Levene's Test

Levene's test is a hypothesis test that checks whether the variance of a continuous outcome is equal across two or more groups, using the absolute deviations of observations from their group center.

Levene 검정은 집단 중심값으로부터의 절대편차를 이용해 두 개 이상의 집단에서 연속형 결과변수의 분산이 같은지를 검정하는 가설검정이다.

How do I check whether the variances of my groups are equal before running a t-test or ANOVA?

t-검정이나 분산분석을 하기 전에 집단 간 분산이 같은지 어떻게 확인하나요?

Levene's test checks the homogeneity-of-variance assumption by testing whether the spread of a continuous outcome differs across groups. It fits before an independent-samples t-test or ANOVA, and a significant result suggests using Welch's correction instead.

When to use Levene's Test

• Checking equal variances before an independent-samples t-test or one-way ANOVA
• One continuous outcome measured across two or more independent groups
• Data that may depart from normality, where the median-centered (Brown–Forsythe) version is robust

When to use something else

• Data are clearly normal and you want a test sensitive only to variance differences: use Bartlett's test
• You only need a robust group comparison: run Welch's ANOVA directly instead of pretesting

Assumptions and requirements

• Independent observations
• Groups are independent of one another
• Continuous outcome variable

What the analysis reports

• Levene's F statistic
• Degrees of freedom (k − 1, N − k)
• p-value
• Group variances or standard deviations

How to report it

Levene's test indicated that the assumption of equal variances was met, F([df1], [df2]) = [value], p = [value].

Related methods

• Bartlett's Test
• Brown–Forsythe Test
• Welch's ANOVA

Run Levene's Test on Quorant