I. Notes for Stats (11-28-06) II. Intro to Correlations III. IV.Correlation – trying to find out the relationship between two variable (Are they related? How much? In what way?) A. Regression – can we make predictions from two correlated variables (using X to predict Y) B. r = ∑(Z x * Z y )/(n – 1) i. Z x = (x – X)/s x ii. Gives us the degree (direction) and extent (magnitude) of a linear relationship V. Covariance – an unstandardized measures of the relationship between two variables A. Different from correlation in that we have no range B. The measures have not been standardized with each other (converted to z scores) C. C xy = ∑(x – X)(y – Y)/(N – 1) i. Similar to ∑(x – X) 2 /(N – 1), which you would use if you wanted to get the variance of s 2 x (variance). If you were then to divide that by s 2 x , you would have 1, but this only holds true on C xy depending on how close ∑(x – X) is to ∑(y – Y) ii. N = number of scores in x. This should be the exact same number of scores in
This is the end of the preview. Sign up
access the rest of the document.
This note was uploaded on 04/17/2008 for the course PSC 204A taught by Professor Emelio during the Fall '07 term at UC Davis.