Chi-Square Test of a Frequency Distribution

From Q
Jump to navigation Jump to search

Tests the null hypothesis that the observed values of a categorical variable diff from the expected values. The test statistic is

[math]\displaystyle{ X^2 = \sum^s_{k=1} \frac{(o_{kj} - e_{kj})^2}{e_{kj}} }[/math]


[math]\displaystyle{ o_{k} }[/math] are the observed values,
[math]\displaystyle{ e_{k} = }[/math] are the expected values,
[math]\displaystyle{ p \approx \Pr(\chi^2_{(s-1)} \ge X^2) }[/math]

See also