Practice quiz: Samples and uncertainty

4 questions. Pick an answer to see if it is right. You can try as many times as you like. These are for practice: the final test has different questions.

1. 1.A poll of 2 million people chosen from car-owner lists predicts an election wrongly. Why?

   The sample was too smallThe sample was biased, and size does not fix biasPolls never workToo many people answered
2. 2.To halve the standard error of a mean, you need about...

   twice as much datafour times as much datahalf the dataten times as much data
3. 3.Which result is more informative?

   +3 points, p < 0.05+3 points, 95% CI 2.5 to 3.5'A significant improvement'+3 points
4. 4.A 95% CI for a difference runs from −4 to +10. What can you conclude?

   There is definitely a gainThere is definitely no effectThe data is consistent with a loss, no effect, or a gainThe study is fraudulent