Most natural experiments do not require any “fancy” statistical analyses. Therefore, all basic statistical programmes (e.g., SPSS, Stata, SAS, R, Mplus etc.) can be used for data analyses. However, some users have written very helpful programmes that facilitate data analyses (my sincere thanks to everyone who has contributed to the great programmes!). Below, you find a list of user-written programmes I use in my research. Please note (1) that I am not the author of these programmes (just a thankful user :-)) and (2) that I only know programmes for Stata (but please feel free to send me some recommendations for other statistical programmes).
Regression Discontinuity Design
Calonico, Cattaneo, Farrell and Titiunik have create the very helpful programme “rdrobust” for Stata and for R. The programme provides, among others, point estimators, confidence intervals estimators, and regression discintinuity plots. To install the programme in Stata, just type “ssc install rdrobust”. You will find more information – including references to Calonico’s, Cattaneo’s, Farrell’s and Titiunik’s work – on their website.
Difference-in-Difference Design
Villa created the programme “diff” for Stata that helps to estimate treatment effects in difference-in-difference designs. There is a nice presentation by Villa on a Stata Meeting in which he explains the programme in more detail. To install the programme in Stata, just type “ssc install diff”. For more information about the programme, see Villa’s paper in the Stata Journal:
Villa, J. M. 2016. diff: Simplifying the estimation of difference-in-differences treatment effects. The Stata Journal, 16(1): 52-71.