MVDA - SPSS
Step by Step Guide
For the SPSS test on June 17th, 2021
,Content
CHAPTER 1 MRA – WEEK 1 4
ASSIGNMENT 1 4
PEARSON CORRELATIONS 4
LINEAR REGRESSION 4
GET RID OF THE OUTLIER(S) 5
REMOVE THE NON-SIGNIFICANT PREDICTORS 5
LINEAR REGRESSION WITH SCATTERPLOT OF THE REMAINING PREDICTORS 5
ASSIGNMENT 2 6
LINEAR REGRESSION 6
LINEAR REGRESSION WITH SCATTERPLOT 7
LINEAR REGRESSION WITH 2 BLOCKS 8
LINEAR REGRESSION WITH 3 BLOCKS 8
CHAPTER 2 ANOVA - WEEK 2 10
ASSIGNMENT 1 10
CROSS TABULATION 10
ANOVA 10
GRAPHICAL INSPECTION OF THE NORMALITY ASSUMPTION 11
ASSIGNMENT 2 12
CROSS TABULATION 12
ANOVA 12
GRAPHICAL INSPECTION OF THE NORMALITY ASSUMPTION 13
ANOVA WITH POST HOC TESTS 13
CHAPTER 3 ANCOVA - WEEK 3 15
ASSIGNMENT 1 15
ANOVA 15
ANOVA WITH TUKEY POST HOC TEST 16
ANOVA TO CHECK USEFULNESS OF COVARIATE 16
CORRELATION TO CHECK USEFULNESS OF COVARIATE 16
SCATTERPLOT 17
ANCOVA TO CHECK PARALLEL REGRESSION SLOPES 17
ANCOVA TO ASSESS EFFECT WITH COVARIATE 17
ASSIGNMENT 2 18
ANOVA 18
ANOVA TO ASSESS DIFFERENCE BETWEEN GROUPS 19
CORRELATION TO CHECK CORRELATION BETWEEN COVARIATE AND DEPENDENT VARIABLE 19
SCATTERPLOT 19
ANCOVA TO CHECK PARALLEL REGRESSION SLOPES 20
ANCOVA TO ASSESS EFFECT WITH COVARIATE 20
CHAPTER 4 WEEK 4 22
2 MVDA SPSS | Step By Step
,ASSIGNMENT 1 22
LINEAR REGRESSION ANALYSIS 22
LOGISTIC REGRESSION 22
REMOVE THE NON-SIGNIFICANT PREDICTORS FROM THE LOGISTIC MODEL 23
ASSIGNMENT 2 24
LINEAR REGRESSION ANALYSIS 24
LOGISTIC REGRESSION WITH 1 BLOCK 24
LOGISTIC REGRESSION WITH 2 BLOCKS 25
LOGISTIC REGRESSION WITH 2 BLOCKS 26
CHAPTER 5 WEEK 5 27
ASSIGNMENT 1 27
MANOVA 27
MANOVA ON ONLY THE SIGNIFICANT VARIABLES 28
ASSIGNMENT 2 28
DISCRIMINANT ANALYSIS 29
ASSIGNMENT 3 30
CROSS TABULATION 30
MANOVA 30
CHAPTER 6 WEEK 6 33
ASSIGNMENT 1 33
RM ANOVA 33
ASSIGNMENT 2 34
RM ANOVA 35
CHAPTER 7 WEEK 7 37
ASSIGNMENT 1 37
LINEAR REGRESSION - MEDIATION ANALYSIS: X à Y (TOTAL EFFECT) 37
LINEAR REGRESSION - MEDIATION ANALYSIS: X à M (INDIRECT EFFECT A) 37
LINEAR REGRESSION - MEDIATION ANALYSIS: M [CORRECTED FOR X] à Y (INDIRECT EFFECT B & C’) 37
SOBEL TEST FOR MEDIATION ANALYSIS 38
ASSIGNMENT 2 38
LINEAR REGRESSION - MEDIATION ANALYSIS: X à Y (TOTAL EFFECT) 39
LINEAR REGRESSION - MEDIATION ANALYSIS: X à M (INDIRECT EFFECT A) 39
LINEAR REGRESSION - MEDIATION ANALYSIS: M [CORRECTED FOR X] à Y (INDIRECT EFFECT B & C’) 39
SOBEL TEST FOR MEDIATION ANALYSIS 40
MVDA SPSS | Step By Step 3
, Chapter 1 MRA – Week 1
Assignment 1
Use the file MVDA_MRA_1.sav. In a study with middle-class children (11-13 years old),
educational psychologists investigated whether academic performance (GPA) can be
predicted from IQ, age, gender and/or self-concept (SC).
Pearson correlations
Calculate the Pearson correlations between the five variables
1. Analyze à Correlate à Bivariate
2. Drag variables (GPA, IQ, Age, Gender, SC) to Variables
3. Correlation Coefficients, select: Pearson
4. OK
a) What is the sample size N ?
Check in Correlations table
Answer: N = 78
b) Does it make sense to perform a linear regression of GPA on IQ, age, gender and/or self-
concept?
Answer: Yes, IQ could be related to GPA and the other variables could also be of influence.
c) Which variable is likely to be a good predictor of GPA?
Check in Correlations table
Answer: IQ has the highest correlation with GPA so would likely be a good predictor of
GPA.
Linear regression
Perform a linear regression of GPA on IQ, age, gender and self-concept. In Statistics, ask for
part and partial correlations, and collinearity diagnostics. In Save ask for Cook’s distances and
Leverage values.
1. Analyze à Regression à Linear
2. Drag dependent (Y) variable (GPA) to Dependent
3. Drag independent (X) variables (IQ, Age, Gender, SC) to Independent(s)
4. In Statistics, select: Part and partial correlations, and Collinearity diagnostics.
5. In Save, select: Cook’s Distances and Leverage values.
6. OK
d) Can the null hypothesis of no relationship between GPA and IQ, age, gender and/or self-
concept be rejected?
Check in ANOVA: significance lower than alpha .05, so can the null hypothesis be rejected?
Answer: Yes, F(4, 73) = 23,117, p < .001
e) How much variance of GPA is explained by IQ, age, gender and SC together?
Check in Model Summary: R Square is .559 = 56%.
Answer: 56% of the variance in GPA is explained by IQ, age, gender and SC together.
4 MVDA SPSS | Step By Step
