The document provides an overview of ANCOVA and MANCOVA, including definitions, applications, and examples in SPSS. ANCOVA is used to examine differences in mean values while controlling for independent variables, whereas MANCOVA extends this analysis to multiple dependent variables. The document also includes research questions and results demonstrating the effectiveness of different diets on weight loss, as well as how to perform these analyses in SPSS.

1. “ ANCOVA AND MANCOVA”
PRESENTED BY: PRUM ROTANA (MS.)
DATE: FEB 17TH , 2019
1
The University of Cambodia
College of Education
Workshop on “ SPSS Analysis for MRP and Thesis”

2. Content
1. COVARIANCE
2. ANOVA
3. MANCOVA
2

3. Covariance
 the mean value of the product of the deviations of two variates
from their respective means.
 provides insight into how two variables are related to one
another.
 A positive covariance means that the two variables at hand are
positively related, and they move in the same direction.
 A negative covariance means that the variables are inversely
related, or that they move in opposite directions.
3

4. Covariance 4

5. 5

6. ANCOVA
 Analysis of covariance (ANCOVA) is used in examining the
differences in the mean values of the dependent variables that are
related to the effect of the controlled independent variables while
taking into account the influence of the uncontrolled independent
variables.
6

7. Example Question of ANCOVA
Does test score effect by the level of Education regarding the
number of hours spent studying?
7
Level of
Education
Number of
Hours Spent
Studying
Test Score
Independent
Variable
Moderating
Variable

8. Why Ancova?
 * In experimental designs, to control for factors which
cannot be randomized but which can be measured on an
interval scale.
 * In observational designs, to remove the effects of
variables which modify the relationship of the categorical
independents to the interval dependent.
 * In regression models, to fit regressions where there are
both categorical and interval independents.
8

9. Practice
Research Question
Does Weight Loss is influenced by Diet controlling by
Height?
DV: Weight Loss
IV: Diet
Moderator: Height
Download: ANCOVA PRACTISE
9

10. ANCOVA IN SPSS 10
Create WEIGTH LOSS Variable
Transform  Calculate Variable

11. 11

12. Analyze General Linear Model Univariate 12

13. Click on ‘Option’ botton 13

14. Click ‘Continue’ Ok 14

15. Interpretation
• Among the 3 types of diet,
Vegetarian Diet is more
effective for Weight Loss.
Mean= 3.153
• Std. is similar between each
type
YET there is an idea that :
15
Height will add more
influence on Weight Loss

16. Tests of Between-Subjects Effects
Dependent Variable: Weight loss
Source
Type III
Sum of
Squares
df Mean
Square F Sig.
Partial
Eta
Squared
Corrected
Model 7.523a 3 2.508 4.165 .012 .258
Intercept
.888 1 .888 1.475 .232 .039
Height
.033 1 .033 .054 .817 .002
Diet
7.118 2 3.559 5.912 .006 .247
Error
21.673 36 .602
Total
303.724 40
Corrected
Total 29.195 39
a. R Squared = .258 (Adjusted R Squared = .196)
• Since p= .006, Diet has
significant effect on weight
loss.
• Moreover, it has moderate
effect on weight loss as
Partial Eta Squared is only
.247= 24.7%.
Note: 0.2  Small effect
0.5 Moderate effect
0.8  Large effect
16

17. Since ANOVA is
Significance, Post Hoc test
are carried out
Analyze Compare Mean
One-way Anova
17

18. Estimated Marginal Means
Diet
Dependent Variable: Weightloss
Diet Mean Std. Error
95% Confidence Interval
Lower Bound Upper Bound
Atkin Diet 2.658a .221 2.210 3.105
Vegetarian Diet 3.141a .222 2.691 3.590
Zone Diet 2.101a
.207 1.680 2.522
a. Covariates appearing in the model are evaluated at the following values:
Height = 161.98.
• Inserting Height:
Vegetarian Diet is still more effective
than others with Mean= 3.141
• No Height:
Vegetarian Diet proceed others with
Mean= 3.153
 Conclusion:
18
The Effect of Height had been removed
or not account for weight loss

19. MANCOVA
19

20. Research Questions
 Do the various school assessments vary by grade level
after controlling for gender?
IV: grade level
DV: various school assessments
Moderator: gender
20

21.  Which diseases are better treated, if at all, by either
X drug or Y drug after controlling for length of
disease and participant age?
IV: X drug or Y drug
DV: Diseases
Moderator: length of disease and participant age
21

22.  Do the rates of graduation among certain state
universities differ by degree type after controlling
for tuition costs?
IV: Degree type
DV: the rates of graduation
Moderator: Tuition costs
22

23. MANCOVA
 Multivariate Analysis of Covariance an extension of
analysis of covariance methods to cover cases where there
is more than one dependent variables and where the
control of concomitant continuous independent variables –
covariates/Moderator – is required.
23

24. MANCOVA 24

25. Type of MANCOVA
Do gender and the outcome of the final exam
influence the standardized test scores of math,
reading, and writing?
IV: the outcome
DV: test scores (math, reading, and writing)
Moderator: : gender
25
 One-way MANCOVA

26. Two-way MANCOVA
2 IVs + 1 Moderator + 2 or more DVs
26

27. One-way MANCOVA IN SPSS
Research Question
Are the number of working hours and income
influenced by the age group within the level of
education?
Download: MANCOVA File Practice
27

28. 28
Analyze General Linear Model Multivariate

29. Data Interpretation
29

30. 30

31. 31

32. Thanks for Your Attention…!!!
32