Multiple Linear Regression
PERFORMING MULTIPLE REGRESSION ON SPSS
MULTIPLE LINEAR REGRESSION: INTRODUCTION
Regression Analysis is the
estimation of the linear relationship between a dependent variable and one or more independent variables or covariates.
MULTIPLE LINEAR REGRESSION:
INTRODUCTION
Market researchers puts forward Structural Equation Model (SEM) while determining the causal relationship of variables.
There are two types of analyses including
3
factor analysis Multiple regression analysis
The
causal relationship is modified by two types of variables.
Y
Dependent Endogenous variable Explained variable Response variable Predicted variable
x1, x2,x3 xk
Independent Exogenous variable Explanatory variable Control variable Predictor variable
4
Regressand
Regressor
EXAMPLE:
Assume
that the job satisfaction is the function of wage, occupation and time flexibility: Y (job satisfaction) = x(Wage)+x(Occupation)+ x(time flexibility) + e(others) Y = 0 + 1 X1+ 2 X2+ 3X3++
5
Copyright 2000, Julia Hartman
MULTIPLE LINEAR REGRESSION:
INTRODUCTION
Independent variables
matyr: Matriculation year totcum: Undergraduate GPA (Total Cumulative) nmtot1: MCAT Total 1992-present, most recent Dependent variable usmle2: USMLE Step 2
Next
MULTIPLE LINEAR REGRESSION:
STARTING THE PROCEDURE
In the menu, click Analyze
MULTIPLE LINEAR REGRESSION:
STARTING THE PROCEDURE
In the menu, click on Analyze Point to Regression
9
MULTIPLE LINEAR REGRESSION:
STARTING THE PROCEDURE
Point to Regression
Point to Linear
10
MULTIPLE LINEAR REGRESSION:
SELECTING VARIABLES
Choose the variables for analysis from the list in the variable box.
11
Move MATRICULATON DATE - YEAR, which is already highlighted, to the box labeled Independent(s) by clicking the arrow.
MULTIPLE LINEAR REGRESSION:
SELECTING VARIABLES
Move totcum to the Independent(s) box by clicking the arrow.
12
MULTIPLE LINEAR REGRESSION:
SELECTING VARIABLES
Move usmle2 to the box labeled Dependent by clicking the arrow.
13
MULTIPLE LINEAR REGRESSION:
REQUESTING STATISTICS
Request descriptive statistics by clicking the button labeled
Statistics
14
MULTIPLE LINEAR REGRESSION:
STATISTICS
Click Estimates Descriptive Model fit Part and partial correlations.
15
MULTIPLE LINEAR REGRESSION:
REQUESTING PLOTS
You can also request several different plots. Click the Plots button.
16
MULTIPLE LINEAR REGRESSION:
STANDARDIZED RESIDUAL PLOTS
Click in the box labeled Standardized Residual Plots Histogram Normal probability plot.
17
MULTIPLE LINEAR REGRESSION:
METHOD OF VARIABLE ENTRY
The independent variables can be entered into the analysis using five different methods. This tutorial includes the Enter and Stepwise methods.
18
MULTIPLE LINEAR REGRESSION OUTPUT:
DESCRIPTIVE STATISTICS
The labels and format of your output may be somewhat different.
19
MULTIPLE LINEAR REGRESSION OUTPUT:
CORRELATIONS
20
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
VARIABLES ENTERED
21
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
MODEL SUMMARY
22
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
ANOVA
23
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
COEFFICIENTS AND CORRELATIONS
24
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
RESIDUALS STATISTICS
25
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
RESIDUALS HISTOGRAM
26
MULTIPLE LINEAR REGRESSION ENTER METHOD OUTPUT:
PLOT OF STANDARDIZED RESIDUALS
27