Clixsense 2014

vampire-girl

Katy, PDF E has let me do more, max is 40

System

I can't say ANY PDF jobs. The one I tried to get in earlier, from your copy & paste on e-mail, told me I had completed all of my work, even when I've done 0 jobs on one of them, so I've e-mailed them, hoping it is a bug that gets sorted and maybe it is linked to the levels thing, who knows!

Very buggy day for me with tasks.

vampire-girl

katykicker wrote: »

I can't say ANY PDF jobs. The one I tried to get in earlier, from your copy & paste on e-mail, told me I had completed all of my work, even when I've done 0 jobs on one of them, so I've e-mailed them, hoping it is a bug that gets sorted and maybe it is linked to the levels thing, who knows!

Very buggy day for me with tasks.

Sorry, was hoping they would have fixed them for you

System

No news yet but I guess they've got a lot of people messaging them, that'll be it until Monday now as it is the weekend isn't it Nevermind hey

vampire-girl

that's a real pain then Katy, maybe you'll have better luck tomorrow

System

Its fine, sure it'll get sorted eventually I think, but not sure, that I had done 50 on that one already. I've earned my target for Monday-Friday and working on my Friday - Monday one anyway, $3/$20 so far, so not too bad

vampire-girl

well done, enjoy the rest of your morning

I'm actually tired now surprisingly so going to bed,

good night

System

I am going to do the same soon!

Night x

RachelBel

Morning all
I spent some time yesterday reading everything on the client side of the Crowdflower website- with all the tutorials telling them how to build a task etc. I think it really gave me an insight into a lot of things. Just wanted to share what I learned, as it might put some people's minds at ease with regards accuracy & those test questions.

First off, they describe 'trusted contributors' as being those with an accuracy of over 70%. So 'good' is much lower than you might think. Their default accuracy on any given task is set at 70% - that being the point where you get bumped off a task if you drop below. The client who is setting the task can change that default to a higher % if they want though.

The client can also select who to target their tasks at according to a very wide range of demographical info... not just accuracy, so don't panic if you're not seeing a task that seems like it 'should' be open to your accuracy level. They can aim it at certain ages, sex, whether you're married or not, have kids or not, education level, location, employed/part-time/full-time/unemployed... and even the reason you use CF, whether its to top up your income, sole income, or 'for fun'...
Those questions/demographics seem to have come from a skills test at some point which was based on personal info. I've never had to answer any questions to give that kind of info to CF so I presume, like me, a lot of you newer folk would be 'uncategorised' as far as demographics is concerned. I'm sure the questions will pop up for everyone at some point.
I'm guessing the majority of tasks don't need to be aimed at certain demos, but maybe it comes into play for some of the survey type ones.

When you raise a ticket - unless CF actually put that task up, (some of them, they do) they don't know anything at all about the task, they just pass it on to the client - and its the client's job to check test questions etc & improve the way the task is going, which they are able to do in real time. CF state on their website that the tickets are helpful to make the task run better and show the client what they're doing wrong... so don't worry if you're sending in a load of tickets across a wide range of tasks- those tickets are all being forwarded on for the attention of different people. (of course, if you're raising multiple tickets for the same task/type of task- like the casino ones, they ARE going to the same person, so maybe be aware of that?)

Test Questions: There's actually two ways that test questions are made.
The first way is that the 'correct' answer is provided by the client who built the task.
The second way is more complicated & might be used by the client to save time on their part - Test questions are built through 'the wisdom of the crowd, so to speak. So if, for example, four people, who have a high 'trusted' accuracy level opt for answer A, but one person opts for B... the test question becomes "Correct answer=A". They use a formula to ascertain confidence levels based on the aggragate of answers and the accuracy levels of the people involved. So essentially this means that a task can accumulate more test questions dotted throughout the task as it proceeds.
But its not infallible at all... because those 'trusted' contributors might become 'untrusted' over the course of the task by getting other test questions wrong, which adjusts their % in real time, and therefore adjusts the 'confidence level' of their answer in those test questions. So accuracy get adjusted automatically at the end of the task if you were marked wrong by what turned out to be inaccurate responders.
It's important to note, 'trusted' contributor on a task is not based on global accuracy, but only your performance on that task... so someone might do well on the quiz mode questions, are regarded 'trusted' and then only do a handful of the task questions, and never stay in the task long enough to be properly judged for their accuracy which might not be all that great overall - thats why its worth raising tickets to draw attention to tasks that might be using bad judgement as a benchmark. And some jobs on a task can be tricky and trip up the majority of accurate people too... like those rooftop ones where the pool has been drained & not obvious, or the googlemap view shows solar panels that aren't on the main pic.

Sorry for the long post... but thought some of you might find all that interesting. And it explains why some of the test questions can be so bafflingly wrong.

I haven't been 'sorted' into a Level category yet, no box for me around any of the levels, but I'm not worried... I'm pretty certain if 70% is 'trusted' then all of us who are in the 80-90% will be happily regarded as high accuracy. Just a hunch, maybe 70-80 is high, 80-90 is very high, and 90-100 will be near perfect??
I imagine they will be sorting the levels for those who are already on a higher banding first, since they have the majority of tasks open to them already, and get around to us uncategorised newbies last.

Edit: Oh, and it is totally possible that if you are on a 'lower' level (like us newbies) that a 'new' task appears when the work is almost finished on it. The client can set their tasks to be done by those who are on the higher levels/who have unlocked the task, but they can change their mind during the task and make it available to all... which they might do if the task is not getting finished quickly enough and they want to speed it up & are already happy that the bulk of the work has been done by high accuracy/trusted contributors on the higher levels. Apparently the downside of aiming a task at the more experienced/trusted contributors is that they have more options and might not be prioritising a clients task, particularly if its low pay. Whereas lower levels get the task done speedily due to lack of other options. (so i'm not going crazy... I"m always only seeing those moderation picture tasks when there's only 1 or 2 left on it... so i presume they get mostly done by people at higher levels, then opened to all to get them finished up)

Northern_Lassy

Hi, thanks for all the info, it has really helped me get started on this site