America has a dysfunctional hate-love relationship
standardized tests. In private, students and their
parents are terrorized by SAT and ACT college admission
test scores. In public, the federal No Child Left Behind
legislation adores tests—it requires that each state
test its students and publish the results as part of the
mandate that every single student in America reach the
"proficient" level in math and reading just seven
years from now.
The lessons we draw about testing make little sense.
For instance, one of the many
fibs Americans tell each other about
schooling is that it`s crucial to attend a
prestigious college in order to get a good education. As
proof, we point to the high test scores that students at
the most famous colleges earn—such as a mean of 1500 out
of 1600 at
the SAT, the best-known college entrance exam.
Yet Yale doesn`t systematically test its seniors to
determine how much their four years in New Haven have
taught them. Instead, Yale`s prestige depends in
large part upon the stratospheric SAT scores earned by
its freshmen while they were
in high school.
In reality, we use the average test scores of the
college that somebody attends to
estimate his intelligence and diligence. The
predictable result: an arms race, as ever more students
try to worm their way into the most
exclusive colleges—a mania that allows colleges to
raise tuition relentlessly.
Worse is what our lack of rigorous thought about the
realities uncovered by testing does to children in the
bottom half of the intelligence curve. Public
make policies that show they don`t have a clue just
how clueless millions of young people are. People who
below average in intelligence have enough problems
as it is, without being persecuted further by ignorant
Public discourse about test scores is also retarded
by a technical problem. There are such a proliferation
of school achievement tests across the 50 states (the
NCLB refused to institute a national test), that few
people understand what the various scores mean. The
states test scores are just not as familiar as SAT
scores, which tens of millions of Americans understand
at least roughly.
Recently, I recently stumbled on a
database on the
LA Almanac website listing the average SAT
scores at every Los Angeles County public high school.
The results were quite startling. They say a lot about
public policy—and, indeed, about the future prospects
for America. Because, perhaps more than anywhere else,
our future is being test-driven in our most populous
county, Los Angeles, with its 10 million residents.
Before going over what Los Angeles area students
average on the SAT, allow me to review the basics of SAT
Each section of the SAT is scored on a 200 to 800
scale. A third section, Writing, was recently added, but
most colleges at present are not paying much attention
to it yet, so I`ll look just at the better known Verbal
"Critical Reading") and Math scores.
To give us oldsters (i.e., anybody over 30) some
perspective on what current SAT scores mean, let me
first point out that scoring was made considerably
easier in the "recentering"
of 1995. The SAT was originally normed on
Northeastern prep school students. As the percentage of
U.S. students who took it each year grew, the average
score fell. The goal of the recentering was to put the
mean of each test back up to around 500.
So, all you oldtimers out there need not be astounded
by how high SAT scores are at elite colleges these days.
Recentering means that test takers who would have scored
420 under the old Verbal SAT scoring system had their
scores raised to 500. Those scoring 470 on the Math SAT
had theirs raised to 500. So, an 890 before 1995 would
today score 1000 (500 V plus 500 M = 1000).
Somebody who had an 1120 (520 V plus 600 M) in the
past would receive a 1200 today. And a 1350 (640 V and
710 M) would garner a 1400 today.
Currently, 1000 is the
46th percentile among test-takers. The mean is now
at 1021 (503 V, 518 M). For your reference, 1200 is the
79th and 1400 the 96th percentile.
An important point: 1021 is just the mean among
students who are considering college seriously enough
to take the SAT. A sizable minority of students
don`t bother. So a 1000 would probably be somewhere
around the 60th or higher percentile nationally among
If you are more familiar with the ACT—an entrance
exam widely taken in the Midwest—a quick
conversion has a 1000 SAT (new style) equal to about
21 on the ACT, while 1200 = 26, and 1400 = 31.
Ambitious parents agonize over whether their scions
can measure up at, say,
Cal Tech, where the mean score among enrolled
freshmen three years ago was 740 Verbal and 790 Math for
a total of 1530.
Duke`s was 1420.
Even public colleges better known for sports often
boast impressive SAT scores. The
U. of Florida, for instance, which is the reigning
NCAA champion in both football and basketball, averages
1267. Football runner-up
Ohio State is at 1190, as is perennial powerhouse
Oklahoma, with the
Alabama at 1107.
