Wash. Post , Fox cite flawed CMPA study on Social Security coverage

A recent study by the Center for Media and Public Affairs purports to show that television news coverage of President Bush's proposal to overhaul Social Security has been “overwhelmingly negative,” but the study contains flaws and omissions that call into serious question the validity of the findings. The study has been cited by Washington Post media reporter Howard Kurtz and Fox News anchor Brit Hume.

The study covers stories on the three network news broadcasts during a two-month period, from the beginning of March to the end of April. Though this period included only 43 stories, it coincided with a drop both in public approval of Bush's handing of the Social Security issue and support for his plan itself.

Because the full data of the study was not available on the CMPA web site, Media Matters for America requested the full set from the CMPA. While the CMPA did not provide the full data set, it did provide Media Matters with raw totals, which upon examination make clear why the study's conclusions are so misleading.

The study's title is “Networks Pan Bush Social Security Plan: 83 Percent of Coverage Critical During 60-Day Push.” But the data tell a different story.

First, and most importantly, the CMPA press release reports only results based on comments made by “nonpartisan sources,” including “reporters, experts and ordinary citizens.” By excluding partisan sources, it hid results that would have undermined its case that the networks have been overwhelmingly negative toward the Bush plan. The CMPA's data show that 58 percent of the comments in news by partisan sources -- most often Bush himself -- were positive toward the plan.

In total, 72 partisan sources and 47 nonpartisan sources were quoted:

Partisan sources, favorable to Bush plan: 42 (35 percent of total)
Partisan sources, unfavorable to Bush plan: 30 (25 percent of total)

Nonpartisan sources, favorable to Bush plan: 8 (7 percent of total)
Nonpartisan sources, unfavorable to Bush plan: 39 (33 percent of total)

Total sources favorable to Bush plan: 42 percent
Total sources unfavorable to Bush plan: 58 percent

While this is something of an imbalance, it is hardly as lopsided as the 83 percent figure the CMPA touts. But why would such an imbalance emerge? One clear answer is that the plan for which Bush was stumping was rapidly losing public support during March and April, a fact that the media coverage reflects.

The report notes: “When covering whether President Bush was effectively persuading Americans of his plan, 98 percent of the comments broadcast said that he was not succeeding.” But this hardly constitutes evidence of imbalance or bias, since public opinion polls at the time showed unambiguously that in fact, Bush wasn't succeeding.

This fact -- not noted in the CMPA press release -- is the explanation for the CMPA's finding that person-in-the-street interviews showed more people opposed to the plan than favoring it. Indeed, it would be a distortion of reality if the media endeavored to precisely balance their person-in-the-street quotations unless public opinion itself was precisely balanced. If, for example, two out of three Americans opposed the Bush plan, one would expect approximately two out of every three citizens quoted in the news to express doubts or opposition.

This is not the first time the CMPA has selectively sliced its data in order to arrive at the conclusion that the media favor a liberal position. As Fairness and Accuracy in Reporting (FAIR) has documented, this is something of a pattern with the CMPA and its founders, S. Robert and Linda Lichter:

The Lichters' tendency to generalize from a narrow sliver of data is the main way that their studies end up supporting their preconceived conclusions of left bias. Take the Center's report on Gulf War coverage (Media Monitor, 4/91) and its widely cited claim that “nearly three out of five sources (59 percent) criticized U.S. government policies during the [Gulf] War.” This, of course, is not 59 percent of all 5,915 sources, but of those 249 sources (4.2 percent) who in the Lichters' judgment stated an explicit position. This leaves only 148 sources, or 2.5 percent of all sources, who made explicit criticisms of U.S. policy (from the left, right or center).

On what basis can you generalize from the 4 percent of sources who supposedly expressed overt opinions to the 96 percent who didn't? Doing so results in absurd claims, such as, “Surprisingly, the U.S. government fared little better than its Iraqi counterpart in the soundbite battle.” That would be surprising, considering that 44 percent of total news sources were from the U.S. government, according to the Center's own research.

In 1988, the CMPA published a study of PBS programming concluding that the network had a liberal bias -- but once again, it did so by excluding the network's conservative-leaning programming from its analysis. As FAIR reported:

The Lichters' study of PBS is notable for what it leaves out: It excluded talkshows such as William F. Buckley's Firing Line and Morton Kondracke's American Interests, news reports like the MacNeil/Lehrer NewsHour, and business programs like Louis Rukeyser's Wall $treet Week. The Center claims this is to ensure “a group of programs that were similar in style and content, to maximize the comparability of judgments.”

The study's focus, however, removes those PBS shows most often criticized for having a conservative slant -- programming that takes up more of the PBS schedule than the documentaries that the Center's study is limited to. Firing Line and American Interests -- programs underwritten by the Center's biggest funders -- provided approximately 50 hours of programming a year between them.

As Media Matters has previously noted, though the CMPA insists that its work is nonpartisan, it has received funding from the conservative Sarah Scaife Foundation and John M. Olin Foundation, Inc. -- two of the top four conservative foundations known as the Four Sisters. While the CMPA has also received grants from more mainstream foundations, Scaife and Olin are highly ideological in their giving, supporting projects to advance the conservative movement. When the CMPA launched in 1985, it sent out fund-raising letters with endorsements from conservatives including then-President Ronald Reagan; future Republican presidential candidate Pat Buchanan; then-attorney general under Reagan Ed Meese; and Christian Coalition founder Rev. Pat Robertson.

From Kurtz's May 16 Washington Post “Media Notes” column:

The network evening newscasts haven't exactly been rooting for President Bush's Social Security plan. During his two-month blitz for the proposal, says the Center for Media and Public Affairs, 83 percent of the nonpartisan sources (from experts and ordinary citizens to the journalists themselves) made negative comments about the proposal.

Perhaps that's because the plan has gone nowhere fast? No, the 83 percent involves comments on the substance of the plan. Coverage of the politics surrounding the plan has been 98 percent negative.

From the May 16 edition of Fox News' Special Report with Brit Hume:

HUME: A new survey shows that network evening newscasts these past two months have been overwhelmingly hard on President Bush's idea for Social Security reform. The survey, by the Center for Media and Public Affairs, shows that 83 percent of comments made by so-called “nonpartisan sources” on Social Security -- from experts to reporters themselves -- were negative. Specifically, ABC had 92 percent negative comments, CBS had 89 percent, and NBC had 69 percent.

As for whether the President was successfully rallying Americans behind his idea, 98 percent of those who weighed in on the three networks said “no.”

Neither Kurtz nor Hume addressed the weaknesses in the study.