iSixSigma

Data analysis

Six Sigma – iSixSigma Forums Old Forums General Data analysis

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #49256

    boettler
    Member

    What’s a good tool to determine if there is a statistical significant difference for defects data types?
    Example;
    2005: 20 Rejects
    2006 16 Rejects
    2007 12 Rejects
    Want to determine any statisctical difference betweeen these years.

    0
    #168115

    JimmyJam
    Participant

    Need more detail on the data…but just off the top, from the little info you provide….I would point you toward a chi square….and I say that with some reservation…not knowing the full picture…

    0
    #168130

    6Sigma Below The Belt
    Participant

    You can use chi-square for this BUT try to consider the following:
    1. Get more samples and ensure that the rejects in your samples can capture at least 5% of the volume
    2. Is it necessary to compare year to year or can you trim it down to quarter to quarter or monthly or even daily?
    3. If #2 is ok with you, try plotting first then see any variations before deciding if there’s really a significant difference.

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

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