U.S. News cherry-picked polling data to misleadingly suggest political danger for Democrats

In a July 30 U.S. News & World Report article titled "Democrats and Defense ," Dan Gilgoff explored what he described as "a head-on collision between the Democrats' staunchly antiwar base and party leaders worried about the security and political implications of an immediate troop withdrawal from Iraq, not only in the fall but also in 2008." In doing so, Gilgoff painted a misleading picture of the Democratic Party's standing with the public on security issues. Gilgoff wrote:

After their losses in 2002 and 2004 were attributed largely to the GOP advantage on national security, most Democrats have been critical of Iraq policy without calling for a deadline for troop withdrawal, as [Senate primary candidate Ned] Lamont [CT] is. In a CBS/New York Times poll last week, 36 percent of Americans said Democrats made better decisions regarding terrorism, compared with 42 percent who said Republicans did.

Gilgoff is guilty of a curious comparison of apples to oranges: He led into his mention of the CBS/New York Times poll by writing about the Democrats and Iraq. Oddly, he then cited poll results about terrorism, not about Iraq. The very same poll (PDF) asked respondents which party they believe "is more likely to make the right decisions about the war in Iraq"; 42 percent of respondents answered that Democrats are more likely to do so, while 36 percent favored Republicans. Further, the poll found that 56 percent think the United States should "set a timetable for the withdrawal of U.S. troops from Iraq," while 40 percent oppose such a deadline. In short, Gilgoff cherry-picked poll results that seemed to make his point that Democrats are perceived as being weak on security. But had he cited the results that actually dealt with the issue he was writing about -- Iraq -- he would have found something quite different: The public agrees with and trusts Democrats more than Republicans when it comes to handling Iraq.

Gilgoff also omitted any context from his recitation of current polling data. Gilgoff connected the fact that Democrats currently trail Republicans by 6 points when it comes to handling terrorism with Democrats' "losses in 2002 and 2004 [which] were attributed largely to the GOP advantage on national security." But this ignores a key fact -- perhaps the key fact: The Democrats' current 6-point disadvantage is actually a significant improvement over previous years. That would seem to be, from a political and electoral perspective, a very positive sign for Democrats, not the millstone Gilgoff portrayed it as.

When the CBS/New York Times poll asked the terrorism question in late October 2002, just days before that year's midterm elections, Republicans had a 48-23 lead. In other words, far from indicating that Democrats are about to relive the electoral defeats of 2002, their current 6-point deficit on the issue constitutes a 19-point gain in four years on a key security question.

The CBS/New York Times poll asked the question three times in 2003, finding GOP advantages ranging from 17 points to 28 points. The question was asked once in 2004 -- in mid-July, on the eve of the Democratic national convention, finding an 8-point GOP advantage.

Other CBS/New York Times poll questions relating to national security showed a similar improvement for Democrats and erosion of support for Republicans.

• In the last CBS/New York Times poll before the 2002 elections, 54 percent approved of President Bush's handling of foreign policy; only 34 percent disapproved. In the last poll before the 2004 election, his approval on foreign policy was down to 45 percent, with 49 percent disapproving. The most recent poll found that only 35 percent approve, while 54 percent disapprove. Bush has gone from a 20-point net positive rating to 19-point net negative.
• Bush's handling of "the situation with Iraq" has gone from a net approval of +11 points in February 2003 (the first time the question was asked) to -5 points right before the 2004 election to -30 points today.
• Bush's handling of the "campaign against terrorism" has gone from a net approval of +49 points in September 2002 (the last time the question was asked before the 2002 elections) to +15 points right before the 2004 election to +9 points today. And for much of 2006, he's been in even worse shape, with results ranging from +10 to -7.

Nor are these trends unique to the CBS/New York Times poll. A quick glance at the polling results archived at PollingReport.com shows a steady erosion of support for Bush's handling of security issues and a steady gain by Democrats. On terrorism, for example:

• Fox News/Opinion Dynamics finds Republicans with a 12-point lead on terrorism -- compared with a 34-point lead in May 2003 and a 23-point lead in February 2004.
• ABC News/Washington Post finds net approval of Bush's handling of terrorism down from +51 points in October 2002 and +17 points in July 2004 (the last times that question was asked before the elections in those years) to +4 points in June 2006.
• In December 2002, shortly after that year's elections, ABC/Washington Post found that, by a 36-point margin, people trusted Republicans to handle terrorism more than Democrats. In June 2006, the GOP advantage was down to 7 points.
• CNN/USA Today/Gallup found a 29-point lead for Republicans on terrorism days before the 2002 elections -- a lead that had shrunk to 4 points by March 2006.

Gilgoff's article is only the latest in a long line of news reports that, despite all evidence to the contrary, continue to suggest that the parties' relative political standing on national security issues remains what it was in 2002. As Media Matters for America explained on July 7:

Put simply, the media are stuck in a pre-2005 mindset. They continue to assume, despite a growing body of evidence, that security issues are a political strength for the GOP. What's worse, they assume that it is an inherent strength -- that is, that it will remain a GOP strength no matter how badly the Bush administration (and a Republican-controlled Congress that obediently marches two steps behind) screw up. The problem is that myths repeated often enough become reality. By repeating -- relentlessly, constantly, unthinkingly, glibly -- the myth of Republican strength on national security, while downplaying the facts that should obliterate that myth, the media are doing everything possible to ensure that the myth will ultimately prevail in a war of attrition against reality. They hear each other repeat it, and they are reinvigorated to continue the cycle of repetition and reinforcement.

[...]

Journalists and pundits seem convinced that it is still 2002. What else could explain their stubborn insistence that security issues will naturally benefit the GOP politically? In 2002, that belief had some justification. But this isn't 2002.

[...]

Four years of lies and failures, of botched and bungled foreign policy have taken a toll. People no longer overwhelmingly support Bush and the GOP on security matters. In many cases, they strongly oppose the Republicans. It's long past time for journalists to realize that, and stop acting like these matters are still sure winners for the GOP. Otherwise, their constant repetition of that assumption could have the effect of making it true.