Unit 3

🧠 Sample, Sampling Techniques, and Sample Size Determination

1. Meaning of Population and Sample

Population

 The population refers to the entire group of individuals, objects, or events that the
researcher is interested in studying.

 It can be finite (countable) or infinite (uncountable).

 Example:

o All college students in India (finite population).

o All possible human behaviors in stressful situations (infinite population).

Sample

 A sample is a subset or smaller group selected from the population to represent it in the
research.

 Since studying the whole population is often impractical (too large, expensive, or time-
consuming), researchers use samples to draw conclusions that can be generalized to the
population.

Definition (Kerlinger, 1986):

“A sample is a small proportion of a population selected for observation and analysis.”

Example

If your study focuses on “the relationship between academic stress and mental health among college
students,”

 Population: All college students in India.

 Sample: 200 students selected from 5 colleges in Delhi.

2. Sampling

Meaning

Sampling is the process of selecting a representative subset of the population to obtain information
and make inferences about the entire group.

Definition (Best & Kahn, 2006):

“Sampling is the process of selecting a small number of individuals from a larger population to serve
as subjects for observation or experimentation.”

3. Purpose of Sampling
 1. Practicality: Easier, faster, and more cost-effective than studying the entire population.

2. Accuracy: Proper sampling can provide highly reliable and valid results.

3. Efficiency: Saves time, money, and effort.

4. Feasibility: Some populations are too large or inaccessible to study fully.

5. Statistical Inference: Allows generalization of results to the larger population.

4. Sampling Techniques

Sampling methods are broadly classified into two categories:

🔹 A. Probability Sampling

In probability sampling, every individual has a known and non-zero chance of being selected.
It ensures representativeness and allows statistical generalization.

1. Simple Random Sampling

 Each individual has an equal chance of being selected.

 Selection is done through random methods (lottery, random number generator, etc.).

 Example: Randomly selecting 100 students’ names from a list of 1000 using a computer-
generated random number.

✅ Advantages: Unbiased and representative.

❌ Disadvantage: Requires a complete list of the population.

2. Systematic Sampling

 Every kth element is selected from a list after a random starting point.

 Formula:
[
k = \frac{\text{Population Size (N)}}{\text{Sample Size (n)}}
]

 Example: Selecting every 10th student from a list of 1000 to get a sample of 100.

✅ Simple to implement.
❌ Risk of periodicity bias if data has hidden patterns.

3. Stratified Sampling

 The population is divided into subgroups (strata) based on certain characteristics (e.g.,
gender, age, income), and random samples are drawn from each group proportionally.
  Example: Dividing students by gender (male/female) and selecting randomly from both
groups proportionally.

✅ Ensures representation of key subgroups.

❌ More complex to organize.

4. Cluster Sampling

 The population is divided into clusters (e.g., schools, cities), and a few clusters are randomly
selected for study.

 All individuals within selected clusters are studied.

 Example: Randomly selecting 3 schools in a city and surveying all students in those schools.

✅ Useful for large, geographically dispersed populations.

❌ Less precise than stratified sampling.

5. Multistage Sampling

 Involves selecting samples in multiple stages using combinations of methods.

 Example:

o Stage 1: Select districts randomly.

o Stage 2: Select schools from each district.

o Stage 3: Select students from each school.

✅ Cost-effective for large-scale research.

❌ Complexity increases with each stage.

🔸 B. Non-Probability Sampling

In this type, not every individual has a chance of being selected.

Used when randomization is difficult or impossible — often in exploratory or qualitative research.

1. Convenience Sampling

 Selecting participants who are easily available or willing to participate.

 Example: Using psychology students from your own college as participants.

✅ Quick and inexpensive.

❌ High risk of bias, low generalizability.

2. Purposive (Judgmental) Sampling

 Researcher selects participants based on specific characteristics or criteria relevant to the

study.
  Example: Selecting only students with high social media use for a study on internet
addiction.

✅ Useful for studying specific groups.

❌ Subjective and potentially biased.

3. Quota Sampling

 Researcher divides the population into subgroups and fills quotas for each (similar to
stratified sampling but non-random).

 Example: Ensuring 50% males and 50% females in a survey without random selection.

