Easy2Siksha.com
GNDU QUESTION PAPERS 2025
BA/BSc 4
th
SEMESTER
QUANTITATIVE TECHNIQUES – IV
Time Allowed: 3 Hours Maximum Marks: 100
Note: Aempt Five quesons in all, selecng at least One queson from each secon. The
Fih queson may be aempted from any secon. All quesons carry equal marks.
SECTION–A
1. Find the mulple regression equaon of
on
and
, from the data relang to three
variables given below :
X₁
4
6
7
8
9
13
5
X₂
15
12
8
6
4
10
3
X₃
30
24
20
14
10
4
2
2. Write about Gernpertz curves. Discuss the general shapes and methods to t Gernpertz
curves.
SECTION–B
3. (a) State and prove addion law of probability.
(b) There are three machines A, B and C in a factory. Their daily outputs are in the rao of
2 : 3 : 5. Past experience shows that 2%, 4% and 5% of the items produced by A, B and C
respecvely are defecve. If an item selected at random is found to be defecve, nd the
probability that it was produced by A or B.
4. Dene Mathemacal expectaon. Explain the properes of Mathemacal expectaon.
Easy2Siksha.com
SECTION–C
5. What is Normal Distribuon? Discuss the properes of Normal Distribuon.
6. (a) Obtain the Mean and Standard Deviaon of a Binomial Distribuon for which
and
(b) If X is Poisson variate such that :
Find the Mean and Variance of the Distribuon.
SECTION–D
7. (a) Write about complete enumeraon sample surveys.
(b) What are the features of a Good Sample?
8. Explain the various methods of Sampling. Also discuss their relave merits and
demerits.
Easy2Siksha.com
GNDU ANSWER PAPERS 2025
BA/BSc 4
th
SEMESTER
QUANTITATIVE TECHNIQUES – IV
Time Allowed: 3 Hours Maximum Marks: 100
Note: Aempt Five quesons in all, selecng at least One queson from each secon. The
Fih queson may be aempted from any secon. All quesons carry equal marks.
SECTION–A
1. Find the mulple regression equaon of
on
and
, from the data relang to three
variables given below :
X₁
4
6
7
8
9
13
5
X₂
15
12
8
6
4
10
3
X₃
30
24
20
14
10
4
2
Ans: What is this question really asking?
You are given data for three variables:
• X₁ → the dependent variable (the one we want to predict)
• X₂ and X₃ → independent variables (the predictors)
The question asks:
Find the multiple regression equation of X₁ on X₂ and X₃
That means we want to build an equation that helps us estimate or predict X₁, using the
values of X₂ and X₃ together.
Given Data
Observation
X₁
X₂
X₃
1
4
15
30
Easy2Siksha.com
2
6
12
24
3
7
8
20
4
8
6
14
5
9
4
10
6
13
10
4
7
5
3
2
Understanding Multiple Regression (in simple words)
In simple regression, we predict one variable using one predictor.
In multiple regression, we predict one variable using two or more predictors.
Here, we are predicting X₁ using X₂ and X₃.
So the general multiple regression equation looks like this:
Where:
• a = constant (intercept)
• b₁ = regression coefficient of X₂
• b₂ = regression coefficient of X₃
Our goal is to find the values of a, b₁, and b₂.
How are these values found?
In theory, we use a system of equations called normal equations, which are derived using
the least squares method.
The idea behind least squares is simple:
Choose the values of a, b₁, and b₂ such that the total error between actual and predicted X₁
values is as small as possible.
In real exams and practical work, this process involves:
• Calculating means
• Finding deviations
• Computing sums like ΣX₂², ΣX₃², ΣX₂X₃, ΣX₁X₂, etc.
• Solving three simultaneous equations
Easy2Siksha.com
This can be long and messy by hand, but the logic is systematic and mathematical.
Final Computation Result
After correctly applying the multiple regression method to the given data, we obtain the
following values:
• a ≈ 8.38
• b₁ ≈ 0.41
• b₂ ≈ −0.29
The Multiple Regression Equation
How to interpret this equation (very important!)
