Chapter 12: Survey Studies: Design and Sampling

What are different survey types?

mailed questionnaires, telephone interviews, face-to-face interviews, or even online pop-ups

Selection:

In survey sampling, the process that guides the choice of respondents. Methods of selection play a major role in the extent to which survey results may be generalized to a larger population

Census:

A survey or enumeration involving all the potential units of the target population (the universe of interest)

Sampling Error:

 The expected typical discrepancy between the estimates calculated from a sample and the value that would be obtained if the entire targeted population (i.e census) has been used)

Precision

 A statistical value computed from members of a sample that serves as inference within some degree of sampling error, of a population value

Sampling Frame:

A complete listing of the population from which members of a sample are drawn and used

Random sampling

In this form of selection, a random mechanism is used to obtain a sample intended to be representative of the underlying population

Nonrandom sampling:

In this form of selection, a non-random mechanism is used to obtain a sample from a population

Simple random sampling:

 A sampling technique in which every member of the population in question has an equal (and nonzero) probability of being selected every time a unit is drawn for inclusion in a sampl

Systematic sampling:

 A form of sampling requiring  sampling every predetermined nth member from a population, after randomly choosing a starting point for the sampling process:

Stratified Sampling:

A technique in which a population of interest is divided into theoretically meaningful or empirically important strata. In pre-stratification, members are randomly drawn from each stratum (subpopulation) before the sample is drawn. In post-stratification, the classification process is done after data are collected from respondents.

Oversampling:

In survey research, a process in which respondents who are underrepresented in the population are selected at a greater rate than more commonly encountered respondents. This is done to ensure a stable representation of the oversampled group in the analysis.

Cluster sampling:

In this form of sampling, geographic locations (or segments) are randomly sampled, and all members of the selected segment are used in the sample

Multistage sampling:

In this method, clusters of locations are sampled from a geographical sampling frame (as in cluster sampling) and then (unlike cluster sampling) units within each cluster are sampled as well.

Panel Survey:

 A survey in which the same respondents are used repeatedly across survey administrations

Convenience Sampling:

In this weak form of sampling, the respondent sample consists of people most readily accessible or willing to be part of a study

Snowball Sampling:

A sampling technique in which initially selected respondents contact and recruit others in their social network to participate in the research.

Quota Sampling:

A method in which respondents are sampled non-randomly until a predefined number of participants for each of a specified set of subgroups (racial, ethnic, religious) are achieved

Random digit dialing:

A sampling approach that uses banks of phone numbers, randomly sampled from specified area codes to construct a random sample of eligible respondents