Data Collection

Population

Full set of observations of interest (denoted as N)

Sample

Subset of the population (denoted as n)

Simple Random Sample

Every possible sample if size n has an equal chance of being selected

Stratified Sampling

divide the population into groups and sample from each

Cluster sampling

Divide the population into clusters, select a cluster, and include all members

Systematic sampling

Select every kth observation from a list of all N

Convenience Sampling

Choose observations easiest to access

Voluntary sampling

Observations volunteer to participate

Sampling Bias

When a subset of population has a lower/higher chance of being selected

Measurement Bias

Collected data is inaccurate

Non-Response bias

Some selected observations fail to provide data, or response is not recorded

Observer Bias

The person recording data allows their expectations, beliefs, or subjective judgement influence the results