9486 in the collection
FCAT bombshell
Everyone in Florida who cares about public education shouldl be a member of FCAR.
Gloria Pipkin and John L. Perry
In the four years we've been publishing F-TREND, we've never seen FCAT news as explosive as what came to light this week. FCAT reading, math, and science scores in all grades in which they're administered were released this week, but they were overshadowed by the Florida Department of Education's admission--a year after the fact--that the 2006 grade 3 reading FCAT scores were artificially inflated because the test was easier than those in previous years.
What did the DOE know, and when did they know it?
When the 2007 scores for grade 3 FCAT were released two weeks ago, the percentage of students scoring at or above levels deemed "proficient" on the reading test dropped from 75% to 69% statewide, with 60 of 67 districts showing declines after six years of steady improvement.
Superintendents and assessment coordinators in many districts began pressing for answers, and one of the first responses from Jeanine Blomberg, acting commissioner of education, was that last year's test might have been easier. Shortly after, former commissioner of education John Winn acknowledged that these concerns had arisen in 2006 when the grade reading scores came in, but, according to Winn, the DOE and Harcourt, the test developer, were unable to detect any significant differences in the tests using traditional methods of statistical equating.
Common Item Equating
How do we know when a grade level test is easier or more difficult than one previously administered? Bob Schaeffer, national education director for Fair Test, resident of Sanibel, and member of FCAR, explained the procedure for equating tests from year to year this way: "The process they appear to have been using is called 'common item equating.' The way it works is that a set of identical questions (often called a 'testlet') is included in the exams for consecutive years. The rationale is that if average performance on these common items increases by a certain percentage, it is likely that student mastery of whatever it is the test measures has also improved by a similar proportion.
Unfortunately, that assumption holds only if the items are embedded in exactly the same place in the test year after year--item order can significantly alter performance." As Bob noted, there have been published reports that the placement of the "testlet" items for the 2006 test varied wasn't the same as the previous test.
Life-altering decisions made on the basis of a single flawed and fallible test
A constant theme running through all of FCAR's advocacy efforts over the last seven years has been that a single test is not a fair and reliable measure of any child's strengths and needs. The latest FCAT debacle only deepens our conviction. Even when there was strong evidence that something was rotten in FCATland, the politicians and bureaucrats kept us in the dark during an election year when the FCAT was a major issue.
Redressing the wrongs of FCAT, assessing the assessors, and supporting constructive assessment
The Florida Coalition for Assessment Reform will shortly propose a plan for digging deep into the flawed FCATs, assessing the assessors, and holding the State accountable to Florida children and families. FCAR's mission statement remains a worthy goal: open, broad-based, constructive assessment that reflects the complexity of learning and respects the diversity of learners.
— Gloria Pipkin and John L. Perry
F-Trend: Florida Test Reform EMail News Digest
2007-05-26
INDEX OF OUTRAGES
Pages: 380
[1] 2 3 4 5 6 Next >> Last >>
