06/04/06 — What's in a test score? GHS under scrutiny but what do numbers really mean?

View Archive

What's in a test score? GHS under scrutiny but what do numbers really mean?

By Phyllis Moore
Published in News on June 4, 2006 2:07 AM

Standardized tests, achievement tests, end-of-course and end-of-grade tests.

The list of benchmarks used to measure student progress grows longer and, depending on which one is used, can determine anything from whether one passes a course, a grade level or goes on to college. Most recently, they also became a consideration of whether or not a school will remain open.

Test scores brought Goldsboro High School under even more scrutiny in recent months, when Judge Howard Manning placed it on a list of 19 schools around the state he said could be closed in the fall if drastic improvements are not made.

Dr. Craig McFadden, assistant superintendent for accountability and student services with Wayne County Public Schools, said it is understandably frustrating for educators these days.



Student achievement tests are re-standardized every few years to adapt to changes in curriculum. Tests are traditionally designed to measure growth and performance. The two might not be synonymous, however.

McFadden likens it to measuring the physical growth of a child. It is easy to stand that child next to a wall and see how much he has grown from one year to the next. The challenge comes in measuring whether the student lives up to his potential, or performance level, he said.

"From a statistician perspective, we have this wonderful scale of scores that we have. There is a lot of precision to it, and we have broken it into a pass/fail," he said. "That's how performance level is measured. All the performance measures are based like that."

In addition to how well a student scores on a test, teachers must assess whether they are considered a "grade level child," McFadden said. If they are determined to be at level 1 or 2, that means they are performing below what is required. Level 3 and 4 are considered to be more proficient.

North Carolina uses three components, all which begin with individual student scale scores.

The ABCs performance composite uses achievement levels for the whole school. AYP, or adequate yearly progress, is a part of the federal No Child Left Behind program. No Child Left Behind measures achievement levels in subgroups and ABCs Growth is a new formula that determines bonus money for educators whose students score well.

ABC and AYP are very similar models, McFadden said. The difference in the AYP is the breakdown of the measure by 10 subgroups -- all students, American Indian, Asian, African American, Hispanic, multi-racial, white, economically disadvantaged, limited English proficient and students with disabilities.

To be considered a subgroup, there must be at least 40 students in that category. For every subgroup, the school must test at least 95 percent of students in that group. The outcome is very basic, McFadden said -- pass or fail.

The subgroup causing the most controversy is students with disabilities, "a group that's typically defined by not doing well on tests," McFadden said.

Put more simply, he said, "there's a danger in giving every test the same weight."

To get a high school diploma, students must pass end-of-course tests in 10 subject areas -- algebra I and II, biology, chemistry, civics and economics, English I, geometry, physics, physical science and U.S. history -- at a level 3 or 4. Last year, because U.S. History and civics tests were being re-standardized, there were only eight subject areas tested, officials said.

The total number of level 3 and 4 scores for the end-of-course tests, with some students counted more than once, are divided by the number of scores for the tests, resulting in a percentage of students passing.




Judge Manning set the standard of 60 percent passing with a level 3 or 4 performance rate. Last year, Goldsboro High School was at 53.5 percent, placing the school on the targeted list Manning announced in March.

When McFadden did some digging, though, he said he found that Manning had originally ruled the baseline would be 55 percent.

"He's used both 55 percent and 60 percent, but decided he would set the closure at 60 percent," he said. "So he's going to close any school that's less than 60 percent, which is interesting because the ABC program uses 50 percent as the cut score for low-performing schools."

Goldsboro High is not considered low-performing according to the state's ABC program, McFadden said, but "they're low-performing according to Judge Manning. It's all kind of arbitrary."

With three different accountability programs, McFadden said that basically, "the poor principal at the school is trying to meet three different standards -- No Child Left Behind, ABC and Judge Manning."

No Child Left Behind is set up as a graduated plan, geared to build up scores each year so that by 2013-14, schools will be at 100 percent performance rates.

And while Goldsboro High's scores might be lagging behind their counterpart schools in the county, that does not negate the fact that progress has been made, McFadden said.

At 53.5 percent a year ago, six years ago it was at 30.2 percent, he said.

"The question is, is it where we want it to be? Of course not. We want it to be where the other schools are. But it's going to take some time, it's going to take some effort. But the school's come a long way in a short time."

For 2004-05, Goldsboro only failed in one subgroup area of the 13 it qualified for, McFadden said.

"With all the little Band-Aids that Washington put on this program, Goldsboro High had a total of 13 targeted goals. They made 12 out of 13. The bottom line is they failed because they failed (the one subgroup)," he said.




Under the area of 10th grade reading, or English I, for example, 138 of the 148 students were tested. Ninety-five percent of the target group met with Safe Harbor, which means the group did 10 percent better on the test than in the previous year.

The target goal at or above grade level is set at 35.4 percent, McFadden said. Goldsboro High had 36.5 percent, so it met AYP, he said.

By contrast, at Southern Wayne High School, all 13 students enrolled in physics tested as proficient, or 100 percent. Percentages were also high at Charles B. Aycock High, where 11 of 12 tested as proficient in the class, or 91.7 percent; likewise at Rosewood, 10 of 11 scored well, for 90.9 percent

To study the chart, Goldsboro High appears to be weaker in that area, with 36.7 percent proficiency.

"Test scores may be lower but at the same time, Goldsboro High School had more than anyone" taking physics, McFadden pointed out.

Comparatively, rankings across the county for 2005 showed 81.6 percent at Charles B. Aycock; 84.2 percent for Eastern Wayne; 53.5 percent at Goldsboro; 84.5 percent at Rosewood; 76.1 percent at Spring Creek; and 75.3 percent at Southern Wayne.



The school system is on the threshold of another round of testing as the school year comes to a close. End of grade tests for students in grades three through eight were given last week. End of course tests at the high schools take place the last two days of school, Wednesday and Thursday.

Local results will be tabulated overnight in time for graduating students to know the outcome. Overall findings for the state, however, could take several months, McFadden said.

"Math tests are new this year. The results, I understand, we won't get until October," he said. "I think we're getting the results before for other areas, but they will not tabulate the final ABC areas until October."

How that will affect Judge Manning's determination remains to be seen.


As far as the local school system is concerned, the important thing is that students receive the education they need so that when testing time comes, they will do well.

"It's an achievement test, a final exam administered by the state, not the teacher," he said. "It's up to the teacher to teach everything that will be on the test, so it's really a good system as far as testing goes.

"When somebody tells me that teachers are teaching to the test, I say, 'good'. You shouldn't teach the kids one thing all year and then test them on something else."

School officials say they are working to make provisions for educators and students alike. Dr. Ralph Smith, a 35-year science teaching veteran, was hired as lead science teacher last year to shore up staff development and training for high school students as well as those at the feeder schools.

Goldsboro High School is also revamping programs to advance performance levels of students. Next year, there will be a full scale Freshman Academy for incoming ninth graders, said Principal Patricia Burden.

Freshmen will be isolated in their own section of the school, with efforts to transition the students from middle school to the high school experience. There will also be a session this summer for incoming students and their parents to better understand what students will face and the importance of home study, she said.

There is also grant money available for something Ms. Burden called a "family academy." Whatever is necessary to get parents at the school and involved, she said she is anxious to try.

"If parents understand what's required and what they have to meet, (they) will hopefully be more supportive in completing homework assignments and preparing for tests," she said.