Shapiro_IC2009_Feature_Blur_and_the_break_of_the_Curveball

Shapiro_IC2009_Feature_Blur_and_the_break_of_the_Curveball...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
More on Correlation Topics . . . • A look at the definitional r computation • Four typical uses for correlations • Problems resulting from: –Restricted range –Outliers r 2 — coefficient of determination • Hypothesis testing –does the correlation exist in the population? –hypotheses, computation, df • Point-Biserial Correlation • Phi-Coefficient Pearson Product-Moment Correlation Coefficient: Pearson r A look at how the equation works Why does the sign of r match the slope? • Positive slope / –More positive products •(+)(+) and (-)(-) –Positive sum –Positive mean ( r ) • Negative slope \ –More negative products •(+)(-) and (-)(+) –Negative sum –Negative mean ( r ) How is the strength of r explained? r is greatest (close to 1 or -1) when the z -score products are greatest • The products are greatest when the points fall on along a straight line •the line must be tilted, at least slightly –a horizontal or vertical line makes r = 0 z Y z X shrinks as line becomes less perfect 1 2 3 4 5 6 7
Background image of page 1

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

View Full DocumentRight Arrow Icon
Y X The point of correlation • Prediction –the stronger the correlation, the more the two variables predict each other –knowing one tells you about the other –the closer r is to -1 or +1, the more accurate the prediction The point of correlation • Validity –Do two tests measure the same thing? –Suppose you develop a new, faster way to measure math ability than the SAT •If it correlates with the standard SAT math test, then a reasonable conclusion is that it measures the same thing (math ability) The point of correlation • Reliability –Reliability of a measurement procedure refers to its ability to produce stable and consistent measurements •with the same individuals, under the same conditions –With reliability, such scores should be positively, and strongly, correlated The point of correlation • Theory Verification –We can also, of course, theorize about relationships between variables •e.g., introverted individuals are more depressed
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 08/22/2011 for the course PSY 207 taught by Professor Pfordesher during the Fall '07 term at SUNY Buffalo.

Page1 / 9

Shapiro_IC2009_Feature_Blur_and_the_break_of_the_Curveball...

This preview shows document pages 1 - 3. Sign up to view the full document.

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