We want to know the mean winning score at the US Open golf championship. An internet search gives us all the scores for the history of that tournament, and we create a 95% confidence interval based on a t-distribution. This procedure was not appropriate. Why?
A) The entire population of scores was gathered so there is no reason to do inference.
B) Tiger Woods' recent record-setting score is probably an outlier.
C) Since these are the best players in the world, the scores are probably skewed.
D) The population standard deviation is known, so we should have used a z-model.
E) In big golf tournaments the players are not randomly selected.

Question

Quizplus · Accepted Answer

Since the entire population of scores was gathered, there is no sampling error, and thus no need to estimate parameters with a confidence interval. Confidence intervals are used to infer about a population parameter based on a sample, not the entire population.

We Want to Know the Mean Winning Score at the US