Friday, April 28, 2023

Polls’ Representative Samples Often Merit Skepticism

Polls’ Representative Samples Often Merit Skepticism 
"When people read about a poll, they are often told something like this: 
  • Pollsters call a random group of people. 
  • Though small, we can infer this group is a representative sample of the U.S. population because it was selected randomly. 
  • Then, if we know the sample size, we can calculate the margin of error, which helps us understand how likely the survey results are to capture reality.
  • This Statistics 101 version frequently carries a disclaimer to the effect: “These findings come from a nationally representative sample of 2,000 Americans with a margin of error of plus or minus 2.2%.” 
Here is the uncomfortable truth
Though the disclaimer remains, nearly nobody does the textbook version of polling anymore...

No comments: