iSixSigma

Which test is more powerful or recommended for comparing mean between 2t and Anova

Six Sigma – iSixSigma Forums Operations Call Centers Which test is more powerful or recommended for comparing mean between 2t and Anova

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #53661

    rohilla
    Participant

    Which test is more powerful or recommended for comparing mean between 2t and Anova for contineus data ?
    Assumption data is normal and test for equal variance has passed for both set of data.

    Regards
    Navin

    0
    #191020

    Robert Butler
    Participant

    You will have to give some more details – if it is just two populations then it really doesn’t matter As for normality and equal variance both are robust with respect to normality. The t-test (I’m assuming that’s what you mean when you write “2t” ) can handle unequal variance (and unequal sample size) as can ANOVA. With ANOVA it’s called Welch’s test.

    0
    #191026

    Schwarz
    Participant

    ANOVA (analysis of variance) and the two independent sample t test are mathematically identical. One must pay attention to the assumptions involved, however, and particularly the equal variance criteria.

    If the variances are unequal, then a transformation may remedy that. The Box-Cox diagnostic procedure will help to find the best transformation.

    Also, the use of Welch’s test, as Robert has suggested, is a good approach..

    0
Viewing 3 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic.