Page 1

• Lit reviews due now

• Grading this afternoon and tomorrow

• Final paper due 4/8

• Grades posted Sunday after grading papers and peer reviews

• To do: CITI Certificate

• Next class on April 1, 2 weeks off

Page 3

• Generalizability, population, sample, strata defined

• Probability vs. nonprobability sampling explained

Page 5

• Sampling involves studying a sample to infer about the population

• Target population, accessible population, sample defined

Page 6

• Bias and error in sampling discussed

• Ways to minimize sampling error

Page 7

• Inclusion and exclusion criteria in sample selection

• Careful consideration of discriminatory language

Page 8

• Detailed criteria for breast cancer study

• Inclusion and exclusion criteria listed

Page 9

• Subject recruitment methods discussed

• Various ways to recruit subjects mentioned

Page 10

• Sampling methods: probability and nonprobability

• Different sampling techniques listed

Page 13

• Simple random sampling explained

• Every member has an equal chance of selection

Page 15

• Systematic random sampling described

• Every nth person chosen from a list

Page 17

• Stratified random sampling defined

• Ensures representation of important subgroups

• Oversampling explained

Page 19

• Example of stratified sampling in Head Start program

• Stratification based on state (rural and non-rural)

• Sample represented 5% of Head Start sites and 3% of enrolled children

Cluster Sampling

• Select sub-groups (clusters) instead of individuals randomly.

  • Useful when the population consists of clusters or units (e.g., class).

• Disadvantages include homogeneity within clusters requiring a large number of clusters.

• Can employ stratified cluster sampling.

• Falls under probability sampling if the number of clusters is known, otherwise under non-probability sampling.

Cluster Random Sampling

• Example: Candies divided into 4 clusters of 5 candies, randomly choose 2 clusters.

Multistage Sampling

• Used in large national surveys.

• Involves drawing a stratified random sample, then selecting units within the stratification, and finally randomly choosing participants.

Nonprobability Sampling

• Total number of subjects is unknown.

Purposive Sampling

• Selects individuals believed to provide the best information.

• Not random, similar to focus groups.

Snowball Sampling

• Useful for locating hard-to-find participants.

• Participants refer the researcher to others.

• Based on trust and confidentiality.

Quota Sampling

• Subjects selected based on characteristics without randomization.

• Sample may be stratified.

• Example: Selecting the first 50 women who walk through a cafeteria.

Convenience Sampling

• Non-random with limited representation.

• Often involves a captive audience like a class or team.

Role of Subjects in Sample

• Subjects play a role in experiencing the intervention and providing information for observations and outcomes.

Sample Size

• Larger samples reduce sampling error.

• Guidelines for sample sizes in different types of studies.

Increasing Sample Size

• Consider increasing sample size for greater variability between groups or small differences.

• Email surveys may require a 40-50% increase in sample size.

Comparison Chart

• A comparison of different sampling methods and their outcomes.

Conclusion

• Understanding various