Correlation

Project Management Project Sem - 6

Submitted To : Mr. Paramjeet Singh

Submitted By:
Poornima Mehra - 2k21/BBA/99
Prashasti Verma - 2k21/BBA/102
Syeda Saba Quasar - 2k21/BBA/153
Yashita Gupta – 2k21/BBA/178
 Introduction to Correlation
Definition of Correlation
Correlation measures the relationship between two variables, showing the strength
and direction of their linear connection using a correlation coefficient ("r").

Importance of Correlation in Data Analysis

Correlation helps understand relationships between variables, aiding in predictions
and decision-making. It uncovers patterns and trends in data, valuable in various fields
like economics, psychology, and medicine.

Correlation Does Not Imply Causation

Correlation indicates a statistical relationship but doesn't prove cause and effect.
Other factors may influence both variables, requiring further investigation
to establish causation.
 Types of Correlation
Positive Correlation

Definition: Indicates a direct relationship

where both variables increase
or decrease together.
Example: As the temperature increases,
so does ice cream sales.
Scatter Plot: Show a scatter plot
graph with points trending upward.
 Types of Correlation
Negative Correlation

Definition: Shows an inverse relationship

where one variable increases while
the other decreases.
Example: More hours spent commuting results
in fewer hours available for
leisure activities.
Scatter Plot: Display a scatter plot graph with
points trending downward.
 Zero Correlation
Definition of Zero Correlation:
Zero correlation indicates no linear relationship
between two variables. Changes in one variable do not
predict changes in the other.

Example of Zero Correlation:

An example could be the relationship between shoe
size and intelligence. There is no logical connection
between these variables, so they are likely to have zero
correlation.

Scatter Plot Illustrating Zero Correlation:

how a scatter plot graph with points scattered
randomly, indicating no clear trend or relationship
between the variables.
 Calculating Correlation Coefficient
Formula for Pearson Correlation Coefficient (r):

Explanation of Terms:
• X, Y: Represent the variables being correlated.
• Mean (, X, Y): Refers to the average value of X and Y, respectively.
• Covariance: Denoted by ∑(X−Xˉ)(Y−Yˉ), it measures the relationship between X and Y.
• Range of Values:
• The Pearson correlation coefficient (r) ranges from -1 to 1.
• r=1indicates a perfect positive correlation.
• r=−1 indicates a perfect negative correlation.
• r=0indicates no linear correlation; the variables are unrelated.
 Interpreting Correlation Coefficient
Interpretation of the Sign (+/-):
• Positive (+) Correlation: Indicates that as one
variable increases, the other variable tends to
increase as well.
•Negative (-) Correlation: Indicates that as one
variable increases, the other variable tends to
decrease.

Strength of Correlation:
•Weak (0.1 - 0.3): A weak correlation where the
relationship between variables is minimal.
•Moderate (0.3 - 0.7): A moderate correlation
where there is a noticeable relationship between
variables.
•Strong (0.7 - 1): A strong correlation where the
relationship between variables is substantial.
 Real-World Examples
• In economics
Higher unemployment rates tend to coincide with lower consumer spending, indicating economic downturns can
reduce consumer confidence. Policymakers and businesses use this correlation to assess the economy's health and
make decisions on policies, marketing, and inventory management.

• In science
Rising carbon dioxide levels often correspond with increased global temperatures. Scientists use this correlation to
forecast future climate patterns and advocate for actions to mitigate climate change, like reducing emissions and
adopting renewable energy.

• In Social Science
Higher education levels correlate with lower crime rates, suggesting that investing in education can reduce crime
by providing alternatives to criminal behavior. Conversely, areas with lower education levels often have higher
crime rates due to limited opportunities. Addressing educational disparities can help create safer communities.
 COMMON MISCONCEPTIONS
• "Clarifying Correlation Misconceptions"
1. One major misconception is that correlation implies causation, meaning if two things are correlated,
one must cause the other.
2. Another misconception is that the strength of correlation indicates the size of the effect. While a
correlation coefficient of 1 signifies a perfect relationship, it's rare to find such a scenario in real-life
data. Additionally, a correlation coefficient of 0 doesn't necessarily mean there's no relationship
between the variables; it simply means there's no linear relationship. It's important to recognize that
correlation doesn't always capture the full complexity of relationships between variables. Sometimes,
seemingly strong correlations may be due to chance, while weak correlations may still be meaningful
in certain contexts.

By clarifying these misunderstandings, we can use correlation more effectively to analyze and interpret
data.
CAUSATION VS CORRELATION
CAUSATION CORRELATION
• Causation means proving one
variable directly causes a change in • Correlation quantifies the degree of
another. For example, in studying association between two variables,
smoking and lung cancer, it's crucial indicating if they change together
(positive), opposite (negative), or
to show smoking causes an increase not at all (zero). However,
in lung cancer cases, not just a correlation doesn't imply causation;
correlation between the two. This a correlation between variables
often requires controlled experiments doesn't mean one causes the other to
change. It can arise from
or well-designed studies. coincidence, confounding factors,
• For example, in a study examining or indirect relationships.
the effect of a new drug on a • For instance, a positive correlation
medical condition, causation would between umbrella sales and rainfall
doesn't mean umbrellas cause rain;
be established if administering the it merely reflects the observation
drug resulted in a significant that people buy more umbrellas
improvement in the when it rains
condition compared to a control
group that did not receive the drug.
EXAMPLES OF MISLEADING CORRELATION
• Ice Cream Sales and Drowning Incidents
• During the summer months, both ice cream sales and drowning incidents tend to increase.
However, it would be incorrect to conclude that increased ice cream consumption causes more

1. drownings. The correlation between these two variables is due to a third factor—hot weather—
rather than a direct causal relationship between ice cream and drownings.

• Number of Firefighters and Fire Damage:

• In some areas, there might be a correlation between the number of firefighters present at a fire
and the amount of fire damage incurred. However, this doesn't mean that having more

2. firefighters causes more damage. Instead, larger fires may require more firefighters to be
dispatched, leading to a correlation between the two variables.

• Spending on Education and Crime Rates:

• It's commonly observed that areas with higher education spending tend to have lower crime
rates. However, this correlation doesn't necessarily mean that investing more in education

3. directly reduces crime. Other factors, such as socioeconomic status and community resources,
may also influence crime rates independently of education spending.
 Case Study on relationship between Advertisi
Spending And Sales Revenue
In a recent study, the company examined the relationship between advertising spending and sales revenue over
the past year.
Using correlation analysis, they calculated the Pearson correlation coefficient (r) and discovered a strong
positive correlation (r = 0.80) between advertising spending and sales revenue.
The finding implies that as advertising spending increases, sales revenue tends to rise in tandem. Consequently,
this insight offers valuable implications for our marketing strategy and budget allocation. By focusing their
resources on channels or campaigns that demonstrate the highest returns, they can maximize the impact of our
advertising efforts. Moreover, adjustments to their advertising strategies can be made based on the correlation
analysis, ensuring that their initiatives are aligned with their revenue generation goals.
In conclusion, correlation analysis has provided the company with a deeper understanding of the relationship
between advertising spending and sales revenue, empowering them to make more informed decisions in both
marketing budget allocation and strategy development.
Benefits of Understanding Correlation for Decision-Mak

03
02 04

01 Improved Marketing 05
Effectiveness
Enhanced Optimized Product
Forecasting Development
Accuracy
Informed Strategic
Customer
Planning
Relationship
Lorem Ipsum Management (CRM)
 References

• Britannica.com
• Ncl.ac.uk
• Investopedia.com
• Basic econometrics by Damodar N Gujarati
THANK YOU