✅ Ensures diversity.
❌ Lacks randomness.

4. Snowball Sampling

 Existing participants recruit future participants from among their acquaintances.

 Example: Used for hard-to-reach populations like drug users or survivors of abuse.

✅ Effective for hidden populations.

❌ High bias and lack of representativeness.

Diagram: Sampling Techniques Overview

SAMPLING

┌────────────────┴────────────────┐

│ │

Probability Sampling Non-Probability Sampling

│ │

Simple Random Convenience

Systematic Purposive

Stratified Quota

Cluster Snowball

Multistage

5. Determining Sample Size

Selecting the correct sample size is critical for accuracy, reliability, and generalization of research
results.

Factors Affecting Sample Size

1. Population Size (N): Larger populations generally require larger samples.

2. Margin of Error: Smaller error requires larger sample.

3. Confidence Level: Usually 95% or 99%; higher confidence → larger sample.

4. Variability in Population: More variation → larger sample needed.

5. Purpose of Study: Experimental studies may use smaller samples; surveys need larger ones.

6. Resources Available: Time, cost, and accessibility.

Sample Size Formula (for large populations)

[
n = \frac{{Z^2 \cdot p(1-p)}}{{e^2}}
]

Where:

 ( n ) = Sample size

 ( Z ) = Z-value (1.96 for 95% confidence)

 ( p ) = Estimated proportion of population (0.5 for maximum variability)

 ( e ) = Margin of error (e.g., 0.05)

Example:

If a researcher wants to study attitudes toward online learning among college students (with 95%
confidence and 5% margin of error):

[
n = \frac{{1.96^2 \times 0.5(1-0.5)}}{{0.05^2}} = 384.16
]

✅ So, the researcher should select around 384 participants for reliable results.

6. Examples by Type

Sampling
Research Study Example
Technique

Divide by department (surgery, pediatrics, etc.)

Studying stress among doctors Stratified Sampling
and sample proportionally

Studying college students’ Simple Random Randomly pick 100 names from college registry
 Sampling
Research Study Example
Technique

views on AI Sampling

Studying drug abuse in hidden Participants refer others they know who use
Snowball Sampling
populations drugs

Studying online behavior of

Purposive Sampling Select teens with high social media use
teenagers

Large-scale survey across Multistage

Select states → districts → schools → students
states Sampling

7. Importance of Sampling in Research

1. Ensures Representativeness – Reflects the diversity of the population.

2. Improves Accuracy – Provides reliable and valid results if done properly.

3. Saves Time and Money – Avoids studying the entire population.

4. Facilitates Data Collection – Makes research manageable.

5. Allows Statistical Inference – Enables generalization to larger populations.

8. Summary

Concept Explanation Example

Population Entire group of interest All university students

Sample Subset of population 200 selected students

Sampling Technique Method of selecting participants Random or purposive

Sample Size Number of participants e.g., 384 (based on formula)

✅ In Summary

Sampling is the bridge between theory and data collection.

A well-chosen sample ensures the validity, reliability, and generalizability of research findings.
The right sampling technique depends on the research purpose, population characteristics, and
available resources.

🧠 1. Introduction
Sampling is the process of selecting a subset (sample) from a larger population to study and make
generalizations.
It ensures that researchers can collect meaningful data efficiently while maintaining accuracy and
validity.

There are two main categories of sampling methods:

 Probability Sampling – based on random selection

 Non-Probability Sampling – based on researcher judgment or convenience

🔹 2. Probability Sampling Methods

In probability sampling, every individual in the population has a known, non-zero chance of being
selected.
This ensures that the sample is representative, allowing researchers to generalize results statistically.

2.1 Simple Random Sampling

 Each member of the population has an equal chance of selection.

 Selection tools: Lottery method, random number table, computer randomizer.

 Example: Selecting 100 students randomly from a list of 1,000 university students.

 Advantages: Minimizes bias; highly representative.

 Disadvantages: Requires complete population list.

2.2 Systematic Sampling

 Selecting every kth individual from a population list after a random start.

 Formula:
[
k = \frac{N}{n}
]
where (N) = population size and (n) = sample size.

