Lecture 9: Sampling Methods and Sample Size
Estimation
Part 1: Definitions
Target Population: The population about which an investigator wishes to draw a conclusion.
Study Population: The population from which the sample is actually drawn and about which a
conclusion can be made.
Sample: A subset of the study population selected for investigation.
Study Unit: The unit on which observations are collected (e.g., persons or households).
Sampling Frame: The list of units from which the sample is selected.
Census Sample: A study in which data is gathered from every member of the population.
Part 2: Types of Sampling Methods
Probability Sampling Methods:
- Random Sampling
- Systematic Sampling
- Stratified Sampling
- Cluster Sampling
Non-Probability Sampling Methods:
- Convenience Sampling
- Judgment Sampling
- Quota Sampling
- Snowball Sampling
Part 3: Errors
Sampling Error (Random Error): Random deviation from the true value. It can be minimized by
increasing the sample size.
Non-Sampling Error (Bias): Systematic deviation from the true value in the same direction. It can be
reduced by careful design of the sampling procedure.
Part 4: Sample Size Formulas
1) Qualitative Variable (One Proportion):
n = (Z^2 * p * (1 - p)) / e^2
2) Single Population Mean:
n = (Z^2 * σ^2) / E^2
3) Comparison of Two Proportions:
Uses Zα/2 and Zβ with expected difference between proportions.
4) Comparison of Two Means:
Uses Zα/2, Zβ, population variance (σ^2), and expected difference (d).
Part 5: Calculations
Exercise 1:
Given p = 0.42, Z = 1.96 (95% confidence), e = 0.05
n = (1.96^2 * 0.42 * 0.58) / (0.05^2)
n ≈ (3.8416 * 0.2436) / 0.0025
n ≈ 0.937 / 0.0025
n ≈ 375
Required sample size ≈ 375 participants.
Exercise 2:
Given Z = 1.96 (95% confidence), σ = 20, E = 5
n = (1.96^2 * 20^2) / (5^2)
n = (3.8416 * 400) / 25
n = 1536.64 / 25
n ≈ 61.47
Required sample size ≈ 62 children.