georgeutsin

Those skills are typically built through training anyway, but MTurk doesn't seem to have the necessary training pipeline.

How do you deal with respondents who just created their account?

Even with more pay, there aren’t any real repercussions for doing bad work. Sure there are different “tiers” of turkers, but realistically anyone could recreate their account once their rating gets low enough.

The problem of providing quality control is that there are a lot of edge cases; even known “high accuracy” turkers may have bad judgement sometimes, which means that every piece of data needs to be validated anyway, whether it be the researcher themselves or another paid contractor.

My undergrad thesis was to build https://tagbull.com, where we tried to have turkers validate the work of other turkers by breaking up a label into sub tasks, and getting multi-turker consensus on those before moving forward.

The main issue we ran into is that the incentive system is incredibly misaligned with the responsibility that the turkers have. It’s very difficult to build trust, especially with a crowd of people who haven’t signed contracts, and who face virtually no repercussions for doing bad work, whether intentionally or unintentionally.

Relevant xkcd https://xkcd.com/2341/

Looking forward to using this tool! Are there plans to make this open source?

This sounds like an echo chamber