There's nothing standard about standarized scores--or why you can't trust the scores schools report

Seeing Geoffrey Canada featured in the pro-charter schools documentary, “Waiting for Superman”, and then reading about his richly-endowed, superbly staffed, 11 month a year operation, Harlem Children’s zone schools so-so educational achievements, (http://www.nytimes.com/2010/10/13/education/13harlem.html New York Times, Oct 1s) left this ex-Harlem public school teacher wondering what was wrong with his approach and what more needs to be done to improve our schools.

We can all agree that we have some serious school problems, particularly when it comes to Black, Hispanic and immigrant students: 50% dropout rates in urban high schools and standardized scores in reading and math 50% below that of white and Asian students. While many can agree that there is a problem, there is no agreement as to what the solution may be. Better teachers? More charters? Less unions? Or what?

What makes schools successful? Good teachers are a necessary but not sufficient part, despite all the trashing teacher unions get, there is some evidence that unions seem to be important (southern public schools generally do not have unionized teachers and score lower compared to northern schools that do have unionized teachers. Moreover, Finland, the country everyone raves about as having the number one educational system has both teacher unions and tenure.) Charters -in general, 4 out of 5 charters do not outperform public schools and here in Boston, the 4 charter high schools have dropout rates of 50%.

Harlem Children’s Zones poor to average standardized test results raises questions about standardized tests and their reliability. Previous to the 2010 New York State test, (http://www.nytimes.com/2010/10/11/education/11scores.html New York Times, Oct. 10^th) the zone did fabulously well-researchers claimed that the Zone’s Harlem students even outperformed the tony students in Scarsdale, NY. In 2010 the state changed the test—raised the number of correct answers to pass the test and, more dramatically, no longer distributed the test questions, which remained basically the same year after year. The implication is that teachers had the test questions and simply taught them to the students and that’s why the students did so well. Moreover, there is a long history of educational research that suggests that high-stakes testing leads to massive massage of the data. In other words, cheating. If your job is on the line or bonus might be coming your way as a result of a test, there is all the reason to change test results in your favor.

How about affordable housing and stable jobs for parents? It's no small thing that there is a high correlation between SAT scores and family income, heck, there is high correlation between zip codes and SAT scores. The richer the parents, and hence the better neighborhood they live in, the better the score. Many a college admissions officer has followed this simple rubric.

For the most part, our so-called "failing schools" occur in impoverished minority/immigrant parts of the urban landscape. That should tell you something; poverty can have a debilitating impact on educational achievement ! Students of poverty move often, have poor job prospects and, as a result, often drop out. High school drop-out rates in urban high schools in often 50%.

Give parents jobs, provide affordable housing, provide a career path for impoverished students and you have the means of solving the problem. Who will pay for it? Many well to do philanthropists are now pouring money into “failing schools.” So far, they have not come up with the right solution to the problem. A jobs and housing program might be just such a solution.