The Elsmar Cove Wiki More Free Files The Elsmar Cove Forums Discussion Thread Index Post Attachments Listing Failure Modes Services and Solutions to Problems Elsmar cove Forums Main Page Elsmar Cove Home Page
Go Back   The Elsmar Cove Forum > Common Quality Assurance Processes and Tools > Six Sigma
Forum Username


Search the Elsmar Cove
Custom Search
Monitor the Elsmar Forum
Follow Marc & Elsmar
Elsmar Cove Forum RSS Feed  Marc Smith's Google+ Page  Marc Smith's Linked In Page   Marc Smith's Elsmar Cove YouTube Page  Marc Smith's Facebook Page  Elsmar Cove Twitter Feed
Elsmar Cove Groups
Elsmar Cove Google+ Group  Elsmar Cove LinkedIn Group  Elsmar Cove Facebook Group
Sponsor Links





Donate and $ Contributor Forum Access

Courtesy Quick Links

Links that Elsmar Cove visitors will find useful in your quest for knowledge:

Howard's
International Quality Services
Marcelo Antunes'
SQR Consulting
Bob Doering's
Correct SPC - Precision Machining

NIST's Engineering Statistics Handbook
IRCA - International Register of Certified Auditors
SAE - Society of Automotive Engineers
Quality Digest Portal
IEST - Institute of Environmental Sciences and Technology
ASQ - American Society for Quality

Related Topic Tags
data analysis, non-normal data distribution, t-test (statistical analysis), variance (data analysis)
Reply
 
Thread Tools Search this Thread Rating: Thread Rating: 1 votes, 5.00 average. Display Modes
  #1  
Old 5th February 2006, 04:04 AM
jackylpt's Avatar
jackylpt jackylpt is offline
Involved - Posts

 
Registration Date: Aug 2004
Location: Canada / Toronto
 
Posts: 38
Thanks Given to Others: 0
Thanked 0 Times in 0 Posts
Karma Power: 44
Karma: 10
jackylpt has less than 100 Karma points so far.
Question Is a t-test used in non-normal data analysis?

As Homogeneity of Varirance( Mintab) can test normal and non-normal data for variance.
My question is does t -test can do it? if not, how we can compare two non-normal data mean value? thanks

Sponsored Links
  #2  
Old 5th February 2006, 11:42 AM
Steve Prevette's Avatar
Steve Prevette Steve Prevette is offline
Deming Disciple

 
Registration Date: Feb 2004
Location: Aiken, SC
 
Posts: 2,318
Thanks Given to Others: 193
Thanked 1,350 Times in 734 Posts
Karma Power: 302
Karma: 14341
Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.
Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.Steve Prevette is appreciated, and has over 1700 Karma points.
Quote:
In Reply to Parent Post by jackylpt

As Homogeneity of Varirance( Mintab) can test normal and non-normal data for variance.
My question is does t -test can do it? if not, how we can compare two non-normal data mean value? thanks
Yes, the t-test is based upon an assumption of normality. If you are concerned about the normality of your data, here are some options:

1. Assume the Central Limit Theorm will get you to "good enough". Generally if you have more than 10 samples you probably are getting close.

2. Ignore the issue. With only 5 samples (for example) it is pretty hard to prove your data are not normal anyway.

3. If you know what distribtution the data are, or theoretically "ought to be", run a mathematical transform to shift them to normality. An example is time to failure data can tend to be exponential.

4. Run your own tests using random numbers and check distributions like what you have been seeing to see how good the t-test is at avoiding false alarms and avoiding failure to detects.

