TYPES OF SAMPLING METHODS

Probability Sampling
In probability sampling, every member of the
population has a known, non-zero chance of being
selected in the sample. This type of sampling is
essential for statistical inference because it reduces
bias and allows for the calculation of sampling
error.
Types of probability sampling include:
1. Simple Random Sampling:
• Every individual in the population has an equal
chance of being selected. This is often done
using random number generators or lottery
systems.

• Example: Drawing names from a hat or using a

random number table to select students from a
school.
Types of probability sampling include:
2. Systematic Sampling:
• Selecting individuals systematically from a
population. You start with a randomly chosen
individual and then select every individual.

• Example: If you want to survey every 10th

customer entering a store, you start with a
random number between 1 and 10 and then
select every 10th customer thereafter.
Types of probability sampling include:
3. Stratified Sampling:
• The population is divided into subgroups (strata)
based on certain characteristics (e.g., age,
gender), and then a random sample is taken
from each stratum. This ensures that the sample
accurately represents different segments of the
population.

• Example: A survey could divide a population by

age group and randomly select individuals from
each age group.
Types of probability sampling include:
4. Cluster Sampling:
• The population is divided into clusters (usually
geographic areas or institutions), and a random
sample of clusters is selected. Then, all
individuals in the chosen clusters are surveyed.

• Example: A researcher might divide a city into

neighborhoods (clusters) and randomly select a
few neighborhoods, surveying every resident in
those areas.
Non-Probability Sampling
In non-probability sampling, not every
individual in the population has a chance of being
selected. This type of sampling is often used in
qualitative research or exploratory studies where
generalizability is less of a concern.
Types of non-probability sampling include:
1. Convenience Sampling:
• Participants are selected based on their
availability or convenience to the researcher.

• Example: A professor surveys their own

students for a research study simply because
they are easily accessible.
Types of non-probability sampling include:
2. Quota Sampling:
• Researchers divide the population into specific
categories (quotas) and then use convenience
sampling to fill those quotas.

• Example: A researcher may want to survey 100

people, ensuring that 50 are women and 50 are
men. They stop surveying when each quota is
met.
Types of non-probability sampling include:
3. Purposive (Judgmental) Sampling:
• Participants are chosen based on the
researcher’s judgment of who will provide the
best information for the study.

• Example: Interviewing only experienced workers

to understand workplace challenges.
Types of non-probability sampling include:
4. Snowball Sampling:
• Participants recruit other participants for the
study. This is often used in research involving
hard-to-reach populations.

• Example: A researcher studying underground art

movements might ask one artist to refer them to
others in the scene.
DETERMINING SAMPLE SIZE
FOR DIFFERENT RESEARCH
DESIGNS
A. Experimental Research Design
Experimental research involves manipulating
variables to test causal relationships. Researchers
typically compare outcomes between two or more
groups (e.g., control and experimental groups).
A. Experimental Research Design
Sample Size for Experimental Designs
• When determining the sample size for an experimental design, the
goal is often to detect a significant difference between groups,
and factors such as effect size and power are important.

•Key Factors:
• Effect Size - The magnitude of the difference you expect between
the groups. Smaller expected differences require larger samples to
detect statistically significant differences.
• Power - The probability of correctly rejecting the null hypothesis
when a true effect exists (commonly set at 80% or 90%).
• Type I and Type II Error - The risk of rejecting a true null
hypothesis (Type I error) and the risk of failing to reject a false null
hypothesis (Type II error). Type I error is commonly set at .
A. Experimental Research Design
Common Experimental Designs:
1.Randomized Controlled Trial (RCT):
1.Participants are randomly assigned to either a treatment or
control group. Sample size is calculated based on the number of
participants needed to detect a treatment effect.
2.Example: Testing the effect of a new medication on a sample of
patients.

2.Quasi-Experimental Design:
1.Participants are not randomly assigned to groups, but the
researcher still compares outcomes. Larger sample sizes may be
needed to account for potential confounding variables.
2.Example: Evaluating the impact of an educational program in two
different schools where random assignment is not feasible.
B. Descriptive Research Design
Descriptive research aims to describe
characteristics or phenomena without making
causal inferences. These designs often involve
surveys or observational studies.
B. Descriptive Research Design
Sample Size for Descriptive Designs
• In descriptive research, the sample size is often determined by the
desired precision of estimates, such as population proportions or
means.

•Key Factors:
• Margin of Error - The degree of precision desired (e.g., ±3%).
Smaller margins of error require larger samples.
• Confidence Level - The certainty you want in your estimate (e.g.,
95% confidence level).
• Population Size - If the population is finite, the sample size is
adjusted accordingly.
B. Descriptive Research Design
Common Descriptive Designs:
1.Cross-Sectional Surveys:
1.Involve collecting data at a single point in time to estimate
population characteristics, such as proportions or averages.
Sample size is often calculated using formulas for estimating
proportions.
2.Example: A survey measuring voter preferences just before an
election.

