Questions: Sampling and Populations in Psychological Research

Question 1

A psychology study recruits 500 participants exclusively from introductory psychology courses at a U.S. university. The study finds that social comparison increases anxiety. What is the primary threat to the study's external validity?

• A. The sample size is too small to detect true effects
• B. The sample is a convenience sample that may not represent the broader population
• C. The study lacks random assignment to conditions
• D. The anxiety measure may not be reliable

Answer: (B)

The core issue is that the sample is drawn from a narrow, self-selected pool — undergraduate psychology students at a U.S. university — who differ systematically from the general population in age, education, cultural background, and many psychological characteristics. This threatens external validity (the ability to generalize findings). The other options address power, internal validity, and measurement reliability — real concerns, but secondary to the representativeness issue here.

Question 2

A researcher surveys 50,000 people using an online opt-in panel. Because the sample is very large, the results can be generalized to the general population with confidence.

• T. True
• F. False

Answer: False

Sample size and sample representativeness are independent. A 50,000-person opt-in online panel is a convenience sample biased toward people who are online, willing to take surveys, and aware of the platform — all of which correlate with many psychological variables. The famous 1936 Literary Digest poll predicted Roosevelt would lose using 2.4 million respondents; he won by a landslide. Systematic bias in who is sampled cannot be corrected by collecting more of the same biased data.

Question 3

What is the WEIRD problem in psychology research, and why does it matter for the field?

Model answer: WEIRD stands for Western, Educated, Industrialized, Rich, Democratic — the narrow slice of humanity that most psychology studies sample from. It matters because findings from WEIRD samples are often assumed to be universal but may not replicate across cultures.

Henrich, Heine, and Norenzayan (2010) showed that the typical psychology sample — North American undergraduates — is a statistical outlier on many cognitive, social, and perceptual variables, not a representative human. Studies on visual perception, fairness norms, individualism/collectivism, and many other topics show large cross-cultural variation. Treating WEIRD findings as universal laws of human psychology risks exporting cultural particulars as scientific universals.