5. Shift to non-Parametric tests.

__________________

Steve Prevette
"A Passionate Statistician", ASQ CQE, ASQ Fellow
Sponsored Links

  #3  
Old 5th February 2006, 02:11 PM
Statistical Steven's Avatar
Statistical Steven Statistical Steven is offline
Statistician

 
Registration Date: May 2005
Location: New Jersey
Age: 49
 
Posts: 973
Thanks Given to Others: 240
Thanked 360 Times in 250 Posts
Karma Power: 143
Karma: 4229
Statistical Steven is appreciated, and has over 1700 Karma points.
Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.
Send a message via Yahoo to Statistical Steven
Homogenity of variance tests like the t-test assume normality. If your homogenity test shows no statistical difference between the two variances, the using a t-test to compare means is conservative. If the variances are not equal, using the Sattherwaite's approximation might help.

Again, you should first test the data for normality using Wilks-Shapiro or Kolmogorov-Smirnov.

__________________

Steven Walfish
When in doubt, ask your company statistician!
  #4  
Old 10th March 2011, 01:21 PM
Mexicanquality's Avatar
Mexicanquality Mexicanquality is offline
Involved in Discussions

 
Registration Date: Dec 2010
Location: Mexico
 
Posts: 22
Thanks Given to Others: 6
Thanked 6 Times in 5 Posts
Karma Power: 17
Karma: 40
Mexicanquality has less than 100 Karma points so far.
Re: Is a t-test used in non-normal data analysis?

Statistica steven:
Why not to use Anderson test instead of Kolmogorov?

thanks
  #5  
Old 10th March 2011, 07:14 PM
Miner's Avatar
Miner Miner is offline
Forum Moderator

 
Registration Date: Apr 2002
Location: Greater Milwaukee area, Wisconsin
 
Posts: 3,502
Thanks Given to Others: 603
Thanked 2,043 Times in 1,299 Posts
Blog Entries: 16
Karma Power: 438
Karma: 18659
Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.
Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.Miner is appreciated, and has over 1700 Karma points.
Look! Re: Is a t-test used in non-normal data analysis?

These tests are sensitive to departures from normality in different areas.

The Anderson-Darling is sensitive to departures from normality in the tail areas.

Wilks-Shapiro and Kolmogorov-Smirnov are sensitive to departures from normality in the center.

You should select the test that is most appropriate for your specific situation.

__________________

"A fool can learn from his own experiences; the wise learn from the experience of others." - Democritus, 460-370 B.C.
Thank You to Miner for your informative Post and/or Attachment!
  #6  
Old 11th March 2011, 10:23 AM
Statistical Steven's Avatar
Statistical Steven Statistical Steven is offline
Statistician

 
Registration Date: May 2005
Location: New Jersey
Age: 49
 
Posts: 973
Thanks Given to Others: 240
Thanked 360 Times in 250 Posts
Karma Power: 143
Karma: 4229
Statistical Steven is appreciated, and has over 1700 Karma points.
Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.Statistical Steven is appreciated, and has over 1700 Karma points.
Send a message via Yahoo to Statistical Steven
Re: Is a t-test used in non-normal data analysis?

Quote:
In Reply to Parent Post by Miner View Post

These tests are sensitive to departures from normality in different areas.

The Anderson-Darling is sensitive to departures from normality in the tail areas.

Wilks-Shapiro and Kolmogorov-Smirnov are sensitive to departures from normality in the center.

You should select the test that is most appropriate for your specific situation.
Miner, excellent point!

My general rule of thumb is to stick with a single test (I tend to use W-S) to avoid the "shopping" for a good p-value. Having said that, if you know that you might have a long tail, Anderson-Darling is perferred.

__________________

Steven Walfish
When in doubt, ask your company statistician!
  #7  
Old 11th March 2011, 10:51 AM
AdamP's Avatar
AdamP AdamP is offline
Involved in Discussions

 
Registration Date: Jun 2010
Location: Massachusetts, USA
 
Posts: 97
Thanks Given to Others: 24
Thanked 61 Times in 41 Posts
Karma Power: 28
Karma: 620
AdamP is appreciated, and has over 500 Karma points.AdamP is appreciated, and has over 500 Karma points.AdamP is appreciated, and has over 500 Karma points.AdamP is appreciated, and has over 500 Karma points.AdamP is appreciated, and has over 500 Karma points.AdamP is appreciated, and has over 500 Karma points.
Re: Is a t-test used in non-normal data analysis?