 Example: Selecting every 10th patient from hospital records.

 Advantages: Easy to implement; quick.

 Disadvantages: May introduce bias if data has a periodic pattern.

2.3 Stratified Sampling

 Population divided into strata (subgroups) based on key characteristics (e.g., gender, age,
income).

 Random samples drawn from each stratum, either proportionally or equally.

  Example: Selecting 40% males and 60% females from a population where those are actual
proportions.

 Advantages: Ensures representation of subgroups.

 Disadvantages: Requires detailed population data.

2.4 Cluster Sampling

 Population divided into clusters (e.g., schools, hospitals, villages).

 A few clusters are randomly chosen, and all or random individuals within selected clusters
are studied.

 Example: Selecting 5 schools randomly and surveying all students in those schools.

 Advantages: Cost-effective for large areas.

 Disadvantages: Less accurate than stratified sampling due to cluster homogeneity.

2.5 Multistage Sampling

 Combines several sampling methods in stages.

 Example: Select districts → randomly choose schools → randomly select students.

 Advantages: Feasible for large-scale research.

 Disadvantages: Increases complexity and sampling error.

🔸 3. Non-Probability Sampling Methods

In non-probability sampling, the probability of selection for each individual is unknown.

Used when randomization is difficult, costly, or when exploring new or hard-to-reach populations.

3.1 Convenience Sampling

 Selecting participants who are easily available or willing.

 Example: Distributing surveys to students in your own classroom.

 Advantages: Quick, inexpensive.

 Disadvantages: Biased; cannot generalize results.

3.2 Purposive (Judgmental) Sampling

 Researcher intentionally selects participants based on characteristics relevant to the study.

 Example: Choosing only clinical psychologists for a study on therapy effectiveness.

  Advantages: Focused and purposeful.

 Disadvantages: Subjective and prone to bias.

3.3 Quota Sampling

 Population divided into categories, and researcher fills predetermined quotas (similar to
stratified but non-random).

 Example: Surveying 50 males and 50 females without randomization.

 Advantages: Ensures representation of groups.

 Disadvantages: May lack randomness.

3.4 Snowball Sampling

 Existing participants recruit others from their network.

 Useful for hidden or sensitive populations.

 Example: Research on substance users or victims of domestic violence.

 Advantages: Effective for hard-to-reach groups.

 Disadvantages: Non-representative and biased toward social networks.

3.5 Theoretical Sampling (Used in Qualitative Studies)

 In grounded theory research, participants are selected based on emerging concepts during
data collection.

 Example: Interviewing participants who can elaborate on newly identified themes.

 Advantage: Helps develop robust theories.

 Disadvantage: Requires flexible planning.

⚖️Comparison Table: Probability vs Non-Probability Sampling

Feature Probability Sampling Non-Probability Sampling

Selection Basis Random Non-random / researcher’s judgment

Representativeness High Low

Generalizability Possible Limited

Examples Simple Random, Stratified Convenience, Purposive

Use Quantitative research Qualitative / exploratory research

🔍 4. Determining Sample Size

The sample size refers to the number of participants or units included in a study.
It determines the accuracy, reliability, and generalizability of results.

4.1 Factors Influencing Sample Size

1. Population Size (N): Larger populations may need larger samples.

2. Margin of Error (e): Smaller error → larger sample.

3. Confidence Level (Z): 95% or 99% confidence increases sample size.

4. Variability (p): Greater diversity in population → larger sample.

5. Research Purpose: Exploratory studies need smaller samples; surveys and experiments need
larger ones.

6. Available Resources: Budget, time, and accessibility.

4.2 Formula for Large Populations

[
n = \frac{{Z^2 \times p (1 - p)}}{{e^2}}
]

Where:

 ( n ) = Sample size

 ( Z ) = Z-value (1.96 for 95% confidence level)

 ( p ) = Estimated proportion (0.5 if unknown)

 ( e ) = Margin of error (e.g., 0.05)

Example

For a population with unknown variability, 95% confidence, and 5% margin of error:

[
n = \frac{{1.96^2 \times 0.5(1-0.5)}}{{0.05^2}} = 384.16
]

✅ Recommended sample size = 384 participants.