Let’s understand what this equation is telling us in plain language.
Constant (8.38)
If both X₂ and X₃ were zero, the predicted value of X₁ would be about 8.38.
(This is theoretical—it just gives a baseline.)
Coefficient of X₂ ( +0.41 )
This means:
If X₂ increases by 1 unit, X₁ increases by about 0.41 units,
assuming X₃ remains constant.
So, X₂ has a positive relationship with X₁.
Coefficient of X₃ ( −0.29 )
This means:
Easy2Siksha.com
If X₃ increases by 1 unit, X₁ decreases by about 0.29 units,
assuming X₂ remains constant.
So, X₃ has a negative relationship with X₁.
Why is this useful?
This equation allows us to:
• Predict X₁ when X₂ and X₃ are known
• Understand how multiple factors together influence one variable
• See which independent variable increases or decreases X₁
This kind of analysis is widely used in:
• Economics
• Business forecasting
• Social sciences
• Data analysis and research
2. Write about Gernpertz curves. Discuss the general shapes and methods to t Gernpertz
curves.
Ans: What is a Gompertz Curve?
Think of a Gompertz curve as a special kind of growth curve. It’s used to describe situations
where growth starts fast but then slows down as it approaches a maximum limit.
Imagine planting a tree:
• At first, it grows quickly because it’s young and full of energy.
• Later, as it matures, the growth slows down.
• Eventually, it reaches a stable height—it doesn’t grow forever.
That slowing-down pattern is exactly what the Gompertz curve captures. It’s widely used in
biology, medicine, economics, and even marketing to describe growth processes that have
natural limits.
General Shape of the Gompertz Curve
The Gompertz curve has a very distinctive “S-shape,” but it’s not perfectly symmetrical like
some other growth curves. Let’s break it down:
Easy2Siksha.com
• Starting Point: Growth begins slowly. For example, a new product in the market
doesn’t sell much at first.
• Middle Stage: Growth speeds up rapidly. Suddenly, everyone wants the product, or
the population of bacteria multiplies quickly.
• Later Stage: Growth slows down again as it approaches a maximum limit. Maybe the
market is saturated, or the bacteria run out of food.
So, the Gompertz curve looks like a stretched-out “S” that leans more heavily on one side.
It’s steeper in the middle but flattens out at the top.
Why is the Gompertz Curve Important?
This curve is powerful because it reflects real-life growth patterns better than simple
straight lines or even some other curves.
• In Biology: It describes how tumors grow, how animals gain weight, or how
populations expand.
• In Business: It models how new technologies or products spread in society.
• In Demography: It’s used to study human mortality and aging patterns.
Basically, whenever growth has a natural ceiling, the Gompertz curve is a great tool to
understand it.
Methods to Fit Gompertz Curves
Now, how do scientists or researchers actually “fit” a Gompertz curve to real data? Imagine
you have a bunch of points on a graph showing how something grows over time. You want
to draw a smooth Gompertz curve that matches those points. Here’s how it’s usually done:
1. Mathematical Formula: The Gompertz curve is expressed with an equation that
looks like this:
• = the upper limit (the maximum value the growth will reach).
• = controls how far the starting point is from the maximum.
• = controls the growth rate (how fast it rises in the middle).
Don’t worry if the formula looks intimidating—it’s just a way to capture the “slow-fast-slow”
growth pattern.
2. Curve Fitting Techniques:
o Least Squares Method: Researchers use statistical tools to minimize the
difference between the actual data points and the curve.
o Nonlinear Regression: Since the Gompertz curve isn’t a straight line, special
regression techniques are used to fit it.
o Software Tools: Programs like R, Python, or even Excel can be used to fit
Gompertz curves to data.
Easy2Siksha.com
3. Practical Example: Suppose you’re studying how a new social media app gains users.
At first, only a few people join. Then, suddenly, millions sign up. Finally, growth slows
because almost everyone who wants the app already has it. By fitting a Gompertz
curve, you can predict when growth will slow and what the maximum number of
users might be.
A Relatable Analogy
Think of the Gompertz curve like baking bread:
• At first, the dough rises slowly.
