Welcome to researchpy’s documentation!

Researchpy produces Pandas DataFrames that contains relevant statistical testing information that is commonly required for academic research. The information is returned as Pandas DataFrames to make for quick and easy exporting of results to any format/method that works with the traditional Pandas DataFrame.

Researchpy is essentially a wrapper that combines various established packages such as pandas, scipy.stats, numpy, and statsmodels to get all the standard required information in one method. If analyses were not available in these packages, code was developed to fill the gap.

Formulas are provided with citations if the code originated from researchpy. All output has been tested and verified by comparing to established software packages such as Stata, SAS, SPSS, and/or R.

Moving forward, Researchpy will be using formula_like approach and is being streamlined during this process. For an example, see the new -difference_test- method.

Note

researchpy is only compatible with Python 3.x. Download using either:

  • pip install researchpy
    • For standard install

  • conda install -c researchpy researchpy
    • For installation through conda