Nov 1 1998

Polls Apart

Manufacturing an Anti-Bilingual Latino Majority

Prior to the June 2, 1998 vote on bilingual education in California, mainstream news polls–local and national–predicted that not only would the anti-bilingual education Proposition 227 pass, but that it would do so with the overwhelming support of Latinos, blacks and other ethnic and racial groups.

Time (5/18/98), citing pre-election polls, reported that “a majority of Latinos actually support the initiative.” U.S. News and World Report (5/25/98) made the same claim, adding that the only reason bilingual education was able to garner support in the Latino community was because “Latino activists” were placing their own “narrow vested interests” above those of the Latino community.

The results of the actual vote, however, defied much of the conventional wisdom reflected in pre-election, poll based projections. As expected, Proposition 227 did pass by a wide margin (63 percent to 37 percent). But a CNN/L.A. Times exit poll, based on interviews with 5,000 voters, indicated that, contrary to pre-election projections, Latinos overwhelmingly voted against proposition 227 by a 63 percent to 37 percent margin, as did a slim 52 percent majority of African-Americans.

Surely overwhelming Latino rejection of an anti-bilingual measure that received the overwhelming support of whites is news, since Spanish-speaking children would be the group most affected by the measure. Moreover, this marked the third time in five years that Latinos had proved pre-election polls dramatically wrong, and where a majority of Latinos and a majority of whites voted against each other–the earlier occasions being the anti-immigration Proposition 187 in 1994 and the anti-affirmative action Proposition 207 in 1996.

On affirmative action in 1996, Latinos, blacks and Asians ended up on one side, with whites on the other. On immigration and bilingual education, blacks split fairly evenly, but a slim majority of African-Americans did side with Latinos on these issues. (It’s not unlikely that media projections that Latinos were supporting these initiatives may have influenced some African-Americans to also support them.) For their part, 57 percent of Asian-Americans sided with 67 percent of whites against bilingual ed, while most Asian-Americans sided with blacks and Latinos on immigration and affirmative action.

A Story Too Good to Check?

But one would never have known any of this if one relied on the post-vote news reporting in the predominant media. For this media, the vote merely confirmed the conventional wisdom conveyed in pre-election forecasts. The Associated Press (6/8/98), USA Today (6/4/98) and NBC News (6/3/98) simply reported that the initiative “passed overwhelmingly,” leaving out the important information of what groups did or did not support ending bilingual education in California. For U.S. News & World Report (6/15/98), the point was that “California spurns bilingualism,” and that America’s “harbinger state” had decreed that “bilingual education is on its way out.” U.S. News and Time felt no need to correct their misleading pre-vote statements about Latinos supporting 227.

While it was bad enough to report the Prop 227 story without any reference to where Latinos–and other groups in the population–came down on the issue, many news outlets actually misreported the story and claimed that Latinos had ended up supporting 227.

According to the Christian Science Monitor (6/4/98), Washington Times (6/4/98) and Chicago Tribune (6/3/98), a “majority” or “most” Latinos supported the initiative–despite the absence of any exit polling data to support these claims. Other outlets reported that the bilingual education was defeated with “wide” (ABC News, 6/3/98), “heavy” (Dallas Morning News, 6/4/98) or “strong backing from Hispanics” (King Features Syndicate, (6/7/98). The Christian Science Monitor (6/4/98), quoted a conservative spokesperson, without contradiction, proclaiming that “this is the first time that those most affected by bilingual programs have spoken so clearly that they don’t want them.”

At first glance, coverage by the New York Times and Washington Post seemed to follow this generalized pattern of mass media misinformation. The New York Times (6/4/98) reported that initiative 227 was “apparently supported by many Latino parents” while the Washington Post (6/4/98) noted that “the measure passed easily, gaining more than 60 percent of the votes in California, and with what exit polls suggest was sizable support from Hispanics–the very group that the state’s bilingual education programs are designed to help the most.”

But further on in both paper’s coverage of the vote, the CNN/L.A. Times exit poll showing Latino opposition to 227 was cited, information that contradicted the thrust of the above leads. As if to compensate, the Times (6/5/98) did a follow-up story on the Latino vote that emphasized that “fewer than 4 in 10” Latinos voted for 227. But the word play in the article’s headline–“The Reply, It Turned Out, Was Bilingual: No”–only served to reinforce the misleading message that dominated media coverage.

