CORRELATION

CORRELATION: Meaning and Uses of Correlation

Correlation – is a measure of relationship between two variables. Coefficient of correlation - determines validity, reliability,
and objectivity of an examination prepared. It also indicates the amount of agreement or disagreement between groups of
scores, measurements, or individuals.

Correlation ranges in value from +1.00 through 0.00 up to -1.00.

Interpretation of ranges is shown below.

±0.00 to ± 0.20 – slightly correlation, almost negligible relationship
±0.21 to ± 0.40 – slightly correlation, definite but small relationship
±0.41 to ± 0.70 – moderate correlation, substantial relationship
±0.71 to ± 0.90 – high correlation, marked relationship
±0.91 to ± 1.00 – very high correlation, very dependable relationship

Computing the coefficient of correlation by the Spearman’s Formula:

6(∑ G)
R=1- N2 −1

Where:
R = is the coefficient of correlation by Spearman’s Formula
1 = is constant
6 = is constant (The constant 1 and 6 in the formula do not change)
ƩG – is the sum of the column G
N – is the number of pairs of scores or measures

The following steps are as follows:

1. Write the sores of the 12 female students in English under column Test x and in Mathematics under column Test y.
2. Rank the scores under column Test x with the highest score as rank 1 and the lowest scores as rank 12
(considering that there are only twelve female students involved). Write the ranks of the scores under Test x in
column Rx which means rank of x
3. Rank the scores under column Test y with the highest score as rank 1.5 and the lowest score as rank 12.
4. Subtract the Ry values from Rx values.
5. Write the difference between the values under the Rx and Ry columns under column G, which means gain. In this
case, consider only the positive values.
6. Get the sum of the values under column G and call this ƩG.
𝟔(∑ 𝐆)
Finally, substitute the values in the original formula R=1- 𝐍𝟐 −𝟏
1. Compute the coefficient of correlation by the use of the Spearman’s formula by using the table showing the scores of
18 female students on two short quizzes in English (X) and Mathematics (Y).

Students Test x Test y Rx Ry G

1 83 71 5 8
2 67 68 12.5 10.5 2
3 46 60 18 17 1
4 68 81 11 6 5
5 91 82 2.5 4.5
6 91 87 2.5 1.5
7 50 52 16.5 18
8 75 77 6 7
9 86 83 4 3 1
10 71 82 8.5 4.5 4
11 71 68 8.5 10.5
12 50 68 16. 10.5 5.5
13 67 64 12.5 14.5
14 55 64 15 14.5 .5
15 71 68 8.5 10.5
16 59 61 14 16
17 94 87 1 1.5
18 71 66 8.5 13
ƩG = 23.5

Solution:

𝟔(∑ 𝐆)
R=1- 𝐍𝟐 −𝟏

6(23.5) 141
𝑅 = 1 − (18)2−1 = 1 − 323 = 1 − 0.436 = 0.564– moderate correlation, substantial relationship

The R in this problem indicates the relationship between scores of 18 female students in English and Mathematics.
Since R=0.564 is moderate and substantial relationship exist between the two sets of scores.
 The coefficient of correlation by Product Moment Method

Computing the coefficient of correlation by Product Moment Method

Ʃ𝒅𝒙 𝒅𝒚
𝒓𝒙𝒚 =
√(Ʃ𝒅𝟐𝒙 )(Ʃ𝒅𝟐𝒚 )

Where:
𝑟𝑥𝑦 = is the coefficient of correlation by the product-Moment Method

Ʃ𝒅𝒙 𝒅𝒚 - is the sum of column 𝑑𝑥 𝑑𝑦

Ʃ𝑑𝑥2 – is the sum of column 𝑑𝑥2

Ʃ𝑑𝑦2 – is the sum column of 𝑑𝑦2

The following steps are as follows:

1. Get the total data under Test x and Test y.
Ʃ𝑥
2. Get the mean of test x: 𝑥̅ = 𝑛
̅̅̅̅
Ʃ𝑦
3. Get the mean of test y: 𝑦̅ = 𝑛

