StellarUndercoverage and Nonresponse
Undercoverage
Examples of undercoverage:
- Phone surveys miss people without phones
- Internet surveys miss people without internet access
- Surveys of households miss people who are homeless
Nonresponse bias
Example: A health survey mailed to 1,000 people gets only 200 responses. If sicker people are less likely to respond, the survey underestimates illness rates.
Reducing nonresponse:
- Follow-up contacts (calls, emails)
- Incentives for participation
- Multiple contact methods
- Making surveys shorter and easier