Based on research by the American Psychological Association. (2016), written by the American Psychological Association, adapted by Juanita N. Baker, Ph.D.
During elections, surveys make claims about how people will vote in upcoming elections. Be skeptical. Only sound research methods lead to valid conclusions
When behavioral scientists want to assess attitudes of very large groups (e.g., American voters), they face an insurmountable problem. It’s impossible to ask every citizen what they will do. However, behavioral scientists have solved this tricky problem by developing a technique called random sampling.
Researchers select a very small proportion of the people from a much larger sample (e.g., 1,000 from all registered voters). They estimate what the entire population is like on the basis of the responses of those sampled. The key to an accurate estimate is the use of random sampling. Random sampling selects people from a population so that everyone in the entire population has an equal chance of being selected. This turns out to be an incredibly powerful technique. If every person of 100 million voters really does have an equal chance of being selected into a national survey, then the results based on 1,000 people will almost always prove to resemble the results for the total population.
Be skeptical. Ask, “How did they sample the population?”
American Psychological Association. (2016). How to Be A Wise Consumer of Psychological Research, Evaluate research-based claims to become a better consumer of products and services that shape your daily life. Research in Action. http://www.apa.org/research/action/mer.aspx
Bellhouse, D.R. (1988). “A Brief History of Random Sampling Methods,” in (P.R. Krishnaiah and C.R. Rao, Editors) Handbook to Statistics: Sampling, Vol. 6, New York: Elsevier Science Publishers B.V., 1-14.
Laplace, P.S. (1786). “Sur Les Naissances, Les Mariages Et Les Morts,” In Histoire de L’Academie Royale des Sciences, 1783, Paris, 693-702.
For more details, see:
How to Be A Wise Consumer of Psychological Research,
American Psychological Association, © 2016