|
This thread is carried over and continued in the Current Elsmar Cove Forums 
|
|
|
This thread is carried over and continued in the Current Elsmar Cove Forums
|
The New Elsmar Cove Forums
|
 The New Elsmar Cove ForumsThe Elsmar Cove Forums
 Statistical Techniques and 6 Sigma
 about a test of significance example
|
 UBBFriend: Email This Page to Someone! |
next newest topic | next oldest topic |
| Author | Topic: about a test of significance example |
|
Doctor Zaius unregistered |
posted 16 August 2001 10:52 AM
 
Two populations, before and after. The response shows statistical significance, Null hypothesis rejected, @ an alpha of 0.05 (t-test). Pretty standard stuff I guess. The above response is actually calculated by taking two measurements and dividing one by the other. The same tests of significance applied to the two component measurements shows no difference between populations for ONE of the measurements (significant at alpha = 0.8 or so). The OTHER component measurement shows significance (rejects the Null hypothesis) at an alpha of 0.15. Here's the question: How can I have two component measurements, one showing 'no difference' before and after, and the other showing difference at a less-than-spectacular alpha, BUT the test of significance on the CALCULATED response (one divided by the other) shows significance @ a very low alpha?? How can I get "good significance" from two components that offer fair and no significance? IP: Logged |
All times are Eastern Standard Time (USA) | next newest topic | next oldest topic |
  |
|
Please Visit the new Elsmar Cove Forums! All these threads are there and much more!
Main Site Search 
