Download Research Methods in Psychology: Understanding Correlations - Prof. Jonathan T. Mordkoff and more Study notes Psychology in PDF only on Docsity! 031:010 - Research Methods in Psychology – Spring, 2009 – March 24th 1 some (more) definitions Experiment (strict) must have at least one manipulated variable (IV) Correlational Study all of the variables are measured - one is treated as the “predicted” variable - the others are treated as the “predictor” variables the difference is important for two reasons - different methods of analysis - different threats to interpretation 2 Correlations can be calculated between any two variables when both variables are naturally quantitative (i.e., both are interval or ratio scales; no orders) (e.g., height and respectability) the “standard” correlation coefficient is used the same goes when one of the variables is dichotomous (i.e., can be coded as 0s and 1s) (e.g., height and gender) but now it’s called a “point-biserial correlation” 5 Correlations can be calculated between any two variables when both variables are naturally quantitative (i.e., both are interval or ratio scales; no orders) (e.g., height and respectability) the “standard” correlation coefficient is used and the same goes when both of the variables are dichotomous (e.g., gender and asking for directions [tee hee]) but now it’s called a “phi coefficient” 6 Correlations can be calculated between any two variables when at least one of the variables is qualitative and takes on more than two values (e.g., race, religion, or hair color) then a different procedure must be used, called multiple regression (not covered in this course) but it’s still a type of correlation 7 Correlations look at (regular) correlations using a scatterplot 10 80 90 100 110 120 130 56 60 64 68 72 76 Height (in inches) E st im at e d I Q these data fit inside a tighter oval, so this correlation is stronger Correlations look at (regular) correlations using a scatterplot 11 80 90 100 110 120 130 56 60 64 68 72 76 Height (in inches) E st im at e d I Q these data form a nearly-straight line, so this correlation is nearly perfect ( ≈ +1.00 ) Correlations look at (regular) correlations using a scatterplot 12 80 90 100 110 120 130 16 20 24 28 32 Body Mass Index E st im at e d I Q Here’s a moderate negative correlation Correlations are greatly affected by the range of values 15 80 90 100 110 120 130 56 60 64 68 72 76 Height (in inches) E st im at e d I Q Full-range: +0.70 Restricted range: 0.00 5’6” and 5’7” only Correlations are greatly affected by the range of values 16 80 90 100 110 120 130 0 10 20 30 40 50 60 70 Estimated Age E st im at ed S tr en g th Full-range: 0.00 Restricted range: +1.00 minors only Correlations can be calculated between any two variables provide a measure of the linear relationship (only) - symbol: r - varies between -1.00 and +1.00 also provide a measure of how much of the variance in one variable is “explained” by the other variable - symbol: r 2 name: “coefficient of determination” - varies between 0.00 and 1.00 are greatly affected by the range of values - cannot generalize outside of the measured range 17
