Vous êtes sur la page 1sur 20

CORRELATION DEFINITION

AND TYPES OF
CORRELATION

CORRELATION
CORRELATION STEPS

MANUAL CALCULATION IN
CORRELATION
CORRELATION DEFINITION

Correlation is one of the analytical techniques in the statistics used to


find the relationship between two variables that are quantitative. The
relationship of these two variables can occur because of a causal
relationship or it could happen by chance alone. Two variables are
said to be correlated if changes in one variable are followed by
changes in another variable regularly in the same direction (positive
correlation) or opposite (negative correlation)
TYPES OF CORRELATION

 Bivariate Correlation
Simple correlation analysis (Bivariate Correlation) is used to
determine the closeness of the relationship between two variables
and to know the direction of the relationship that occurred. The
simple correlation coefficient shows how big the relationship is
between two variables

 Partial Correlation
Partial correlation analysis (Partial Correlation) is used to
determine the relationship between two variables where other
variables considered influential are controlled or made fixed (as
control variables)
BIVARIATE STEPS

1 2

3
4 5 6

6
8 9 10
PARTIAL STEPS

1. first, turn on your computer and open your spss program application.

2. then, you must type the variables’s data in variable view for the first . So, the variable will be input in data vie

3. Next , type the data of each variables in data view.


4. From menu bar, click analyze , then click sub menu correlate.

5. then, click partial, so it will be shown like picture below .


6. - For “variable” , click prestasi variable and motivation’s variable.
- For “controlling for” click IQ’s variables
- For test of significant correlate click the symbol √ for homogenity

7. Click option , then in “statistics” . Click Zero – order correlations and in “missing values” click exclude
case pairwase.
8. Click ok and continoue, so the output will be shown like this :
Manual calculation of
bivariate correlation
1. STATE HYPOTHESIS AND α
H0 : THERE IS NO RELATION BETWEEN BOTH VARIABLES
HA : THERE IS RELATION BETWEEN BOTH VARIABLES
Α : 1 – CI = 1- 0,95 = 0,05= 5%
2. Decision Rule
H0 IS ACCEPTED IF SIG > 0,05
H0 IS REJECTED IF SIG < 0,05

3. RESULTS
PAIRWISE :

a. TILANG – MOBIL  MOBIL – TILANG


0,015 > 0,05 0 < 0,05
H0 IS ACCEPTED H0 IS REJECTED
Litswise :
a. Tilang – Mobil  Mobil – Tilang
0,015< 0,05 0 < 0,05
H0 is rejected H0 is rejected
4. Conclusion

Pairwise
a. Tilang – Mobil  Mobil – Tilang
H0 is accepted, so there is H0 is rejected, so there is
Relation between both relation between both
Variables variables

Listwise
a. Tilang – Mobil  Mobil – Tilang
H0 is rejected, so there is H0 is rejected, so there is relation
Relation between both between both Variables
Variables
Manual calculation of partial
correlation
1. STATE HYPOTHESIS

H0 : THERE IS NO RELATION BETWEEN BOTH


VARIABLES
HA : THERE IS RELATION BETWEEN BOTH VARIABLES
2. The analysis used is Pearson correlation analysis is useful to measure the
level of linear relationship between variables. The variables used in this study
are the variables (X1) and (X2) as independent variables and (Y) as the
dependent variable. This correlation value can be calculated using the
following formula:

The partial correlation coefficient between Y and X1 when X2 is


constant:

The correlation between X1


and X2 together with Y:
The partial correlation coefficient between Y and X2 when X1 is
constant:

The partial correlation coefficient between X1 and X2 when Y is


constant:
3. Fcount

4. Ftable

α : 1 – CI = 1- 0,95 =
0,05= 5%
Df1 = df pembilang = k
Df2 = df penyebut = n-k
5. Hypothesis Testing Criteria

Accept Ho Fcount < Ftable


Reject Ho Fcount > Ftable

6. Conclusion

If Fcount < Ftable so Ho accepted, It meansThere is no


relation between both variables

If Fcount > Ftable so Ho rejected and Ha accepted, it


means There is relation between both variables

Vous aimerez peut-être aussi