This article originally appeared in the LA Times on July 14, 1998
© Los Angeles Times

How to interpret the scores

Complete LAUSD Stanford 9 test scores
|
Test Scores Yield Mixed Results for L.A. Students
Though figures vary widely, third-graders fare poorly in English. Higher numbers are posted in math.
LOS ANGELES SCHOOL STUDENTS SCORED SUBSTANTIALLY BELOW the national
average in reading at every grade level, plunging to the bottom fourth in
third grade, a Times analysis of state standardized test scores shows.
Reading scores improved through the eighth grade, then dipped again into
the bottom quarter in the ninth and 10th grades.
While the average was low, figures released Monday by the Los Angeles
Unified School District showed tremendous variation between schools.
The third-grade scores were in the 21st percentile, meaning that the
average Los Angeles student is below 78% of all students nationally.
Ninth- and 10th-grade scores fell at the 22nd percentile.
Los Angeles students posted stronger math scores, but showed the same
pattern as in reading--scores dipped to the bottom quarter in the fourth
grade, then rose to the 37th percentile by ninth grade.
The scores are the first to include all of the district's
limited-English students. Consequently, the results were significantly
lower than those released two weeks ago by the Los Angeles Unified School
District, which showed that the district as a whole scored in the 32nd
percentile. At that time, the district factored in only a selected number
of students with limited English.
Meanwhile, the district Monday released school-by-school scores for
the fourth, eighth and 10th grade levels in reading and math. Not
surprisingly, the scores showed dramatic disparities in the performances
of schools. Five schools, topped by the San Jose Highly Gifted Magnet in
Mission Hills, scored at the 90th percentile or above in math at the
fourth grade. At the low end, six central Los Angeles schools scored at
the 10th percentile or below. In reading, two schools--including San
Jose--scored in the 90th percentile or better at the fourth grade and 24
schools scattered across the city were in the 10th percentile or below.
Meaning of Scores Criticized
The state planned to release all the results of the first standardized
tests in decades June 30 via the Internet, but that plan was blocked by a
lawsuit. As a result, many school districts have been slowly rolling out
the scores.
The Los Angeles school results released Monday, which are published in
today's Times, represent the information the state Department of
Education intended to publish June 30.
Supt. Ruben Zacarias said he decided to release the school
scores--although the district may not be required to--because of the
intense public and media interest.
He and other officials have complained that the scores are not an
accurate measure of the district's performance because children with
limited English were required to take the test in English.
"It is important to understand that the scores do not accurately
reflect the quality of instruction occurring in this school district,"
Zacarias said. "With the large number of [limited English] students in
our classrooms, comparisons with other districts and even between LAUSD
schools can be misleading."
To show the effect of testing limited-English students, the school
district calculated its scores three ways, with limited-English only,
fluent English and both groups combined.
Overall, limited English students scored only about half as well as
the combined group in the lower grades and about a third as well at high
school. Limited-English students make up about 46% of the district's
enrollment.
Zacarias said the district Information Technology Division is
preparing another school-by-school report that will compare this year's
test with one administered by the district last year. In that previous
test, only a portion of limited-English students participated.
These results will be used to evaluate the success of district reform
efforts such as LEARN and Zacarias' personal attention to the 100 lowest
performing schools, he said.
In another slice of test results, the Los Angeles County Office of
Education last week released a summary of the performance of all 82
school districts in the county. The summaries included a percentile score
for grades 2-11 in various subjects.
City Schools Perform Poorly
The Times analysis compared Los Angeles school scores in reading and
math with the combined scores of every other district in the county. The
analysis showed that Los Angeles schools performed more poorly at all
grade levels than the remainder of Los Angeles County.
But the other county schools showed the same pattern of scores dipping
in the third or fourth, ninth and 10th grades.
Reading scores at other Los Angeles County schools were lowest in
third grade at the 33rd percentile, 12 points above Los Angeles schools.
County reading scores peaked at the 43rd percentile in the eighth grade,
before falling back to the 29th percentile in 10th grade.
Math scores at the other schools fell to the 39th percentile in fourth
grade, also 12 points above the Los Angeles district, then climbed to the
49th percentile in ninth grade, just a point below the national average.
In both Los Angeles and the other schools, the students who performed
the highest at most grade levels were those once classified as
limited-English but who have been redesignated as English fluent.
Although few in number, the redesignated students generally scored
near the national average in the elementary grades.
Los Angeles officials said they are still analyzing the data, but
could not explain the declines in third and fourth grades and high
school, despite a modest gain for the district overall.
"We don't really have answers," said Liliam Castillo, deputy
superintendent of instruction.
* * *
Comparing the Results
Reading and math test scores of L.A. Unified students fell
significantly below all other students in the county at all grade levels.
Math scores were higher than English scores. The bards mark the average
percentile score for students at each grade level.
Source: Harcourt Brace Educational Measurement
* * * Times data analyst Sandra Poindexter contributed to this report.
|
|