Understanding

Statistics and Its

Real-World
Applications
From Basics to Practical Use in Your Career
What You Will Learn (Revise)
Statistics and its importance.
• Types of statistics.
• Types of data.
• Population and sampling.
• Frequency distribution.
• Graphical representation.
• Measures of central tendency (mean, median, mode).
What is Statistics?
• The science of collecting, analyzing, interpreting, and presenting data
to make informed decisions.
• Example:
• Deciding on a new product launch based on customer survey data.

• Real-World Use:
• Marketing: Analyzing customer preferences.
• Healthcare: Tracking disease trends.
Real world examples

Collecting Data: A bakery owner wants to know which type of cake sells the most.
•They record the number of each type of cake sold daily over a month.
Example: Chocolate cakes: 50, Vanilla cakes: 30, Red velvet cakes: 20.

Analyzing Data: The owner calculates the average daily sales for each cake type and compares the
numbers.
•Example: Average daily sales: Chocolate - 5, Vanilla - 3, Red velvet - 2.

Interpreting Data: The owner observes that chocolate cakes are the most popular.
•Based on this, they decide to stock more chocolate cakes and promote them.

•Conclusion: "Chocolate cakes are the top-selling item; increasing production can boost sales.
Descriptive vs. Inferential
Statistics
• Descriptive Statistics: Summarize and describe data.
• Example: Average marks of students in a class.
• Inferential Statistics: Make predictions or generalizations.
• Example: Predicting election outcomes from sample surveys.

Use Case:
• Descriptive: Reporting monthly sales figures.
• Inferential: Forecasting future sales trends.
 Types of Data
Categorical vs. Numerical Data
• Categorical Data (Qualitative): Describes categories (e.g., colors,
brands).
• Example: Preferred car brands (Toyota, BMW, Tesla).
• Numerical Data: Quantitative values (e.g., height, weight).
• Example: Age of participants in a survey.
1. Categorical Data

Describes categories or qualities that do not have numerical meaning.

• Nominal Data:
• Definition: Categories with no logical order.
• Example: Favorite car brands (Toyota, BMW, Tesla).
• Real-World Example: Types of cuisines (Italian, Chinese, Indian).
• Ordinal Data:
• Definition: Categories with a meaningful order but no consistent difference
between them.
• Example: Movie ratings (Excellent, Good, Average, Poor).
• Real-World Example: Education level (High school, Bachelor's, Master's, PhD).
2. Numerical Data
• Quantitative values that can be measured or counted.
Discrete and CONTINOUS
• Interval Data:
• Definition: Ordered data with meaningful intervals, but no true zero point.
• Example: Temperature in Celsius or Fahrenheit.
• Real-World Example: Calendar years (e.g., 1990, 2000, 2020).
• Ratio Data:
• Definition: Ordered data with meaningful intervals and a true zero point.
• Example: Weight of a person (in kilograms or pounds).
• Real-World Example: Distance traveled by a car (e.g., 0 km to 100 km).
Population and Sampling
• Population: The entire group you're studying.
• Example: All residents of a city.

• Sample: A smaller, representative group from the population.

• Example: 500 residents surveyed about city amenities.

• Surveying customer feedback without contacting every customer.

Types of Sampling techniques
Simple Random Sampling:
A company randomly selects 100 employees from a list of 1,000
to participate in a survey on job satisfaction. Every employee
has an equal chance of being chosen.

Stratified Sampling: A university selects students from different

departments (e.g., science, arts, business) to ensure that each
department is adequately represented in a survey on student
satisfaction.

Systematic Sampling: A researcher chooses every 10th person

on a city’s voter registration list to participate in a study on
election preferences.

Cluster Sampling: A school district selects a few schools at

random, then surveys all the students within those selected
schools to assess educational performance.
 Frequency Distribution
• A table or chart showing how often data values occur.
Graphical Representation
•Why Use Graphs: Graphs simplify complex data
making it easier to understand and compare.

•How They're Used: Graphs display trends,

relationships, and patterns visually, helping users
analyze data effectively.

•Usefulness in Analysis: Graphs highlight key

insights, making it easier to communicate findings
and make data-driven decisions.

