iSixSigma

Capability sixpack in Minitab

Six Sigma – iSixSigma › Forums › Old Forums › General › Capability sixpack in Minitab

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • March 16, 2007 at 9:44 am #46440

    Maciek
    ★ 10 Years ★
    Participant

    I have some problem with analysing my data. I’m trying to do capability analyse with Minitab – capability sixpack. The normality test done in capability sixpack graph showed p-value 0,100 (normality distribution), Kolmogorov-Smirnov test, p-value >0,15 (normality distribution). So my result are strange. Two from three tests showed normality distribution.
    I would like to change the normality test in capability sixpack in MInitab. Is there any options to change that? I was looking for change this option but without success.

    0
    March 16, 2007 at 12:37 pm #153458

    Caliandro
    ★ 10 Years ★
    Participant

    I never got results as you describe .. and I’m curious to see.
    If you want send me the  .MPJ file .
    emilio (dot) caliandro (at) fastwebnet (dot)  it 
    Emilio

    0
    March 16, 2007 at 3:36 pm #153478

    Erik L
    ★ 10 Years ★
    Participant

    Maciek,
    Without the data I can only hypothesize.  But the effect that you are experiencing (flipping null and alternate states for normality) could be a function of sample size.  Are you looking at a large data set (say g.t. 5,000 data points?).  If so, this could be the culprit.  JMP has a default that when it senses large amounts of data, it will migrate from the Shapiro-Wilk Normality test to the the KS test due to Shapiro-Wilk overrejecting the null with large amounts of data…
    Regards,
    Erik

    0
  • Author
    Posts
Viewing 3 posts - 1 through 3 (of 3 total)

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