An elementary school principal wants to know the mean number of children in families whose children ...

An elementary school principal wants to know the mean number of children in families whose children attend this school. He checks all the families using the school's registration records, and we use the computer to create a 95% confidence interval based on a t-distribution. This procedure was not appropriate. Why?

▸ The recent record-setting family with twelve children is probably an outlier.

▸ At a given school families are not randomly selected.

▸ The entire population of families was gathered so there is no reason to do inference.

▸ The population standard deviation is known, so he should have used a z-model.

▸ Since these families are from only one school, the family sizes may be skewed.