•Example: A bar chart is used to compare market

share percentages of different brands, guiding
marketing strategies.
Bar Chart
Definition: A chart with rectangular
bars representing data values.
• Real Example: Sales of different
products over a quarter.
• Why Use It: To compare different
categories or groups.
• When to Use: When you need to
show comparisons between different
groups.
Line Graph
Definition: A graph showing data
points connected by straight lines,
often used to show trends.
• Real Example: Sales trends over
a year.
• Why Use It: To show changes
over time.
• When to Use: When you need
to illustrate the trend or
changes of data points over
time.
 Pie Chart
Definition: A circular chart divided into
segments to show proportionate data.
• Real Example: Market share of
different brands in a sector.
• Why Use It: To show parts of a whole.
• When to Use: When you want to
show percentage or proportional data.
 Histogram
Definition: A bar graph representing the
frequency distribution of data.
• Real Example: Distribution of ages in a
survey.
• Why Use It: To show the distribution of
data and understand its spread.
• When to Use: When you want to display
the frequency of continuous data.
Measures of central tendency
• What are Mean, Median, and Mode?
• Mean: The average value, sensitive to all data points.
• Median: The middle value, less sensitive to outliers.
• Mode: The most frequent value, useful for identifying trends.
• Importance: Helps analyze real-world data like salaries, exam scores, and
customer purchases.
• 75,72,54,92,90
• 54,72,75,90,92 4,3,4,4.6,4.8,4.2 mean 4.1
• Median 75 4,3,4,4.6,4.8,4.2,2,2 mean 3
• south Indian 21 Chinese 15 Indian 34 itlaian 12
• Mode 34
Mean
• Mean is the average of all the numbers in a dataset.
• =4,8,10,12,16
• =adding up all the values in a dataset/number of values
• =4+8+10+12+16/5
• =50/5=10 average value (mean)
• =20,20,30,40,50 32 is the centre of my dataset
• =20+30+20+40+50/5
• =160/5=32
Real estate
• House prices=[72,000,75,000,56,000,2,000000]
• =5,50,750
• Outlier (abnormal behaviour)
Mean
• Average- most of the people have given it 4.2
• Mean is the average of all the numbers in a dataset
• Steps- add up all the numbers/number of values
• Ages=[4,8,10,12,16]
• Calculate the mean of my ages
• 4+8+10+12+16= 50
• 50/5= 10
• 10,20,30 mean of these numbers =20+30+10/3 =60/3= 20
Business and Sales analysis
• A company wants to know mean revenue per month to assess overall
performance
• Example if a shop earns around 10,000, 15,000, 12,000 over 3
months
• Calculate?
• Mean=add all the values/number of values
• [10,000, 12,000, 15,000]
• =10,000+15,000+12,000/3=37,000/3=12,333
• On an average my company earns around 12,333
Healthcare
• Doctors calculate the avg bp or heartrate of a patient over time to
assess health
• Bp readings are
• [120,125,130]
• =120+130+125/3=125
• Bp of 125 suggests stable readings over time
Education
• [92,45,67,87,92]
• 92+45+67+87+92/5=383/5=76.6
Sports
• The mean score or performance shows a players or teams overall
consistency
• A mean score of 60 runs suggests that a player typically scores around
60

Transportation
The mean travel time represents the typical time it takes to complete a
route
Customer behavior
• The mean spending or rating reflects the typical behavior
• A mean rating of 4.5 stars suggests more customers are satisfied even
if a few left low ratings
Median
• It gives me central or the middle point
• Ages= 8,4,10,12,13
• Step 1 – arrange them in an ascending order (lowest to highest)
• - 4,8,10,12
• Step 2- count the number of values values= 4 (even)
• Step 3 if it is an even number just calculate the average of the middle
2 elements
• 8+10/2= 18/2 = 9
Median
• 8,4,10,12,14
• Step 1- arrange it in an ascending order (lowest to highest)
• 4,8,10,12,14
• Step 2 count the no. of values = 5 (odd number)
• Step 3 pick the central element median=10
Median
• Step 1 arrange it
• Wheter the number of values are even/odd
• If it is even just average the middle 2 elements
• If it is odd just pick the central element
•
Mode – most frequent value
• Cuisines marks
• Indian 24 Rajani,abhishek,Rajani,Rajani,Rajani,danush,
• South Indian 23 Rajani 4
• Italian 19
• Chinese 26

• [2,4,6,6,8,10] 34,38,39,40,40,48
• 2 1 34 1
• 39 1
• M 38 1 40 2
• 4 1
• 6 2
Q1 Mean income of the employees
in a dataset
• [2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• Calculate the mean
• =sum of all incomes/number of employees
• =42100/10
• =4210 (mean)
Q2 If an employee earning 4000$ is removed
from the dataset what is the new mean income?
• 2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• Updated dataset=[2500,3000,3200,3200,4500,5000,5200,5500,6000]
• =add up all the values/number of values
• =38100/9
• =4233.33
Q3 If a new employee with an income of 7000
is added what will be the new mean income
• 2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]

