Molasses, Snails, and the ELA Achievement Gap

When the 2009 state English Language Arts test scores were released by New York City last month, the news seemed quite good.  The percentage of students judged proficient increased in every grade, and proficiency rates were higher for Black, Hispanic and white students, leading Mayor Mike Bloomberg and Chancellor Joel Klein to proclaim great pleasure at the closing of the achievement gap.  Students in NYC were “closing the shameful achievement gap faster than ever,” said Mayor Mike.  Chancellor Klein reiterated this, saying, “”I’m especially pleased that we are closing the shameful achievement gap between black and Hispanic students and their white peers faster than ever.”

Just how much progress has New York City made in closing the achievement gap?  The figure below tells the story.  In 2003, the average white student in grades three through eight scored .74 standard deviations higher than the average Black student and the average Hispanic student, and the average Asian student scored .70 standard deviations higher than the average Black student and the average Hispanic student.  Put differently, in 2003, the average Black student and the average Hispanic scored at the 43^rd percentile of the citywide distribution, whereas the average white student scored at the 71^st percentile, and the average Asian student at the 70^th percentile.

By 2009, the Black-white gap had shrunk by 14%, with the average white student scoring .64 standard deviations higher than the average black student.  The Hispanic-white gap had shrunk by 10%, a bit less, with a remaining gap of .67 standard deviations between the average Hispanic student and the average white student.  There was little shrinkage between 2003 and 2009 in the gap between Asian students and Black and Hispanic students—8% and 3%, respectively, over this six-year period of prime-time Bloomberg/Klein reforms.  In 2009, the average Asian student scored .65 standard deviations above the average Black student, and .68 standard deviations above the average Hispanic student.

Where would we locate Black, Hispanic, White and Asian students in the 2009 citywide distribution of ELA performance?  In 2009, the average Black student scored at the 43^rd percentile of the citywide distribution, and the average Hispanic student at the 42^nd percentile.  In contrast, the average white student and the average Asian student scored at the 68^th percentile of the citywide distribution.  Contrasted with the relative position of these racial and ethnic groups in 2003, you can see just how slow the progress in closing the achievement gap has been.

Reducing the gaps in test scores among students of differing racial and ethnic backgrounds has proven to be extremely difficult in American society, and even small reductions are to be celebrated.  It’s unfortunate that Mayor Bloomberg and Chancellor Klein feel compelled to distort the magnitude of the progress made in New York City in order to garner support for the school reform initiatives they have championed.