Undercoverage vs Convenience Sample
Both terms below come up in the same part of the course, and students mix them up. Here is each one defined on its own, side by side, so you can see where they part company.
Undercoverage
Collecting data and study design
Undercoverage is a form of sampling bias that occurs when some groups in the population are left out or underrepresented by the sampling method.
Undercoverage happens when the method used to reach people systematically misses part of the population. For example, a telephone survey conducted only on landlines undercovers younger adults, who mostly use cell phones. Because the missing groups may differ from those included, the sample statistic can be pushed away from the true parameter. Using a sampling frame that covers the whole population helps reduce it.
Convenience sample
Collecting data and study design
A convenience sample includes whichever individuals are easiest to reach, rather than selecting them through any random process.
A convenience sample takes the members who are simplest to access, so it usually fails to represent the population and tends to be biased. For example, surveying only your own friends about a school policy is a convenience sample. Because the selection is not random, there is no way to know the direction or size of the bias, and standard error formulas do not apply. Ease of collection does not make up for a lack of randomness.