External Packages (page 25 of 38)

Most of the methods and functions in researchpy take pandas data frames or series as arguments. The output of most researchpy methods and functions is also a pandas data frame, which makes it easy for us to use our results with any function or method that works with a traditional Pandas DataFrame object. We can get descriptive information about our data frame using the codebook (e.g., researchpy.codebook(my_df)) or univariate descriptive statistics using summary (e.g., researchpy.summary(my_df['sales']) or researchpy.summarize(my_df['sales']))

Other useful functions include: 1) researchpy.summary_cat() to summarize categorical Series, 2) researchpy.summary_cont() to summarize continuous Series, and 3) researchpy.crosstab() to build a cross tab report.

With researchpy, we can perform t-tests, a variety of difference tests, correlation matrices, ANOVAs, sign-rank tests, chi-square tests of cross tabulated data, and OLS regressions.