• Then, it suddenly expands quickly in the oven.
• Finally, it stops rising once it reaches its limit.
That’s the same “slow-fast-slow” rhythm the Gompertz curve describes.
Key Takeaways
• The Gompertz curve is a growth curve that starts slow, speeds up, and then slows
again.
• Its shape is an asymmetric “S,” leaning more on one side.
• It’s used in biology, economics, marketing, and demography to describe growth with
limits.
• Fitting Gompertz curves involves using mathematical formulas and statistical
methods to match real-world data.
Final Thoughts
The beauty of the Gompertz curve lies in its realism. Life rarely grows in a straight line.
Whether it’s populations, products, or even personal progress, growth tends to follow that
“slow-fast-slow” rhythm. The Gompertz curve gives us a mathematical lens to see and
predict that journey.
SECTION–B
3. (a) State and prove addion law of probability.
(b) There are three machines A, B and C in a factory. Their daily outputs are in the rao of
2 : 3 : 5. Past experience shows that 2%, 4% and 5% of the items produced by A, B and C
respecvely are defecve. If an item selected at random is found to be defecve, nd the
probability that it was produced by A or B.
Ans: Part (a): Addition Law of Probability
What is Probability? (Quick Reminder)
Easy2Siksha.com
Probability is simply a way to measure how likely something is to happen.
• Probability of any event lies between 0 and 1
• 0 means impossible
• 1 means certain
Understanding Events
Imagine two events:
• Event A: It rains today
• Event B: It is cloudy today
These two events can happen together. This idea is very important.
Statement of Addition Law of Probability
General Addition Law
For any two events A and B:
Where:
• = Probability that A or B or both occur
• = Probability that both A and B occur
Special Case: Mutually Exclusive Events
If events A and B cannot happen together, then:
So the formula becomes:
Easy2Siksha.com
Why Do We Subtract ?
Think of this like counting people:
• You count people wearing red shirts
• You count people wearing blue shirts
• Some people are wearing both
If you just add both groups, you count those people twice.
So, you subtract the overlap once. That’s exactly what happens in probability.
Proof of the Addition Law
Let:
• Event A has probability
• Event B has probability
Now, event A can be divided into:
• A only
• Both A and B
Similarly, event B includes:
• B only
• Both A and B
So if we add and , the common part gets added twice.
To correct this, we subtract it once.
Hence,
Thus, the Addition Law of Probability is proved.
Part (b): Factory Machines and Defective Items
Now let’s move to the real-life application.
Easy2Siksha.com
Understanding the Situation
There are three machines in a factory:
Machine
Output Ratio
Defective Percentage
A
2
2%
B
3
4%
C
5
5%
Total ratio = 2 + 3 + 5 = 10
Step 1: Convert Ratios into Probabilities
Since total production is 10 parts:
• Probability item is from A:
• Probability item is from B:
• Probability item is from C:
Step 2: Probabilities of Defective Items
Given:
• Defective from A = 2% = 0.02
• Defective from B = 4% = 0.04
• Defective from C = 5% = 0.05
Step 3: Find Total Probability of Getting a Defective Item
Easy2Siksha.com
This is where Addition Law is used.
Substitute values:
So, the probability that a randomly selected item is defective is 0.041.
Step 4: Probability That Defective Item Came from A or B
We are asked:
or
That means:
“Given that the item is defective, what is the chance it came from A or B?”
Using Conditional Probability
Now calculate numerator:
•
•
So:
Easy2Siksha.com
4. Dene Mathemacal expectaon. Explain the properes of Mathemacal expectaon.
Ans: What is Mathematical Expectation?
Mathematical expectation, often called expected value, is a way of predicting the average
outcome of a random event if you repeated it many times.
Imagine tossing a fair coin:
• You know there are two possible outcomes—heads or tails.
• If you toss it thousands of times, you’d expect about half to be heads and half to be
tails.
That “average” result you expect is what mathematicians call expectation. It’s like the long-
term prediction of what will happen, even though individual outcomes may vary.
Formally, mathematical expectation is defined as the weighted average of all possible
outcomes, where each outcome is multiplied by its probability.
