iSixSigma

ANOVA

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #32197

    ROSS
    Member

    As we know, the meaning of the Stat > ANOVA > One-way > Comparisons is as below, but I still have some confusion in the error rate , can somebody give a brief  introduction

    No
    Method
    Purpose
    error rate

    1
    Tukey’s, family error rate;
    all pairwise difference
    family

    2
    Fisher’s, individual error rate;
    all pairwise difference
    individualy

    3
    Dunnett’s family error rate;
    comparison to a control
    family

    4
    Hsu’s MCB, family error rate
    comparison to a best
    family

    0
    #85646

    RR
    Member

    Tony –
    Are you looking for a textbook response to help you in your own application or are you looking for something (examples, etc.) that will help you speak to it?   
    RR

    0
    #85647

    ROSS
    Member

    RR:
       Thanks for your feedback.
       I am studying the ANOVA, and I do not know the meaning the family error rate and individually error rate in the setting of Stat > ANOVA > One-way > Comparisons ,  can you give me a brief  introduction for it , some sample is better, thank you very much.

    0
    #85650

    Scott
    Member
    #85665

    Zilgo
    Member

    In a basic sense, you are setting a limit on the probability that the confidence intervals will not contain the true difference between means.  Individual error rates are equal to the probability that any one confidence interval does not contain the true difference while family error rates are equal to the probability that any one or more of the confidence intervals does not contain the true difference. 

    0
    #85713

    ROSS
    Member

    RR Rick Zilgo:
     Thanks for your help, I will try next step.

    0
Viewing 6 posts - 1 through 6 (of 6 total)

The forum ‘General’ is closed to new topics and replies.