• Updated
dataset=[2500,3000,3200,3200,4000,4500,5000,5200,5500,6000,700
0]
• Adding up all the values/number of values
• 49100/11=4463.63
Q4 calculate the mean income of
employees earning less than 5000
• [50,80,2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• New dataset=50,80,2500,3000,3200,3200,4000,4500
• mean=20400/6
• =3400

20530/8
2566
Q5 what is the mean income if the
incomes are increased by 10%?
• Dataset= 2500,3000,3200,3200,4000,4500,5000,5200,5500,6000
• updated
dataset=[2750,3300,3520,3520,4400,4950,5500,5720,6050,6600]
• =46310/10
• =4631
Median
Q6 what is the median income of the employees in
a dataset
• [2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• Step1 – arranging them in an ascending order
• Step 2- counting the number of items (even or odd)
• Step 3- 10(even)

• 4000+4500/2
• 8500/2=4250 (median value).
Q7 if an employee earning 7000 is
added, how does the median
income change?
2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]

2500,3000,3200,3200,4000,4500,5000,5200,5500,6000,7000]

11(odd)
4500
Q8 Remove the highest income from
the dataset what is the new median
income?
• [2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• Updated dataset=[2500,3000,3200,3200,4000,4500,5000,5200,5500]
• 4000
Q9 find the median income of
employees earning between 3000
and 5500
• [2500,3000,3200,3200,4000,4500,5000,5200,5500,6000
• Updated dataset=[3000,3200,3200,4000,4500,5000,5200]
• median- number of values (7)
• Middle element /median 4000
mode
Q11 What is the mode of the dataset
[2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
3200
Q12 if the income 3200 is changed to 3500 for one of the employees
what is the new mode
2500,3000,3200,3500,4000,4500,5000,5200,5500,6000]
Mode- no mode
Q13 if all the incomes are multiplied by 2 does the
mode change?

• 2500,3000,3200,3200,4000,4500,5000,5200,5500,6000]
• Updated
dataset=[5000,6000,6400,6400,8000,9000,10000,10400,11000,12000
]
• mode 6400 (2 times)
Q14 find the mode of the subset of
incomes greater than 3000$
• 2500,3000,3200,3200,4000,4500,5000,5200,5500,6000
• Updated dataset=3200,3200,4000,4500,5000,5200,5500,6000
Mode 3200
Q15 if an additional employee with an income of 3200 is added what
happens to the mode?
Updated
dataset=2500,3000,3200,3200,3200,4000,4500,5000,5200,5500,6000
Mode 3200 frequency (3)
Household prices
• 30,000 , 35,000, 40,000, 5,00,000 , 55,000
• 5,00,000 – outliers
• Mean- 1,32,000 we do not use mean (sensitive to outliers)
Median- 40,000 (you will use median in case of outliers)

Mode – use to deal with categorical data

- To calculate the frequency
• 30,000 , 35,000, 40,000, 55000, 5,00,000 outlier
• Outlier – a value which behaves abnormally
• Mode no - number of highest frequency (categorical data)
• Mean 1,32,000
• Median 40,000
Mean for Grouped Data
Grouped distribution

f- frequency
x-middle point (lower+upper)/2

• Steps:
1.Compute midpoints x.
2.Multiply f by x for each class.
3.Add f⋅x and divide by total f:
Real-World Example:

Problem:
A company collects data on employee salaries (in thousands).

• Mean=1590/38
• The mean of the grouped distribution is 41.8421
Median for Grouped Data
Where:
•L: Lower boundary of the median class
•N: Total frequency
•CF: Cumulative frequency before the median class
•fm: Frequency of the median class
•h: Class width
Real-World Example:

• Problem: Using the same salary data:

 Median for grouped distribution
Where:
•L: Lower boundary of the median class
•N: Total frequency
•CF: Cumulative frequency before the median class
•fm: Frequency of the median class
•h: Class width
Mode for Grouped Data

Where:
•L: Lower boundary of the modal class
•f1​: Frequency of the modal class
•f0​: Frequency before the modal class
•f2: Frequency after the modal class
•h: Class width
Mode for grouped distribution
Overall results:
•When to use each measure:
• Mean: For overall trends.
• Median: When data has outliers.
• Mode: To find the most common
category.

•Mean = 41.8k
•Median = 42.6k
•Mode = 44k