22-02-2023

Statistical
Correlation
(Part 1)
Basics Concepts
Types of Correlation
Methods

Biostatistics & Research Methodology

B Pharm 8th Sem | M. Pharm. | PhD

Correlation
Correlation- Relationship between two variable
Correlation is a statistical measure that expresses the degree of relationship
between two variables
Useful statistical measure in bivariate or multivariate data
If two quantities vary in such way that movement in one are accompanied by
movement in the other, these quantities are correlated
SN Concentratio Absorbance Conc vs Abs
n (ug/ml) 4

1 10 0.5 3
ABSORBANCE

2 20 0.8 2

3 30 1.4 1

4 40 2.3 0
0 10 20 30 40 50 60
5 50 3.6 CONCENTRATION

1
 22-02-2023

Correlation
Correlation- Relationship between two variable
The measure of correlation called the correlation coefficient or correlation
index, which is summarized in one figure the direction and degree of correlation

SN Concentratio Absorbance Conc vs Abs

n (ug/ml) 4

1 10 0.5
ABSORBANCE

2 20 0.8 2

3 30 1.4 1

4 40 2.3 0
0 10 20 30 40 50 60
5 50 3.6 CONCENTRATION

Correlation

Correlation

Simple, Partial, & Linear or Non-

Positive/Negative
Multiple Linear

2
 22-02-2023

Correlation
Positive- If one variable increase/decrease then 2nd
variable is also increase/decrease
Negative- If one variable increase/decrease then 2nd
variable is decrease/increase

Positive Negative
25 100
20 NO OF PARACITE 80
ABSORBANCE

15 60
10 40
5 20
0 0
0 1 2 3 4 5 0 20 40 60
CONCENTRATION CONCENTRATION

Correlation
Linear- Change in variables in constant ratio
Non-Linear- Doesn't Change in variables in constant
ratio Conc. % Response
2 8
Student Teachers 4 20
2019 1Y 100 5 8 50
2020 2Y 200 10 16 88
2021 3Y 300 15 32 100
2022 4Y 400 20 64 100

Linear Non Linear

400 100 100
88
300
200 50
100 20
5 10 15 20 8
2019 2020 2021 2022 2 4 8 16 32 64
Student Teachers % Response

3
 22-02-2023

Correlation
Simple- Study of two variables, e.g., Abs. Vs Conc
Partial- Study of three variables, e.g., Rice production depends
on- soil, rainfall, fertilizer
Multiple- Study of multiple variables- e.g., Marks depends on-
Study, Knowledge, Questions, Health, evaluator etc

Correlation

Methods of Studying
Correlation

Karl Pearson’s Spearman's

coefficient of Rank correlation Others
Correlation Coefficient ✓ Scattered Diagram
✓ Graphic
✓ Concurrent Deviation
✓ Method of Least Squares

4
 22-02-2023

Correlation

Statistical
Correlation
(Part 2)

Karl Pearson’s coefficient of

correlation

Biostatistics & Research Methodology

B Pharm 8th Sem | M. Pharm. | PhD

5
 22-02-2023

Correlation
Correlation- Relationship between two or more variables
Correlation is a statistical measure that expresses the direction and degree of
relationship between variables

SN Concentratio Absorbance Abs

n (ug/ml) 4
y = 0.076x - 0.58
1 10 0.5 3 R² = 0.9316
AXIS TITLE

2 20 0.8 2
Abs
3 30 1.4 1 Linear (Abs)

4 40 2.3 0
0 20 40 60
5 50 3.6 AXIS TITLE

Correlation

Methods of Studying
Correlation

Karl Pearson’s Spearman's

coefficient of Rank correlation Others
Correlation Coefficient ✓ Scattered Diagram
✓ Graphic
✓ Concurrent Deviation
✓ Method of Least Squares

6
 22-02-2023

Karl Pearson’s coefficient of Correlation

Used to measure the strength of relationship between two variables
Denoted by “r”. It is the Pearson’s Product
r = -1 to +1
Applied when deviation is taken from actual mean value not to the
assume men
𝛴𝑥𝑦 𝑥 = x − xത
r=
σ 𝑥 2 x 𝛴𝑦 2 𝑦 =Y−ഥ Y

r Inference
+1 perfect positive
correlation
0 No correlation
-1 perfect negative
correlation