One thing to keep in mind -if you're using a software package such as Minitab, is what normality test they use by default. You can use several specific normality tests within Minnie, but it defaults to the A-D test for the basic reports. Just FYI. Also - I believe some of the tests are sample size sensitive, so you might take a look at how much data you have as an input to which test you select.

You can use a non-parametric test like Mood's median to see if any differences exist without worrying about distribution though, but take a look at what the data is. If it's something like a hole diameter then you might expect that to be normally distributed. However, if you're evaluation cycle times or arrival times, you would look more toward a lognormal, so a t-test might not be what you choose. Make sure you have continuous or interval data, not just numbers.

Cheers,

Adam
  #8  
Old 11th March 2011, 12:55 PM
Bev D's Avatar
Bev D Bev D is offline
Heretical Statistician

 
Registration Date: Jan 2003
Location: New England
 
Posts: 2,941
Thanks Given to Others: 1,229
Thanked 2,084 Times in 1,239 Posts
Karma Power: 374
Karma: 16645
Bev D is appreciated, and has over 1700 Karma points.
Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.Bev D is appreciated, and has over 1700 Karma points.
Re: Is a t-test used in non-normal data analysis?

all of the above complications and caveats are true for statistical tests that are based on the assumption of Normality.

on the other hand in the real world most processes, experiments and data are relatively straighforward. As long as the experimental structure is correct* the only necessarry analysis is to plot the raw data and look for the differences. The Tukey-Duckworth approach is typically sufficient for a statistical sanity check.

I've solved hundreds of problems and performed thousands of analyses of various types and with the exception of screenign experiments and thsoe that were looking for really small differences - I've never used - or needed a t-test or ANOV. (although I do admit that at times I will run the test after I've completed the analysis just for those individuals who 'have to see the p value' because graphs elude them) I have used ANOM on occassion but not lately. Why get all fancy when you don't have to?

*the biggest error I've seen in experimental analyses is not an incorrect statistical test, but a poorly designed experimental structure.

__________________

Statistics without physics is gambling; Physics without statistics is psychics
Thank You to Bev D for your informative Post and/or Attachment!
Reply

Lower Navigation Bar
Go Back   The Elsmar Cove Forum > Common Quality Assurance Processes and Tools > Six Sigma

Do you find this discussion thread helpful and informational?


Bookmarks


Visitors Currently Viewing this Thread: 1 (0 Registered Visitors (Members) and 1 Unregistered Guest Visitors)
 
Thread Tools Search this Thread
Search this Thread:

Advanced Forum Search
Display Modes Rate Thread Content
Rate Thread Content:

Forum Posting Settings
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Similar Discussion Threads
Discussion Thread Title Thread Starter Forum Replies Last Post or Poll Vote
Non-Normal Data F test and T test optomist1 Capability, Accuracy and Stability - Process, Machine, etc. 19 17th August 2011 10:47 AM
Is Cpk Analysis only for Data that came from Normal Distribution? Tavoludo Capability, Accuracy and Stability - Process, Machine, etc. 7 30th March 2011 01:27 PM
Capability Analysis of Non-Normal Data Seyed Statistical Analysis Tools, Techniques and SPC 23 14th October 2009 12:50 AM
Analysis of non-normal stratified data for cpk/ppk? Rupture test nwillnow Capability, Accuracy and Stability - Process, Machine, etc. 6 28th June 2006 12:34 PM
Normal Distribution - How do I test for normality? All the data or just the averages? DJN Statistical Analysis Tools, Techniques and SPC 8 13th February 2003 06:04 PM



The time now is 04:46 AM. All times are GMT -4.
Your time zone can be changed in your UserCP --> Options.


   


Marc Timothy Smith - Elsmar.com
8466 LeSourdsville-West Chester Road, Olde West Chester, Ohio 45069-1929
513 341-6272