{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Correlation, Regression

# Correlation, Regression - α vs p Alpha is the cut off...

This preview shows page 1. Sign up to view the full content.

-what does it mean if a coefficient is significant? -regression: predicting something from someone - use one to predict the variation in another -variation: how spread out the scores are -predicting how spread out a score is by looking at how spread out the other is SS total = SS predicted + SS error -of y (criterion) σ 2 total = σ 2 predicted + σ 2 error -“expect a positive correlation” – there is a tendency -above average on one, probably above average on the other (covariance) - how to tell which variable is a better predictor
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: α vs. p Alpha is the cut off determined for p (.05= take a 5% of being wrong) P value is based on the number calculated Choose the smallest p value – least likely to be wrong-coefficient will be closer to +1 or -1-closest to correlation line-eliminated most error-compare actual to predicted (tests if the prediction is good)-which one is closer? – less error- SS predicted : how spread out the scores would be based on the regression line...
View Full Document

{[ snackBarMessage ]}