|
|
9486 in the collection
TAKS predictions leave lots of room for error
Ohanian Comment: Who came up with the idea of "standardized test score as destiny?"
Reader Comment [about Aldine, the Broad prize winner for 'best district in the USA,' the district that banned my website--until I complained.:
I taught in Aldine ISD (Houston) for seven years; I found it interesting during campaign season when all employees were sent an email from a group of candidates who wanted to unseat a few of the existing school board members. The emails talked about how district administrators had been disproportionally paying themselves giant TAKS bonuses, while teachers received meager bonuses. The emails contained links that directed you a website that showed all Aldine ISD employees and their TAKS bonuses.
I was shocked to see then-superintendent, Nadine Kujawa, already making around $200K per year, received $25K in TAKS bonus, while Aldine ISD's general counsel, Jonathan Hantke, received a $15K bonus, yet he does not teach one class. My principal, Milo Ortiz, who preached to the staff about not being absent during the school year, yet he was gone the Friday before and the Monday after, the Super Bowl in Houston so he could be apart of Super Bowl pre-game and halftime shows by holding a rope so the float would not fly away, was paid $12K because our school was "recognized" by TEA.
A middle school principal, Nancy Blackwell, received $20K extra in TAKS bonus money because her school, Hambrick, was "exemplary." ... Each teacher's bonus was $471, yet we are the ones in the classrooms, battling each day.
These are just a few examples of school district administrators who DO NOT want the "Texas Projection Measure" to go away, because it makes their schools look great, while they employ the same teachers, basically do the same things, while educating kids from the same neighborhoods. Yet, when you water down the test-passing standards so badly, no wonder it appears the schools are improving. Meanwhile, upper administrators pocket huge amounts of tax-payer money.
I'm very thankful to Rep. Scott Hochberg for exposing this corruption and to Houston Chronicle columnist, Rick Casey, for diligently reporting these stories in the last few months.
By Holly K. Hacker and Jeffrey WeissOLLY
TAKS report cards sent this spring to Texas parents included a new piece of information: predictions of whether their kids are on track to be ready for college.
For some parents, the projections brought the disturbing news that their children might lose their "commended" ratings. But the state didn't tell parents that some of those warnings could be false.
School districts were left in the same bind: They received projections that for some grades and subjects could be off by as much as 20 percentage points.
Some critics see the commended predictions as one more reason to scrap the Texas Projection Measure altogether.
"If you don't have confidence that the measures you're getting from the state are accurate, then even the legitimate successes are subject to question. And that's not fair to the educators and the parents who work so hard," said Rep. Scott Hochberg, D-Houston.
The Texas Projection Measure was implemented last year to allow schools to count as passing any student who failed a TAKS test but was projected to pass future tests. Doing so helped more than 2,000 schools earn higher school performance ratings.
Projections about commended ratings were added this year to give parents and educators an early look at how students may perform on harder tests scheduled to start in three years. A commended score on the TAKS indicates mastery of a subject.
Raising concerns
Hochberg, chairman of a House subcommittee on education, raised doubts about the accuracy of the predicted TAKS passing rates during a hearing last month. That hearing did not touch on the commended projections. But The Dallas Morning News examined those commended predictions and found that many show a troubling decline.
On Thursday, Education Commissioner Robert Scott told the state Board of Education that the Texas Projection Measure is "statistically accurate, valid and reasonable." But he has said that he'll consider scrapping it or scaling back its use.
Other top agency officials say the predictions are inevitably imperfect and shouldn't cause alarm.
"You would not want to draw the conclusion that we believe performance in the state of Texas is going to plunge," said Gloria Zyskowski, TEA's deputy associate commissioner for assessment. Nor should parents panic if their children are predicted to fall from "commended" status, she said.
"I would just interpret it with a grain of salt," she said.
Zyskowski said any statistical measure used to predict future test scores won't be totally accurate, especially for children already achieving at high levels. And officials say they selected the best model available with feedback from testing experts, teachers and others.
State officials knew about limitations of the Texas Projection Measure before the data was distributed to parents and school districts. "Previous analyses indicate that underprojecting commended performance by 10 percent to 20 percent is not unusual," TEA officials wrote in response to questions from the News. That information, however, was not included on the report card sent to parents or disclosed to school districts.
One education policymaker said the predictions are too inaccurate to help families or educators.
"It's sending mixed signals to schools, to parents and to kids themselves," said Daria Hall, director of K-12 policy for the Education Trust, a nonprofit organization that advocates for poor and minority students. Hall served on a federal panel that reviewed Texas' new model, which the U.S. Department of Education approved in January 2009.
"These data at their core ought to be providing useful and quite clear information to parents, teachers and students about where they are and where they need to go," Hall said.
Major drops
The Texas Projection Measure predicts student scores on future Texas Assessment of Knowledge and Skills tests based on the students' current scores and average scores at their schools. Yet some predicted commended rates depart radically from recent trends.
For example, 46 percent of Texas students who took the eighth-grade reading TAKS this spring scored commended. Next spring, only 27 percent of students are expected to score commended on the same test – an unprecedented 19-point drop.
But because the system may be underprojecting by up to 20 percentage points, it's possible the commended rate could actually go up by 1 point.
State officials say that commended rates will have less accurate predictions than the passing rates. That's because of a phenomenon called "regression to the mean," a tendency for students who scored below average to do better the next time, and for those who scored above average to do worse.
They say the measure's precision also depends on how high the "commended" bar is set on each test. The higher the bar, the lower the precision.
Despite the lack of precision, Zyskowski said the projections are a valuable tool when assessing student progress. "It's a powerful piece of information, but it's just one piece of information about the student," she said.
School ratings
The Texas Education Agency developed the projection model for the federal accountability system called No Child Left Behind, then adopted it to meet a state legislative mandate to measure student gains on state exams.
Last year, the agency began using the projection model to help schools improve their state ratings. Many campuses are expected to benefit again when the state releases the 2010 ratings next week.
Schools receive credit for students who failed the TAKS but are predicted to pass in a year or two or three. The measure can only help campuses, even when the predictions don't pan out. Say that a student fails the fourth-grade math test but is predicted to pass in fifth grade. Even if that student fails that second test, meaning the prediction was wrong, the school pays no penalty.
Skeptics have concerns about false predictions. A Texas Education Agency analysis found that, depending on the grade and subject, anywhere from 19 percent to 48 percent of students who failed the TAKS in 2008 were predicted to pass in 2009 but did not – a fact that surfaced during Hochberg's subcommittee hearing.
Agency officials say that next week they will disclose how projections made in 2009 stacked up against 2010 TAKS results.
— Holly K. Hacker and Jeffrey Weiss
Dallas Morning News
2010-07-25
http://www.dallasnews.com/sharedcontent/dws/dn/latestnews/stories/0722410dnmettpm.3d316cb.html
INDEX OF OUTRAGES
Pages: 380
[1] 2 3 4 5 6 Next >> Last >>
|
