PG Regression Mediation
Use for regression-style questions that Pingouin supports directly.
Load
Read:
../../references/supervision-gates.md../../references/pingouin-api-quickref.md../../references/apa-output-template.md if writing results.
Function Choice
- Continuous outcome, additive linear predictors ->
pg.linear_regression.
- Binary outcome ->
pg.logistic_regression with X, y.
- Single or multiple mediator path model ->
pg.mediation_analysis.
Required Inputs
- Outcome variable and scale.
- Predictor list.
- Covariates and whether they are theoretical controls.
- Binary coding for logistic regression.
- Mediation paths:
x, m, y.
- Bootstrap count and seed for mediation.
Code Patterns
Linear regression:
vars_needed = ["outcome", "x1", "x2"]
tmp = df.dropna(subset=vars_needed)
res = pg.linear_regression(tmp[["x1", "x2"]], tmp["outcome"],
add_intercept=True).round(3)
pg.print_table(res)
Logistic regression:
vars_needed = ["binary_outcome", "x1", "x2"]
tmp = df.dropna(subset=vars_needed)
res = pg.logistic_regression(tmp[["x1", "x2"]], tmp["binary_outcome"],
remove_na=False).round(3)
pg.print_table(res)
Mediation:
res = pg.mediation_analysis(data=df, x="x", m="mediator", y="outcome",
covar=["age"], n_boot=5000,
seed=42).round(3)
pg.print_table(res)
Reporting
- Regression: report coefficient, SE, statistic, p, CI, and model R-squared/adjusted R-squared if returned.
- Logistic: report coefficient and convert to odds ratio when useful with
np.exp(coef).
- Mediation: report total, direct, indirect paths, bootstrap CI, seed, and bootstrap count.
Guardrails
- Pingouin linear regression is OLS; it is not a mixed model or SEM package.
- For Pingouin 0.6.1,
logistic_regression takes X, y; verify the signature before using examples from older wrappers.
- Check residual assumptions and influential cases when substantive conclusions depend on them.
- Do not present mediation as causal unless temporal order, design, and assumptions support it.
- Do not hide automatic missing-data removal; report effective n.
- Mediation bootstrap p-values vary with seed and bootstrap count.
- End result-bearing answers with one compact S0-S5 audit line.
1---2name: pg-regression-mediation3description: Run or generate Pingouin code for linear regression, binary logistic regression, and mediation analysis with reproducible reporting.4---56# PG Regression Mediation78Use for regression-style questions that Pingouin supports directly.910## Load1112Read:1314- `../../references/supervision-gates.md`15- `../../references/pingouin-api-quickref.md`16- `../../references/apa-output-template.md` if writing results.1718## Function Choice1920- Continuous outcome, additive linear predictors -> `pg.linear_regression`.21- Binary outcome -> `pg.logistic_regression` with `X, y`.22- Single or multiple mediator path model -> `pg.mediation_analysis`.2324## Required Inputs2526- Outcome variable and scale.27- Predictor list.28- Covariates and whether they are theoretical controls.29- Binary coding for logistic regression.30- Mediation paths: `x`, `m`, `y`.31- Bootstrap count and seed for mediation.3233## Code Patterns3435Linear regression:3637```python38vars_needed = ["outcome", "x1", "x2"]39tmp = df.dropna(subset=vars_needed)40res = pg.linear_regression(tmp[["x1", "x2"]], tmp["outcome"],41 add_intercept=True).round(3)42pg.print_table(res)43```4445Logistic regression:4647```python48vars_needed = ["binary_outcome", "x1", "x2"]49tmp = df.dropna(subset=vars_needed)50res = pg.logistic_regression(tmp[["x1", "x2"]], tmp["binary_outcome"],51 remove_na=False).round(3)52pg.print_table(res)53```5455Mediation:5657```python58res = pg.mediation_analysis(data=df, x="x", m="mediator", y="outcome",59 covar=["age"], n_boot=5000,60 seed=42).round(3)61pg.print_table(res)62```6364## Reporting6566- Regression: report coefficient, SE, statistic, p, CI, and model R-squared/adjusted R-squared if returned.67- Logistic: report coefficient and convert to odds ratio when useful with `np.exp(coef)`.68- Mediation: report total, direct, indirect paths, bootstrap CI, seed, and bootstrap count.6970## Guardrails7172- Pingouin linear regression is OLS; it is not a mixed model or SEM package.73- For Pingouin 0.6.1, `logistic_regression` takes `X, y`; verify the signature before using examples from older wrappers.74- Check residual assumptions and influential cases when substantive conclusions depend on them.75- Do not present mediation as causal unless temporal order, design, and assumptions support it.76- Do not hide automatic missing-data removal; report effective n.77- Mediation bootstrap p-values vary with seed and bootstrap count.78- End result-bearing answers with one compact S0-S5 audit line.
Exekiel179