A sampling method is biased if it tends to produce estimates that are systematically too high or too low. Bias comes from how the data are collected, and a bigger sample does not fix it. Survey a simulated town about a new skate park six different ways and see which methods miss, in which direction, and why.
The town has 2,000 adults. Younger adults are much more likely to support the skate park. The true proportion who support it is the parameter p.
Bias shows up as a center far from p. Variability shows up as spread (SD). Bigger n shrinks the SD, not the bias.
For each survey, choose the main source of bias.
A method is biased when it tends to over- or underestimate the parameter, survey after survey. Random chance error goes both ways; bias pushes one way.
Part of the population has little or no chance of being in the sample, like people without landlines in a landline phone survey.
Individuals selected for the sample can't be reached or refuse to respond, and the people who don't respond differ from those who do.
People give inaccurate answers because of question wording, the interviewer, a sensitive topic, or wanting to look good. It happens even with a perfect random sample.
An open call (online poll, call-in show) attracts people with strong opinions, usually negative ones. The sample chooses itself, so it is not random.
Choosing whoever is easy to reach, like customers at one store, gives a sample that often differs from the population in ways related to the question.
Full credit names the bias, explains who is missing or misrepresented, and says which way the estimate is pushed.
| Survey | Bias | Why | Direction |
|---|---|---|---|
| Skate park poll on the parks department website | Voluntary response | Supporters care more and are more likely to vote online | Overestimates support |
| Interviewing customers at the local skate shop | Convenience | Shop customers are mostly young skaters who want the park | Overestimates support |
| Random landline phone numbers | Undercoverage | Few young adults have landlines, and they support the park most | Underestimates support |
| Random mail survey; mostly older adults mail it back | Nonresponse | Young adults who were chosen rarely respond, and they differ in opinion | Underestimates support |
| "Do you support wasting tax money on a skate park?" | Response bias (wording) | The loaded words push people toward "no" | Underestimates support |
AP-style multiple choice. Pick an answer to see the explanation.
The vocabulary for topic 1.12. Read the list, flip through flashcards, or quiz yourself. For every term in the unit, see the Unit 1 terminology guide.