A radio show asks its listeners to text in, and nine in ten say the same thing.
3 lessons, about 1 h 30 min3 simulations10 problems3 practice questionsPython in every lesson
A cook tastes one spoonful to judge the pot. In statistics the pot is the population and the spoonful is the sample. A number that describes the population is a parameter, and the same number worked out from a sample is a statistic. This lesson shows why we sample and why two fair samples give different statistics.
A sample is biased when the way it was chosen makes it miss the truth in the same direction, again and again. This lesson meets the usual kinds: the convenience sample, the voluntary response sample, undercoverage and survivorship bias. It also measures a bias in percentage points.
Even a fairly chosen sample can go wrong if many people do not answer, or if the question pushes them. This lesson works out the effect of non-response, shows how the wording of a question changes the answers, and proves by simulation that a huge biased sample is still biased.
Log in to join the discussion.
Loading…