Email discussion lists for the UK Education and Research communities

## allstat@JISCMAIL.AC.UK

#### View:

 Message: [ First | Previous | Next | Last ] By Topic: [ First | Previous | Next | Last ] By Author: [ First | Previous | Next | Last ] Font: Proportional Font

#### Options

Subject:

Regression Modelling with inclusion of sample sizes

From:

Date:

Wed, 18 Nov 2009 08:45:02 +0000

Content-Type:

text/plain

Parts/Attachments:

 text/plain (24 lines)
 ```Hi allstaters I am a 2nd year PhD student and I am working on elasticity of rubber bands. I have been running an experiment with rubber bands of different shapes and sizes which I have grouped based on these features. Next, I measure, for each rubber band, their maximum elasticity. I have got 6 groups of rubber bands with various sample sizes e.g., n1=5, n2=10, n3=8 and so on. It is stipulated by the manufacturers that the expected mean for all of these 6 groups should be, say, 10. However, I'd like to show that, in fact, they have a mean elasticity of 8. I'd like to conduct a regression modelling in R, SPSS or Minitab so to capture the level of these groups whilst including their sample sizes as weights. I have looked through the literature but with no success. The only weights referred to are coeffcients weights. Has anyone got any idea please? Also, out of curiousity only, could someone point me in the right direction for this other problem. If n=10 and the expected mean is 50, what would be the expected mean for n=6, assuming normality. I've always wanted to try to solve this problem. Many thanks and Best Regards, Chris Roberts ```