Here:
• = possible outcome
• = probability of that outcome
• = expected value
A Relatable Example
Let’s say you play a simple dice game:
• If you roll a 6, you win ₹60.
• If you roll anything else, you win nothing.
What’s your expected earning?
• Probability of rolling a 6 = .
• Reward for rolling a 6 = ₹60.
• Expected value =
₹.
So, even though you might win ₹60 sometimes and nothing most of the time, on average,
each roll is worth ₹10. That’s the power of expectation—it tells you the “fair value” of a
random game.
Properties of Mathematical Expectation
Easy2Siksha.com
Now let’s explore the key properties that make expectation so useful.
1. Linearity of Expectation
This is one of the most important properties. It means:
In simple words: if you combine two random variables (like two dice rolls), the expectation
of the sum is just the sum of their expectations.
Example: If one dice roll has an expected value of 3.5, and another dice roll also has 3.5,
then the expectation of rolling both and adding them is .
2. Expectation of a Constant
If a random variable always takes the same value (say 5), then its expectation is just that
constant.
It’s like saying: if you always get ₹100 no matter what, the expected value is obviously ₹100.
3. Multiplication by a Constant
If you multiply a random variable by a constant, the expectation also gets multiplied by that
constant.
Example: If the expected value of your dice roll is 3.5, and you decide to double the reward
for each outcome, the new expectation is .
4. Additivity
Expectation is additive. If you have two independent random variables, the expectation of
their sum is the sum of their expectations.
This is why expectation is so useful in probability and statistics—it allows us to break down
complex problems into smaller, manageable parts.
5. Non-Negativity (for Non-Negative Variables)
If a random variable can only take non-negative values (like the number of goals scored in a
match), then its expectation is also non-negative.
6. Expectation and Probability Distribution
Easy2Siksha.com
Expectation depends directly on the probability distribution of the random variable. If
probabilities change, the expectation changes too.
For example, if a dice is biased to land on 6 more often, the expected value of a roll will be
higher than 3.5.
Why is Mathematical Expectation Important?
Expectation is everywhere in real life:
• Insurance companies use it to calculate premiums.
• Casinos use it to design games where the expected value favors the house.
• Economists use it to predict average incomes or market behaviors.
• Scientists use it to model outcomes in experiments.
It’s like a crystal ball that doesn’t tell you exactly what will happen, but gives you the
average picture over time.
A Simple Analogy
Think of expectation like the “center of gravity” of a random event. Even though outcomes
may swing left or right, the expectation tells you where the balance point lies.
Or imagine playing a game over and over—the expectation is the score you’d expect if you
averaged all your results.
Final Thoughts
Mathematical expectation is more than just a formula—it’s a way of thinking about
uncertainty. It helps us see patterns in randomness, predict long-term outcomes, and make
fair decisions in games, business, and science.
SECTION–C
5. What is Normal Distribuon? Discuss the properes of Normal Distribuon.
Ans: Imagine you are standing in your class and looking at the heights of all students. You’ll
notice something interesting:
• Most students are of average height
• A few students are very tall
• A few students are very short
Easy2Siksha.com
If you draw a graph of these heights, the shape you get looks like a bell —wide in the
middle and tapering on both sides.
This bell-shaped curve is called the Normal Distribution.
So, in simple words:
Normal Distribution is a statistical distribution in which most of the data values are
concentrated around the average (mean), and the rest of the values spread symmetrically
on both sides, forming a bell-shaped curve.
It is one of the most important concepts in statistics, because many natural, social, and
economic phenomena follow this pattern.
Understanding Normal Distribution Through Daily Life
To really feel what normal distribution means, let’s look at some everyday examples:
• Marks in an exam
Most students score average marks, fewer students score very high or very low
marks.
• IQ scores
Most people have an average IQ, while very high or very low IQ scores are rare.
• Errors in measurement
Small errors occur more often than large errors.
• Blood pressure, body weight, reaction time
These biological traits usually follow a normal distribution.
All these examples show the same idea:
Average values occur most frequently, extreme values occur rarely.
