PhD Courses in Denmark

Reproducible data analysis in Stata (4/3 +11/3 2025)

Graduate School of Health Sciences at University of Southern Denmark

A scientific article typically provides the first table presenting characteristics for the study population. The characteristics include for example frequencies for categorical variables, mean and standard deviations for continuous variables, p values etc. There should also include tables typically presenting results out of several statistical analyses. The ideal reproducible data analysis requires that the whole working process from data cleaning to the published results should be reproducible without any copy-and-paste because it is well-known that copy-and-paste Stata output can be not only tedious and time-consuming but also error prone. Therefore, workout a sequential Stata do-files automatically export both the descriptive and the analytical output is crucial not only to align with the principles of reproducible data analysis but also make the working process much more efficient.