Question

8. Each of the following surveys could result in a potential non- sampling bias. Determine the type of bias and suggest a remedy.

(a) A researcher wishes to study tax issues for all city residents. She randomly selects recipients of an exclusive literary magazine to participate in the study.

Answer 1

In this example, we collect data only from 1 type of sample.

Ideally, the data that is collected must be sufficiently large and should be a good representation of the population. In this example, this is not the case. Thus, we can say that sampling method is suffering from coverage bias or coverage error because it is not covering the entire population effectively, but only a smart portion of it.

This can be resolved by collecting data randomly from recipients of multiple magazines, or in generally, having a wider coverage than just a single, exclusive magazine.