Just a Little Statistical Competence…Please

Quotes that make me cringe:

In 2014, the state test will switch to a new, more rigorous exam that will align with the Common Core, a set of curriculum standards adopted by states across the country to better prepare students for college.

Donoso, who is replacing Charles Payne, the interim chief education officer, is responsible for developing the district’s curriculum strategy and working with school leaders to carry it out. Her main focus in the coming years will be implementing the Common Core curriculum, which is designed to develop analytical skills beyond those currently tested on the ISAT.

Radner said CPS “needs to step it up” or scores are going to crash when the new test is given in 2014, calling the change “the biggest shift I’ve ever seen.”

“We can’t be complacent,” she said “This is a whole different generation of standards and assessment.”

 

Scores won’t crash.  They will change and a bunch of statistically illiterate people will be outraged because the numbers meeting expectations will be lower.  That has nothing to do with achievement levels falling–it has to do with using different standards.  In fact, scores won’t be directly compared at all.  What will be compared is the percent of students at level, below level, and above level.  This is a pretty much pointless exercise in comparing the old standards to new standards given they are very different. It is extremely likely the CPS will do poorly on such measures not because of anything the CPS does, but who the students are in the CPS.

Look at the PISA (international tests comparing achievement across countries) scores comparing US schools of a poverty level to nations of particular poverty levels:

 

Country Poverty Rate PISA score
United States < 10% 551
Finland 3.4% 536
Netherlands 9.0% 508
Belgium 6.7% 506
United States 10% – 24.9% 527
Canada 13.6% 524
New Zealand 16.3% 521
Japan 14.3% 520
Australia 11.6% 515
United States 25-49.9% 502
Estonia 501
Switzerland 501
Poland 500
United States 50-74.9% 471
Austria 471
Turkey 464
Chile 449
United States >75% 446
Mexico 425

Of course, only Mexico fits in the final category, but notice that the US schools in each level of poverty perform at the very top of the world. The problem is that few other countries allow high percentages of their children to live in poverty as the US does.  The United States, adjusted for poverty level, has students perform better than counterparts in other countries in every grouping.
That doesn’t mean we don’t need to provide the best education we can for the kids in high poverty schools, but the reality is those efforts will only be mildly successful because the chaos that accompanies poverty in those kids’ home lives will limit how the average student performs.  More time in the classroom may be a good thing for many of these kids because it is more time in a structured environment that kids need and crave.

The students in poverty are able to achieve, but poverty itself is a limiting factor.  Poverty breeds chaos in the home and that lack of structure will lead to poorer performance for those students. Some students overcome that and that’s wonderful, but the reality is that on average, socioeconomic status is the biggest predictor of student performance.  We can make marginal improvements in high poverty populations, but ultimately, existing in poverty will put a limit on how much improvement can take place.

Any outrage at the system that our great paternalistic and condescending news organizations want to throw out would be much better focused on alleviating poverty first. The CPS needs improvement in many ways, but all of these pseudo privatized solutions are not going to solve the core problems and the data already show that they are not significantly better than the CPS.

Leave a Reply

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