Mathematical Definition (Very Simple)
A normal distribution is completely described by two values:
Easy2Siksha.com
1. Mean (μ) – the average
2. Standard Deviation (σ) – how much the data spreads from the mean
Once you know these two, the entire curve is fixed.
Shape of the Normal Distribution Curve
The graph of a normal distribution has some very clear features:
• It looks like a bell
• It is symmetrical
• The highest point is at the mean
• The curve never touches the horizontal axis (it comes very close but never meets it)
This unique and smooth shape makes the normal distribution easy to recognize.
Properties of Normal Distribution
Now let us discuss the important properties of the normal distribution in a simple and easy-
to-remember way.
1. Bell-Shaped Curve
The most obvious property is its bell shape.
• The curve rises gradually, reaches a peak, and then falls gradually
• The highest point represents the most frequent value
This bell shape shows that most observations cluster around the average, and very few are
far away from it.
2. Symmetrical About the Mean
The normal distribution is perfectly symmetrical.
• Left side of the curve = Right side of the curve
• Values above the mean behave the same way as values below the mean
Easy2Siksha.com
For example:
If 10% of students score much higher than average, then about 10% score much lower than
average too.
3. Mean = Median = Mode
This is a very important property.
In a normal distribution:
• Mean (average)
• Median (middle value)
• Mode (most frequent value)
All three are equal and lie at the center of the distribution.
This tells us that the data is balanced, with no skewness to the left or right.
4. Maximum Frequency at the Mean
The highest point of the curve is at the mean.
This means:
• The mean value occurs most frequently
• Most observations are close to the average
As you move away from the mean in either direction, the frequency of observations
decreases.
5. The Curve Extends Infinitely on Both Sides
The normal curve:
• Extends endlessly to the left and right
• But never touches the X-axis
This means:
• Extreme values are possible
• But their probability is very very small
Easy2Siksha.com
So, extremely high or extremely low values exist, but they are rare.
6. Total Area Under the Curve is Equal to 1
This property is about probability.
• The entire area under the normal curve equals 1 (or 100%)
• This represents the total probability of all possible values
Half of the area lies on:
• The left side of the mean (50%)
• The right side of the mean (50%)
7. Follows the Empirical Rule (68–95–99.7 Rule)
This is one of the most useful and easy rules of normal distribution.
According to this rule:
• 68% of data lies within ±1 standard deviation from the mean
• 95% of data lies within ±2 standard deviations
• 99.7% of data lies within ±3 standard deviations
This helps us quickly estimate probabilities without complicated calculations.
8. Defined Completely by Mean and Standard Deviation
Once you know:
• Mean (μ)
• Standard Deviation (σ)
You can:
• Draw the curve
• Calculate probabilities
• Compare distributions
No other information is needed.
Easy2Siksha.com
9. No Skewness (Perfect Balance)
Normal distribution has:
• Zero skewness
• No long tail on either side
This means the data is perfectly balanced, unlike skewed distributions where data is pulled
more toward one side.
10. Widely Used in Statistics and Research
Because of its natural occurrence and useful properties, normal distribution is used in:
• Education (exam scores)
• Economics (income distribution)
• Psychology (intelligence tests)
• Medical science (health parameters)
• Quality control and research
It forms the foundation of many statistical techniques.
Why Normal Distribution Is So Important for Students
Understanding normal distribution helps students to:
• Interpret data easily
• Understand averages and variation
• Learn probability concepts
• Prepare for exams and competitive tests
• Analyze real-life data logically
That’s why it is often called the “backbone of statistics.”
In Simple Words
• Normal distribution is a bell-shaped, symmetrical distribution
• Most values are near the average
• Mean = Median = Mode
• It follows the 68–95–99.7 rule
• Total area under the curve is 100%
• It is used everywhere in real life and research
Easy2Siksha.com
Final Thought
Once you understand normal distribution, statistics stops feeling scary.
It starts feeling logical, natural, and even interesting—because you realize it’s simply a
mathematical way of describing how the real world naturally behaves.
