Sampling Methods in Research: Random, Stratified, Systematic, and Opportunistic Techniques

Judgmental Sampling

Sampling based on the researcher's judgment rather than random selection.

Random Sampling

Process that gives each sample unit in the population the same probability of being selected.

Random Sampling Assumption

Assumes that variability of a sampled medium is insignificant and/or homogenous (e.g., aquarium, lagoon, lake).

Random Sampling Limitation

Not good for heterogeneous populations/phenomena.

Statistical Basis of Random Sampling

A random number table can be used.

Random Sampling Analysis

Simple downstream statistical analysis.

Efficiency of Random Sampling

Often efficient and lower cost than other methods.

Random Sampling Use Case

Good for new/unknown sites where little information is known.

Stratified Random Sampling

Divides the sampling population into several non-overlapping zone/strata (mutually exclusive).

Stratified Random Sampling Method

Simple random sampling within the zones/strata.

Stratified Random Sampling Requirement

Requires some prior knowledge of the system.

Stratified Random Sampling Flexibility

Able to vary the sample size according to project need.

Systematic Sampling

Selection according to a specified pattern in time and space (e.g., equal distance intervals along a line/grid).

Systematic Sampling Process

Often the first sample is random and others follow the pre-determined intervals.

Systematic Sampling Benefits

Can make extrapolation from the same period to future periods easier.

Opportunistic & Haphazard Sampling

Include 'ships of opportunity' and crowd-sourced data like iNaturalist.

Opportunistic Sampling Nature

May appear to be 'random' but is not statistically random.