iSixSigma

ANOVA analysis for Attribute Data

Six Sigma – iSixSigma Forums Old Forums General ANOVA analysis for Attribute Data

This topic contains 4 replies, has 4 voices, and was last updated by  Bob Peterson 18 years, 6 months ago.

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #27119

    B. Peterson
    Participant

    I would like to do ANOVA analysis on a population of data collected in a survey where men and women were asked to rate eight sample items ranging in score from 1 (low) to 10 (high). This was not a forced ranking, participants could rate multiple items with the same score.

    ANOVA analysis is normally used with Continuous data, whereas this survey data is Attribute. I’ve run the table and charts, the analysis appears reasonable.
    Can anyone tell me if using ANOVA in this way is acceptable?

    0
    #65973

    Cone
    Participant

    Treating the attribute data as variable is okay as long as at least 4 distinct categories are available and USED. Minimum count in each used cell should be at least 5. Variable R&R using a well defined criteria for each number, 1-10, would also be appropriate (make sure you are measuring the response, not you poorly defined scale).

    0
    #65983

    Neil Polhemus
    Participant

    What questions do you want to ask of the data? If your desire is to compare the men with the women, there are much better ways to handle it.

    0
    #65984

    B. Peterson
    Participant

    I intend using this attribute data for instructional purposes training others within our company, no actual decisions will be made based on the results of the analysis. I prefer using real “in-house” data to demonstrate Six Sigma tools, it makes for a more meaningful discussion.

    One use might be to compare Male & Female responses, although a 2-T test would probably be better used for that purpose. A comparison of responses for the eight different styles seems to me to be a better use for ANOVA and this data.

    Bob P

    0
    #65991

    Bob Peterson
    Participant

    Thanks Gary, your advise is appreciated.

    Bob

    0
Viewing 5 posts - 1 through 5 (of 5 total)

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