regression III

Regression III - Quantitative Methods Regression III March 6 2008 John M Ackroff 2008 Where are we Correlation the degree to which two variables on

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
  © John M. Ackroff 2008 Quantitative Methods Regression III March 6, 2008
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
  © John M. Ackroff 2008 Where are we? Correlation – the degree to which two  variables on a single individual are  related. Regression – the ability to predict the  value of one variable for an individual  based on the value of the other.
Background image of page 2
  © John M. Ackroff 2008 Where did we leave off? How does being able to predict Y from  X help?
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
  © John M. Ackroff 2008 What’s is all about? We do statistics to see if we can find  differences. We have ideas about how things work  in the world, and we want to find out if  our ideas are correct. I have a treatment for X I believe people will behave this way in  situation X but this other way in situation Y.
Background image of page 4
  © John M. Ackroff 2008 So I have to do statistics If I want to claim that I’ve really found  something, I have to show that my  results aren’t due to random chance. By convention, the likelihood of my  getting the results I got by chance alone  must be less than 5% in order for me to  claim I’ve found something.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
  © John M. Ackroff 2008 Finding differences More specifically, the chance of my  finding a difference between my  experimental and control group as large  as I did by chance alone must be < 5%. The problem is that there’s variability in  my measurements.
Background image of page 6
  © John M. Ackroff 2008 The Problem of Variability Suppose I have a drug that I think promotes  weight loss. I administer it to 30 subjects, who lose 10  pounds each; SS = 3000 (30 * 10 2 ). If all Ss weigh exactly the same before and  after, and all lose exactly the same amount,  the SS is all attributable to my drug. How likely is that ?
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
© John M. Ackroff 2008 Variability  In fact, μ B  = 210, s B  = 10; μ A  = 180, s A  =  8 So now, some of my variability is within  subjects before treatment, some is  within subjects after treatment, and  some is between before and after  conditions. How much is where?
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/05/2008 for the course PSYCH 200 taught by Professor Ackroff during the Spring '08 term at Rutgers.

Page1 / 30

Regression III - Quantitative Methods Regression III March 6 2008 John M Ackroff 2008 Where are we Correlation the degree to which two variables on

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online