# questions

Six Sigma – iSixSigma › Forums › Old Forums › General › questions

- This topic has 6 replies, 5 voices, and was last updated 16 years, 8 months ago by Kat.

- AuthorPosts
- November 13, 2002 at 5:00 pm #30769
I have two quick questions, if someone could help, it would be greatly appreciated. I have a dataset that I need to apply some statistical measurement tools with

What is the best tool to yield a prediction equation for my factors, would it be R&R or the Anderson-Darling?

Also what would be the most beneficial test if I was testing for a difference in central tendency between various variables?

Please advise, thanks!

Student

0November 13, 2002 at 5:48 pm #80605Student,

on question (1) my answer is neither. Regression or DOE are your best bets for a prediction equation.

on (2) “most beneficial” test is subjective…I would vote for Anova.

pablo0November 13, 2002 at 7:52 pm #80610Someone at the office suggested the Levene’s Test to test for a difference in central tendencies does someone have a clear definition for this test?

0November 13, 2002 at 8:12 pm #80611Student:

I agree with Pablo on number 1, Regression analysis or DOE will help you obtain your equation.

On question number 2, I disagree just because it depends in the nature of your data. If your data shows a Normal behavior then you can use one way anova or a 2 sample T test to compare means for both groups of data (assuming you have 2 groups). If the data is Not normal, then you want to test the medians by using a non parametric test such as Mann-Whitney or Kruskall-Wallis.

The Levene test is used to test for differences in the standard deviation of non-normal data sets.

Hope this helps,

Fernie0November 13, 2002 at 9:01 pm #80616

Mike CarnellParticipant@Mike-Carnell**Include @Mike-Carnell in your post and this person will**

be notified via email.Fernie,

I agree with your post. Small point – Levenes tests variances.0November 13, 2002 at 9:04 pm #80617

Mike CarnellParticipant@Mike-Carnell**Include @Mike-Carnell in your post and this person will**

be notified via email.Actually if you are using Minitab you will get Bartletts and Levenes at the same time. You need to understand if you are dealing with normal or non normal data to know which to use.

0January 27, 2004 at 12:14 am #94659Mike –

Just a couple of follow up questions to your response below. As you indicated, Minitab does produce Bartlett and Levene at the same time. If you do not know the distribution of the data or the data is non-normal, it is suggested to use Levene’s test.

Have you ever encountered a situation in Minitab where Bartlett’s test is less than .05 and Levene’s is greater than .05? If so, what would you recommend as a next step?

Thanks.0 - AuthorPosts

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