CLUSTER

SAMPLING

STATISTICAL THEORY
BY: CLARO, KIM
TEOFILO, YHANNY
 Lesson Outline

Definition of Cluster Sampling

Usage of Cluster Sampling

Steps in Conducting Cluster Sample

Applications and Examples of

Cluster Sampling

Advantages and Disadvantages

I. Definition:
A Cluster Sampling is obtained by
dividing the population into sections or
clusters and then selecting one or more
clusters and using all members in the
cluster(s) as the members of the
sample.
 A cluster sampling is obtained by dividing

Cluster Sampling
the population into sections or clusters and
then selecting one or more clusters and
using all members in the cluster(s) as the
members of the sample.

II. Usage of Cluster Sampling

This method is particularly useful when it is
impractical or costly to sample every individual in a
population spread over a wide area. It simplifies data
collection but may
25% increase sampling error compared to
25%
simple random likely
sampling
Not
Not because the individuals within
likely
clusters tend to be more similar to each other.
Example: A health researcher
wants to study the
nutritional habits of
schoolchildren in
Tagum City
 III. Steps on using
Cluster Sampling

1. Define the population

2. Divide the population into clusters
3. Select clusters randomly
4. Collect data from each selected cluster
(i) One-stage cluster sampling, or
(ii) Two-stage cluster sampling
5. Analyze the data
6. Consider weighing (if necessary)
 III. Steps on using A health researcher wants to
study the nutritional habits of
Cluster Sampling schoolchildren in Tagum City

POPULATION
1. DEFINE the population
Since the researcher aims to
study the nutritional habits of the Schoolchildren
schoolchildren of Tagum City—we of Tagum City
shall define our population as
such.
 III. Steps on using
Cluster Sampling 1

2. DIVIDE the population into

clusters

Suppose we divide it into 4 clusters:

School 1
 III. Steps on using
Cluster Sampling

2. DIVIDE the population into 1 2

clusters

Suppose we divide it into 4 clusters:

School 1
School 2
 III. Steps on using
Cluster Sampling
1 3

2. DIVIDE the population into

clusters

Suppose we divide it into 4 clusters:

School 1 2
School 2
School 3
 III. Steps on using
Cluster Sampling 4
1
2. DIVIDE the population into
clusters 3
Suppose we divide it into 4 clusters:
School 1 2
School 2
School 3
School 4
 III. Steps on using
Cluster Sampling 4
1

3. SELECT clusters randomly

3
Let us choose
School 1 2
School 3
 III. Steps on using
Cluster Sampling
1
4. COLLECT data from each
selected cluster and use:

(i) One-stage cluster sampling, or

3
(ii) Two-stage cluster sampling
 III. Steps on using
1
Cluster Sampling

4. COLLECT data from each

selected cluster and use:
(ii) Two-stage cluster sampling 3
In two-stage cluster sampling, the process
involves two levels of sampling. First, a random
sample of clusters is selected, and then a
random sample of individuals is chosen within
each selected cluster.
 III. Steps on using
1
Cluster Sampling

4. COLLECT data from each

selected cluster and use:
(ii) Two-stage cluster sampling 3
In two-stage cluster sampling, the process
involves two levels of sampling. First, a random
sample of clusters is selected, and then a
random sample of individuals is chosen within
each selected cluster.
 1 3
III. Steps on using
Cluster Sampling

5. ANALYZE the data

Clearly present findings, highlight
and discuss whether the POPULATION
nutritional habits of the sampled
population are representative of
the broader population of Schoolchildren
of Tagum City
schoolchildren in the city.
 III. Steps on using 1
Cluster Sampling

6. Consider WEIGHTS 3
Suppose you selected 2 schools
(clusters) out of 4 total schools, the
weight for each school would be 35 selected out of 48
4/2 = 2. If one school had 48
students but you only sampled 35,
each student from that school
would initially have a weight of 2,
reflecting the cluster selection. 24 selected out of 36
III. Steps on using Cluster Sampling
III. Steps on using Cluster
Sampling
Applications:
Public Health Studies:
Useful in large populations,
such as assessing the
prevalence of diseases.
Applications:
Public Health Studies: Education Research:
Useful in large populations, Studying the effectiveness
such as assessing the of teaching methods
prevalence of diseases. across different schools.
Applications:
Public Health Studies: Education Research: Market Research: Analyzing
Useful in large populations, Studying the effectiveness consumer behavior by
surveying specific stores or
such as assessing the of teaching methods
regions rather than
prevalence of diseases. across different schools. individual customers.
ADVANTAGES of Cluster Sampling

Cost-Effective: Reduces travel and administrative

costs because data collection can occur in one
location rather than multiple sites.

Easier Logistics: Simplifies the process of data

collection, especially in geographically dispersed
populations.

Feasibility: Makes it easier to conduct studies where a

complete list of the population is hard to obtain.

Quick Implementation: Sampling from clusters can

often be completed more quickly than simple random
sampling.
DISADVANTAGES of Cluster Sampling

Higher Sampling Error: Clusters may not perfectly

represent the population, leading to increased
sampling error if clusters are not homogeneous.

Less Precision: The results may be less precise

compared to simple random sampling, particularly if
clusters differ significantly from each other.

Bias Risk: If the clusters themselves are not randomly

selected or are inherently biased, the findings may not
be generalizable.

Dependence on Cluster Size: If the sizes of the

clusters vary greatly, it can lead to unequal
representation of certain segments of the population.
 Summary
Some Applications
A cluster sampling is
Education Research Public Health Market Research
obtained by dividing the
population into sections or
clusters and then
selecting one or more
clusters and using all
members in the cluster(s) Advantages Disadvantages
as the members of the
Cost-Effective Higher Sampling Error
Easier Logistics Less Precision
sample. Feasibility Quick
Implementation
Bias Risk
Dependence on Cluster Size
 Summary
Some Applications
Steps of Cluster Sampling
Education Research Public Health Market Research
1. Define the population
2. Divide the population into clusters
3. Select clusters randomly
4. Collect data from each selected
cluster
(i) One-stage cluster sampling, or
(ii) Two-stage cluster sampling Advantages Disadvantages
5. Analyze the data Cost-Effective Higher Sampling Error
6. Consider weighing (if necessary) Easier Logistics Less Precision
Feasibility Quick Bias Risk
Implementation Dependence on Cluster Size