BIO 3211– Biometry and Biostatistics

Lecture 3 – Sampling Methods

-Aarif
Baksh

1
 Population and Sample
Population
 A complete collection of all entities (people,
plants, inanimate object) that are of interest
to a statistical study.

Sample
 A sub-collection of entities from a population,
intended to represent the population from
which it was drawn.

2
Big Picture – Observational Units
Example
Variable 1
of Interest: Height
Variable Type: Quantitative Cens
us

Heights
5.58, 5.28, 5.22, 5.35, 5.48, 6.0, 6.04, 5.15,
5.94, 5.05, 5.45, 5.48, 5.05, 6.1, 5.35, 5.77,
5.45, 5.94, 5.51, 5.54, 6.07, 5.51, 5.35

Average Height

Population,
Parameter 3
Big Picture –
Example
Variable 1
of Interest: Height
Observational Units
Variable Type: Quantitative Surve
y

Sampling Heights
5.48, 6.0, 5.05, 6.1

Sample,

Inference Average Height

𝝁
Population, Statistic 4
Big Picture – Observational Units
Example
Variable 2 Color
of Interest: Eye
Cens
Variable Type: Categorical
us

Eye Color
Gray, Blue, Brown, Green, Brown, Green,
Blue, Blue, Hazel, Gray, Green, Hazel, Gray,
Brown, Hazel, Blue, Hazel, Brown, Green,
Green, Hazel, Gray, Blue

Proportion of Blue Eyes

Population,
Parameter 5
Big Picture –
Example
Variable 2 Color
of Interest: Eye Observational Units
Variable Type: Categorical Surve
y

Sampling Eye Color

Blue, Brown, Green, Brown

Sample,

Proportion of Blue Eyes

𝝅 Inference

Population,
6
Statistic
 Types of Sampling
Random Sampling Non-Random Sampling

 Also known as probability sampling.  Also known as non-probability sampling.

 Every unit of the population has the  Every unit of the population does not

same probability of being included in have the same probability of being

the sample. included in the sample.

 Eliminates selection bias.  Open to selection bias.

 The aim is to make inference about  The aim is to gather specific insights

the population from which the sample about the sample (exploratory research).

was drawn.
7
 Selection Bias
Selection Bias – occurs when certain individuals/groups have a higher/lower
chance of being included in the sample.

1. Sampling Bias – The method of selecting participants systematically excludes some

groups.

2. Self-Selection Bias – Individuals choose whether to participate.

3. Survivorship Bias – Only successful cases are considered.

4. Attrition Bias – Participants drop out of a study in a non-random way.

5. Nonresponse Bias- Certain groups are less likely to respond.

8
 Selection Bias
Example: Survivorship
Bias

9
Q. Identify the type of selection bias in each scenario below:

1. A vaccine trial where participants with severe side effects drop out, leaving only those who

tolerate the vaccine well in the final analysis.

2. Recruiting participants for a fitness supplement study through a gym's newsletter, where

only highly motivated health enthusiasts are likely to volunteer.

3. A wildlife conservation survey about endangered species sent to local communities, but only

receiving responses from a small, select group of environmentally active residents.

4. A research team studies bird populations only in national parks, completely missing birds in

urban, agricultural, or private land ecosystems.

5. Analyzing the genetics of successful long-distance marathon runners without considering

those who stopped running due to injuries or health complications. 10

 Types of Sampling

Type of
Sampling

Random Non-Random

Simple Systematic Stratified Cluster Convenience Purposive Snowball Quota

11
 Types of Sampling
Rando
m

12
 Types of Sampling
Non-
Random

13
Q. Identify the sampling technique in each scenario shown below:
1. Researchers studying ant colonies in a savanna randomly select 5 areas of 1 km² each, then
study all ant colonies within those selected areas.
2. A marine biologist studying coral bleaching selects every 10th coral colony along a 2-kilometer
reef to assess damage.
3. To study a viral infection in prairie dogs, researchers first find one infected colony, then trace
and study all colonies that have had contact with the first group, then study all colonies that had
contact with those groups.
4. An ornithologist needs to quickly gather data about urban bird feeding habits and decides to
only observe birds at the three parks closest to their lab.
5. Entomologists studying insect diversity use a random number generator to select coordinates in
a meadow where they will place their insect traps.
6. Marine biologists studying shark behavior set targets to observe 20 great whites, 15 tiger
sharks, and 25 reef sharks before concluding their study.
7. Researchers studying coral bleaching divide the Great Barrier Reef into sections based on
14
distance from shore (near-shore, mid-reef, outer-reef), then randomly sample 15 coral colonies