Big Drop in Pittsburgh PSSA Scores: Why?
Last week the state reported that more stringent security measures had been implemented for 2012 PSSA testing as a result of evidence that scores were being enhanced by cheating in some Pennsylvania districts-either by the students or by school personnel. The state report suggested that several schools districts, including Pittsburgh, would likely see test scores drop significantly as a result of the tightened security.
This week the 2012 results for each school in Pittsburgh were released. Of the 59 schools taking the tests, 43 saw declines in the math scores from 2011 with 29 schools having 5 percent or more of students failing to show grade level proficiency in math. Meanwhile, 14 schools had double digit declines in the number of students reaching grade level proficiency. Manchester and Grandview posted the sharpest drops with over 22 percent fewer students scoring at the proficient level in math compared to 2011 results. For African-American students, 35 schools saw fewer students scoring at the proficient level, with 17 of those declines over 10 percent.
On the reading test, 41 schools saw declines in the numbers of students scoring at grade level proficiency. Of those declines, 26 showed 5 percent fewer students scoring at proficiency level while 15 had decreases of 10 percent or greater in the number of students achieving proficiency or better. African-American student results were somewhat worse than the overall with 36 schools having lower percentage proficient level scores. Ten of those were double digit declines. Grandview (-22.5), Beechwood (-22.1), and Brookline (-20.2) posted the biggest test result drops by African-American students. On the positive side, Arlington (+18.6) and Concord (+12.9) showed nice gains, albeit from very low percentage proficiency scores in 2011.
Statistically, one can expect year to year variations in student scoring since there are different students each year in each grade level and there are always random fluctuations in any complex process with large numbers of variables. However, declines of the magnitude and in the number of schools experienced by the Pittsburgh School District point to some structural change in the testing process. In this case, the most likely change contributing to the large one year drop in scores was the tightening of the security to ensure unimpeachable testing and results.
Most worrisome, the largest average 2011 to 2012 declines in the number of proficient students in reading occurred in the K-5 grades for African-American students. Assuming the 2012 scores are closer to being correct owing to the new security measures, it suggests that the 2011 scores, and perhaps earlier years’ scores as reported, were too high possibly as a result of inappropriate actions by students or school personnel.
Beyond the loss of overall achievement as indicated by the 2012 test scores, there is a very troubling situation regarding the enormous falloff in academic achievement the longer students are in Pittsburgh schools-and most likely many other Pennsylvania districts as well. Consider the math results for African-American students in kindergarten through fifth grade compared to the math proficiency of 11th graders. On average, for the 21 schools for kindergarten through 5th grade, 60 percent of all students scored at the proficient level or higher-ranging from a low of 32 percent at Woolslair to a high of 87 percent at Fulton. Meanwhile, for 11th graders at seven high schools only 19 percent of students scored at the proficient level or higher-Carrick was highest at 30 percent, Westinghouse lowest at 7.7 percent. The drop from 60 percent proficient in K-5 to 19 percent in 11th grade is fairly strong evidence that the intervening years between elementary and high school have been extraordinarily unproductive in terms of students’ educational advancement. Beyond the remarkable failing of the 11th graders to score well, one should not brag too much about the 60 percent proficiency scores posted by the K-5 group-especially in light of the big drop in scores at half of the K-5 schools from 2011 to 2012.
On the reading exam, the gap between K-5 and 11th grade scoring proficient or better by African-American students was not as dramatic as math but still appreciable. In K-5, the 21 school average was 48 percent proficient. Scores ranged from 29 percent at Woolslair to a high of 71 percent at Phillips-which saw its 2012 percentage proficient rise by 13 point from the 2011 result. In 11th grade the percent proficient was 34 percent, 14 percent below K-5. The best performing high school for African-American students in reading was Allderdice (44 percent proficient) and the worst school was Perry (28 percent). Even worse for Perry, it saw a hefty 18 point decline from 2011’s score.
In an otherwise very bleak picture for the Pittsburgh School District as regards its academic performance there were few positives in the 2012 PSSA test results. A handful of schools did manage to post respectable gains from 2011. But beyond those few schools the rest of the story is uniformly dreary. Major declines in 2012 that call into question the validity of results from earlier years along with the possibility that in earlier years mischief was afoot in the test taking or the post-test handling of the exams-a very unsettling development if true. Further, there are the perennial problems of extraordinarily weak performance in the high schools and the so-so performance in the earlier grades at many schools.