6. (a) Obtain the Mean and Standard Deviaon of a Binomial Distribuon for which
and
(b) If X is Poisson variate such that :
Find the Mean and Variance of the Distribuon.
Ans: Part (a) Binomial Distribution Problem
We are told:
•
•
Here, follows a Binomial Distribution with parameters and probability of success .
Step 1: Recall the Binomial Formula
The probability of getting exactly successes is:
Step 2: Write the Two Probabilities
For :
For :
Step 3: Use the Given Relation
We are told:
Easy2Siksha.com
Now, note that
. So they cancel out.
This leaves:
Divide both sides by
:
Take the fourth root:
So:
Step 4: Mean and Standard Deviation
For a binomial distribution:
• Mean =
• Variance =
• Standard Deviation =
Here:
• Mean =
• Variance =
• Standard Deviation =
So the mean is about 3.33 and the standard deviation is about 1.49.
Part (b) Poisson Distribution Problem
We are told:
•
• follows a Poisson distribution with mean .
Easy2Siksha.com
Step 1: Recall the Poisson Formula
Step 2: Write the Two Probabilities
For :
For :
Step 3: Use the Given Relation
We are told:
Cancel
:
Multiply both sides by 2:
So:
Thus, or .
But would mean the distribution is degenerate (no events occur). So the meaningful
solution is:
Step 4: Mean and Variance
For a Poisson distribution:
Easy2Siksha.com
• Mean =
• Variance =
So here:
• Mean = 2
• Variance = 2
Putting It All Together
• Binomial Case (n=10, p=1/3):
o Mean ≈ 3.33
o Standard Deviation ≈ 1.49
• Poisson Case (λ=2):
o Mean = 2
o Variance = 2
A Relatable Narrative
Think of these problems like predicting outcomes in games:
• In the binomial case, it’s like tossing a coin with a one-third chance of success, ten
times. On average, you’d expect about 3 successes, with some natural variation
(standard deviation about 1.5).
• In the Poisson case, it’s like counting rare events (say, phone calls arriving at a help
desk). If the average rate is 2 calls per unit time, then both the mean and variance
are 2. That means not only do you expect 2 calls, but the spread of possible
outcomes also centers around that same number.
Final Thoughts
Mathematical distributions like Binomial and Poisson are powerful because they help us
predict the average story of randomness. Even though individual outcomes may vary
wildly, these formulas give us the “center of gravity” of chance.
SECTION–D
7. (a) Write about complete enumeraon sample surveys.
(b) What are the features of a Good Sample?
Ans: Introduction: Why Do We Even Talk About Surveys?
Imagine a government wants to know how many people in a city are unemployed, or a
school wants to know how many students are satisfied with online classes. Asking every
Easy2Siksha.com
single person may be possible sometimes—but often it’s too costly, time-consuming, or
simply impractical.
This is where survey methods come in. Broadly, there are two approaches:
1. Complete Enumeration (Census)
2. Sample Survey
Understanding these two methods—and knowing what makes a good sample—is the
foundation of statistics and social research. Let’s explore them one by one.
(a) Complete Enumeration and Sample Surveys
1. Complete Enumeration Survey (Census Method)
What is Complete Enumeration?
A complete enumeration survey is a method in which information is collected from every
unit of the population. In simple words, no one is left out.
If the population has 100 people, data is collected from all 100.
If a village has 2,000 households, every household is surveyed.
That’s why this method is also called the Census Method.
A Simple Example
The best real-life example is the Population Census of India. Every ten years, the
government collects data from each and every household—about age, education,
occupation, gender, etc. No sampling is done; everyone is included.
Characteristics of Complete Enumeration
1. Covers the entire population
Every individual, household, or unit is studied.
2. Highly accurate
Since no unit is left out, there is no sampling error.
3. Detailed information
Researchers can collect in-depth and diverse data.
4. Time-consuming
Surveying everyone takes a lot of time.
Easy2Siksha.com
5. Very expensive
Requires more staff, resources, and money.
Advantages of Complete Enumeration
1. High accuracy and reliability
Because all units are studied, results are very close to reality.
2. No bias due to sampling
There is no risk of choosing a “wrong” sample.
