MODULE 2: DISTRIBUTIONS, SAMPLING TECHNIQUES, AND CENTRAL
LIMIT THEOREM
1. Normal Distribution
Definition
A normal distribution is a continuous probability distribution in which the
values of a variable are distributed symmetrically about a central value,
known as the mean. The distribution forms a bell-shaped curve where most
observations cluster around the mean, and fewer observations occur as we
move away from it.
Explanation
In many real-world situations, data tends to concentrate around an average
value. This happens because multiple small and independent factors
influence the outcome. When these factors combine, the resulting
distribution often becomes symmetric and bell-shaped.
For example, human height is influenced by genetics, nutrition, and
environment. Since these factors vary slightly for each individual, most
people fall near the average height, while very short and very tall individuals
are less common.
The central point of the distribution is called the mean (μ), and the spread of
the data is measured by the standard deviation (σ). A smaller standard
deviation indicates that the data points are closely packed around the mean,
while a larger standard deviation indicates greater variability.
Key Properties
The distribution is symmetric about the mean.
Mean, median, and mode are equal.
The total area under the curve represents total probability and is equal
to 1.
The distribution extends infinitely in both directions but never touches
the horizontal axis.
Empirical Rule
A useful property of the normal distribution is the empirical rule:
Approximately 68% of values lie within one standard deviation of the
mean.
Approximately 95% lie within two standard deviations.
Approximately 99.7% lie within three standard deviations.
This rule helps in quickly understanding how data is spread.
Example
Consider examination marks in a large class. Most students score around the
average marks, while only a few score extremely high or extremely low. This
creates a bell-shaped distribution.
2. Binomial Distribution
Definition
A binomial distribution is a discrete probability distribution that describes the
number of successes in a fixed number of independent trials, where each
trial has exactly two possible outcomes and a constant probability of
success.
Explanation
The binomial distribution applies to situations where an experiment is
repeated multiple times under identical conditions. Each trial results in either
success or failure.
For instance, when a coin is tossed, the outcome is either head or tail. If the
coin is tossed multiple times, we may be interested in counting how many
times a particular outcome (such as heads) occurs.
To apply the binomial distribution, four conditions must be satisfied:
1. The number of trials must be fixed in advance.
2. Each trial must have only two possible outcomes.
3. The probability of success must remain constant across trials.
4. The trials must be independent of each other.
Parameters
n: number of trials
p: probability of success in each trial
q = 1 − p: probability of failure
Interpretation
The binomial distribution answers questions of the form: “What is the
probability of obtaining a certain number of successes in a given number of
trials?”
Example
Suppose a coin is tossed 10 times. Each toss has a probability of 0.5 of
resulting in a head. The binomial distribution can be used to determine the
probability of getting exactly 6 heads out of the 10 tosses.
Applications
Quality control (defective or non-defective items)
Surveys (yes or no responses)
User behavior in digital systems (clicked or not clicked)
3. Poisson Distribution
Definition
A Poisson distribution is a discrete probability distribution that models the
number of events occurring within a fixed interval of time or space, given
that these events occur independently and at a constant average rate.
Explanation
In many situations, we are interested in counting how many times an event
occurs rather than whether it occurs or not. The Poisson distribution is
particularly useful when events happen randomly but with a known average
frequency.
For example, consider the number of customer arrivals at a service center in
one hour. While the exact number of arrivals may vary from hour to hour, we
can estimate an average number. The Poisson distribution helps determine
the probability of observing a specific number of arrivals.
Parameter
λ (lambda): the average number of events in the given interval
Key Properties
The mean of the distribution is equal to λ.
The variance is also equal to λ.
Events occur independently.
The probability of more than one event occurring in a very small
interval is negligible.
Example
If a website receives an average of 4 requests per second, the Poisson
distribution can be used to calculate the probability that exactly 6 requests
occur in a given second.
Applications
Network traffic analysis
Number of system failures
Customer arrivals in queues
Event counts in time intervals
4. Random Sampling
Definition
Random sampling is a method of selecting a subset of individuals from a
population such that each individual has an equal and independent chance
of being selected.
Explanation
In practical situations, it is often not feasible to study an entire population
due to constraints of time, cost, or resources. Instead, a sample is selected to
represent the population.
Random sampling ensures fairness and eliminates bias because every
member of the population has the same probability of inclusion. This
increases the likelihood that the sample accurately reflects the
characteristics of the entire population.
Example
If a class has 100 students and 10 students are selected by randomly
drawing names from a list, each student has an equal chance of being
chosen. This constitutes random sampling.
Importance
Produces unbiased results
Provides reliable representation of population
Forms the basis of statistical inference
5. Cluster Sampling
Definition
Cluster sampling is a sampling technique in which the population is divided
into groups, known as clusters, and a random selection of these clusters is
made. All elements within the selected clusters are included in the sample.
Explanation
In some cases, it is difficult or expensive to perform random sampling on
individuals spread across a large area. Cluster sampling provides a more
practical alternative by grouping the population into clusters and selecting
entire groups instead of individual elements.
Each cluster is intended to represent the population as a whole. Once
clusters are selected, every member within those clusters is studied.
Example
A city is divided into different regions. Instead of surveying individuals from
all regions, a few regions are randomly selected, and all individuals within
those selected regions are surveyed.
Advantages
Reduces cost and time
Easier to implement in large populations
Limitation
May introduce bias if clusters are not representative
6. Central Limit Theorem (CLT)
Definition
The Central Limit Theorem states that the distribution of sample means
tends to follow a normal distribution as the sample size becomes sufficiently
large, regardless of the original distribution of the population.
Explanation
The Central Limit Theorem is one of the most important results in statistics. It
explains why the normal distribution appears so frequently in practice.
Even if the original data is not normally distributed, if we repeatedly take
samples from that data and compute their means, the distribution of those
sample means will approximate a normal distribution, provided the sample
size is large enough.
Key Idea
Individual observations may vary widely.
However, averages of samples tend to stabilize and form a predictable
pattern.
Example
Consider a population with a highly irregular distribution. If we repeatedly
take samples of size 30 and calculate their averages, the distribution of
these averages will form a bell-shaped curve.
Importance
Forms the foundation of statistical inference
Enables estimation of population parameters
Used in hypothesis testing and confidence intervals
Widely applied in data science and analytics