KARL PEARSON’S CORRELATION COEFFICIENT
Procedure:
❖ From the menu, choose:
Analyze → Correlate → Bivariate…
❖ Select two or more variables and move it to “Variable(s)” box.
❖ Choose the option “Pearson”.
❖ Now click “OK” to generate the Pearson’s Correlation coefficient
between the entered variables.
2
SPEARMAN’S RANK CORRELATION COEFFICIENT
Procedure:
❖ From the menu, choose:
Analyze → Correlate → Bivariate…
❖ Select two or more variables and move it to “Variable(s)” box.
❖ Choose the option “Spearman”.
❖ Now click “Ok” to generate the Spearman’s Rank Correlation coefficient
between the entered variables.
3
4
PARTIAL CORRELATION COEFFICIENTS
Procedure:
❖ From the menu, choose:
Analyze → Correlate → Partial…
❖ Select two variables and move it to “Variable(s)” box.
❖ Select another one variable and move it to “Controlling for” box.
❖ Now click “Ok”, to generate the partial correlation coefficients for the
entered data.
5
6
MULTIPLE CORRELATION AND MULTIPLE REGRESSION
Procedure:
❖ From the menu, choose:
Analyze → Regression → Linear…
❖ Select one variable and move it to “Dependent” box.
❖ Select all other variables and move it to “Independent(s)” box.
❖ Now click “Ok” to generate the multiple correlation coefficients and to fit
multiple regression equation.
7
8
FITTING OF STRAIGHT LINE
Procedure
❖ From the menu, choose:
Analyze → Regression → Curve Estimation…
❖ Select one numerical variable and move it to “Dependent(s)” box.
❖ Select another numerical variable and move it to “Independent variable”
box.
❖ Choose the option “Linear” from the given models.
❖ Now click “Ok”, to generate fitting of straight line.
9
10
FITTING OF SECOND DEGREE PARABOLA
Procedure
❖ From the menu, choose:
Analyze → Regression → Curve Estimation…
❖ Select one numerical variable and move it to “Dependent(s)” box.
❖ Select another numerical variable and move it to “Independent variable”
box.
❖ Choose the option “Quadratic” from the given models.
❖ Now click “Ok”, to generate fitting of second degree parabola.
11
12
FITTING OF EXPONENTIAL CURVE
Procedure
❖ From the menu, choose:
Analyze → Regression → Curve Estimation…
❖ Select one numerical variable and move it to “Dependent(s)” box.
❖ Select another numerical variable and move it to “Independent variable”
box.
❖ Choose the option “Exponential” from the given models.
❖ Now click “Ok”, to generate fitting of second degree parabola.
13
14
Regression Analysis
Fitting a linear regression model
Procedure:
➢ From the menu, choose, Analyze → Regression → Linear…
➢ Select the dependent variable ‘Y’ and move it to ‘Dependent’ box and select all
the independent variables ‘X1, X2, X3’ and move them to ‘Independent(s)’
list.
➢ Click ‘Statistics’ and select the options, ‘Estimates’ and ‘Model fit’ and click
‘Continue’.
➢ Verify that whether the option ‘Include constant in equation’ was selected by
clicking ‘Options’ button and then click ‘Ok’.
Step 1
15
Step 2
Output:
Model Summary
Model Adjusted R Std. Error of the
R R Square Square Estimate
1 .730a .533 .406 8.220
a. Predictors: (Constant), X3, X2, X1
ANOVAb
Model Sum of Squares df Mean Square F Sig.
1 Regression 848.777 3 282.926 4.187 .033a
Residual 743.223 11 67.566
Total 1592.000 14
a. Predictors: (Constant), X3, X2, X1
b. Dependent Variable: Y
16
Coefficientsa
Model Standardized
Unstandardized Coefficients Coefficients
B Std. Error Beta t Sig.