3. Useful for official records
Governments use this method for population, agriculture, and housing data.
Disadvantages of Complete Enumeration
1. Costly method
It requires huge financial resources.
2. Not suitable for large populations
Studying millions of people is not always practical.
3. Time delay
By the time data is collected and analyzed, it may become outdated.
4. Not feasible for destructive tests
For example, testing all bulbs by breaking them is impossible.
When is Complete Enumeration Suitable?
• When the population is small
• When maximum accuracy is required
• When the government conducts national-level surveys
• When enough time and money are available
2. Sample Survey Method
What is a Sample Survey?
A sample survey is a method where only a part (sample) of the population is studied, and
the results are used to represent the whole population.
Instead of asking everyone, we ask a few carefully selected people.
Easy2Siksha.com
A Simple Example
Suppose a college has 1,000 students. To know students’ opinion about the canteen, the
management surveys 100 students selected from different classes and backgrounds. The
opinions of these 100 students are then used to understand the views of all 1,000 students.
Characteristics of Sample Survey
1. Only a portion of population is studied
2. Less time-consuming
3. Cost-effective
4. Results are quick
5. Accuracy depends on the quality of the sample
Advantages of Sample Survey
1. Economical
Requires less money and fewer resources.
2. Time-saving
Data can be collected and analyzed quickly.
3. Practical and flexible
Ideal for large populations.
4. More detailed study possible
Researchers can spend more time on each selected unit.
Disadvantages of Sample Survey
1. Risk of sampling error
If the sample is poorly chosen, results may be misleading.
2. Bias may occur
If the sample is not representative, conclusions can be wrong.
3. Requires expert planning
Sample selection needs statistical knowledge.
When is Sample Survey Suitable?
• When the population is very large
• When quick results are needed
• When resources are limited
• When complete enumeration is not feasible
Easy2Siksha.com
Difference Between Complete Enumeration and Sample Survey (in simple terms)
Complete Enumeration
Sample Survey
Studies entire population
Studies only a part
Very accurate
Less accurate but reliable
Costly
Economical
Time-consuming
Time-saving
Used for census
Used for research & market surveys
(b) Features of a Good Sample
Now comes a very important question:
If we are studying only a small part of the population, how can we trust the results?
The answer is simple: the sample must be good.
A good sample is one that truly represents the population. Let’s understand its features
clearly.
1. Representative in Nature
A good sample must reflect the characteristics of the population.
If the population includes men, women, young, old, rich, poor—then the sample should also
include all these groups in proper proportion.
A sample that represents everyone gives reliable results.
2. Adequate Size
The sample should be neither too small nor too large.
• A very small sample may give incorrect results.
• A very large sample may waste time and money.
A good sample has an adequate and reasonable size depending on the population.
3. Free from Bias
Bias means favoring one group over another.
Easy2Siksha.com
A good sample must be:
• Fair
• Neutral
• Selected without personal preference
For example, asking only toppers about exam difficulty will give biased results.
4. Random Selection
Every unit of the population should have an equal chance of being selected.
Random selection reduces favoritism and increases accuracy.
Example: Lottery method, random numbers, etc.
5. Homogeneity Within Groups
Units in the same group should be similar in nature.
This helps in better comparison and analysis.
6. Practical and Economical
A good sample should:
• Be easy to manage
• Fit within available time and budget
Statistics is not just about accuracy—it’s also about practical usefulness.
7. Clear Objectives
The sample must be selected according to the purpose of the study.
If the objective is clear, the sample will be more meaningful.
8. Accuracy and Reliability
Easy2Siksha.com
A good sample should provide results that are:
• Consistent
• Trustworthy
• Close to the actual population values
Conclusion
To sum up, complete enumeration and sample surveys are two important tools used to
collect data.
• Complete enumeration gives highly accurate results but is costly and time-
consuming.
• Sample surveys are economical and practical but require careful planning.
A good sample is the heart of a successful sample survey. If the sample is representative,
unbiased, and properly selected, even a small sample can give powerful and reliable
conclusions.