Correlation

7
 22-02-2023

Karl Pearson’s coefficient of Correlation

Calculate the r value from given data 𝛴𝑥𝑦 𝑥 = x − xത
r=
σ 𝑥 2 x 𝛴𝑦 2 𝑦 = Y−ഥ Y
Student A B C D E
Pathology 5 6 7 9 8 8
r=
Pharmacolog 4 7 9 8 7 10 x 14
y
8
r=
140
Student X x (X-7) x2 Y y (Y-7) y2 xy
8
A 5 -2 4 4 -3 9 6 r=
11.8
B 6 -1 1 7 0 0 0
C 7 0 0 9 2 4 0 r = 0.67
D 9 2 4 8 1 1 2
E 8 1 1 7 0 0 0
Sum 35 10 35 14 8

Karl Pearson’s coefficient of Correlation

Calculate the r value from given data 𝛴𝑥𝑦 𝑥 = x − xത
r= ഥ
σ 𝑥 2 x 𝛴𝑦 2 𝑦 = Y−Y
Patient A B C D E
Blood 120 140 200 240 300 10800
Sugar (X) r=
21600 × 5800
BP (Y) 80 100 110 130 180
10800
r=
125280000
Patient X x (X-200) x2 Y y (Y-120) y2 xy
10800
A 120 -80 6400 80 -40 1600 3200 r=
11192.8
B 140 -60 3600 100 -20 400 1200
C 200 0 0 110 -10 100 0 r = 0.96
D 240 40 1600 130 10 100 400
E 300 100 10000 180 60 3600 6000
Sum 1000 21600 600 5800

8
 22-02-2023

Karl Pearson’s coefficient of Correlation

https://www.ahajournals.org/doi/full/10.1161/01.hyp.30.6.1511

Statistical
Correlation
(Part 3)

Multiple correlation

Biostatistics & Research Methodology

B Pharm 8th Sem | M. Pharm. | PhD

9
 22-02-2023

Multiple Correlation
Correlation- Relationship between two or more variables
Correlation is a statistical measure that expresses the direction
and degree of relationship between variables
Multiple
Correlation
Variable

Independent Dependent
Variable Variable
1 variable 2,3,4,……

Multiple Correlation Coefficient establishes/indicates the

Relationship between one variable with multiple other variables

Multiple Correlation
Multiple Correlation Coefficient establishes/indicates the
Relationship between one variable with multiple other variables
Denoted by ‘R’
Karl Pearson’s coefficient of Correlation
S Students BMI BP Blood
𝛴𝑥𝑦 𝑥 = x − xത
N (A) (B) Sugar r=
(C) σ 𝑥2 x 𝛴𝑦 2 𝑦 = Y−ഥ Y
1 Sham 15 80 120
𝛴𝑥𝑦
2 Ram 20 100 140 r=
𝑁. σ𝑥 σ𝑦
3 Raju 25 110 200
4 Ramu 30 130 240
5 Shetu 35 180 300

10
 22-02-2023

Multiple Correlation
Multiple Correlation Coefficient establishes/indicates the Relationship
between one variable with multiple other variables
Denoted by ‘R’
The multiple correlation coefficient is denoted as Ra.bc……
A (independent variable) which correlate with dependent Variable
(B,C,D…k)
S Students BMI BP Blood
N (A) (B) Sugar
(C)
1 Sham 15 80 120
2 Ram 20 100 140
3 Raju 25 110 200
4 Ramu 30 130 240
5 Shetu 35 180 300

Multiple Correlation
S Students BMI a (A- a2 BP b (B- b2 Blood c (C- c2 ab ac bc
N (A) Mean) (B) Mean) Sugar Mean)
(C)
1 Sham 15 -10 100 80 -40 1600 120 -80 6400 400 800 3200
2 Ram 20 -5 25 100 -20 400 140 -60 3600 100 300 1200
3 Raju 25 0 0 110 -10 100 200 0 0 0 0 0
4 Ramu 30 5 25 130 10 100 240 40 1600 50 200 400
5 Shetu 35 10 100 180 60 3600 300 100 10000 600 100 6000
Sum 125 250 600 5800 1000 21600 1150 1400 10800
mean 25 120 200

