Opinion poll results can vary widely
depending on the way a question is posed. How policy alternatives are presented
and how key words are emphasized tend to have a major influence on respondents.
A question on the Unz initiative that – (1) fails to mention that it would
outlaw bilingual instruction and (2) stresses the importance of learning
English – is likely to elicit strong support, as demonstrated by a poll
reported by the Los Angeles Times on October
reported just two weeks later by the Times, asked a different
question and got a very different result. Meanwhile, other pollsters have
been highly critical of the newspaper's surveys on the Unz measure, according
to the North County Times (Oceanside).
The more voters know about the antibilingual
Proposition 227, the less likely they are to support it, according to a
February 1998 poll by the University of Southern
California School of Education. James Crawford, Stephen Krashen, and Haeyoung
Kim analyze the results in a commentary for the Hispanic
Link News Service.
According to Los Angeles Times Poll
of October 15, California's registered voters favored the initiative by
more than 4 to 1, while Latinos were even more likely
to support it than Anglos. A dramatic – and counterintuitive – result,
with major implications for public policy. This is the kind of public opinion
survey that newspapers love to report. It suggests they have their finger
on the pulse of the electorate and it gets quoted widely by other media.
But in this case, how many respondents
really knew the details of the initiative on
which they were being polled, which have thus far received little exposure
in California newspapers? Certainly they learned little about the Unz initiative
from the Times Poll. Here's the question it posed:
"There is a new initiative trying to qualify
for the June primary ballot that would require all public school instruction
to be conducted in English and for students not fluent in English to be
placed in a short-term English immersion program. If the June 1998 primary
election were being held today, would you vote for or against this measure?"
Curiously, the question itself was omitted from the Times
story on the poll. To find it – to determine what
potential voters were responding to – readers had to dig into the deepest
recesses of the newspaper's web site. Did the Times have
second thoughts about the clarity or fairness of its question? Certainly,
it failed to inform repondents about the following provisions of the Unz
initiative, any one of which could provoke opposition:
- a ban on bilingual instruction
for the vast majority of limited-English-proficient (LEP) students in California
- no accountability for schools,
no standards for teachers, and no follow-up of students placed in experimental
"sheltered English immersion" classrooms;
- limited parental choice – "waivers"
of the English-only policy might be granted for older and "special
needs" students, but there would be no guarantees;
- restricted access to foreign-language
instruction for all students, LEP or otherwise;
- threats of lawsuits and monetary
damages against teachers and other school personnel if a language other
than English is used in the classroom;
- no flexibility for local educators
and school boards to make their own decisions in teaching LEP students;
- a 2/3 "super-majority"
rule for legislative amendment – which would make an Unz law virtually
immune to change or repeal.
When an email correspondent raised
some of these points with Ray Enslow, a spokesman for the Los Angeles Times
poll, here's what he said:
"Thank you for your comments
on our recent California survey. It is certainly true that had we asked
a different question about bilingual education we might have gotten different
results. The question we asked was an attempt to gauge response to a
possible ballot initiative that has not yet been written. No doubt
we will be polling on this issue again between now and the election; perhaps
a future question will provide the information you are interested in [emphasis
It was then pointed out to Mr. Enslow
that the initiative has not only been written, but reportedly has been
endorsed by enough registered voters to appear on the June 2 ballot. He
"I misspoke. ... I appreciate
your point of view, but it seems to me that our survey question fairly
described the essential point of the initiative."
And journalists wonder why readers
often find them arrogant in exercising the powers entrusted to them by
the First Amendment.
The Times Poll does not merely "gauge"
public opinion; it helps to shape it. Its dramatic findings
(summarized below) have been cited widely – by politicians, pundits, reporters,
editorial writers, and of course by Unz proponents – to argue that "even
Latinos" oppose bilingual education. Many are suggesting that the
issue has already been decided, more than seven months before election
day – indeed, before opponents have launched a campaign to present their
case. If voters believe the Times Poll, why should they pay any attention
to the other side of the issue? Unfortunately, they remain largely uninformed
about what the ballot measure actually says.
Los Angeles Times Poll
reported on October 15.
Source: Los Angeles Times poll of 1,396 adults statewide, including
1,092 registered voters, conducted by telephone, Oct. 4-7, 1997; sampling
error of + or - 3%.
But wait a minute – here's a
2, the Times published a poll of Ventura and Los Angeles
County residents that showed considerably stronger support for bilingual
education. The story, which highlighted the persistence of anti-immigrant
feeling in conservative Ventura, nevertheless found views evenly
split on bilingual education (the 2% difference was within the survey's
margin of error). It noted even stronger support among Latinos and among
Los Angeles County residents.
How can we account for the difference
between the two polls? Characteristicallly, the Times story
did not address this issue – that might cast some doubt on its procedures.
But it doesn't take a Ph.D. to see the role of the pollster's question.
This time the survey simply asked:
Here are the responses:
Los Angeles Times Poll reported on November
|Favor Bilingual Ed
||Oppose Bilingual Ed
Source: Los Angeles Times poll of 1,286 adults in Ventura County
conducted by telephone Sept. 20-Sept. 23 (no information was provided on
when the Los Angeles survey was conducted); sampling error of + or - 3%.
How does the Times reconcile these differing
In a November 23 story, reporter
Fred Alvarez writes:
According to a Los Angeles Times Poll, 80% of California voters--including
84% of Latinos--support the initiative.
However, a separate Times Poll of Ventura
County residents found quite a different perception of the program. Only
49% of poll respondents opposed bilingual education in local schools, including
28% of Latinos.
Times Poll Director Susan Pincus said the
difference in response is attributable to the way the question was asked.
Statewide, the poll question mirrored the
language of the initiative, asking whether respondents would support an
initiative that would require all public school instruction to be conducted
in English. The Ventura County poll was more specific, mentioning bilingual
education by name and asking respondents whether they favor that program
in public schools.
Bilingual education advocates believe the
statewide results are an inaccurate reflection of attitudes surrounding
And they are emboldened by results of the
local poll, saying they believe that as the initiative is more thoroughly
explained, voters will be less likely to support it.
At least somebody on the Times is getting
the message. Too bad this article was confined to the paper's Ventura County
Yet another Times Poll – of
Orange County residents, conducted in June 1997 –
asked another poorly phrased question about bilingual education and
yielded yet another verdict, as analyzed in a letter to the editor by Stephen
If the Times keeps
trying, maybe some day it will come up with a fair and accurate survey
of voter sentiment on this issue.
1997 by James Crawford. Permission is hereby granted to reproduce this
page for free, noncommercial distribution, provided that credit is given
and this notice is included. Requests for permission to reproduce in any
other form should be emailed to firstname.lastname@example.org.
But before writing, please read my permissions FAQ.