I am working on undergrad research in econometrics and am seeking some advice/help in terms of finding data sets and picking variables. I have an idea chosen for my topic, but I need to make a official proposal including a rough linear regression equation to turn in to my teacher soon. My topic idea is to observe the causes for recent decline in female labor force participation in the US. I am hoping this topic is not too broad. Initially, I was going to look at whether campaign spending impacted voter share in Senate races from 2012 data sets, but I find this idea a tad boring and would like to make my new idea work if possible. My ideas for variables include the obvious: age & teen participation rates (16-24), race/ethnicity, educational attainment, children, martial status, avg weekly hours worked, etc. However, I wanted to also see if the increase in opioid addiction in the US could also be having an impact on labor force participation specifically for women, as well as paid family leave benefits in the US vs other countries and labor force participation of immigrant women. So with all said, does anyone have any suggestions for me both as to whether this topic is viable as a project in my course, as well as where I could find the data for such a topic? I have been reading a lot of literature on the topic, but having a hard time finding actual data sets that I could use for regression. Thanks!


Can you provide your research question in one concise sentence? It seems like you've proposed half a dozen topics some of which are peripherally related. You need a quantifiable dependent and independent variable as well as essential control variables.

The question needs to be quantifiable and have a wealth of data already available. I'd highly recommend for your undergraduate paper instead of asking a question and then searching for the data to back up the answer, you instead peruse some interesting data sources and ask a question based on what is available.

Check out sites like https://www.data.gov/open-gov/ which have a wealth of easily downloadable datasets that you can explore. Find datasets related to topics that you find interesting and go through the available data and begin to think of what can act as an independent or dependent variable. Base your question so that the dataset you have already found and think is interesting perfectly fits instead of the other way around.

Unless you plan to do original research it will be quite difficult to simply find data to fit your question unless you already knows it exists.

  • $\begingroup$ +1, a very short statement of the thesis of the paper and preferably a form for the regression would be appropriate to lay out for us @Haley $\endgroup$ – Kitsune Cavalry Sep 24 '18 at 19:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.