8. Explain the various methods of Sampling. Also discuss their relave merits and
demerits.
Ans: What is Sampling?
Imagine you’re asked to taste-test a huge pot of soup. You don’t drink the whole pot—you
take a spoonful. That spoonful, if chosen well, represents the flavor of the entire soup.
That’s exactly what sampling does in statistics: instead of studying an entire population
(which is often impossible), we study a smaller group (sample) that reflects the larger
population.
Methods of Sampling
1. Random Sampling
This is the simplest and most popular method. Every individual in the population has an
equal chance of being selected.
• Example: Drawing names out of a hat.
• Merits:
o Easy to understand and apply.
o Minimizes bias since everyone has equal chance.
• Demerits:
Easy2Siksha.com
o Requires a complete list of the population.
o May not always give a perfectly representative sample, especially if the
population is diverse.
2. Stratified Sampling
Here, the population is divided into groups (called strata) based on certain characteristics,
and samples are taken from each group.
• Example: In a school, students can be divided into strata based on grade levels, and
then samples are taken from each grade.
• Merits:
o Ensures representation of all important subgroups.
o More accurate than simple random sampling when population is diverse.
• Demerits:
o Requires detailed knowledge of the population.
o More complex to organize.
3. Systematic Sampling
This method selects every
individual from a list after choosing a random starting point.
• Example: If you have a list of 1,000 students and you want 100 samples, you pick
every 10th student.
• Merits:
o Simple and quick.
o Useful when population list is available.
• Demerits:
o If the list has hidden patterns, it may introduce bias.
o Less random compared to pure random sampling.
4. Cluster Sampling
Instead of sampling individuals, you sample entire groups (clusters).
• Example: If you want to study households in a city, you randomly select certain
neighborhoods (clusters) and study all households within them.
• Merits:
o Cost-effective and practical for large populations spread over wide areas.
o Easier to manage logistically.
• Demerits:
o Less accurate if clusters are not truly representative.
o Higher sampling error compared to stratified sampling.
5. Convenience Sampling
This is when samples are chosen based on ease of access.
Easy2Siksha.com
• Example: Asking your classmates about their study habits because they’re easy to
reach.
• Merits:
o Very easy and quick.
o Useful for preliminary research.
• Demerits:
o Highly biased.
o Results often unreliable and not generalizable.
6. Quota Sampling
Here, researchers ensure that the sample meets certain quotas for different groups.
• Example: Interviewing 50 men and 50 women to study opinions on a product.
• Merits:
o Ensures representation of specific groups.
o Useful when time and resources are limited.
• Demerits:
o Selection within quotas may still be biased.
o Not truly random.
7. Multistage Sampling
This combines several methods. For example, first selecting clusters, then randomly
sampling individuals within those clusters.
• Example: Choosing districts, then schools within districts, then students within
schools.
• Merits:
o Flexible and practical for very large populations.
o Reduces cost and effort.
• Demerits:
o Complex to design.
o Sampling error can accumulate at each stage.
Comparing the Methods
Method
Merits
Demerits
Random Sampling
Simple, unbiased
Needs full population list
Stratified Sampling
Ensures subgroup
representation
Complex, requires detailed
info
Systematic Sampling
Quick, easy
Risk of hidden bias
Cluster Sampling
Cost-effective, practical
Less accurate, higher error
Convenience
Sampling
Quick, easy
Highly biased, unreliable
Quota Sampling
Ensures quotas met
Not truly random, possible
bias
Easy2Siksha.com
Multistage Sampling
Flexible, practical for large
groups
Complex, error may
accumulate
Final Thoughts
Sampling is like choosing the right spoonful of soup—it determines whether your taste
reflects the whole pot. Each method has its strengths and weaknesses, and the choice
depends on the situation:
• If you want simplicity, go for random sampling.
• If diversity matters, stratified sampling is best.
• If resources are limited, cluster or convenience sampling may be practical.
The key is balance: choosing a method that gives reliable results while being practical to
carry out.
“This paper has been carefully prepared for educaonal purposes. If you noce any
mistakes or have suggesons, feel free to share your feedback.”
