Survey methodology

Survey methodology is the study and practice of designing surveys to collect reliable data. It addresses sampling, question wording, measurement, and data collection procedures.

Probability sampling: A sampling method in which each population member has a known, nonzero chance of selection. Known selection probabilities support estimates of population characteristics and sampling error.
Total survey error: A framework that treats survey error as the combined effect of multiple sources across the survey process. It organizes design choices around their effects on the accuracy of survey estimates.
Census: A systematic attempt to enumerate every member of a defined population. Unlike sample surveys, a census seeks full enumeration but remains vulnerable to coverage and response error.
Administrative data: Data collected by organizations during routine operations, such as records of enrollment, payments, or service use. Unlike surveys, these records can cover behavior directly but often lack needed attitudes or consistent population coverage.
Nonresponse: Failure of selected survey units or individuals to provide the requested information. Falling participation raises uncertainty about when adjustment methods can recover population estimates.
Sampling frame: A list or operational representation of units from which a sample is drawn. Its coverage determines which population members can be selected.
Coverage error: Error caused by differences between the target population and the population represented by a sampling frame. Frame omissions or duplicates can systematically exclude or overrepresent people.
Household survey: A survey that collects information from people or households in a population, often using a household sampling frame. Household selection and within-household respondent rules are central design decisions.
Experimental design: A plan for assigning conditions and measuring outcomes to estimate causal effects. Surveys usually describe or measure associations; experiments manipulate conditions to support causal inference.
Survey satisficing: A response strategy in which people give minimally acceptable answers rather than carefully optimizing their responses. Long or difficult questionnaires can increase satisficing and weaken measurement quality.

Linked from 8 pages