Chapter 11: Sample Surveys

Sample

A smaller group selected from a population to examine when studying the entire population is impractical.

Population

The entire group of individuals or cases you want to learn about.

Sample Survey

A study that asks questions to a sample in order to estimate information about a whole population.

Bias

Any systematic failure of a sampling method to represent the population. It invalidates conclusions. Random selection helps avoid it.

Randomization

Protects against bias and allows for valid inferences from the sample to the population.

Sample Size Matters

It’s the size of the sample, not the population and that determines the accuracy of estimates. Exception: If sampling more than 10% of a small population.

Census

A survey that attempts to include the entire population. It’s often impractical due to time, cost, and population changes.

Parameter

A numerical value that describes a characteristic of a population. Denoted with Greek letters.

Statistic

A numerical value that describes a characteristic of a sample. Denoted with Latin letters.

Simple Random Sample (SRS)

Every group of a given size has an equal chance of being selected. The gold standard for sampling.

Sampling Frame

The list of individuals from which the sample is drawn.

Sampling Variability

Natural differences in results from different random samples of the same population.

Stratified Sampling

The population is divided into homogeneous groups (strata), and an SRS is taken from each group.

Cluster Sampling

The population is divided into similar, heterogeneous groups (clusters), and entire clusters are randomly selected.

Multistage Sampling

Combines several sampling methods (e.g., stratified + cluster + SRS).

Systematic Sampling

A sample drawn by selecting every individual from a list. Must start from a random position.

Who in Surveys

• Population of Interest: The group you want to learn about.

• Sampling Frame: The group you actually sample from.

• Target Sample: The group you intend to get responses from.

• Actual Respondents (Sample): The people who actually respond.

Valid Survey Design

• Ask clear, specific, and relevant questions.

• Use an appropriate sampling frame.

• Consider question and answer wording.

• Conduct a pilot survey to test for errors.

Voluntary Response Sample

Only people who feel strongly respond. Leads to voluntary response bias.

Convenience Sample

Uses individuals who are easiest to reach. Often not representative.

Bad Sampling Frame

Using an incomplete or outdated list introduces bias.

Undercoverage

When a portion of the population is not included or is underrepresented.

Nonresponse Bias

People who don’t respond may differ significantly from those who do.

Response Bias

Anything in the survey design (e.g., question wording, interviewer influence) that affects answers.