Flashcards
FIELD ch 8: correlation
Cards in this deck (11)
- 1how to (conceptually) calculate the covariancemultiply the deviation for one variable by the deviation of another variable (this is the cross-product deviation). calculate the average value by dividing it by N-1
unstandardised - 2what is the name of the standardised covariance?(Pearson's) correlation coefficient r
- 3values of r that represent small, medium and large effectsmall 0.1
medium 0.3
large 0.5 - 4what is the correlation coefficient squared? name + explanationcoefficient of determination, a measure of the amount of variability in one variable that's shared by the other, so it tells you how much of the total variability is explained by this variable
- 5non-parametric version of the correlation coefficient + concept + how to calculate in SPSSSpearman's correlation coefficient (Rs), Spearman's Rho. ranking data and then applying Pearson's equation. useful when you have outliers or a non-normal distribution
correlate; bivariate; spearman - 6what is Kendall's tau?non-parametric correlation coefficient when you have a small data set with a large number of tied ranks, often better estimate than Spearman's
- 7point-biserial correlation coefficient and biserial correlation coefficient + how to do in SPSSwhen investigating relationship between two variables when one is dichotomous. when discrete dichotomy (e.g. pregnancy) use point-biserial correlation. when continuous dichotomy, use biserial correlation coefficient.
correlate; bivariate; pearson - 8semi-partial correlation (/part correlation)a measure of the relationship between two variables while adjusting for the effect that one or more additional variables have on one of those variables (so expresses the unique relationship)
proportion of the TOTAL variance of Y explained by X - 9how to calculate the part correlation (simple, not in SPSS)taking the square root of R square (makes sense because that's the strength of the unique relationship between two variables)
- 10partial correlationexpresses the unique relationship between x and y as a function of the variance in y LEFT OVER WHEN OTHER VARIABLES HAVE BEEN CONSIDERED
proportion of Y's variance that is explained by X when leaving out all of the parts that are explained by other variables - 11how to calculate partial correlation in SPSScorrelate; partial
variables you want to correlate in box variables and the variables you want to adjust for in 'controlling for;
Information
- Description
- university ssr exam 3 Flashcards on FIELD ch 8: correlation, created by Sabine - on 14/11/2020.
- Created by
- Sabine -
- Views
- 0
- Rating