Scribd
Upload
16 views

Mini - Lecture (Pearson CorrelationShort)

The document discusses correlation and Pearson correlation coefficient. It provides objectives and definitions of correlation and Pearson correlation. It also uses a dataset to demonstrate calculating correlation using R and discusses correlating time spent on an exam to exam grade, finding a strong positive correlation.

Uploaded by

Sami Alikacem
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
16 views

Mini - Lecture (Pearson CorrelationShort)

The document discusses correlation and Pearson correlation coefficient. It provides objectives and definitions of correlation and Pearson correlation. It also uses a dataset to demonstrate calculating correlation using R and discusses correlating time spent on an exam to exam grade, finding a strong positive correlation.

Uploaded by

Sami Alikacem
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Castleton University

Welcome to
Mini-Lecture
CORRELATION
(Pearson Correlation)
Pearson Correlation
Data Set:
Two Quantitative Variables

What Is
Graphical Correlation Numerical
Presentation Coefficient? Presentation

Use (R-Studio or R ) to
Compute and Graph the
Correlation Practice
Correlation measures the linear relationship between two quantitative
variables.
Objectives : Pearson correlation (r)
Pearson correlation (r) is used to assess the relationship between two
continuous variables (for each subject). The Pearson is the most commonly
used form of correlation; hence, it will be discussed first.

NOTE: If the term “correlation” is used, presume that it is Pearson, unless


stated otherwise
Determines the relationship between 2 continuous variables
• Correlation range: −1…+1
• Correlation sign (+ or −) indicates the direction of the correlation

Correlation r Variable directions

Positive 0…+1 (X Y)

Positive 0…+1 (X Y) or (X Y)

Correlation r Variable directions


Negative −1…0 (X Y) or (X Y)

−1 0 +1

Strong Weak Strong


Research Problem: Dr. Rajia wants to learn if there is a correlation between the length
of time spent taking an exam and the grade on that exam. He gathers two pieces of
data from each student; minutes spent taking the exam (time)and grade on the exam.

The Data are published in Canvas as csv file under the name “TimeandGrade Data”
name time grade
Brown 75 61
Campbell 107 83
DiNitto 80 60
Frank 91 67
Gray 91 78
Harrett 89 64
Jarvis 95 85
Kirk 110 85
Kretzschmar 118 90
Lindsey 115 92
Martin 102 81
McHugh 83 63
Montgomery 97 73
Moon 105 77
O'Connell 94 78
Prince 115 86
Reed 82 78
Roseman 96 80
Rumball-Peter 103 84
Shaw 98 70
Sheldon 71 52
Smith 76 79
Swatt 73 55
Torcivia 120 95
Vanmeter 109 77
Waddell 97 62
Wambach 96 74
Webb 89 66
Whitaker 100 89
Wilsoncroft 120 91
Use data set: Enter Data into R or R Studio
name time grade
Brown 75 61
Campbell 107 83
DiNitto 80 60
Frank 91 67
Gray 91 78
Harrett 89 64
Jarvis 95 85
Kirk 110 85
Kretzschmar 118 90

Lindsey 115 92
Martin 102 81
McHugh 83 63
Montgomery 97 73
Moon 105 77
O'Connell 94 78
Prince 115 86
Reed 82 78
Roseman 96 80
Rumball-Peter 103 84

Shaw 98 70
Sheldon 71 52
Smith 76 79
Swatt 73 55
Torcivia 120 95
Vanmeter 109 77
Waddell 97 62
Wambach 96 74
Webb 89 66
Whitaker 100 89
Wilsoncroft 120 91
Create a Scatter Plot Using R: First Open the data

Scatter Plot for (time, grade)

8
Each dot depicts a pair of variables

83

107
Calculate the Correlation Using R

Results
• The Pearson correlation (r) of .815 with a p < .001
(< .05) indicates a statistically significant strong positive
correlation between the test taking time and grade

Since r = .815 is a positive correlation


 time :  grade
Conclusion: Documenting results
To determine if there is a correlation between the
length of time students spent taking a two hour exam and the
score on it, we recorded the submission time (0…120
minutes) of each exam (n = 30). The exams were then
graded. We discovered a statistically significant positive
correlation between test time and grade (r = .815) (p < .001,
 = .05), wherein students who spent longer on their exams
earned higher grades, and vice versa.
Practice

Use (R-Studio or R ) to Compute and Graph the Correlation

See the Workshop “Correlatio0n Using RStudio

12
Correlation vs. Causation
• Correlation means that two variables move in a predictable fashion
with respect to each other
• In positive correlations, the variables move in the same direction:
(X Y) or (X Y)
• In negative correlations, the variables move in different directions:
(X Y) or (X Y)
• No matter how strong the correlation, it would be
inappropriate to automatically claim that X causes Y, or that
Y causes X
• Example:
• Z (Depression) may be causing:
X (Poor sleep)
and
Y (Low productivity)
Correlation vs. Causation

Three Causality Criteria


Criteria Rule Example

Variable A and variable B must be


Association / empirically related; there must be a
1 correlation (scientific) logical relationship between
Taking a dose of aspirin lowers fever.
A and B.

A (cause [independent variable]) The person took aspirin, and then the
2 Temporality precedes B (effect [dependent fever went down; not the other way
variable]). around.

The drop in fever is not due to the


The relationship between A and B are
3 Nonspurious
not caused by other variable(s).
room getting colder, submerging the
person in an ice bath, or other factors.

You might also like