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
The surveys have a lot of control questions, sometimes half the questions in a normal survey are control questions. They ask people about their race, what their income is, their age, their political affiliation, where they live, how many kids they have, etc. This way they can correct for any inaccuracies in who you surveyed. For example if they find out that 60% of the people they surveyed were female they can weight the male answers higher. Similar with location and income brackets. This makes the surveys far more accurate. We also accept that the results of these surveys are not perfectly accurate. There are ways of calculating the confidence interval for each number in the result, the interval that the researchers are confident the real answer is in. But often the confidence interval is not included when journalists use the result of the survey to write an article except in the most serious news publications. However policy makers and even competent journalists pay close attention to this number as well.
Latest Answers