Friday, November 02, 2012

K-PREP Percentiles Gives Public School Foes a New Hammer

The most serious problem with NCLB was its unrealistic expectations for student achievement under the requirement of Adequate Yearly Progress (AYP) which called for 100 percent proficiency in reading and mathematics by 2014. So it was something of a relief when Kentucky asked for and was granted an NCLB waiver. Kentucky Education Commissioner Terry Holliday was one of the nation's most outspoken opponents of AYP and was the first to jump off the bus. Holliday complained to the Herald-Leader that AYP was a "wrong" measure because schools are punished if they fail to reach even one AYP goal. He was correct. With the federal waiver, he earned the chance to build a better mouse trap.

So it was a big surprise when we learned that the state of Kentucky would establish another wrong measure for its new Kentucky Performance Rating for Educational Progress (K-PREP) tests and the Unbridled Learning Accountability system.  

It was easy to see that almost all schools would fail to meet NCLB requirements. As early as 2003, the National Assessment of Educational Progress (NAEP) showed that no state or large district had anything close to 100% of their students performing at the basic NAEP achievement level, much less the NAEP proficient level. Schools that failed to meet AYP were initially categorized as “failing.” But under pressure from - virtually everywhere - it was recast as “Improvement Status” which called for a series of interventions that become more extensive for the lowest schools. Improvement status schools were required to write improvement plans that were approved by the state. Thousands of schools faced demoralizing targets while the political attack on public schools advanced, and schools funds ebbed away. 

As WKU Public Radio reported, public schools across Kentucky will be scored on a scale of 1-100 under the new Unbridled Learning Accountability system. The top 90 percent of schools will be labeled Distinguished; Proficient for schools in the 70-89th percentiles, and Needs Improvement for the remaining schools. Every year thereafter, 70 percent of Kentucky’s schools will fail to reach proficiency - not because they failed to improve or reach the proficiency cut score - but because the system withholds the designation from all but the top 30%.

“I think there’s a big difference between the ‘Needs Improvement’ category and No Child Left Behind which labeled these schools as failing", says State Education Commissioner Terry Holliday. "These schools are not failing, they just have particular components that they need to work on.”

Holliday added that the new system raises the bar on student achievement for school districts by taking into consideration student growth and college and career readiness. “They will sit down and say where are our priorities to work on and then they’ll set about writing a plan that involves parents and teachers.” Holliday said schools in the “Needs Improvement” category will have follow-up reviews from the Department of Education. 

In terms of incentivising schools to produce the outcomes we hope to see for Kentucky school kids, is there really such a difference between AYP and K-PREP?

Richard Innes, education analyst for the Bluegrass Institute for Public Policy, told Associated Press reporter Roger Alford that people had lost confidence in the former [Kentucky] test because many of the students who performed well on it still were ending up in remedial classes in their first year of college...

Holliday encouraged educators and parents to consider this year's scores a starting point for improvement. "Although more than two-thirds of schools and districts are in the "needs improvement" category, this is not an indicator of failure," Holliday said.

Innis didn't see it in such a positive light. "I think Kentucky parents have every right to be concerned at this point," he said.

Of course, Innes is unconcerned by the cut score. His only regret is that the arbitrary choice of the 70th percentile as the criterion point isn't higher so that he could declare even more public schools to be failures - and so that BIPPS could argue further for moving public funds into private hands through vouchers. The Unbridled Learning Accountability system just handed Innes a bigger hammer, and he knows how to use it.

As the Courier-Journal reported, the new testing system is a key component of Senate Bill 1, enacted in 2009 by the Kentucky General Assembly, which mandated a new public school assessment and accountability program beginning with the 2011-12 school year. The law also called for more rigorous academic standards, aimed at having students develop a deeper understanding of concepts, and not just the regurgitation of facts or formulas.

Kentucky schools responded by adopting Common Core Standards in reading and math, which are designed to be more rigorous and better aligned with college coursework and 21st-century workplace skills. Kentucky was the first to adopt those standards; since then, 46 other states have followed suit....

The new standards were first taught last school year, and Kentucky was the first to test them last spring. Bob Rodosky, director of testing and accountability for the Jefferson County Public Schools, has called the new accountability system “very ambitious.”

“The stakes are much higher,” Rodosky said. “People are expecting much more. And because we were among the first to implement the higher standards, everyone will be watching us.” The new tests will also be used to judge schools’ performance on reading and math standards that are mandated by the federal No Child Left Behind law. The law has come under increasing fire for holding schools to what some contend are unrealistic expectations. And the Obama administration has called on Congress to overhaul it.

Kentucky is one of 34 states that have been granted a waiver from No Child Left Behind, allowing it to use its own system to determine whether its schools are making sufficient progress. The law requires states to use a single accountability system for public schools to determine whether all students, as well as individual subgroups of students, are making progress toward meeting state academic content standards.

When Danville Superintendent Carmen Coleman gave the Herald-Leader something less than a full-throated endorsement of the system, we're sure the Commissioner was disappointed. Coleman said that while she supports higher standards, “this system is extremely confusing. We've gone from a system that allowed as many to achieve as possible. This system won't allow that. I believe this is going to be devastating for some schools and districts." 

Holliday worried that stories like the one KSN&C posted on Sunday would present a “completely inaccurate portrayal” of the new system. KSN&C is concerned with accuracy and asked KDE to let us know what facts we got wrong. We will pass along whatever response we receive.

We really want to be fair here. We know that Kentucky has a lot at stake in its effort to implement the new K-PREP tests. We are also aware (and support) that the feds required high standards, and that the state had to set cut scores to define the various accountability categories. Further, we are very glad to hear that those cut scores will not be moving. (The single most frustrating and unproductive period of assessment I ever lived through was the 6 or 7 years of the KIRIS test. The schools lacked a curriculum or a stable target, and it was terrible.)

Today’s principals need a way to rally the troops. They need to know that the entire school system (state and local) is fully invested in their success. The teachers have to be shown a meaningful (directly connected to student success) and attainable goal. All schools that meet the proficiency requirements ought to receive the hard-earned designation without also having to do so at the expense of any other schools.

We agree that – theoretically - every school in the state has the opportunity to exceed the proficiency cut score. But it is also true that a full two-thirds of the schools will be prevented from achieving a designation of proficiency. The state’s use of percentiles to rank the schools each year and withhold the designation of proficient from all but the top 30% of schools will be perceived by many as unfair.

Unfortunately, this system will create winners and losers - mostly losers - and it’s not that hard to foresee negative long-term outcomes for many teachers and students – all too similar to No Child Left Behind. 

Will that be devastating for some? Yes, we suspect so - especially for those who care deeply and personally about their work in the schools. Coleman had it about right. It’s not just about numbers. It is about respecting the hard work we have asked teachers to do and encouraging their success - not hammering the schools into submission. There are reasons why so many promising young teachers are leaving the field. 

In the Unbridled Learning Accountability system, what needs to be unbridled is the measure that constitutes proficiency, and the schools that will be short-changed as a result. 


Anonymous said...

Gap score is a screwed up way of looking at a school. The fewer at risk kids you the more disproportionate the percentage is in relationship to school effectiveness and the more inaccurate. How can you reasonably compare gap kids of Oldham County and Owsley County with the same weight in a collective school score?

GAP is one of the most over hyped elements in education reform. From a racial point of view is blantant racist stereotyping. We are to assume that all hispanics, african americans and native americans are all automatically the same academically disadvantages? I know that is not a popular view point, but really, if you are raising both the lower and higher performers in any grouping from some GAP baseline do you honestly expect to accelerate ELL or special needs students to perform at the same level as your highest performing kids. Its ridiculous both in practice and theory to lump kids into these identified pools instead of simply dealing with them as individuals. Why don't we look at single parent families or student households with less HS education or less as at risk groups? It just political cherry picking. Wasn't that long ago that KDE only recognized african americans as the only at risk minority.

Just teach all kids well and quit creating these artificial comparisons

Anonymous said...

Why not create the same mixture of students in each school? That way the mix,ability,minority,socioeconomic etc is the same percentage in each school. Then all schools should have the same score.

Anonymous said...

I think the point which the first contributor is making is that if you are going to rank schools there is going to be limited reliability in the GAP component.

Assume you were comparing two schools each theoretically with 400 assessed students and one school had 75% free and reduced lunch and another had 25% free and reduced lunch. Further, if both schools have roughly equal numbers of non duplicated special education, LEP and minority, then one school's score is going to be based on the performance of 100 students where as another is going to have 300 as its representative group (basically most of the student body). Seems like schools with low GAP populations are having 1/3 of their score being based on a very small percentage of their total school population.