Overview
Today's lesson introduced sampling design, focusing on the methods and terminology used in collecting data through surveys and experiments, as well as common sources of bias.
Methods for Gathering Data
- Data can be gathered via surveys, opinion polls, interviews, observational studies, and experiments.
- Observational studies can be retrospective (past data) or prospective (future data).
Key Sampling Terminology
- The population is the entire group you want information about.
- A census is a complete count of the entire population.
- Destructive sampling destroys or loses the product being tested.
- A sample is a subset of the population used to infer information about the whole.
Types of Sampling Designs
- A sampling frame is the list of all individuals in the population.
- Simple random sample: Each individual has an equal chance of being selected.
- Stratified random sample: Divide the population into groups (strata), then randomly sample from each group.
- Systematic random sample: Randomly choose a starting point, then select every nth individual.
- Cluster sampling: Randomly select entire groups (clusters) and sample all individuals within those groups.
- Multi-stage sampling: Randomly select smaller groups from larger groups in multiple levels.
Advantages and Disadvantages of Sampling Methods
- Simple random sample: Unbiased, easy to do; may have large variance, requires full population list.
- Stratified sample: More precise, less variability; more complex, requires defined strata and population list.
- Systematic sample: Unbiased, efficient; can be affected by trends, less flexible, formulas can be complex.
- Cluster sample: Unbiased, reduced cost; clusters may not be representative.
Sources of Bias in Sampling
- Voluntary response: Individuals choose to respond, often leading to extreme opinions.
- Convenience sampling: Sampling people who are easy to reach; often introduces bias.
- Undercoverage: Some groups are not represented in the sample.
- Nonresponse: Chosen individuals refuse or fail to respond.
- Response bias: Answers are influenced by the respondent or interviewer’s behavior or question wording.
Key Terms & Definitions
- Population — the entire group you want information about.
- Sample — a part of the population actually studied.
- Census — data collection from every individual in the population.
- Sampling frame — the list of all individuals in the population.
- Simple random sample — every member has an equal chance of selection.
- Stratified sample — sample chosen from distinct subgroups.
- Cluster sample — all individuals from randomly selected groups are surveyed.
- Bias — systematic error favoring certain outcomes.
Action Items / Next Steps
- Review types of sampling methods and their properties.
- Practice identifying sources of bias in sample designs.
- Prepare assigned survey questions, ensuring unbiased and clear wording.