2.Longitudinal Studies:
1.Data is collected from the same participants at multiple time
points. Sample size calculations often consider attrition (dropout)
rates, which means starting with a larger sample to ensure enough
participants remain over time.
2.Example: A study tracking changes in health behaviors over
C. Correlational Research Design
Correlational research examines the
relationship between two or more variables
without manipulating them. The sample size in
correlational studies is critical for detecting
meaningful correlations.
C. Correlational Research Design
Sample Size for Correlational Designs
• Larger sample sizes are usually required to detect weak
correlations, whereas smaller samples can detect stronger
correlations.

•Key Factors:
• Effect Size - The expected strength of the correlation (e.g., small,
medium, or large).
• Power - To ensure you have enough participants to detect a true
correlation if it exists.
• Type I and Type II Error - Similar to experimental designs,
sample size is influenced by the acceptable levels of these errors.
C. Correlational Research Design
Common Correlational Designs:
1.Cross-Sectional Correlational Studies:
1.Data is collected at a single point in time to examine the
relationship between variables. For instance, the sample size might
be calculated to detect whether income level is related to life
satisfaction.
2.Example: Investigating the correlation between physical activity
and mental health in a large population.

2.Regression Analysis:
1.Used to assess how one or more predictor variables relate to an
outcome variable. In multiple regression, sample size depends on
the number of predictors and the strength of their relationships
with the outcome.
2.Example: Analyzing how education level, age, and income
D. Qualitative Research Design
Qualitative research focuses on understanding
experiences, perceptions, and meanings through
non-numerical data, such as interviews or focus
groups. Sample size determination in qualitative
research is more flexible and often based on data
saturation — the point at which no new themes or
insights emerge.
D. Qualitative Research Design
Sample Size for Qualitative Designs
•In qualitative studies, sample sizes are typically smaller than in
quantitative research because the goal is in-depth exploration, not
statistical generalization.

•Key Factors:
• Data Saturation - The point at which additional data does not
add new insights. Sample sizes are usually determined as the
research progresses.
• Study Scope - Broader research questions may require larger
samples to capture the diversity of experiences or perspectives.
D. Qualitative Research Design
Common Qualitative Designs:
1.Phenomenology:
1.Focuses on the lived experiences of individuals. Sample sizes are
often small (e.g., 10-20 participants) to allow for deep exploration.
2.Example: Interviewing individuals about their experience of
recovering from a natural disaster.

2.Grounded Theory:
1.Aims to develop a theory based on data collected during the
study. Sample size is determined by data saturation, and it may
evolve as the theory is constructed.
2.Example: Developing a theory about how people cope with
chronic illness.
E. Mixed-Methods Research Design
Mixed-methods research combines both
quantitative and qualitative approaches in a
single study. The sample size determination will
depend on the specific quantitative and
qualitative components used.
E. Mixed-Methods Research Design
Sample Size for Mixed-Methods Designs
•The quantitative part often follows traditional sample size formulas,
while the qualitative part uses smaller, more flexible sample sizes.

•Key Factors:
• Balance - Ensuring the quantitative sample size is sufficient for
statistical analysis, while the qualitative sample allows for rich, in-
depth insights.
• Research Questions - The focus of the study determines how the
qualitative and quantitative samples are balanced.
E. Mixed-Methods Research Design
Common Mixed-Methods Designs:
1.Explanatory Sequential Design:
1.Quantitative data is collected first, followed by qualitative data to
explain or explore the quantitative results.
2.Example: A survey about school performance followed by
interviews with students to explore their experiences.

2.Exploratory Sequential Design:

1.Qualitative data is collected first to explore a phenomenon,
followed by quantitative data to test or confirm findings.
2.Example: Interviews to identify key factors influencing employee
motivation, followed by a large-scale survey to measure these
factors.
THE CONCEPT OF SAMPLING
ERROR
The Concept of Sampling Error
Sampling error is the difference between the
characteristics of the sample and those of the
entire population. This error occurs because we are
only observing a subset of the population, not the
whole population.

• Sampling error can be reduced by increasing

the sample size. As the sample size increases,
the sample becomes more representative of the
population, and the variability decreases.
The Concept of Sampling Error
Key Points about Sampling Error:
1.Random Sampling Error:
1.Random fluctuations due to the fact that we are only using a
sample. These errors are inherent in any sampling process but can
be minimized by using larger, more representative samples.

2.Bias:
1.Systematic errors that occur when the sampling method is flawed
(e.g., convenience sampling might lead to a biased sample that
doesn’t represent the population). Bias is different from random
error because it cannot be reduced by increasing sample size
alone.
The Concept of Sampling Error
Reducing Sampling Error:
• Increase sample size
- Larger samples reduce variability and improve precision.

• Use appropriate sampling methods

- Random and stratified sampling help reduce bias and ensure a
representative sample.

• Ensure proper sampling technique

- Avoid using non-probability methods for statistical inference
unless necessary.