Linear Correlation PDF

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 40

HYPOTHESIS TESTING

“SIGNIFICANT RELATIONSHIP” OR
“SIGNIFICANT CORRELATION”
“SIGNIFICANT RELATIONSHIP”

- Significance of the relationship that


exist between two variables means
that it did not happen by chance
alone.
Suppose you are a researcher and you want to know
whether there is a significant relationship between a
person’s number of hours spent in playing online
games and his/her level of aggressive behavior, how
are you going to establish this relationship?

Submit your answer by going to menti.com and use the


code 7856701

SITUATIONAL ANALYSIS
1. Is there a significant relationship between a business
owners’ level of self-confidence and age?
2. Is there a significant relationship between students’
degree of independence and number of family
members?
3. Is there a significant relationship between teenagers’
number of social media accounts and no. of sleep
hours?

EXAMPLES OF RESEARCH QUESTIONS THAT COULD BE ANSWERED BY TESTING IF


THERE IS A SIGNIFICANT RELATIONSHIP BETWEEN THEM
LINEAR
CORRELATION
USING PEARSON PRODUCT MOMENT
CORRELATION COEFFICIENT
PEARSON PRODUCT MOMENT CORRELATION COEFFICIENT

o also called as Pearson’s r, in honor of the


English mathematician Karl Pearson who
developed the formula in the 1880s.

o a statistical tool that determines the existence,


strength, and direction between two variables.
FORMULA: Pearson’s r
Pearson r Qualitative Interpretation
FOR THE + 1
INTERPRETATION Perfect
OF THE RESULT:

+0.80 – +0.99 Very High


+0.60 – +0.79 Moderately High
+0.40 – +0.59 High
+0.20 – +0.39 Moderately Low
+0.01 – +0.19 Very Low
0 No Correlation
FOR THE INTERPRETATION OF THE RESULT:

 A positive correlation is a relationship between two variables


in which both variables move in the same direction. Therefore,
when one variable increases as the other variable increases, or
one variable decreases while the other decreases.
 A negative correlation is a relationship between two variables
in which an increase in one variable is associated with a decrease
in the other.
 A zero correlation exists when there is no relationship
between two variables.
ASSUMPTIONS IN USING PEARSON'S R
1. The two variables should be continuous variables, they
are measurable.
2. The number of sample size for the two variables are
equal.
3. The variables should be approximately normally
distributed.
4. There are no outliers or extreme values.
EXAMPLE 1
JM, an educational researcher, wants to know whether a student’s physics
grade depends on his math grade. He collects a sample of 5 students and
gathered their grades in math and science. The math and science grade of the
students are in the table below:

Math Grade
(x)
76 82 87 92 95
Physics Grade
(y)
75 83 88 89 93

Can JM conclude a relationship between the math and physics grades using
5% significance level?
STEP 1: NULL AND ALTERNATIVE HYPOTHESIS

 Ho: r=0 ; There is no significant relationship


between math and physics grades.
 Ha: r≠0 ; There is a significant relationship
between math and physics grades.
STEP 2: SIGNIFICANCE LEVEL

Based on the alternative hypothesis:

ɑ=.05
𝑟 ≠ 0 ; 𝑡𝑤𝑜 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant relationship)
𝑟 > 0 ; 𝑟𝑖𝑔ℎ𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant positive relationship)
𝑟 < 0 ; 𝑙𝑒𝑓𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant negative relationship)
STEP 3: CRITICAL VALUES
df= n-2 (n = number of pairs)
df = 5-2 = 3
Refer to the following table
df = 3, a=0.05, two-tailed
𝒕𝜶 = 𝟎. 𝟖𝟕𝟖𝟑
The table contains critical values for two-tail tests.
For one-tail tests, multiply α by 2.
The table contains critical values for two-tail tests.
For one-tail tests, multiply α by 2.
STEP 4: SOLVE FOR THE TEST STATISTIC
 Solve for Pearson’s r
EXAMPLE 1: SOLUTION
Math Physics
Student xy
Grade (x) Grade (y)
1 76 75 5, 700 5, 776 5, 625
2 82 83 6, 806 6, 724 6, 889
3 87 88 7, 656 7, 569 7, 744
4 92 89 8, 188 8, 464 7, 921
5 95 93 8, 835 9, 025 8, 649

TOTAL:
EXAMPLE 1: SOLUTION
STEP 5: DECISION
If the absolute value (without considering the sign +/-) of the
calculated Pearson’s correlation coefficient is greater than the
critical value from the table, then reject the null hypothesis.

