How does the public want to pay for cleaning up after Hurricane Katrina and her disruptive relatives? As politicians grapple with the costs of disaster relief following a string of major hurricanes along the Gulf Coast, there is no clear public consensus over where the money should come from. In particular, while most Americans believe the government should make spending cuts elsewhere to offset the costs of disaster relief, what kind of cuts they support depend heavily on what alternatives are offered.

About a month ago, the Gallup Poll presented the intriguing finding that a 54% majority chose reducing spending on the war in Iraq as the best way to pay for relief efforts following Hurricane Katrina. In the survey, conducted September 16-18, 17% chose raising taxes as the best way to pay, 15% favored increasing the budget deficit, and just 6% supported reducing spending on domestic programs such as education and health care.

This lopsided support for cutting spending on Iraq–more favored this option than the other three options combined– raised the question of whether the response reflects growing public opposition to the war in Iraq, or a broader adjustment in the public’s domestic vs. foreign priorities. Would a question that offered more generic cuts in defense spending, without mentioning Iraq specifically, garner the same support?

In addition, by raising education and health care as the likely areas in which to cut social programs, the Gallup item offered a more specific description than many other polls that just refer to “domestic spending” more generically. This phrasing could also serve to increase the proportion who opt for reducing Iraq expenditures, since education and health care are among the most popular domestic programs.

To answer these questions, a recent Pew survey conducted Oct. 6-10 includes two questions modeled after the Gallup item, but with slight modifications in how the options are described. The first of these retained the option to reduce spending for the war in Iraq just as in the Gallup question, but didn’t mention education and health care as specific areas in which domestic spending might be cut. This small rephrasing made a significant difference. Instead of favoring reductions in Iraq spending over domestic by a nine-to-one margin (54% to 6%) as in the Gallup survey, just 31% saw Iraq as the place to cut, while 20% chose domestic cuts.

Removing the specific mention of Iraq from the question produces yet another pattern of responses. Americans are far more evenly divided over the best way to pay for disaster relief when asked to choose between cutting “defense and military spending” (28%) or “domestic spending” (31%).

The varying responses to these similar survey questions tells us that in questions of this nature–when a broad range of possible answers exists and no clear political consensus has developed on the dimensions of public debate–the options provided in a survey question can have a sizeable impact on how people answer. The political implications of this for Congress, as it struggles with the specifics of slicing billions of dollars from the federal budget are obvious: The public is far more likely to endorse cuts in domestic programs when they are described in general terms than when they are presented with added detail about where the federal dollar goes and where the budget ax may fall.

Still, at the same time, certain attitudes appear to be relatively stable. In this case, regardless of how the domestic and military options are described, the number who see increasing either taxes or the budget deficit as the best solution remains relatively small.