I’ve noticed most U.S. polls query around 1,000 people and sometimes even less. Somehow that qualifies for headlines like “Americans say…” or “Most Americans…” How is it acceptable that 0.0002% of the population is accepted as representative?
In: Other
Take a course in statistics and you will get the real answer. A statistical sample is not random in the way that you can be led to believe. Random polling by voice call preselects:
1. People who will answer a phone call from a stranger
2. People who want to talk
3. People who tend to be older
4. People within more populous geographic areas
5. People with a landline
Random voice voting excludes:
1. People with cellular phones with call filtering
2. People with prepaid cell phones
3. People with limited mobility who cannot reach the phone in time
4. People who do not want to talk
It goes on and on. Significant time is spent in class talking about how to select populations for statistical analysis. When someone puts together a national opinion poll they now have to stitch several methods together to reach the statistically valid sample. They need to hit a cross section of regions, ages, marital situations, educational attainment, and perhaps political affiliation or lack thereof.
Latest Answers