S + M, Chapter 7

Definition of Sampling: The act of selecting a small portion or sample from a larger population to reflect characteristics and allow conclusions about the entire group.
Advantages of Sampling:
- Saves time and resources compared to studying an entire population
- U.S. Census Bureau example: Employed over 1.4 million staff at a cost exceeding $14 billion for the 2010 census
Uses of Sampling: Sampling provides accurate information about larger populations, critical for social scientists who lack massive resources.

Key Questions:
- What constitutes a sample?
- What makes a sample representative?
- What does the sample represent?
Populations: Any group of people or objects from which a sample can be drawn. Examples include voters, residents, documents, etc.
Characteristics of Samples: A sample should capture the diversity of the larger population in proportion to their occurrence.
Microcosm Concept: A representative sample serves as a smaller model of the entire population, helping draw valid conclusions.

Example Study: Analyzing political organization membership levels within the U.S.
- Method: Shaded areas demonstrate sampling versus actual population attributes. Conclusion: A sample should reflect the distribution of the population attribute
Biased Sampling: Unrepresentative samples lead to incorrect conclusions.

Literary Digest Fiasco: Example of biased polling leading to inaccurate predictions due to poorly selected samples based on socioeconomic status.
Representative Sample: Must include all major segments of the population in correct proportions to avoid skewed results.

Definition: Random selection is used to ensure that each member of the population has an equal chance of being included.
Types of Probability Samples:
1. Random Samples
2. Systematic Random Samples
3. Cluster or Multistage Random Area Samples
4. Stratified Samples
Random Samples: Selected through a lottery-like method to guarantee equal opportunity for every individual.

Process:
- Count the population size and divide by the desired sample size (k). Select every kth case based on random starting point.
Biased Risks: Systematic methods may miss diversity if the order contains bias.

Multistage Random Area Sampling: When no centralized lists exist, select based on geographical units instead of individuals.
- Involves mapping, random district selection, and subsequent narrowing down to individuals, maintaining randomness at each stage.

Purpose: Enhances representation of small subgroups within larger populations by drawing separate samples from both subpopulations and larger groups.
Key Considerations: Care in establishing subgroup definitions and statistical validity while generalizing results.

Definition: Samples not selected randomly, resulting in potential bias affecting representative accuracy.
Types of Nonprobability Samples:
1. Convenience Samples
2. Volunteer Samples
3. Purposive Samples
4. Snowball Samples
5. Quota Samples
6. Judgmental Samples

Factors Affecting Sample Size:
- Homogeneity of the population: Uniform characteristics require smaller samples, while diversity requires larger samples.
- Number of categories: More variables lead to a need for larger samples to capture differences.
- Sampling error: The greater accuracy sought, the larger the sample needed.

Definitions of the confidence interval and level of confidence in research context.
Calculation considerations on acceptable sampling error margins and sample sizes (e.g., larger samples yield more accurate estimates).

Necessity of disclosing sample selection methods and precision of results for validation.
Ethical standards: Research professionals are expected to provide a complete description of sampling procedures, population definitions, response rates, and methodological transparency.

Importance of careful sample selection for generalizing research results across populations.