Sampling Techniques: Population and Types

• Sampling techniques are the methods used to select a subset of a population for study.
• Dr. Wilson Cordova, LPT, is associated with this presentation.

Population

• Population refers to a complete set of measurements or data that a researcher is interested in.
• It includes all responses, measurements, or counts that are relevant to the study.

Sample

• A sample is a subset of the population.

Purpose of Sampling

• Sampling is necessary because studying an entire population is often impossible.
• Sampling is done for manageability of data.
• Sampling is done for economic reasons.
• Sampling helps save time and effort.

Types of Sampling

• Probability sampling gives each member of the population an equal chance of being included in the sample.
• Non-probability sampling means that not all members of the population have an equal chance of being included.

Comparison of Probability and Non-Probability Sampling

Probability Sampling

• A complete sampling frame is required.
• Enables the selection of a random sample from the population.
• Allows generalization of results from the random sample to the population
• Can be more expensive and time-consuming

Non-Probability Sampling

• Useful when an exhaustive population list isn't available.
• Selection is not random.
• Effective for generating ideas and gathering feedback.
• More convenient and less costly.

Biased vs. Unbiased Sampling

Example 1: Zoo Visitors and Pet Ownership

• Antonio surveys 100 zoo visitors and finds that 90% own pets, concluding that 90% of all adults own pets.
• The flaw is that zoo visitors are more likely to be pet owners, leading to biased sampling.

Example 2: Internet Use Among Students

• Kathy polls students via email to determine internet usage for learning, finding that 83% use the internet.
• This is biased as it only surveys students with internet access.

Non-Probability Sampling Methods

Convenience Sampling

• Uses readily available subjects or those easy to reach.
• Student volunteers are a common example.

Purposive Sampling

• Involves selecting predefined groups that would serve as samples.
• A researcher seeking advice on graduating summa cum laude might interview individuals who achieved that distinction.

Probability Sampling Methods

• Simple Random Sampling
• Stratified Sampling
• Cluster Sampling
• Systematic Sampling
• Multi-Stage Sampling

Simple Random Sampling (SRS)

• All members of the population have an equal chance of being selected.
• A "fish bowl method" is sometimes used.

SRS Example

• A credit card company wants to gauge satisfaction with its new billing system, where 1,000 random numbers are generated to select cardholders from an alphabetical list.

Stratified Sampling

• The population is divided into smaller groups or strata, and then SRS is applied within each stratum.

Stratified Sampling Example

• Students are grouped by grade level (e.g., grades 7-10), and 100 representatives from each grade are selected for a study on science and math grades.

Cluster Sampling

• Groups or clusters, rather than individuals, are randomly chosen.

Cluster Sampling Example

• All children in five districts of Metro Manila form a sample to study dengue fever occurrence.

Systematic Sampling

• Selects every kth member of the population, with the starting point determined at random.

Sample Size (n)

• Most statisticians agree that a minimum sample size of 100 is needed to get any kind of meaningful result
• In a population of 5000, 10% would be 500..
• In a population of 200,000, 10% would be exceeded 1000, so in this case .20,000 the maximum would be 1000
• A good maximum sample size is usually 10% as long as it does not exceed 1000.

The More Samples, The Better

• In research, the more sampes get, the better
• The opinion of 1,000 people is always better than the opinion of 100 people

Determining Sample Size

Methods

• Using a census for small populations (n <= 100).
• Using a sample size of 10% of N.
• Using published tables.
• Using formulas to determine sample size, such as Slovin’s Formula.

Slovin's Formula

• n = N / (1 + Ne^2)
• n is the sample size
• N is the population size
• e is the margin of error