Correlation and Simple Linear Regression

Term
1 / 16
Pearson's r
Click the card to flip 👆
Terms in this set (16)
Pearson's r
Measures the strength of a linear relationship.
Negative correlation
As one variable increases the other decreases.
Coefficient of determination
Measures percentage of variability in one variable that is predictable by knowledge of another variable
"Chance" or zero correlation
P > .05
Statistically significant correlation
P< .05
Ypredicted = b X + a
General case form of simple linear regression
Homoscedasticity of variances
Normal distribution of error around a regression line.
standard error of estimate
range around predicted score within which you are sure with a specified degree of certainty that the predicted score will indeed fall.
restriction of range
can make the value of a correlation seem lower than in fact in may be
outliers
can make the value of a correlation coefficient higher than it in fact may be
biased standard deviationdivisor is "n" rather than "n-1". must be used with Dr. Simpson's computational formulae for Pearson's r, linear regression, and standard error of estimateGraphs>Legacy>ScatterplotMenu Path on SPSS to get preliminary graph to allow an informed judgment about whether the relationship is linearAnalysis>Correlate>BivariateMenu path on SPSS to determine Pearson r values and whether or not they are statistically significantAnalysis>Regression>LinearMenu path on SPSS to take if Pearson correlation is significant and you wish to develop a prediction formulaMultiple linear regressionmore than one predictor"Dependent" variableIn correlational procedures, the criterion---that which you are trying to predict. The last collected variable