I know I'm not alone in wondering: with 55,000 tenth graders who had their scores raised by the (now caught) error, why is it that not a single one left the "failure" catagory? How is that possible?
I asked the Administration this at Thursday's meeting, and here's the answer: while the chart that converted the raw score (how many a student got wrong) to scaled score (something in the 200's) did change between 2009 and 2010 (thus causing the error when the company used the wrong chart), the number of answers wrong that would give a student a failing grade did NOT change between 2009 and 2010. It DID change for the other categories. Thus, if you took the test in 2009 and got a raw score of either 13, 14, 15, or 16, you got a scaled score of 218; if you took the test in 2010, the same is true.
In every other category--Needs Improvement, Proficient, Advanced--where the break between those fell changed between 2009 and 2010. The error in the chart thus gave kids right on the line the chance to move up.
It seems counterintuitive (and, needless to say, cruel), but this is a feature, not a bug, of any standardized test -- in fact, it is why the test is called "standardized".
ReplyDeleteThat is, a "standardized test" is one such that a certain number of students are guaranteed to fail every year. (The exact number of failures is determined prior to the test, usually as a percentage of the total student body.)
Not coincidentally, the same *number* of students -- tho not the same students, obviously -- are guaranteed to earn the top score (in the original scoring matrix, not the corrected one). The use of a standardized test also explains why no scores decreased, since the test was 'graded on the curve'.
The problem in this case was not the test, but its' interpreters, who made a basic -- and in my view, disqualifying -- error. It is sad that no individual in the lowest category could move up, but it is not mysterious: It is a predictable, required outcome of a standardized test.