Scribd
Upload
17 views

Stastic Chapr 4

This document discusses correlation and regression analysis. It defines correlation analysis as a statistical procedure to determine the relationship between two variables. The variables can be categorized as independent variables, which are usually known or manipulated, and dependent variables, which depend on changes to the independent variable. Scatter diagrams can be used to visualize the relationship between variables by plotting their values along the x and y axes. The correlation coefficient, r, measures the strength of the linear correlation between variables, ranging from +1 to -1. Pearson's product-moment correlation coefficient is used for two quantitative variables, while Spearman's rank correlation coefficient can be used if one or both variables are quantitative.

Uploaded by

nitya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
17 views

Stastic Chapr 4

This document discusses correlation and regression analysis. It defines correlation analysis as a statistical procedure to determine the relationship between two variables. The variables can be categorized as independent variables, which are usually known or manipulated, and dependent variables, which depend on changes to the independent variable. Scatter diagrams can be used to visualize the relationship between variables by plotting their values along the x and y axes. The correlation coefficient, r, measures the strength of the linear correlation between variables, ranging from +1 to -1. Pearson's product-moment correlation coefficient is used for two quantitative variables, while Spearman's rank correlation coefficient can be used if one or both variables are quantitative.

Uploaded by

nitya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

CHAPTER 4

CORRELATION & REGRESSION


INTRODUCTION
• In this chapter, the statistical procedure for determining whether two
independent variables demonstrate a linear relationship is explained.

• For instance, “ Would a student who scores highly in Mathematics also score
highly in Accounting?” or “ Can an athlete who runs fast also jumps further in
the long jump event?”

• CORRELATION ANALYSIS  a statistical procedure to determine the nature


and strength of any relationship between 2 variables.
INDEPENDENT AND DEPENDENT
VARIABLES

• Correlation and regression analyses are based on the association or


relationship between two or more variables.

• These variables can be categorized as independent variable and


dependent variable.
INDEPENDENT VARIABLE

• Usually the known variable


• Refer to the variable that the researcher manipulates or changes in a study.
DEPENDENT VARIABLE
• Depends on the changes that resulted from manipulating the independent
variable.

• In other word, dependent variable is what changes when the independent


variable changes.

• The dependent variable is the usually the predicted variable.


EXAMPLE
• Is there relationship between sugar concentration and the degree of non-
enzymatic browning (Maillard reaction)?

• Question: Identify independent and dependent variable?

• Independent variable 
• Dependent variable 
EXAMPLES
Determine which variable is independent and dependent for the following
cases:

1. The price of fresh free-ranged chicken and the demand for it.
2. The concentration of Noni sp. (mengkudu) extract and the diameter of
zone of antimicrobial activity.
3. The proofing temperature and the rise in sweet bread dough.
4. The supply of nasi briyani and the demand for it.
5. The number of unemployed and the volume of exports.
SCATTER DIAGRAM
• To determine if a relationship between 2 variables (bivariates), a graph can
be plotted by assigning the independent variable to x-axis and dependent
variable to y-axis.

• Data for the two variables are regarded as the x and y coordinates.

• The resulting graph is called a scatter diagram.


EXAMPLE OF SCATTER DIAGRAM

+ve correlation strong +ve correlation perfect +ve correlation, r = +1.000

-ve correlation strong -ve correlation perfect -ve correlation, r = -1.000


NO CORRELATION

No linear correlation but


curvilinear correlation
r=0
CORRELATION COEFFICIENT
• The strength of linear correlation between two variables is measured by
correlation coefficient, r.

• r values range between +1.00 and +1.00.

• If r values are close to +1.00 or -1.00, the two show strong +ve correlation and
strong –ve correlation.

• If r = 0, no correlation.
RULE OF THUMB GUIDE TO DESCRIBE THE
STRENGTH OF CORRELATION COEEFICIENT

CORRELATION COEFFICIENT, r THE STRENGHT OF CORRELATION


+0.91 to +1.00 or -0.91 to -1.00 Very strong
+0.71 to +0.91 or -0.71 to – 0.90 Strong
+0.51 to +0.70 or -0.51 to - 0.710 Moderate
+0.31 to +0.50 or -0.31 to - 0.50 Weak
+0.05 to +0.30 or -0.05 to – 0.30 Very weak
0 or close to 0 No correlation
• Pearson product-moment correlation coefficient, r  used if both variables
are quantitative variables.

• Spearman rank correlation coefficient, ρ  used if 1 or both variables are


quantitative variables.
PEARSON PRODUCT-MOMENT
CORRELATION COEFFICIENT
n Σ xy – (Σ x) (Σ y)
r=
𝑛 Σ 𝑥2 − Σ𝑥 2
{ Σ 𝑦2 − Σ𝑦 2
}

where:
n = the number of pairs of data
Σ = the sum of the item selected
Σx = denotes teh sum of all x−values
Σy = denotes teh sum of all y−values
Σ𝑥2 = denotes the sum of all 𝑥2 –values therefore each x value should be squared and then added together
Σ𝑦2 = denotes the sum of all y2 –values therefore each x value should be squared and then added together
Σ𝑥 2 = indicates that the x values should be added up and the total squared
Σ𝑦2 = indicates that the y values should be added up and the total squared
Σ xy = indicates that each x-value should first be multiplied by its corresponding y-values before being added together
EXAMPLE
1. Using the following data set, find the value of the Pearson product-moment
correlation coefficient, r. comment on the value obtained.

x 1 2 4 5 8
y 8 7 6 3 1
• Solution:
calculate the quantities for n,Σ x, Σ y, Σ xy, Σ 𝑥2and Σ𝑦2
2. Calculate the Pearson product-moment correlation coefficient for the
Mathematics scores and the Food Chemical Engineering scores of the 10
students.

Students A B C D E F G H I J
Course Scores
Mathematics 20 25 37 43 45 56 70 75 80 90
Food
Chemical 15 20 30 35 30 43 50 60 70 78
Engineering
3. 14 different second year medical students took blood pressure
measurements of the same patient and the results are shown below. Is
there a correlation by the systolic and diastolic values?

Systolic 138 130 135 140 120 125 120 130 130 144 143 140 130 150
Diastolic 82 91 100 100 80 90 780 80 80 98 105 85 70 100
SPEARMAN CORRELATION
COEFFICIENT
6 Σ𝑑2
ρs = 1 -
𝑛 (𝑛2 −1)

where:
n = number of data
d = difference in rank order between two variables (x and y)

You might also like