pingouin.intraclass_corr

pingouin.intraclass_corr(data=None, groups=None, raters=None, scores=None, ci=0.95)[source]

Intra-class correlation coefficient.

Parameters
datapd.DataFrame

Dataframe containing the variables

groupsstring

Name of column in data containing the groups.

ratersstring

Name of column in data containing the raters (scorers).

scoresstring

Name of column in data containing the scores (ratings).

cifloat

Confidence interval

Returns
iccfloat

Intraclass correlation coefficient

cilist

Lower and upper confidence intervals

Notes

The intraclass correlation (ICC) assesses the reliability of ratings by comparing the variability of different ratings of the same subject to the total variation across all ratings and all subjects. The ratings are quantitative (e.g. Likert scale).

Inspired from: http://www.real-statistics.com/reliability/intraclass-correlation/

Examples

ICC of wine quality assessed by 4 judges.

>>> import pingouin as pg
>>> data = pg.read_dataset('icc')
>>> pg.intraclass_corr(data=data, groups='Wine', raters='Judge',
...                    scores='Scores', ci=.95)
(0.727526, array([0.434, 0.927]))