK5UYTRC262 I actually took way longer than I needed to on this one. But now I know which questions to ask when I get caught in a game
I agree with you. I have noticed a lot of new workers on here that truly do work hard. I don't think a number makes you any better than anyone else. I have over 69,000 approve hits, but I can honestly say that there are some hits on here that I know someone could do better than me. I normally pass these hits up because I know that I would not do as good as someone else. Transcription hits are a perfect example. I know that these are not my strength and I leave them to someone else who has more experience with them and can do a good job. For all I know a person with 1 approved hit could do better than me at those. A number doesn't make you a good worker. Taking the time to learn how to do a hit properly and knowing when to pass a hit up is what makes you a good worker. I think Rubin is trying to figure out the best way to do it, but I don't think limiting the amount of workers is the way to go. I think it will just take time for him to learn who provides him with good work and who doesn't. I wouldn't be surprised if people start getting blocked for providing bad work. That may be the only way he can truly weed out the people who don't take the time to provide him with correct information.
NP, I just don't understand why they keep posting these. Unless someone has some secret way of doing these, I don't get it!
Title: Short Survey for Sports Fans Requester: Amit Bhattacharjee [A2PYSN7JHNRKMR] (TO) Description: Short survey for sports fans about advertising for sports-related products. Reward: $0.20 Qualifications: HIT approval rate (%) is not less than 95, Location is US Link: https://www.mturk.com/mturk/preview?groupId=2TFJMVHVKCJ0OOPNV8A7FKQ1C6B641 [size=-2]Powered by non-amazonian script monkeys [/size]
Generally, the higher the approval rate based on more HITs, the more reliable worker. But mturk obviously doesn't have the best stat system. We could debate this to no end... How effective raising the approval rate req is. I say it's pretty effective. But quality tests are undeniably the best of determiners, and by far.
If anyone likes transcribing, Speechink has quite a lot of HITs out right now. (I've been raging through these for the past few hours. I though I was gonna end the day with only around $7.00, but should be able to hit my $30 daily goal before bed.) I just did each one with slightly different emphasis.
This this this this this. I don't want to deal with a qual. I don't want to sit around hovering for weeks to catch a qual batch like I had to with MedCrowd, or have to email them because I accidentally screwed up the qual like I did with MedCrowd. I just want access to the work without a ton of nonsense. If that means spending a day or two doing .01-.02 hits until I can hit the required amount, whatevs, the payoff is generally worth it.
I guess I'm going to work on these for a little bit Title: Are These Cell Phones The Same? Requester: CrowdFlower [A2IR7ETVOIULZU] (TO) Description: In this job, you will be shown the specifications for a <b>"Listing" Cell Phone</b> and 1-3 <b>"Option" Cell Phones</b>. What we want to know is which "Option" cell phones, if any, match the "Listing" phone. Reward: $0.05 Qualifications: Link: https://www.mturk.com/mturk/preview?groupId=21WH0OHOVR14QPPRTF02LEDWH9254M [size=-2]Powered by non-amazonian script monkeys [/size]
Took me about a min: Title: Feedback for a product Requester: Eric Posen [A18PUKS47NP0RV] (TO) Description: Visit a single website for a product and give feedback Reward: $0.15 Qualifications: Total approved HITs is not less than 50, HIT approval rate (%) is not less than 80, Location is US Link: https://www.mturk.com/mturk/preview?groupId=2HGWQIHPCGJ2OL142GWXR0JPPYO17J [size=-2]Powered by non-amazonian script monkeys ��[/size]