iSixSigma

ANOVA – To Stack Or Not To Stack Data

Six Sigma – iSixSigma Forums General Forums Tools & Templates ANOVA – To Stack Or Not To Stack Data

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #53666

    Doa
    Participant

    Hi To All,

    Observational Data – I have a stamped part with 10 dimensions; with two factors that are changed, one factor has two levels, the second has three levels. Each setting combination is used to stamp six parts.

    The above combinations give 360 data points, given that all 10 dimensions are on the same part, and I am attempting to identify that combination of settings that produces parts with the mean closest to target, is it a good idea to stack the data into a single column and evaluate the data; confirm equal variances, them a Two-Way ANOVA?

    Thanks for your assistance…

    Marty

    0
    #191047

    MBBinWI
    Participant

    Marty: Looks more like a DOE to me. If you don’t have sufficient data to meet the requirements in Minitab to analyze with the data that you have, you can either add more or use a gen linear model to analyze.

    0
    #191048

    Doa
    Participant

    Thank you for the response….ahhh Minitab is a beautiful thing.

    First it was all about visualizing and structuring the data, then slicing and dicing the data with Two-Way ANOVA, then GLM, then a DOE.

    I am at the end of my BB certification journey, this is part of an extra credit case study…finally clarity of how it all works, how to structure the data and what Minitab
    tool(s) to use.

    In this particular case, I had two factors, one with two levels, the other with three; ran ANOVA….Press Tonnage factor was significant p=0.001, Material factor was not significant, means revealed different; checked for normalcy of residuals, residuals v order etc, all checked good the ran GLM, and DOE all confirmed the same.

    To be for sure for sure, after running Two Way ANOVA, and discovering that at least one mean of the three setting factor means was not the same as the others, I ran an ANOM confirmed again. Are you a BB or a MBB?

    Regards,
    Marty

    0
    #191051

    MBBinWI
    Participant

    Would you mind sending the data set? I’d be interested in looking at it. Just add @wi.rr.com to my name.

    0
    #191052

    Doa
    Participant

    Just the raw (excel) data or minitab output? I don’t think this site supports minitab uploads,

    Regards,
    Marty

    0
    #191055

    MBBinWI
    Participant

    either way. please send to my e-mail by adding @wi.rr.com to my name. Thanks.

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

You must be logged in to reply to this topic.