𝒓 = 𝟎. 𝟗𝟕𝟒𝟔 > 𝒕𝜶 = 𝟎. 𝟖𝟕𝟖𝟑 Reject Ho


STEP 6A: PRESENTATION OF RESULTS

Table 1. Pearson’s r Correlation Result Between Math and Physics Grades


Physics
𝒓 Qualitative Interpretation
Math Very High Positive
0.9746*
Correlation
* significant at 𝛼 = 0.05
STEP 6B: INTERPRETATION

Table 1 shows the Pearson’s r correlation result between math


and physics grades. Based on the analysis, at 5% significance level,
there is a significant relationship between math and physics
grades. Based on the value of r, there exist a very high positive
correlation between math and physics grades. This implies that if a
student has a high grade in math then he or she has a high grade
in physics. Conversely, if a student has low grades in math then he
or she could have a low grade in physics.
USING MS EXCEL (METHOD 1)

1. Formulas
2. More
Functions
3. Statistical
4. CORREL
USING MS EXCEL (METHOD 1)

Select the VALUES of 1st and 2nd variable,


click OK.
USING MS EXCEL (METHOD 2)

1.Data Menu
2.Data Analysis
3.Correlation
4.OK
USING MS EXCEL (METHOD 2)

6. Select the two columns


for INPUT RANGE
7. GROUPED BY columns
8. LABELS are not included
in the range
9. OK
USING MS EXCEL (METHOD 3)

Then use the formula for r


USING MS EXCEL (METHOD 3)

Then use the formula for r


EXAMPLE 2

The table shows the time in hours spent by five students in


playing computer games and scores of these students got on a
Math test. Solve for the Pearson’s r and describe the result. Use
5% significance level.

Time(x) 1 2 3 4 5

Score (y) 25 20 15 10 5
STEP 1: NULL AND ALTERNATIVE HYPOTHESIS

 Ho: r=0 ; There is no significant relationship


between the time spent in playing computer games
and the scores on a Math test.
 Ha: r≠0 ; There is a significant relationship between
the time spent in playing computer games and the
scores on a Math test.
STEP 2: SIGNIFICANCE LEVEL

Based on the alternative hypothesis:

ɑ=.05
𝑟 ≠ 0 ; 𝑡𝑤𝑜 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant relationship)
𝑟 > 0 ; 𝑟𝑖𝑔ℎ𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant positive relationship)
𝑟 < 0 ; 𝑙𝑒𝑓𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 (significant negative relationship)
STEP 3: CRITICAL VALUES
df= n-2 (n = number of pairs)
df = 5-2 = 3
Refer to the table
df = 3, a=0.05, two-tailed
𝒕𝜶 = 𝟎. 𝟖𝟕𝟖𝟑
The table contains critical values for two-tail tests.
For one-tail tests, multiply α by 2.
STEP 4: SOLVE FOR THE TEST STATISTIC
 Solve for Pearson’s r
EXAMPLE 2: SOLUTION
Time Score
Student xy
(x) (y)
1 1 25 25 1 625
2 2 20 40 4 400
3 3 15 45 9 225
4 4 10 40 16 100
5 5 5 25 25 25

TOTAL:
EXAMPLE 2: SOLUTION
STEP 5: DECISION
If the absolute value (without considering the sign +/-)
calculated Pearson’s correlation coefficient is greater than the
critical value from the table, then reject the null hypothesis.

𝒓 = −𝟏 > 𝒕𝜶 = 𝟎. 𝟖𝟕𝟖𝟑 Reject Ho


STEP 6A: PRESENTATION OF RESULTS
Table 1. Pearson’s r Correlation Result Between
Time on Computer Games and Math Scores
Math Scores
Qualitative
Time on 𝒓
Interpretation
Computer
Games Perfect Negative
-1*
Correlation
* significant at 𝛼 = 0.05
STEP 6B: INTERPRETATION
Table 1 presents the Pearson’s r correlation result between time
on computer games and math scores. Based on the analysis, at 5%
significance level, there is a significant relationship between the
time spent in playing computer games and the scores on a Math
test. Specifically, there exists a perfect negative correlation
between the two variables. This implies that the longer time one
spend on playing computer games, the lower is his/her score in
the Math test.
ACTIVITY 8
(SCHEDULED ON WED, NOV 25; WAIT FOR THE GOOGLE FORM LINK TO BE SENT IN THE GC)

Loida studies if age correlates with the average number of hours of sleep,
so she selected a random sample size 6 and surveyed the needed data. The
gathered data are given below. Is there a significant relationship between
a person’s age and the number of hours he or she sleeps?

Age(x) 8 15 22 27 34 40

Sleep (y) 8 8 7 7 5 6

You might also like