Cluster Sampling
Cluster Sampling
A sample of 2,000 was sought to estimate the average achievement in science of fifth graders in a
city’s public schools. The average fifth grade enrollment in the city’s elementary schools is 100 students.
Thus, 20 schools were randomly selected and within each of those schools all fifth graders were tested.
- Explanation:
Cluster sampling involves dividing the population into clusters (in this case, schools) and then randomly
selecting some of these clusters for inclusion in the sample. Within each selected cluster, all individuals
(or a subset) are included in the sample.
In the given example, the population consists of fifth graders in the city's public schools. The schools
serve as the clusters, and 20 schools were randomly selected. Within each of these 20 schools, all fifth
graders were tested. Therefore, each cluster (school) is representative of the population, and all
individuals within the selected clusters contribute to the sample.
2. A researcher is interested in maximum-security inmates. She groups inmates by the type of crime
committed, determines the percentage of the total in each crime category, and uses that percentage to
determine the number of inmates randomly selected from each group.
Explanation:
Stratified random sampling involves dividing the population into homogeneous subgroups, or strata,
based on certain characteristics (in this case, the type of crime committed). Then, a random sample is
selected independently from each stratum.
In the given example, the researcher groups inmates by the type of crime committed, creating strata. The
percentage of total inmates in each crime category is determined, and inmates are randomly selected
from each group based on these percentages. This ensures that each subgroup (stratum) is represented
proportionally in the sample, allowing for more accurate analysis and inference about the entire
population of maximum-security inmates.
3. A researcher has a population of 100 third grade children from a local school district from which a
sample of 25 children is to be selected. Each child’s name is put on a list, and each child is assigned a
number from 1 to 100. Then the numbers 1 to 100 are written on separate pieces of paper and shuffled.
Finally, the researcher picks 25 slips of paper and the numbers on the paper determine the 25
participants.
Explanation:
Simple random sampling involves randomly selecting individuals from the population such that each
individual has an equal chance of being chosen, and each combination of individuals has an equal
chance of being selected as the sample.
In this example, each child in the population of 100 third-grade children has an equal chance of being
selected because their names are all on the list and assigned numbers. By shuffling the numbered slips of
paper and then randomly selecting 25 of them, the researcher ensures that every child has an equal
probability of being chosen for the sample. This method eliminates bias and allows for a representative
sample to be selected from the population.
4. A sociologist conducts an opinion survey in a major city. Part of the research plan calls for describing
and comparing the opinions of four different ethnic groups: African Americans, Asian Americans,
European Americans, and Native Americans. For a total sample of 300, the researcher selects 75
participants from each of the four predetermined subgroups.
Explanation:
Stratified sampling involves dividing the population into homogeneous subgroups, or strata, based on
certain characteristics relevant to the research question. In this case, the researcher divides the
population into four ethnic groups: African Americans, Asian Americans, European Americans, and Native
Americans.
Once the population is stratified, the researcher then selects a sample from each stratum. In this
example, the researcher selects 75 participants from each of the four ethnic groups. By ensuring
representation from each subgroup, the researcher can describe and compare the opinions of each ethnic
group accurately. This method allows for more precise analysis and comparison of opinions across
different demographic groups.
5. A researcher is interested in political attitudes of individuals within a particular state. The researcher
uses the following procedure to obtain a sample: (a) a random sample of counties is drawn, (b) within the
counties selected, districts are randomly drawn, (c) within each district, blocks are randomly sampled, (d)
within blocks selected, building are randomly drawn, and (d) all residents within the buildings selected
comprise the sample.
Explanation:
Multi-stage cluster sampling involves a series of stages in which clusters of units are randomly selected at
each stage until the final sample is obtained.
In this example:
Within each selected county, districts are randomly drawn as the second stage.
Within each district, blocks are randomly sampled as the third stage.
Within the selected blocks, buildings are randomly drawn as the fourth stage.
Finally, all residents within the selected buildings comprise the sample.
This method allows for a representative sample of individuals within the state by incorporating random
selection at multiple levels (counties, districts, blocks, and buildings). It is particularly useful when a
complete list of individuals in the population is not available, as is often the case in large-scale surveys.
6. Instructors teaching research methods are interested in knowing what study techniques their students
are utilizing. Rather than assessing all students, the researchers randomly select 10 students from each
of the sections to comprise their sample.
Explanation:
Stratified random sampling involves dividing the population into homogeneous subgroups, or strata,
based on certain characteristics relevant to the research question. In this case, the population consists of
students enrolled in different sections of the research methods course.
The researchers select a sample from each stratum by randomly choosing 10 students from each section.
By doing so, they ensure representation from each section, allowing them to understand the study
techniques utilized by students across different instructional groups. This method enables the researchers
to draw conclusions that are more reflective of the entire student population taking the research methods
course.
Explanation:
Multi-stage cluster sampling involves a series of stages in which clusters of units are randomly selected at
each stage until the final sample is obtained.
In this example:
Maximum-security prisons are grouped by state, serving as the first stage clusters.
From the grouped states, 10 are randomly selected, constituting the second stage.
Within the selected states, three maximum-security prisons are randomly chosen, forming the third stage
clusters.
Finally, all the inmates within the selected prisons comprise the sample.
This method allows for a representative sample of maximum-security inmates across different states by
incorporating random selection at multiple levels (states and prisons). It simplifies the sampling process
and is particularly useful when a complete list of inmates in all maximum-security prisons is not available.
8. Say you wanted to survey professors at KSU about their publication records. You want to be sure to
include all professors from each department, even the very small ones.
Explanation:
Stratified Sampling: This method involves dividing the population (professors at KSU) into homogeneous
subgroups, or strata, based on certain characteristics relevant to the research question (in this case,
departments).
Including all departments: By using stratified sampling, you ensure that all departments, including the very
small ones, are represented in the sample. This approach allows for a more accurate representation of
the entire population of professors at KSU, as it accounts for the variability across departments.
Random selection within each department: Within each department stratum, a random sample of
professors can be selected to participate in the survey. This ensures that each professor has an equal
chance of being selected, regardless of the size of their department.
By employing stratified sampling, you can obtain a sample that reflects the diversity of departments at
KSU while still maintaining randomness within each department, thereby improving the generalizability of
your findings to the entire population of professors at the university.