Section 1.5 Bias in Sampling
¶Definition 1.5.1. bias.
when sample results are not representative of the populationDefinition 1.5.2. sample bias.
bias because the technique used to select the sample favors some individuals over others (not random)Definition 1.5.3. under-coverage.
type of sampling bias where part of the population has a lower chance or no chance of being in the sample (e.g. if the frame is incomplete)Definition 1.5.4. non-response bias.
bias because individuals in the sample do not respondDefinition 1.5.5. response bias.
bias because responses do not reflect the true feelings or beliefs of the respondent.Response bias can occur in a number of different ways:
- From the interviewer—the nature or wording of the interviewer can affect the truthfulness of responses.
- Misrepresented answers—when the results of a survey misrepresent facts or are flat out lies.
- Wording of questions—the way a question is worded can lead to response bias in a survey.
- Ordering of questions or choices—many surveys will rearrange the order of the questions so that responses are not influenced by prior answers.
Definition 1.5.6. open question.
allows a respondent to choose his or her own response e.g. "What is the most important problem facing youth today?"Definition 1.5.7. closed question.
requires the respondent to choose from a list of predetermined responses.Definition 1.5.8. non-sampling errors.
errors that result from under-coverage, non-response bias, response bias or data-entry error.Definition 1.5.9. sampling errors.
errors that result from using a sample to estimate information about a population.I will leave it for you to read section 1.5 for examples. This section is only three pages but is a must read section!