As coverage of the vote by the Times and Post demonstrates, information in the CNN and LA Times exit poll was readily available to media outlets. But for so many of these outlets, it seemed easier to go with a depiction of voting results that confirmed their own expectations rather than finding out the truth. At least, though, the Dallas Morning News (6/5/98) and The Christian Science Monitor (6/8/98) eventually corrected themselves.

Some of the most accurate reporting on the initiative came, perhaps unsurprisingly, from the outlets that had done the exit polling. CNN‘s Inside Politics (6/3/98) reported that “Proposition 227 won because of heavy support from white voters and Asian-Americans,” while it received a divided black vote and “was rejected by Hispanics.” The program also reported that the main reason 227 opponents gave for opposing the measure was that “they thought it would lead to discrimination,” a message, they noted, “that apparently got through to black voters.” CNN commentator Bill Schneider observed that “this could be another measure, like the 1994 cut-off of benefits to illegal immigrants and the 1996 ban on affirmative action…a measure that passes, but leaves Californians and the country deeply divided.” The L.A. Times coverage–e.g., “Orange County’s Latino Precincts Voted Strongly Against Proposition 227,” 6/5/98–was also on target.

Impoverished Polls

Few mainstream outlets addressed the inaccuracy of the pre-election polls. One program that did was the NewsHour with Jim Lehrer (6/3/98), although the L.A. Times pollster who appeared on the show provided a curious explanation as to why Latinos ended up opposing the anti-bilingual prop 227: “Latinos have a habit of being for some proposition…187…209, right up until the end, and then they switch over,” pollster George Skelton said. “It may be some ethnic pride in the end, but then they overwhelmingly go against it.”

But as Antonio Gonzalez of the William C. Velasquez Institute (formerly the Southwest Voter Research Institute) noted in the LA Weekly (6/12/98), “They’re going to cover it up by saying we decide at the last moment, but it’s really their profoundly bad social science.”

Along with the Weekly, an article in the San Francisco Chronicle (6/5/98) addressed how the fact that “most African-Americans and Latinos–the biggest non-English speaking group in California–resoundingly voted ‘no'” on 227 was a stunning reversal of pre-vote polling predictions. These publications pointed out numerous flaws in mainstream polling methods, including the under-representation of Latinos in polls (resulting in a very high margin of error), the lack of Spanish-speaking pollsters and the ambiguous, if not biased, framing of polling questions.

The electoral polarization in California of whites and Latinos over immigration, affirmative action and bilingual education issues continues to be a story that cries out for media attention. Instead, the focus has been on conflicts between African-Americans and Latinos–as when Time (12/1/97) suggested that the conflict between blacks and Latinos over bilingual education constituted “The Next Big Divide.” The corollary to this theme is the suggestion that the political interests of Latinos and whites are converging, to the detriment of blacks–an argument necessarily made without reference to the results of the three California initiatives (New York Times, 8/16/98; New York Times Magazine, 8/16/98).

No mainstream outlets was found that devoted any in-depth post-ballot analysis to the implications of the brown/white divide made manifest in recent initiative voting in California. The closest the mainstream media have come has been to focus on the difficulties the Republicans are having in getting Latinos to become part of their “new majority” coalition. (Newsweek, 7/13/98; Washington Post, 7/2/98; Associated Press, 7/3/98;

In an exceptional contemporaneous column pointing to instances of media misrepresentation on 227 (Houston Chronicle, “Media Misleading about Hispanic Vote on Prop. 227,” 6/11/98), Charlie Erickson of the Hispanic Links News Service included the following exchange with Prop 227 co-chair Gloria Matta Tuchman on the outcome of the vote: “Tuchman told Hispanic Links editor Joseph Torres that more than 60 percent of Hispanics voted for the initiative. When Torres challenged the figure, Tuchman couldn’t cite a source, but, she assured him, ‘I read it in the papers.'”

Research assistance provided by Yomara Velasquez.

Mikal Muharrar is the Coordinator of FAIR’s Racism Desk. He contributes a regular column on racism in the media to Extra!.