SuperMisleading Graph

I’ve got nothing against home schooling and in fact seriously considered it for our own children, but the Washington Post is presenting what may be good news in a very misleading manner.

Source: How many kids are homeschooled in the U.S.? Growth by school district. – Washington Post

OK, So perhaps no-one would read this as saying that 51% of US kids are now home-schooled and would see that we are talking about %change. But is that year over year or in total since 2017? The labels don’t say, and many readers are used to seeing %change presented in the former sense which might lead to the unthinking impression (once one has absorbed that we are looking at growth rather than absolute numbers) that there continued to be more parents starting to take their kids out of school after the end of the COVID closures than were going back to the public classrooms. But if it’s total change then the number of home schoolers decreased in the last two years (which is more like what one would expect, and is indeed what is described later in the article where it is admitted that “In most states examined by The Post, home schooling has fallen slightly from its peak, while remaining at highs unmatched before the 2020-2021 school year”). Sure, anyone who stops to think (or reads on) may not end up permanently misled, but the false emotional impressions, first of dramatic increase and then of continued growth, may continue to be felt even after being corrected.

Also, when we see the graph shoot up in 2019-20 it looks dramatic because it starts at “0” in 2017-18 when that was the original 100% and having that at the bottom makes the peak height in 2020-21 look like much more than a mere 62% increase. So in fact, even at the height of COVID, the (actually quite small) absolute number of home schoolers never even doubled from what it was before COVID.

If we project the pre-COVID trend though, then it does look as if the number of home schoolers might have increased at a rate more like 5% per year or by about a third between 2017 and 2023. This is indeed respectable growth, but not at the dramatic scale suggested by that graph. And of course when a quantity is small then a large %growth rate is still a small absolute number and only becomes really impressive if sustained for an extended period. So I don’t really think it can fairly be said that growing (during an extended school-closing pandemic) from around 1% to a bit under 2% of the overall school population “demonstrates home schooling’s arrival as a mainstay of the American educational system”.

This entry was posted in uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *