Clixsense, Neobux, Crowdflower 2014/5 thread.

older_wiser

i've done afew of those but i always answered in great detail about why and what shouldn't be included or whats a good search nothing so far but i hope a ban isn't coming my way

Fee77

Just got $17 for completing another MIT task

Checklist looking very good already with the higher percent.

older_wiser

I might give the MIT task ago after christmas if its still around

thanks for the confirmation Fee and well done on the earnings thats a nice bonus with 28%

System

older_wiser wrote: »

I might give the MIT task ago after christmas if its still around

thanks for the confirmation Fee and well done on the earnings thats a nice bonus with 28%

I believe the next one is tomorrow night and then one around New Years Eve sometime.



Well done Fee, I think I thanked it but forgot to say as well. A lovely bonus, bet your sponsor will be happy!

asajj

My CF accuracy isn't seem to be moving very much. I try to do "verify articles, authors" type of tasks to build it and finish couple of them with 92%ish accuracy but my total accuracy is not affected.

Is there any number of tasks to be completed to increase your accuracy for a certain percent?

System

The number of tasks required would be dependant on your number of completed tasks.

For example...

If you've done 100 tasks with 100 test questions then just 1 test question task is needed to change your accuracy by 1%.

If you've done 10000 tasks with 10000 test questions then 100 test question tasks, completed successfully or failed, is needed to change your accuracy by 1%.

Basically, the more tasks you've done the more you need to get correct to raise it up. The accuracy of a specific task doesn't directly affect it as such.

If you had completed 100 test questions successfully with 100% accuracy and then finished on 70% on the 101st task your total accuracy would still be 99% as it is the amount of test questions failed, not the overall accuracy percentages combined from each task.

Hope that explains it a bit better? It's early so apologies if the figures aren't 100%, just trying to break it down for you

MollyDolly

Merry Christmas & a Happy New Year everyone - all the best for 2015.

Clixsense:

Quick n easy Peanut Lab survey (Uni study) 0.67.

System

Thanks Molly, same to you.

Thanks for posting the surveys you complete too as it always prompts me to go and have a look.

MollyDolly

Thanks Katy,

V quick PL survey (on kids) 0.50.

V quick Samp survey (calendar) 0.59

asajj

katykicker wrote: »

The number of tasks required would be dependant on your number of completed tasks.

For example...

If you've done 100 tasks with 100 test questions then just 1 test question task is needed to change your accuracy by 1%.

If you've done 10000 tasks with 10000 test questions then 100 test question tasks, completed successfully or failed, is needed to change your accuracy by 1%.

Basically, the more tasks you've done the more you need to get correct to raise it up. The accuracy of a specific task doesn't directly affect it as such.

If you had completed 100 test questions successfully with 100% accuracy and then finished on 70% on the 101st task your total accuracy would still be 99% as it is the amount of test questions failed, not the overall accuracy percentages combined from each task.

Hope that explains it a bit better? It's early so apologies if the figures aren't 100%, just trying to break it down for you

Many thanks Katy