My last post was a bit of crankiness about the term "nonresponse bias." There is a bit of terminology, on the other hand, that I do like -- "Proxy Y's." We used this term in a paper a while ago.
The thing that I like about this term, is that it puts the focus on the prediction of Y. Based on the paper by Little and Vartivarian (2005), this seemed like a more useful thing to have. And we spent time looking for things that could fit the bill.
If we have something like this, the difference between responders and the full sample might be a good proxy for bias with the actual Y's. I'm not backtracking here -- it's still not "nonresponse bias" in my book. It's just a proxy for it.
The paper we wrote found that good proxy Y's are hard to find. Still, it's worth looking. And, as I said, the term keeps us focused on finding these elusive measures.
The thing that I like about this term, is that it puts the focus on the prediction of Y. Based on the paper by Little and Vartivarian (2005), this seemed like a more useful thing to have. And we spent time looking for things that could fit the bill.
If we have something like this, the difference between responders and the full sample might be a good proxy for bias with the actual Y's. I'm not backtracking here -- it's still not "nonresponse bias" in my book. It's just a proxy for it.
The paper we wrote found that good proxy Y's are hard to find. Still, it's worth looking. And, as I said, the term keeps us focused on finding these elusive measures.
I also like the way Andridge and Little (2011) defines proxy variable as way to reduce dimensionality using the prediction of the outcome variable based on a set of auxiliary variables. There I think it's even more explicit the focus on the prediction of Y that you mention.
ReplyDeleteI was thinking mainly of paradata design. That is, designing auxiliary variables that can function as proxy Y's. These are likely to be interviewer observations. But could be other paradata elements.
DeleteI see, variables such as presence of children at home in NSFG, right? I believe that's a very interesting way to go too. I have suggested that in a survey and they even implemented that, but we didn't use that for anything at the end. They thought that the interviewers observations were not very good, when compared to respondent data. In that sense, I believe we also need more research like what Brady West has been doing in evaluating measurement error in these kind of paradata.
DeleteYour this blog giving us information about subjected topic. Thanks for doing this
ReplyDeleteFileCrop UK proxy