Chapter 5: Data Analysis
Introduction
Data analysis means carefully studying and arranging data so that we can understand it better and
take out useful information from it.
In short (2–3 lines):
Data analysis is the process of examining and organizing data to find patterns, results, and to
make good decisions.
Basic Statistical Concept
Definition:
Statistics is the part of mathematics that deals with collecting, arranging, presenting, and
studying data to make it meaningful and useful.
Important Points:
Helps in changing large amounts of data into simple information.
Used in education, business, sports, health, and research.
Makes decision-making easier by showing facts with numbers.
Example (text):
A school records marks of all students. Instead of checking each student’s marks, statistics help
the teacher find the class average.
Measure of Central Tendency
Definition:
A measure of central tendency is a single value that represents the middle or central position of
the whole group of data.
Important Points:
It summarizes the entire data set with one value.
It shows the most typical or common performance.
The three main types are mean, median, and mode.
Example (text):
If five students get different marks, the central tendency will give one number that shows the
overall performance of the class.
Types of Central Tendency
Mean
Definition:
Mean is the average value of data, calculated by dividing the total sum of all numbers by the
total number of items.
Formula:
Mean (x) = Sum of all values = a + b + c +d
No of values (n) n(4)
Important Points:
It uses every value in the data set.
It is easily affected by very high or very low values.
Example (text):
If marks are 10, 20, 30 → Mean = (10+20+30)/3 = 20.
Median
Definition:
Median is the middle value of data when it is arranged in order, dividing the data set into two
equal halves.
Formula:
If nn is odd:
Median=Value at ( n + 1 )
2
Important Points:
It is not affected by very high or very low values.
It is more reliable for uneven data distribution.
Example (text):
If marks are 10, 20, 30, 40 → Median = (20+30)/2 = 25.
Mode
Definition:
Mode is the value that comes again and again in the data set and shows the most common
observation among all values.
Formula:
Mode = Most repeated value in the data set.
Important Points:
A set can have no mode, one mode, or more than one mode.
It is useful for finding the most common choice.
Example (text):
If marks are 12, 15, 15, 18, 20 → Mode = 15.
Measures of Dispersion
Introduction
Definition:
Measures of dispersion are values that tell us how much the data values differ or spread out from
the average.
Important Points:
It shows the consistency or variation in data.
If dispersion is small → data is consistent.
If dispersion is large → data is spread out.
Common measures: Range, Variance, Standard Deviation.
Example (text):
Two cricket players have the same batting average of 50 runs. But one player usually scores
between 40–60, while the other scores between 0–100. Their averages are the same, but their
dispersion is very different.
Variance
Definition:
Variance measures the average of squared differences between each data value and the mean,
showing how spread out the data is.
Formula:
Where:
xix_i = each value
xˉ\bar{x} = mean of data
NN = number of values
Example (calculation):
Data: 2, 4, 6, 8
1. Mean = (2+4+6+8)/4 = 5
2. Differences from mean = -3, -1, 1, 3
3. Squared differences = 9, 1, 1, 9
4. Variance = (9+1+1+9)/4 = 20/4 = 5
Real-life example:
In a class, two groups have the same average marks (70), but in one group all students scored
near 70, while in the other group some scored 40 and some 100. Variance helps to see which
group has more mixed results.
Standard Deviation
Definition:
Standard deviation is the square root of variance and shows the average distance of values from
the mean.
Formula:
Example (calculation using same data):
Data: 2, 4, 6, 8
Variance = 5 (calculated above)
Standard deviation = √5 = 2.23 (approx.)
Real-life example:
If two factories produce light bulbs, both have average life of 1000 hours. In factory A, most
bulbs last between 990–1010 hours (small standard deviation). In factory B, bulbs last between
800–1200 hours (large standard deviation). This shows factory A is more reliable.
Introduction to Probability
Definition:
Probability is the measure of how likely an event is to happen. It tells the chance of an event.
Formula:
Important Points:
Probability is always between 0 and 1.
0 means impossible, 1 means certain.
More favorable outcomes = higher probability.
Example (text):
If you toss a coin:
Probability of getting Head = 1/2
Probability of getting Tail = 1/2
Real-life example:
Weather forecasting uses probability. If the weather report says 70% chance of rain, it means in
similar conditions, out of 100 days, it usually rains on 70.
Data Collection and Preparation
Data Collection
Definition:
Data collection is the process of gathering information from different sources so it can be studied
and used for making decisions.
Methods of Data Collection:
1. Surveys – Asking people questions to get their opinions or information.
2. Observation – Watching and recording behavior or activities.
3. Experiments – Performing tests in controlled conditions to gather results.
4. Interviews – Talking directly with people to collect answers.
5. Questionnaires – Written forms filled by people with their responses.
Surveys (as a data collection method)
Definition:
A survey is a method of collecting data by asking questions from a group of people.
Customer Preference Surveys:
These surveys are used to find out what customers like, dislike, or expect from a product or
service.
Points about Customer Preference Surveys:
1. Help businesses understand customer needs and choices.
2. Provide feedback for improving products and services.
3. Guide companies in creating better marketing strategies.
4. Support launching new products according to customer needs.
5. Show which features or services customers value most.
Examples of Surveys in Real Life:
A mobile company asking which phone feature (camera, battery, price) customers prefer.
A restaurant asking for feedback about food and service.
An online store asking customers to rate delivery speed.
Observation
Definition:
Observation is collecting data by carefully watching, listening, or recording people’s actions and
behavior.
Examples:
A teacher watching how students focus in class.
A traffic police officer counting vehicles passing in one hour.
Important Points:
Provides real and direct information.
Shows what people actually do, not just what they say.
Sometimes slow and time-consuming.
Experiments
Definition:
An experiment is a method of collecting data by performing tests in controlled conditions and
recording the results.
Examples:
A scientist testing fertilizers on plant growth.
A doctor testing how a new medicine works.
Important Points:
Produces accurate and reliable results.
Results can be repeated and verified.
Often requires more cost and time.
Data Preparation
Definition:
Data preparation is the process of organizing and adjusting collected data so it is ready for
analysis.
Example:
If survey data records age as “twenty,” it must be converted into 20 before analysis.
Practical Life Example:
Before analyzing exam results, a teacher prepares the marks list by correcting mistakes, filling
missing marks, and arranging it properly.
Data Cleaning and Transformation
1. Data Cleaning
Definition:
Data cleaning means fixing errors, removing wrong information, and correcting mistakes in data
to make it accurate and useful.
Example:
If a student’s age is recorded as 200 years, it should be corrected.
Important Points:
Removes typing errors or duplicates.
Fixes missing or wrong values.
Improves data quality.
2. Data Transformation
Definition:
Data transformation means changing data into another format or structure so it is easier to
analyze.
Example with Table:
Raw Data Transformed Data
Male/Female 1 / 0
Twenty-Five 25
High/Low 2 / 1
Important Points:
Makes data suitable for computer programs.
Helps in faster and more accurate analysis.
Converts data into a standard form.
Handling Missing Data
Sometimes some information is not available. This is called missing data.
Ways to Handle Missing Data (with examples):
1. Remove Incomplete Data – Ignore empty or useless forms.
Example: Dropping a blank feedback form.
2. Fill with Estimates – Replace missing values with averages.
Example: If a student’s height is missing, use class average height.
3. Use Last Known Value – Use the most recent available data.
Example: If one month’s meter reading is missing, use the previous month’s reading.
4. Ask Again – Contact the person for missing information.
Example: Call a customer again if they skipped a question.
Flow Diagram
Data Collection → Data Preparation → Data Cleaning → Data
Transformation → Analysis