Poverty Isn’t Destiny?

Quite a few Ed Deformers say that Poverty Isn’t Destiny. They say that it doesn’t matter if a child has been subjected to lead poisoning, separation from parents, violent or otherwise cruel child abuse, inadequate nutrition, and has lacked dental or health care and the love and care of a family during the first, crucial years. All it takes is for a Bright Young Thing fresh out of college to work her butt off for two years before she goes to work for a bank — and all of those handicaps will be overcome, with no extra dollars invested, and maybe even less!

Or maybe not.

Lots of teachers have been working their butts off for many decades, doing their best, believe it or not (for the most part).

Here are two three graphs from Wisconsin that show how close the connection between the poverty rates and student achievement levels, at all of their schools for which they provide data. My data come from here and are for SY 2011-2012. In fact, you can download the entire spreadsheet for the state of Wisconsin if you click on this link:

http:reportcards.dpi.wi.gov/files/reportcards/xls/2011-12reportcarddata.xlsx

In both all three graphs, the percentage of students at the schools is along the horizontal (X) axis. In the first two, the average achievement score at the school is along the vertical (Y) axis.

In this first graph, Wisconsin uses a 100-point scale for overall student achievement.

That is an incredibly strong correlation between poverty levels and student achievement. The fewer the proportion of poor students at a school, the better the achievement scores at that school.

I had Excel compute two correlation “trend” lines – one straight, in black, and one curved, in red following a third-degree polynomial, since it looks like we have a serious “Matthew effect” going on here. In either case, the R-squared and R values are very elevated, showing that, in fact, poverty is in fact destiny for a lot of kids.

The next graph is for reading only, but it shows essentially the same trend. School reading scores go from 0 to 50.

There are very few real-life correlations between two entities stronger than what you see in these two graphs.

This next graph is a little different, for two reasons: the y-axis is math, and it’s the percent of students deemed ‘proficient’ on whatever test Wisconsin is using. It also shows a very strong correlation.

Advertisements

A Look at population trends in DC’s school population

I’m going my best to play it straight.

In the following fiew posts I will try to give you a clear look at how the publicly-funded student enrollment in Washington, DC has been trending over the past decade or so.

I’ll then make a few predictions of how I think things will continue in the next few years.

And then make some judgments on what these records mean.

I think that graphs are often one of the very easiest ways to make things clear, but as Darrell Huff wrote a long time ago in a classic work called “How to Lie With Statistics“, you can still use them in many ways to mislead if you want to.

Here is the very first graph given by DC’s Office of the State Superintendent of Education, or OSSE:

Looks like a huge jump, right? Maybe the latest enrollment is about twice as much as it was at its low point just after Rhee got here to save the day in 2008, right?

No.

It’s not such a huge jump.

And it’s for the regular DC public schools combined with the DC charter schools (privately run tho funded by taxpayers). Not just DCPS, which I’ll look at in a subsequent blog.

The scale is really misleading. The total population is in fact increasing, but (a) most of it seems to have occurred after Rhee left, and (b) from the current high of 80,854 from a nadir of 70,922 is only a 14% increase, or about one-seventh, not a doubling of population.

One’s eye wouldn’t trick one so if one used a scale that went all the way to zero on the vertical axis. (BTW, this is precisely one of Huff’s methods of lying with statistics!) So here is what I think is a fairer way to represent the data, with a scale that goes from 0 to 90,000.

You can see that over the past decade or so, there was a modest drop, followed by a modest rise. These are NOT huge changes, folks!

Is there something special and weird going on? Not really. The population of DC is rebounding somewhat, as well. Take a look at this graph prepared by Google and the US Census Bureau, not by me:

I hope it’s not a surprise that the school enrollment numbers and the total DC population of all ages do not move in lockstep! But, as a general rule, if you get more adults, they have a mysterious way of making babies, and those little’uns eventually do grow up and go to school somewhere!