1 (Constant) 10.441 18.692 .559 .588
X1 .670 .321 .652 2.089 .061
X2 .058 .249 .070 .231 .822
X3 .090 .211 .092 .427 .678
a. Dependent Variable: Y
Result:
✓ The multiple regression model is
Y = 10.441 + 0.670X1 + 0.058X2 + 0.090X3.
✓ The multiple correlation coefficient is RY. X1,X2,X3 = 0.730
✓ The estimate of is given by = 67.566
To test the hypothesis for overall regression parameters and individual parameters.
Hypothesis:
(i) H0: β1 = 0 where β1 = (β1, β2, β3, β4)|
(ii) H0: βj = 0 for j = 1,2,3,4
Procedure:
➢ From the menu, choose, Analyze → Regression → Linear…
➢ Select the dependent variable ‘Y’ and move it to ‘Dependent’ box and select all
the independent variables ‘X1, X2, X3’ and move them to ‘Independent(s)’
list.
➢ Click ‘Statistics’ and select the options, ‘Estimates’ and ‘Model fit’ and click
‘Continue’.
17
➢ Verify that whether the option ‘Include constant in equation’ was selected by
clicking ‘Options’ button and then click ‘Ok’.
Step 1
18
Step 2
Output:
ANOVAb
Model Sum of Squares df Mean Square F Sig.
1 Regression 2520.272 4 630.068 84.540 .000a
Residual 201.228 27 7.453
Total 2721.500 31
a. Predictors: (Constant), X4, X2, X1, X3
b. Dependent Variable: Y
Coefficientsa
Model Standardized
Unstandardized Coefficients Coefficients
B Std. Error Beta t Sig.
1 (Constant) 1.015 1.861 .545 .590
X1 -.029 .091 -.060 -.316 .755
19
X2 .216 .068 .362 3.187 .004
X3 -4.320 2.851 -.670 -1.515 .141
X4 8.975 2.773 1.333 3.237 .003
a. Dependent Variable: Y
Result:
(a) Overall Parameter Test:
✓ The F-value = 84.54 and its significant value = 0.000
✓ Since the p-value is less than 0.05, we reject the null hypothesis.
✓ Therefore, β1 ≠ 0.
(b) Individual Parameter Test:
Parameters t-value p-value compared with 0.05 Conclusion
0 0.545 0.590 0.05 Not Significant
1 -0.316 0.755 0.05 Not Significant
2 3.187 0.004 0.05 Significant
3 -1.515 0.141 0.05 Not Significant
4 3.237 0.003 0.05 Significant
To compute the given diagnostic measures.
Procedure:
➢ From the menu, choose, Analyze → Regression → Linear…
➢ Select the dependent variable ‘Y’ and move it to ‘Dependent’ box and select all
the independent variables ‘X1, X2, X3’ and move them to ‘Independent(s)’
list.
➢ Click ‘Save’ and select the options, ‘Unstandardized’ under ‘Predicted Values’
and select the options ‘Unstandardized, Studentized and Studentized deleted’
under ‘Residuals’ and select ‘Cook’s’ under ‘Distance’ and select whatever
options are necessary.
20
➢ Click ‘Continue’ and then click ‘Ok’.
➢ From the menu, choose, Analyze → Regression → Curve Estimation…
➢ Select ‘Unstandardized predicted value’ and move it to ‘Independent’ box and
select ‘Unstandardized residual’ and move it to ‘Dependent(s)’ box and click
‘Ok’.
Step 1
21
Step 2
22
Step 3
To select the variables using forward selection, backward elimination and stepwise
methods
Procedure:
➢ From the menu, choose, Analyze → Regression → Linear…
➢ Select the dependent variable ‘Y’ and move it to ‘Dependent’ box and select all
the independent variables and move them to ‘Independent(s)’ list.
➢ Choose ‘Forward’ from the given ‘Methods’ for performing forward selection
method, Choose ‘Backward’ from the given ‘Methods’ for performing
backward elimination method and Choose ‘Stepwise’ from the given
‘Methods’ for performing stepwise method.
➢ Click ‘Ok’.
23
Step 1
24
25