The strongest traditionally black college,
Howard U. in Washington D.C., averages around a
quite respectable 1080.
Cal State Long Beach, a somewhat above-average
public commuter college, is about 1040.
What`s much less well-known are the average SAT
scores of high schools.
Nearly matching Harvard-Westlake with a 1382 is the
public high school
Whitney, a mostly Asian public school in Cerritos
that chooses students based on
test scores. (Like the famous New York City science
highs such as
Stuyvesant, which averages 1410.)
The highest scoring neighborhood high school in LA
Country without special admissions tests is
San Marino, in the exquisite township south of
Pasadena founded by old money WASPs. (San Marino`s first
George S. Patton Sr.) It`s been popular with Hong
Kong millionaires for a few decades, and the student
body is now 70 percent Asian. The average SAT score is
Below this top one percent of public high schools,
where almost everybody takes the SAT, a methodological
problem crops up: how to count students who don`t? For
Beverly Hills H.S., probably the most famous public
school in America from the countless movies and
set on the campus, averages 1190. That sounds really
good, but only 74 percent of the students take the test.
The other quarter would likely drag down the school`s
average if they were forced to take it. (Electing
A New People note: 40 percent of the students are
Fortunately, the LA Almanac
data table reports another very useful number: the
percent of students in 12th grade who scored at least
1000. There`s nothing magical about that score, but it`s
reasonable to say that kids with 3-digit SAT scores are
going to have to work exceptionally hard to graduate
from college. (For example, at
Cal State Northridge in the San Fernando Valley, the
average SAT score is 930 and only 25 percent graduate
within six years, even though tuition is free.)
Thus, at Beverly Hills H.S., 62 percent of students
scored 1000 or better. That`s excellent, yet not all
that impressive considering Beverly Hills`
storied wealth—it`s only 9th best in LA County,
behind Whitney, California Math Academy in Long Beach,
San Marino, La Canada, the two Palos Verdes schools,
Arcadia, and Calabasas.
Los Angeles Unified School District, which is 72
percent Hispanic overall, the highest ratio of 4-digit
SAT scores is found at the all-magnet
Los Angeles Center for Enriched Studies (LACES).
Some 57% of its seniors reached at least 1000.
The only other of the 60 LAUSD high schools with over
half of the students exceeding 1000 is the suburban
Granada Hills charter high school, which includes a
subschool. (All LAUSD figures incorporate the scores
of students at magnet subschools within the overall high
Other LAUSD high schools include the SOCES magnet,
with 40 percent scoring over 1000, North Hollywood and
Van Nuys at 28 percent, Birmingham (which was the
subject of an
LA Times series
on the dropout crisis because it`s so average) at eleven
percent, and Hollywood at five percent.
Near the bottom are
Locke, with only two percent of seniors breaking
1000. And last but least is
Jefferson, which is 92 percent Hispanic and eight
percent black, and has been the
scene of brown v. black
race riots. At Jefferson, merely 7 of the 639
twelfth-graders reached 1000 or higher.
In the entire LAUSD, only 14 percent of seniors make
it to 1000.
Among those who took the SAT in LAUSD, blacks
807, Hispanics 829, Asians 1007, and whites 1059.
Did lack of English hold Latino mean scores down?
Unlikely. LAUSD`s Latinos averaged
408 on the Verbal section and 421 on the Math
section, compared to 525 V and 534 M for whites and 477
V and 530 M for Asians. This small gap between the
Verbal and Math scores for Latinos suggests that
unfamiliarity with English is not a severe problem for
those who do take the SAT. Those who
don`t speak English well are less likely to take the
It would be easy to blame the poor test scores in Los
Angeles public schools on the LAUSD, a vast bureaucracy
with a poor reputation for management. Yet
26 of the 56 other school districts in Los Angeles
County have worse test scores—many much worse.
Take the Compton Unified School District … please.