4. Get the deviations 𝑑𝑥 and 𝑑𝑦 by getting the difference between the mean and the score.

5. Square 𝑑𝑥 to obtain 𝑑𝑥2 and 𝑑𝑦 to obtain 𝑑𝑦2

6. Get the summation of 𝑑𝑥2 and 𝑑𝑦2

7. Get the product of 𝑑𝑥 and 𝑑𝑦 to have 𝑑𝑥 𝑑𝑦

8. Get the Ʃ𝑑𝑥 𝑑𝑦

9. Apply the formula

Ʃ𝒅𝒙 𝒅𝒚
𝒓𝒙𝒚 =
√(Ʃ𝒅𝟐𝒙 )(Ʃ𝒅𝟐𝒚 )
 2. Compute the coefficient of correlation by the use of the Product Moment formula by using the table showing the
scores of 18 female students on two short quizzes in English (X) and Mathematics (Y).
Students Test x Test y 𝑑𝑥 (𝑥 − 𝑥̅ ) 𝑑𝑦 (𝑦 − 𝑦̅) 𝑑𝑥2 𝑑𝑦2 𝑑𝑥 𝑑𝑦
1 83 71 12.67 -0.61 160.5289 0.3721 -7.7287
2 67 68 -3.33 -3.61 11.0889 13.0321 12.0213
3 46 60 -24.33 -11.61 591.9489 134.7921 282.4713
4 68 81 -2.33 9.39 5.4289 88.1721 -21.8787
5 91 82 20.67 10.39 427.2489 107.9521 214.7613
6 91 87 20.67 15.39 427.2489 236.8521 318.1113
7 50 52 -20.33 -19.61 413.3089 384.5521 398.6713
8 75 77 4.67 5.39 21.8089 29.0521 25.1713
9 86 83 15.67 11.39 245.5489 129.7321 178.4813
10 71 82 0.67 10.39 0.4489 107.9521 6.9613
11 71 68 0.67 -3.61 0.4489 13.0321 -2.4187
12 50 68 -20.33 -3.61 413.3089 13.0321 73.3913
13 67 64 -3.33 -7.61 11.0889 57.9121 25.3413
14 55 64 -15.33 -7.61 235.0089 57.9121 116.6613
15 71 68 0.67 -3.61 0.4489 13.0321 -2.4187
16 59 61 -11.33 -10.61 128.3689 112.5721 120.2113
17 94 87 23.67 15.39 560.2689 236.8521 364.2813
18 71 66 0.67 -5.61 0.4489 31.4721 -3.7587
∑ 𝑥 = 1266 ∑ 𝑦 = 1289 Ʃ𝑑𝑥 =3654.0002 Ʃ𝑑𝑦 =1768.2778 Ʃ𝑑𝑥 𝑑𝑦 =2098.3334
2 2

Solution:
∑𝑥 1266
𝑥̅ = = = 70.33
𝑛 18

∑𝑦 1289
𝑦̅ = = = 71.61
𝑛 18

𝑑𝑥 (𝑥 − 𝑥̅ ) = 83 – 70.33 =12.67
𝑑𝑦 (𝑦 − 𝑦̅) =71 – 71.61 = -0.61

∑ 𝑑𝑥𝑑𝑦 2098.3334 2098.3334

𝑟𝑥𝑦 = = = =0.826
√(∑ 𝑑𝑥 2 )(∑ 𝑑𝑦 2 ) √(3654.0002)(1768.2778) 2541.906

The 𝑟𝑥𝑦 in this problem indicates the relationship between scores of 18 female students in English and
Mathematics. Since 𝜌 = 0.826 is high correlation and marked relationship exist between the two sets of scores.
Computing the coefficient of correlation by the use of the Spearman’s rho (Rank-difference Method)
The Spearman rank correlation (𝑟𝑠 ) is calculated by ranking the observations for each of the two variables and
then computing the Pearson correlation on the ranks. When ties are encountered, the average rank is used

𝟔(∑ 𝑫𝟐 )
𝒓𝒔 (𝝆) = 𝟏 −
𝒏(𝒏𝟐 − 𝟏)
This produces a correlation coefficient which has a maximum value of 1, indicating a perfect positive association
between the ranks, and a minimum value of -1, indicates no association between the ranks for the observed values
of x and y.
Where:
𝝆( rho) – is the coefficient of correlation by the Rank-Difference Method
1 – is constant
6 – is constant (the constant 1 and 6 in the formula do not change)
Ʃ𝐷2 – is the sum of the column D²
N – is the sum of pairs of scores or measures.

The following steps are as follows:

1. Follow the same steps from 1 to 3.
2. Find the difference between the two sets of ranks or values under columns Rx and Ry. It does not matter if the difference
is found by subtracting the Rx values from Ry values or Ry values from the Rx values as long as you subtract the larger
value from the smaller value.

3. Write the difference between the values under the Rx and Ry columns under d, which means difference.
4. Square the value under column D and write the results under column D².
5. Get the sum of the values under column D² and call this Ʃ𝐷 2 .
3. Compute the coefficient of correlation by the use of the Product Moment formula by using the table showing
the scores of 18 female students on two short quizzes in English (X) and Mathematics (Y).
Students Test x Test y Rx Ry D D²
1 83 71 5 8 3 9
2 67 68 12.5 10.5 2 4
3 46 60 18 17 1 1
4 68 81 11 6 5 25
5 91 82 2.5 4.5 2 2
6 91 87 2.5 1.5 1 1
7 50 52 16.5 18 1.5 2.25
8 75 77 6 7 1 1
9 86 83 4 3 1 1
10 71 82 8.5 4.5 4 16
11 71 68 8.5 10.5 2 4
12 50 68 16. 10.5 5.5 30.25
13 67 64 12.5 14.5 2.0 4
14 55 64 15 14.5 .5 .25
15 71 68 8.5 10.5 2.0 4
16 59 61 14 16 2.0 4
17 94 87 1 1.5 .5 .25
18 71 66 8.5 13 4.5 20.25

ƩD²=129.25

6(∑ 𝐷 2 ) 6(129.25) 775.5

𝜌 = 1 – 𝑁(𝑁2−1) = 1 − =1− = 1 − 0.133
18[(18)2 −1] 18[323]

𝜌 = 0.866 high correlation, marked relationship

The 𝜌 in this problem indicates the relationship between scores of 18 female students in English and Mathematics.
Since 𝜌 = 0.866 is high correlation and marked relationship exist between the two sets of scores.