in the collection
Wisconsin New Plan: Proficiency Is In the Interpretation
The percentage of Wisconsin fourth-, eighth- and 10th-graders considered academically proficient is expected to jump sharply in many instances, particularly in eighth-grade and 10th-grade math, when state test data is released in several weeks.
That will be true whether there is any improvement in the actual performance of students.
State Department of Public Instruction officials argue strongly that they are not lowering the bar for defining students as proficient. They say they are setting it at more realistic and appropriate levels after an in-depth process of redefining what constitutes proficient performance.
However it is described, new tests and new "cut scores" - the minimum scores needed to meet various definitions of a student's performance - have opened the way for Wisconsin student performance to appear decidedly more rosy.
That will help the state and, presumably, many individual schools meet goals set under the sweeping federal education law called No Child Left Behind, and avoid sanctions set by the law. It appears Wisconsin will make its goals for the next several years without difficulty.
Comparing the new cut scores with the ones used since 1997 is a dubious exercise, according to testing experts, because the tests have changed. On the other hand, the scores are listed on a scale that is theoretically intended to be comparable across different tests and different years.
Most scores drop
And comparison of the new and old cut scores suggests a decided trend. The new scores are lower than the old ones for all five areas (reading, language arts, math, science and social studies) reported for fourth-graders and all five for eighth-graders. They are lower than the old ones for two areas for 10th-graders, but higher in two and identical in one.
The differences are biggest in math. Under the prior cut scores, a score for an eighth-grader of 718 on a scale of 300 to 800 was needed to be considered proficient. The new score is 682.
But even if the percentage of students rated proficient goes up, the number of Wisconsin schools labeled as "identified for improvement," as officials diplomatically put it, is expected to go up, perhaps by a substantial number, when that list is released along with the results at the end of June. There were 69 schools on the list this year, when the process of sanctions under the federal law began having some effect.
State School Superintendent Elizabeth Burmaster said the number to be on the list has not yet been determined, but she expects it to go up, given the law's stringent requirements. Students in those schools could be eligible for transfer to other schools or federally funded private help outside of school and, if schools remain on the list for several years, they could face steps such as being turned into charter schools or having their staffs replaced.
The changes in the percentage of students who will be described as proficient in the new round of test results does reflect better actual performance across the state as school reforms have positive effect, Burmaster said.
All in the interpretation
But the reshaping of both the tests students took during this school year and the way the results are interpreted is eye-catching.
State education officials say they were required to use new tests this year because of demands of federal officials that actually go back to before the No Child Left Behind law was passed. And that meant they had to come up with new cut scores for determining which of four categories a student's work falls into. The categories are labeled "advanced," "proficient," "basic" and "minimal."
The general measuring stick for the performance of schools, school districts or other groupings of students is what percentage of students are rated proficient or advanced in each of the five subject areas.
In a plan recently approved by federal education officials, Wisconsin committed itself to goals of at least 37% of students being rated as proficient in math and at least 61% in reading for this year and next year. Those figures are based on the percentage who were proficient in recent years.
The percentage figures are scheduled to rise to 47.5% in math and 67.5% in reading in 2004-'05 and continue rising until 2013-'14, when 100% of students are supposed to be proficient in both areas.
At the heart of the new cut scores and proficiency percentages is a session convened by the Department of Public Instruction in February in Madison. About 240 educators from around the state took part in a complex effort to decide where to set the cut scores for the tests that were given statewide in November.
Proficient category grows
Results of that work indicate that the percentage of students considered proficient or above will be up 25 points or more in some instances - and down a few points in a handful of cases.
For example, in eighth-grade math, 44% of students were considered proficient a year ago. The number for this year is likely to be around 70%. But in language arts, 70% of students were considered proficient last year, and that number is expected to be at least 6 points lower this year.
DPI officials say emphatically that the goal of the Madison sessions was not to lower the bar for Wisconsin students and that no comparison was made in the sessions between current cut scores and those used in prior years.
Burmaster said that the prior cut scores, established in 1997, reflected "kind of an ideal" and did not carry any high-stakes consequences. Educators around the state have complained heavily in recent years, particularly in math, that the bar was set so high that relatively few students were rated as proficient.
Burmaster said the new cut scores "are a more accurate reflection" of what Wisconsin students are expected to master to meet a valid definition of proficiency.
Andrew Porter, a University of Wisconsin-Madison professor who is a national leader in education research, called the new cut scores "not too high and not too low - Goldilocks standards, just right."
The cut scores set in 1997, he said, reflected a period when "everybody wanted to set high standards." He added, "In my mind, they sometimes set standards unrealistically high."
Porter said, "We're a lot closer the second time around than we were before. . . . I think they did a good thing. But I know they're going to be criticized in some quarters for having lowered the bar (in a way) that is not helpful. When I say they lowered the bar, I see it as helpful."
Department of Public Instruction leaders say they do not know yet what percentage of students will end up being rated as proficient in each category.
But rough data released on the impact of the cut scores strongly suggests there will be notable increases in the proficiency picture for 10th-graders in all five subject areas, especially math and science.
For eighth-graders, there are likely to be increases in reading, science, social studies and especially math, but a decrease in language arts.
For fourth-graders, the data suggests there will be jumps in the percentage proficient or advanced in all areas, except perhaps science.
The results to be released in coming weeks are for tests given in November.
Officials hope in future years to return individual test results to parents and schools and to release public summary data much earlier than this year. The new tests and resetting the cut scores slowed the process.
Alan J. Borsuk
Reforms in test scoring may make state students look better But more schools may be targeted for improvement
June 10, 2003
INDEX OF THE EGGPLANT