"The top 20-25% of our 6th grade students will participate in our AC classes, and we place those students in each content area based on criteria. In most recent years, the students that are placed in our AC Language Arts range from about the 95%tile- 99.9%tile."
You know, some days I wish school administrators had to take a comprehensive course in statistics before they were allowed to wave them around. 75th or 80th percentile in his school is 95th percentile for the population as a whole? That seems tremendously statistically unlikely.
Does it, really? Stats is not my strong point, but suppose we're measuring on the usual normal distribution, SD 15 basis. Suppose School A has, because of local population or because of explicit selection, a distribution which is shifted one standard deviation above the population distribution. Then half of School A's students are above their mean and above the 84th percentile of the population (50th percentile in School A is 84th for the population) , and 16% of School A's students are above the 98th percentile of the population (84th percentile for School A is 98th for the population). I don't know how to interpolate, but this seems to be roughly the scenario we need to make the school's statement true.
I couldn't quickly google convincing proof, but I don't think this amount of distribution-shifting as a result of an area having mostly privileged children is implausible, at least in the UK. It's unlikely to arise by chance, but children aren't sorted into schools by chance.