4.3 For Finite Populations

[
n_f = \frac{n}{1 + \frac{(n - 1)}{N}}
]
where ( N ) = population size.
This corrects for smaller populations.

⚙️5. Power Analysis

5.1 Meaning

Power analysis is a statistical technique used to determine the minimum sample size required to
detect an effect of a given size with a specific level of confidence.

It ensures that your study is neither underpowered (too few participants) nor overpowered
(unnecessarily large sample).

Definition:
Statistical power is the probability that a test correctly rejects a false null hypothesis (i.e., detects a
real effect).

5.2 Components of Power Analysis

Parameter Meaning Typical Values / Notes

Magnitude of difference or relationship Small = 0.2, Medium = 0.5, Large =

Effect Size (d)
expected 0.8

Significance Level Probability of Type I error (rejecting true

Usually 0.05
(α) null)

Power (1 – β) Probability of detecting a true effect Usually 0.80 (80%)

Sample Size (n) Number of participants required Calculated via power analysis

5.3 Purpose of Power Analysis

1. To determine minimum sample size needed to detect expected effects.

2. To prevent waste of resources by avoiding unnecessarily large samples.

3. To ensure ethical standards by minimizing participant burden.

4. To increase reliability of statistical conclusions.

5.4 Example of Power Analysis

A researcher expects a medium effect size (d = 0.5), uses α = 0.05, and wants 80% power for a t-test
comparison.

Using power tables or software (e.g., G*Power):

 Required sample size ≈ 64 participants total (32 per group).

✅ This ensures that if a real difference exists, there’s an 80% chance of detecting it.

5.5 Tools for Power Analysis

 G*Power (free software for all test types)

 SPSS SamplePower

 R (pwr package)

 Python (statsmodels module)

💡 6. Summary Table

Concept Definition Example / Note

Random selection; known chance of

Probability Sampling Simple random, stratified
inclusion

Non-random; based on accessibility or

Non-Probability Sampling Convenience, purposive
judgment

Sample Size
Deciding number of participants needed Formula or power analysis
Determination

Statistical method to ensure adequate Uses effect size, α, and

Power Analysis
sample size power

✅ 7. In Summary

 Probability sampling provides representative and generalizable data — ideal for quantitative
studies.

 Non-probability sampling is practical for qualitative or exploratory studies.

 Sample size determination ensures valid results with acceptable precision.

 Power analysis ensures your sample is statistically sufficient to detect meaningful effects.

In short:
Proper sampling and power planning make the difference between a study that discovers truth —
and one that misses it.

🧩 1. Population – Definition and Explanation

Meaning:

In research, a population refers to the entire group of individuals, objects, or events that the
researcher is interested in studying.
It is the complete set from which a sample is drawn and about which the researcher wants to make
generalizations.

Definition (Kerlinger, 1986):

“A population is the total set of individuals or objects having common observable characteristics in
which the researcher is interested.”

Simple Definition:
The population is the whole group about which the researcher wants to draw conclusions.

Example:

 If you are studying the effects of stress on academic performance among college students,

o Population: All college students in India (or your chosen region).

 If you are studying anxiety levels among nurses,

o Population: All registered nurses working in hospitals.

Types of Population:

1. Target Population:
The entire group that the researcher wants to study.
Example: All college students in India.

2. Accessible (Study) Population:

The portion of the target population that is available and accessible to the researcher.
Example: College students in Delhi who can be reached for the study.

🧠 2. Sample – Definition and Explanation

Meaning:

A sample is a subset or a smaller group selected from the population for actual data collection.
The sample should represent the population as accurately as possible so that results can be
generalized.

Definition (Best & Kahn, 2006):

“A sample is a small proportion of a population selected for observation and analysis.”

Simple Definition:
A sample is a small group taken from the population to participate in the study.

Example:

 From the population of 10,000 college students, the researcher may select 300 students to
complete a stress questionnaire.

o Here, the 300 students form the sample.

🧩 3. Relationship between Population and Sample

Concept Explanation Example

Population Entire group you want to study All college students

Sample Small part of the population selected for study 300 selected students

Population → Sample → Data Collection → Inference about Population