Tuesday, October 5, 2021

Stata Tip: Creating Quick and Beautiful Tables of Descriptive Statistics with Asdoc

Open up Stata right now and write this in the command window: 

ssc install asdoc, replace

help asdoc

Upload your favorite data set and start making beautiful tables of descriptive statistics! It's like tabstat, but you will have no additional formatting to do. No need to copy-paste from the Stata window into excel. No need to spend hours making tables beautiful. Impress your coauthors and advisors with these beautiful tables even for just preliminary analyses! Read more here

You're welcome! 

I'll take the credit for discovering this, but thank you StataProfessor for making it.

Wednesday, July 7, 2021

All About Robustness Checks: "Do Not Run a Regression Just Because You Can!"

Tatyana Deryugina provides some excellent tips on how to choose robustness checks for a paper here. Running way too many regressions is a typical mistake I see many really smart and hardworking graduate students make. It's funny that she starts the blog entry with the wise words of her advisor, "You need to think more and do less," because I still remember my advisor saying something very similar: "In general, there is no good substitute for thought." (In case you're wondering, we did not have the same advisor.) 

So the key is to think about which robustness checks (or tests for heterogeneity or placebo tests) are most important for telling your story. I'm all for playing around with the data when you first start a project to get some sense of how things are. Readers of this blog will know that I'm a big advocate of tables of descriptive statistics (with max's and min's--this can help you find coding issues) as well as pretty pictures. But when you are at the end stages of a project, it's important to think carefully about which robustness checks to include in the paper. Tatyana provides an excellent guide to help you through the thought process. I also really like her advice about choosing the preferred specification--not just in terms of which controls to include but also the baseline sample:  

"Your preferred specification should be based on the most natural sample of treated and control units for your study (e.g., counties in hurricane-prone states). This will often be the sample that includes the largest number of treated units and enough high-quality control units to estimate a credible counterfactual. For example, try not to pare down your control units so much that you have 3 times as many treated as control units. At the other extreme, it is also unlikely that having a sample with 10 times as many control as treated units will be more useful than something closer to a 1-to-1 ratio. If you have a panel dataset, your preferred specification should be based on a balanced panel."


Tuesday, June 29, 2021

Cool Data Alert: Refugee Resettlement Data 1975-2018

You can download data on 2.5 million individuals from 121 origin countries that entered the U.S. between 1975 and 2008. Refugees are recorded at the U.S. state and county levels, and several individual characteristics are also provided. Browse through the website here, and let me know if you come up with any brilliant ideas for research! 

Saturday, April 24, 2021

How to Prepare to Meet with a Seminar Speaker

Seminar speakers--regardless of how famous they are or how many papers they've written on your favorite topics--are still just people who want to hear about your work, and help out if they can. Hopefully they conversation will flow naturally. But I still think it's worthwhile to prepare a bit before these meetings, and I recently found a perfect little template to help you do just that. 

Joshua Goodman tweeted about a template he came up with to help his students describe their final projects in five minutes. I think it's excellent! What a great way to think about all of your projects! 


Thursday, April 8, 2021

Cool Data Alert: Immigration and Ancestry Instruments

Hooray for researchers who come up with new instruments. An extra special hoooray for those who make those instruments readily available and easy to use for other researchers. And even more points if the instruments can be used to study immigration. 

See it all here



Sunday, February 28, 2021

Crosswalks between Different Levels of Geography in the U.S.

Imagine you have geographic information at the county level in one data set but you want to merge in some additional information that is only available at the MSA level (or you have MSA in your data but want to merge in state level variables, etc.) What to do? Check out Geocorr 2014: Geographic Correspondence Engine provided by the Missouri Census Data Center. Super easy to use! 

H/T: Eric Brunner

Sunday, February 21, 2021

How to Write an Abstract

David Evans provides a handy dandy formula here. Here's an exercise for you: Find the 5-7 most closely related papers to yours and diagram their abstracts. Do they use any of the formulas David describes? A different formula? Do you see a difference by the quality of the journal they're published in.  

Describe what you find in the comments below!