LESSON 5: CORRELATION
Objectives:
1. Describe the key concepts correlation;
2. Solve correlation coefficients; and
3. Interpret correlation coefficients.
Correlation describes the relationship between two or more variables.
Describes the strength of the relationship in terms of a number from -1.0 to +1.0.
Describes the direction of the relationship as positive or negative.
Types of Correlation:
1. POSITIVE CORRELATION means there is direct relationship that exists positively between two variables. It
exists when:
HIGH scores in one variable and HIGH scores on the other
LOW scores in one variable and LOW scores on the other
Example: 1. the less you study, the lower your test score will be
2. the more you eat, the more weight you will gain
2. NEGATIVE CORRELATION means there is an indirect relationship that exists negatively between two
variables. It exists when
HIGH scores in one variable and LOW scores on the other
LOW scores in one variable and HIGH scores on the other
Example: 1. the older you are, the less flexible your body is
2. the less time you study, the more errors you will make
3. ZERO CORRELATION means that the relationship is difficult to describe.
This can be low or slight in relationship, negligible or zero relationship
We will use the Correlation Value in interpreting the degree of relationship
Value of r Interpretation
0.00 Zero correlation / relationship
±0.01 to ± 0.20 Negligible correlation / relationship
±0.21 to ± 0.40 Low or slight correlation / relationship
±0.41 to ± 0. 60 Moderate correlation / relationship
±0.61 to ± 0. 80 High correlation / relationship
±0.81 to ± 0.99 Very high correlation / relationship
±1 Perfect correlation / relationship
WAYS TO ESTABLISH RELIABILITY
Reliability refers to the consistency and dependability of test results. In other words, tests agree with itself. We will
use the Correlation Value in interpreting the degree of relationship
1. TEST – RETEST – The same test is administered twice to the same group of students and correlation
coefficient is determined.
A Spearman rank correlation coefficient or Spearman rho is the statistical tool used to measure the
relationship between paired ranks assigned to individual scores on two variables, X and Y of the first
administration (X) and second administration (Y). To obtain the value of Spearman rho (r s) consider the
formula below.
6 ∑ D2
r s=1−
N 3−N
Where; (r s) – stands for the Spearman rho, ∑ D 2 – sum of the squared difference of rank, and N – total
number of cases
Prerequisite skill: ranking the scores
RANKING
It is the process of determining the RELATIVE POSITION of values or scores according to some bases such
as magnitude, worth, quality, importance or chronology.
Usually, the highest score may be indicated as rank 1, Second is rank 2, and so and so forth however… for
ties it’s different.
Example: Rank the following test scores in Algebra: 88, 87, 88, 72, 87, 86, and 87
Steps in solving:
1. Arrange the name of the students alphabetically
2. Register the scores of the students for the first test (X) and second test (Y).
3. Rank the scores in X and Y separately.
4. Subtract Rx by Ry then square the difference.
5. Total the value of D 2, and then use the formula.
6. Interpret using the correlation scale.
Students X Y Rx Ry D D
2
1 50 51
2 43 42
3 48 48
4 45 44
5 40 41
6 47 47
7 52 51
8 39 38
9 44 43
10 43 42
11 41 41
12 46 45
13 39 39
14 51 50
15 49 48
Total
2. PARALLEL FORMS OR EQUIVALENT FORMS – Test is administered to a group of students. Two forms of
tests must be constructed but the content, type of test item, difficulty and instruction is similar but not identical
Product Pearson Moment Correlation Coefficient or Pearson r
N ∑ XY −∑ X ∑ Y
r xy =
√[ N ∑ X −(∑ X ) ][ N ∑ Y −(∑ Y ) ]
2 2 2 2
3. SPLIT-HALF – one test is given. The test is divided into two halves: odd and even numbers.
1. Record the number of correct answers for odd numbers (X) and even numbers (Y) for every student
2. Rank X and Y.
3. Subtract Rx by Ry, and square the difference.
4. Compute the total of D 2, and then solve the reliability of the half test using spearman rho.
5. Solve the reliability of whole test using Spearman Brown Formula then make your interpretations.
Spearman – Brown Formula:
2(r ¿¿ h t)
r wt = ¿
1+r h t
Where, r wt isthe reliability of the whole test
r ht is the reliability of thehalf test
ACTIVITY1: Correlation and Test Reliability
Spearman Rank Correlation (Test-Retest Method)
Objective: Compute the Spearman Rank Correlation Coefficient (𝑟ₛ) to determine the reliability of a test administered
twice.
Instructions:
1. Use the test scores from the first (X) and second (Y) administrations below.
2. Rank both sets of scores (1 = highest rank). For ties, assign the average rank.
Compute 𝑟ₛ using the formula:
3. Compute the difference between ranks (D = Rₓ - Rᵧ), then square it (D²).
4.
5. Interpret the result.
6∑ D
2
r s=1− 3
N −N
Student X (Test 1) Y (Test 2 Rₓ Rᵧ D = Rₓ - Rᵧ D²
1 50 51
2 43 42
3 48 48
4 45 44
5 40 41
6 47 47
7 52 51
8 39 38
9 44 43
10 43 42
11 41 41
12 46 45
13 39 39
14 51 50
15 49 48
