SAMPLE SIZE & SAMPLING

TECHNIQUES
26-sep-2024

Dr Sadia Tabassum
Asst Prof
Operative Dentistry
 LEARNING
OBJECTIVES
• Estimating Sample Size
• Describe the importance and methods
of calculating sample size
• Calculate sample size based on
prescribed guidelines
• Justify sample size for their research
project
• Demonstrate the use of appropriate
software for sample size estimation
• Sampling Techniques
• Discuss various common sampling
techniques
• Justify the selection of sampling
technique for their project
WHY CALCULATE SAMPLE SIZE?
Study Validity:
• Study results will be valid, reliable, and generalizable if sample
size is properly calculated. Too small a sample can lead to results
that are not statistically significant, while an excessively large
sample may waste resources.

Avoids Type I and Type II Errors:

• An adequate sample size helps minimize the chances of making a
Type I error (false positive) or Type II error (false negative).
WHY CALCULATE SAMPLE SIZE?
Achieves Sufficient Statistical Power
• Statistical power, is the probability that the study will detect an
effect if there is one, usually set at 80%. Proper sample size
ensures the study has enough power to detect meaningful
differences or relationships.

Resource Optimization
• Correct sample size avoids wasting resources on an
unnecessarily large study and ensures sufficient data for
meaningful analysis.
WHY CALCULATE SAMPLE SIZE
Ethical Considerations
• In clinical research, using the correct sample size prevents
exposing too many participants to potential risks while ensuring
that the study produces valid results.

Generalizability of Results
• A sample that is too small may not represent the population
accurately, leading to biased or non-generalizable results.
WHY CALCULATE SAMPLE SIZE
Supports Publication Standards
• Many journals require a justification of the sample size calculation
during peer review.

Guides Study Design

• Sample size calculation can influence other aspects of study
design, including the selection of statistical tests, data collection
methods, and study timelines.
STEPS OF SAMPLE SIZE CALCULATION
Define the Research Objective
• Clearly identify the primary research question or hypothesis.

Determine the Study Design

• Identify whether your study is a clinical trial, observational study,
cohort study, case-control study, or cross-sectional survey.

Identify the Type of Data and Outcome

• Determine whether your outcome measure is continuous (e.g.,
blood pressure, height) or categorical (e.g., disease present or
absent).
STEPS OF SAMPLE SIZE CALCULATION
Set the Desired Confidence Level:
• 95% confidence level (corresponding to a Z-value of 1.96) is used.
Indicates how confident you want to be that the result should reflect
the true population parameters.
Determine the Acceptable Margin of Error (Precision)
• A margin of error of 5% is common, meaning you are willing to accept a
5% deviation from the true population value.
Estimate the Effect Size
• Effect size is the expected difference between groups or the strength of
the relationship. This could be a difference in means, proportions, or a
correlation. Larger effect sizes typically require smaller sample sizes,
while smaller effect sizes need larger samples to detect.
STEPS OF SAMPLE SIZE CALCULATION
Set the Power of the Study
• Power is usually set at 80% or 90%. It represents the probability of
detecting a true effect if it exists. Higher power increases the likelihood
of detecting an effect but requires a larger sample size.
Estimate the Population Variability (Standard Deviation)
• If you are measuring a continuous variable, you need an estimate of the
population’s standard deviation. This can be obtained from previous
studies, pilot studies, or relevant literature.
Account for Expected Dropout or Non-Response Rates
• Adjust your sample size to account for expected participant dropout or
non-response. For instance, if you anticipate a 10% dropout rate,
increase your calculated sample size accordingly.
Examples of common formulae
OPENEPI
What is Sampling?
• Sampling refers to selecting a
subset of individuals from a
population to estimate
characteristics of the whole
population.

• Population, Sample, Sampling

Frame

• Proper sampling ensures results

are representative and minimizes
bias
Why the need for sampling?
• Cost-Effective: Studying the entire population is often
impractical.

• Time-Efficient: Saves time compared to surveying everyone

• Precision: Allows for more accurate, focused research. Reduces

errors if done correctly.
SAMPLING TECHNIQUES
Simple Random Sampling (SRS)
• Equal chance for every individual to be selected.
• Pros: Reduces bias.
• Cons: Can be difficult to implement in large populations.

Stratified Sampling
Population divided into subgroups (strata), and random samples are
taken from each.
• Pros: Ensures representation of key groups.
• Cons: Requires knowledge of population strata.
Cluster Sampling
• Population divided into clusters, a few clusters are randomly selected,
and all individuals in those clusters are surveyed.
• Pros: Cost-effective for large populations.
• Cons: Less accurate if clusters are not homogeneous.
Systematic Sampling
Selecting every "n-th" individual from a list after randomly selecting a
starting point.
• Pros: Simple and fast.
• Cons: Can introduce bias if the list has a pattern.
Which sampling technique is this?
Non-Probability Sampling Techniques
Convenience Sampling
Selecting individuals who are easiest to reach.
• Pros: Easy to implement.
• Cons: High risk of bias, unrepresentative.

Purposive (Judgmental) Sampling

Selecting individuals based on specific characteristics or purpose.
• Pros: Good for specific, focused research.
• Cons: Potential for bias, not generalizable.
Non-Probability Sampling Techniques
Snowball Sampling
Participants recruit other participants.
• Pros: Useful for hard-to-reach populations.
• Cons: Likely to be biased and non-representative.
Quota Sampling
Ensures the sample meets a set quota of characteristics.
• Pros: Ensures inclusion of diverse groups.
• Cons: Not random, risk of bias.
How to Choose the Right Sampling Method
Research Objective:
Does the study aim to generalize findings or explore specific subgroups?
Population Characteristics:
Large, heterogeneous populations may require stratified or cluster
sampling.
Resources:
Budget and time available can determine the choice.
Potential Bias:
Choose methods that minimize bias and ensure representativeness.
Which of the following is a key characteristic of probability
sampling?
A) Every individual in the population has an equal chance of being
selected
B) Sampling is based on convenience
C) Participants are selected based on the researcher's judgment
D) The sample is gathered from people who volunteer
In which sampling method is the population divided into
subgroups, and random samples are taken from each
subgroup?
A) Simple Random Sampling
B) Stratified Sampling
C) Systematic Sampling
D) Cluster Sampling
What is the main disadvantage of convenience sampling?
A) It is too time-consuming
B) It may not represent the entire population
C) It is too expensive
D) It requires a detailed sampling frame
Which sampling technique involves selecting every "n-th"
individual from a list or population?
A) Cluster Sampling
B) Systematic Sampling
C) Stratified Sampling
D) Simple Random Sampling
Which of the following is a common drawback of non-
probability sampling techniques?
A) They are expensive to implement
B) They always require a complete sampling frame
C) They may introduce bias and limit the generalizability of the
results
D) They are more time-consuming than probability sampling
In which non-probability sampling method do participants help
recruit other participants, often used for hard-to-reach
populations?
A) Convenience Sampling
B) Snowball Sampling
C) Purposive Sampling
D) Quota Sampling
Which of the following factors is essential for determining an
appropriate sample size for a study?
A) The number of researchers involved in the study
B) The confidence level, margin of error, and population variability
C) The number of pages in the study report
D) The time of year the study is conducted