Dawid-Skene annotation model with partial pooling of annotators



I’m very sorry to hear that.

Could you send the link to the paper on measurement error in content analysis?


You can find a free preprint and als materials here: https://osf.io/e7m9z/. The official paywalled version is here: http://dx.doi.org/10.1080/19312458.2017.1305103.
Don’t expect anything special, we just cover a very simple method in this paper. The Dawid Skene Stan model will be much cooler, because it provides a straightforward way to propagate the errors and uncertainty through all stages of the model - something which is not easy to do in the methods which we discussed in the paper.


Thanks—I’ll take a look. And thanks for sending a non-paywalled link!