𝛴𝑥𝑦
r=
σ 𝑥 2 x 𝛴𝑦 2

11
 22-02-2023

Multiple Correlation
S Students BMI a (A- a2 BP b (B- b2 Blood c (C- c2 ab ac bc
N (A) Mean) (B) Mean) Sugar Mean)
(C)
1 Sham 15 -10 100 80 -40 1600 120 -80 6400 400 800 3200
2 Ram 20 -5 25 100 -20 400 140 -60 3600 100 300 1200
3 Raju 25 0 0 110 -10 100 200 0 0 0 0 0
4 Ramu 30 5 25 130 10 100 240 40 1600 50 200 400
5 Shetu 35 10 100 180 60 3600 300 100 10000 600 100 6000
Sum 125 250 600 5800 1000 21600 1150 1400 10800
mean 25 120 200

𝛴𝑎𝑏
r=
σ 𝑎2 x 𝛴𝑏 2

𝛴𝑎𝑏
r=
𝑁. σ𝑎 σ𝑏

Multiple Correlation
S Students BMI a (A- a2 BP b (B- b2 Blood c (C- c2 ab ac bc
N (A) Mean) (B) Mean) Sugar Mean)
(C)
1 Sham 15 -10 100 80 -40 1600 120 -80 6400 400 800 3200
2 Ram 20 -5 25 100 -20 400 140 -60 3600 100 300 1200
3 Raju 25 0 0 110 -10 100 200 0 0 0 0 0
4 Ramu 30 5 25 130 10 100 240 40 1600 50 200 400
5 Shetu 35 10 100 180 60 3600 300 100 10000 600 100 6000
Sum 125 250 600 5800 1000 21600 1150 1400 10800
mean 25 120 200

𝛴𝑎𝑏
r=
σ 𝑎2 x 𝛴𝑏 2

𝛴𝑎𝑏
r=
𝑁. σ𝑎 σ𝑏

12
 22-02-2023

Multiple Correlation
S Students BMI a (A- a2 BP b (B- b2 Blood c (C- c2 ab ac bc
N (A) Mean) (B) Mean) Sugar Mean)
(C)
1 Sham 15 -10 100 80 -40 1600 120 -80 6400 400 800 3200
2 Ram 20 -5 25 100 -20 400 140 -60 3600 100 300 1200
3 Raju 25 0 0 110 -10 100 200 0 0 0 0 0
4 Ramu 30 5 25 130 10 100 240 40 1600 50 200 400
5 Shetu 35 10 100 180 60 3600 300 100 10000 600 100 6000
Sum 125 250 600 5800 1000 21600 1150 1400 10800
mean 25 120 200

𝛴𝑎𝑏
r=
σ 𝑎2 x 𝛴𝑏 2

𝛴𝑎𝑏
r=
𝑁. σ𝑎 σ𝑏

Multiple Correlation
S Students BMI a (A- a2 BP b (B- b2 Blood c (C- c2 ab ac bc
N (A) Mean) (B) Mean) Sugar Mean)
(C)
1 Sham 15 -10 100 80 -40 1600 120 -80 6400 400 800 3200
2 Ram 20 -5 25 100 -20 400 140 -60 3600 100 300 1200
3 Raju 25 0 0 110 -10 100 200 0 0 0 0 0
4 Ramu 30 5 25 130 10 100 240 40 1600 50 200 400
5 Shetu 35 10 100 180 60 3600 300 100 10000 600 100 6000
Sum 125 250 600 5800 1000 21600 1150 1400 10800
mean 25 120 200

rab2 =0.92 rab = 0.96

rac2 =0.36 rac = 0.6
rbc2 =0.92 rbc = 0.96

13
 22-02-2023

Multiple Correlation
Calculate R2.13
r12 = 0.8 0.64
r13 = 0.7 0.49
r23 = 0.6 0.36

Thanks for Watching

Subscribe my YouTube
Channel

14