Statistics for Data Science

Bias sample selection

The process of sample selection or sampling is also subject to the introduction of bias. Sample bias occurs when the sample does not accurately represent a population. The bias that results from an unrepresentative sample is called selection bias.

Issues that can result in introducing bias to a statistical sample include:

• The timing of taking the sample
• The length or size of the sample
• The level of difficulty of the question 
• Undercoverage (of the population)
• Nonresponses incorrectly used in a sample
• Voluntary responses incorrectly used within a sample
• The manner (in which the subjects in the sample were contacted (phone, mail, door-to-door, and so on), or how the observation data was split)

Enough about bias. Let's ...