psystats 0.1.0

Basic statistics and biostatistics for psychologists with an R-like API

psystats

Basic statistics and biostatistics for psychologists, with an R-like functional API. Part of the ANOVA Methods family (psystats, psymetrics, psyreport).

Free and open (MIT). Every function returns a result object that prints an R-style summary and can be handed to psyreport.report() for APA-7 output.

Install

pip install psystats

What's inside

Descriptives and tables — describe, freq, corr_matrix (correlation matrix with significance stars), and table1 (group comparison with automatic test selection by variable type and distribution).

Inferential tests with effect sizes — ttest (Cohen's d), anova (η² and ω²), chisq (Cramér's V).

Regression — linreg (OLS with standardized betas, R²/adjusted R², F test) and logreg (odds ratios with confidence intervals, McFadden pseudo-R²).

Risk factors — riskratio, oddsratio, and attributable_risk (risk difference with CI, attributable risk percent, population attributable risk and PAR%, number needed to treat).

Example

import pandas as pd
from psystats import table1, corr_matrix, linreg, riskratio, attributable_risk

df = pd.read_csv("study.csv")

print(table1(df, group="condition"))            # auto t-test / chi-square / etc.
print(corr_matrix(df, ["age", "score", "rt"]))  # r with * significance stars
print(linreg(df, "score", ["age", "condition"]))

rr = riskratio(df, exposure="smoker", outcome="disease", exposed=1, positive=1)
ar = attributable_risk(df, exposure="smoker", outcome="disease",
                       exposed=1, positive=1)
print(rr); print(ar)

All estimates are validated in the test suite against scipy and statsmodels.