Chapter 8 Flashcards
regression
use the assoication to predict
regression equation
y=mX+b, y=B0+B1x
goal of regression analysis
to develop a regression equation from which we can predict 1 outcome variable on the basis of 1 or more other predicitor variables
regression line
the line of best fit
linear regression
when variables are lineraly related we can describe their relationship with the equation for a straight line
Y
the variable we would like to predict, outcome
X
the variable we are using to predict 1, predictor variable
B0
y-intercept of the line that best fits the data, also called the regression constant
B1
indicates strength of relationship, regression coefficient
linear relationship
correlation, means a straight line can be drawn through the data in the scatter plot
multiple regression analysis
more than one predictor variable, each predictor has a coefficient
more predictors means
better prediction, account for more systematic variance
coefficent of determination
R2
simultaneous
equation with all predictors
stepwise
add 1 at a time, note how much increase in R with each predictor
hiearchical
add 1 at a time
multilevel modeling
intended to analyze data sets with a nested structure, groups within groups
variance accounted for
the percent of systematic variance in the outcome variable accounted for by variation in the predictor variable
factor analysis
analyze the relationships among a large number of variables
factors
related, correlated variables