Only one percent of
Compton`s seniors score over 1000. While Compton was
the home of
West Coast gangsta rap, its school district is now
69 percent Latino. At Compton`s
Centennial High, whose red school color was adopted
by the notorious Bloods gang when it was founded at
Centennial in 1972, only 22 percent of 12th graders took
the SAT. And this cream of the Compton crop recorded an
average SAT score of 715.
For Los Angeles County as a whole, which includes
some of the ritiziest suburbs in America, just 18
percent of public school students reached the 1000
But the news is actually worse. These
statistics overestimate the average LA County
That`s because a huge percentage of Southern
California students drop out before 12th grade. At a
typical LAUSD high school like
Birmingham in the San Fernando Valley, there are
1442 9th graders and only 532 12th graders. At South
Locke, which is 5/8ths Hispanic and 3/8ths black, a
UCLA study estimated the graduation rate at 24 percent.
LA Times reported:
study by the Civil Rights Project at Harvard University
calculated that only 45% of students were graduating in
four years from Los Angeles schools. The rate was even
lower for Latino students, and much higher for white and
Asian American students. African Americans were close to
the districtwide average." (L.A.
Mayor Sees Dropout Rate as `Civil Rights Issue`
By Mitchell Landsberg, Times Staff Writer March 2, 2006)
If we assume that few of the dropouts would have
broken 1000, then, as a rule of thumb, we can divide by
two. Thus, only about seven or eight percent of the
students who start 9th grade in the LAUSD will break
For all Los Angeles County public high school
freshmen, only about ten percent will exceed 1000 by the
time they leave high school.
What about private school students? Do they lighten
this dismal picture? The
Council for American Private Education reported that
at the national level:
"Combined scores for public
schools, religious schools, and independent schools
were, respectively, 1012, 1057, and 1119."
Students in religious and independents schools,
however, make up less than
nine percent of all LA County ninth-graders. So even
adding them in, it`s unlikely that much more than 16
percent of all freshmen in America`s most populous
county will ultimately break 1000.
It`s time for our elites to face up to the fact:
millions of young people just aren`t all that bright by
the standards of the upper middle class. Passing laws
based on the assumption that we live in
Lake Wobegon where
all the children are above average just makes life
worse for kids on the
left half of the bell curve.
Duke Helfand wrote an important investigative report
in the Los Angeles Times last Jan. 30, 2006
A Formula for Failure in LA`s Schools:
Los Angeles Board of Education approved tougher
graduation requirements that went into effect in 2003,
the intention was to give kids a better education and
groom more graduates for college and high-level jobs.
For the first time, students had to pass a year of
algebra and a year of geometry or an equivalent class to
(I wonder how many members of the Board of Education
can pass an algebra test?)
"The policy was born of a
worthy goal but has proved disastrous for students
unprepared to meet the new demands.
"In the fall of 2004,
48,000 ninth-graders took beginning algebra; 44%
flunked, nearly twice the failure rate as in English.
Seventeen percent finished with Ds… Among those who
repeated the class in the spring, nearly three-quarters
"The school district
could have seen this coming if officials had looked at
the huge numbers of high school students failing basic
(Yes, but looking up numbers would have been
“insensitive”—and that`s the gravest sin these days.
Better to make hundreds of thousands of people go
through life without a high school degree than publicly
to notice that some people aren`t as smart as others.)
Sacramento didn`t ask questions either. After Los
Angeles Unified changed its policy, legislators turned
algebra into a statewide graduation requirement,
effective in 2004.
"Now the Los Angeles
school board has raised the bar again. By the time
today`s second-graders graduate from high school in
2016, most will have to meet the University of
California`s entry requirements, which will mean passing
a third year of advanced math, such as algebra II …"
(Oh, great! Algebra II!)
By law, admission to the University of California is
reserved for the top 1/8th of California high school
students, as measured by test scores and grade point
average. Yet the Los Angeles Unified School District (LAUSD)
is now on course to deny a diploma to the bulk of its
students simply because they aren`t
bright enough to master Algebra II.
A large fraction of LA high school students should be
working on finally mastering fractions and percentages,
skills they`ll actually use in their careers—not banging
their heads against the Algebra II wall of abstraction
until they drop out.
Why turn the rest of the country into Los
Angeles—without the sunshine?