Request: Datasets with measurement error and missing data?
Summary: I’m currently working on a PhD in statistics looking at Bayesian models for data with measurement error, rounded values and missing data. I am interested in finding datasets, either just to use as examples to illustrate my method, or as a more serious project (I would also be open to collaborating with others!), and I think it would be very interesting to work with applications in global health or animal welfare.
My PhD is very open (scheduled to finish in August 2024), and I’m currently thinking about how I should spend the last 1.3 years of my degree in the best way. I’m also currently struggling a bit creatively with deciding what to do after recently finishing my first paper, which was really the most important part of my PhD, plus it would be very cool if my work could be used for something somewhat useful. I got some advice to ask here, since I’m sure there are a ton of interesting datasets out there that may qualify, and maybe some of you could point me towards them.
My topic is just Bayesian models for data with measurement error and missing data in the explanatory variables, and so any dataset where we suspect there to be some kind of measurement error and missing data is of interest! I have also been looking a bit into the effects of rounding/data heaping/digit preference and would be super interested in data that suffers from any of these as well (although I don’t really know how to “solve” that problem, but that would be something to look into!) If there is already research published using the data, that is just fine (provided I am allowed to use the data), it could be very interesting to see if the conclusions might change if measurement error/rounding/missing data is accounted for, if it was not considered in the original analysis.
So specifically I am interested in the following:
Do you know of any datasets with variables that suffer from measurement error, rounded or missing data, that are open to use?
Do you have this kind of data and would be open to collaborating with me somehow?
More meta: Do you have suggestions for people/organizations to contact or otherwise advice on how to move forward to find these types of data?
If anyone is interested, my paper is available as a preprint on arXiv: https://arxiv.org/abs/2303.15240
I’ve reached out to some people doing impactful data-driven work in global development, who may be interested in collaborating.
It’s great that you are looking for impactful applications in your PhD, and I wish you the best of luck! I also really appreciate the amount of details and openness in this post :)
Thank you! If you don’t mind me asking, how do you tend to go about asking for collaborations with people, if you don’t know them from before? I tend to find this “social” part of research very difficult, because I feel like I’m intruding too much or asking too much of people.
The JPAL and IPA Dataverses have data from 200+ RCTs from development economics and the 3ie portal has 500+ studies with datasets available (and you can further filter by study type if you want to limit to RCTs). I can’t point you to particular studies that having missing or mismeasured covariates, but from personal experience, a lot of them have lots of missing data.
Thank you for this, these are both new to me! I will definitely take a look through them.