Comparison Testing Questions

Six Sigma – iSixSigma Forums Old Forums General Comparison Testing Questions

Viewing 1 post (of 1 total)
  • Author
  • #47187


    I am studying comarison testing and have a couple quick ( i hope) questions for the SMEs:

    It talks of using essentially two types of approaches:  randomization tests and t tests.  I understand the latter, but not the former…it talks of using historical data to build a distribution between differences of means, plotting them, and calculating a probability level….got that…but then it speaks of randomization and exchangability theory…i understand the concept (kinda), but have no idea how i would apply it, or even if i would want to… this a practical technique or one buried in the workings of MTB, etc?  Why not just use the t test if appropriate assumptions hold true? 
    I dont truly understand the layout of the t table…it appears to be used for caluculating the prob level as well as Control Limits….what are the row, colomn, and matrix numbers referring to and how do you run through them to find CL and probability level from the t stat?

Viewing 1 post (of 1 total)

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