It is hardly a trivial question. How long do the American people think our troops should stay in Iraq? You would think polls could give us a clear-cut answer.

As mentioned in passing last week, they don’t. That does not stop sponsors from making extravagant claims about the insights their polls yield. Examining the varying responses provides some insight into opinion on Iraq but tells us even more about the limitations of polling.

The polls are not quite as contradictory as their sponsors might think, but public views are hardly clear-cut.

In recent weeks, 10 polls have asked in some form how long we should remain in Iraq. A simple reading suggests that in six at least a plurality want to bring home troops now, while in four voters are willing to keep our forces there until their mission is actually accomplished.

Anchoring one end of the spectrum is an AP/Ipsos poll that asked, “Should the United States keep troops in Iraq until the situation has stabilized, or should the United States bring its troops home from Iraq immediately?” Fifty-nine percent favored keeping our troops in place, while only 37 percent supported withdrawal.

On the other end of the continuum was Harris, which found 33 percent in favor of “keeping a large number of U.S. troops in Iraq until there is a stable government there” but 63 percent saying we should bring “most of our troops home in the next year.”

Of course, a number of responses were arrayed in between. Pew found 50 percent thinking that “the U.S. should keep military troops in Iraq until the situation has stabilized” and 46 percent believing instead that “the U.S. should bring its troops home as soon as possible.”

Are some of these polls “right” and others “wrong”? Probably not. Each is “correct,” but each question’s wording evokes a slightly different context, which can lead to radically different results.

First, there is a difference between “withdrawing” on the one hand and “reducing” the number of troops on the other. Every question that asks about reducing our presence finds majority support for that option. Questions that focus on complete withdrawal reveal majorities willing to stay.

The scope of the stated goal also affects responses. Voters are more likely to be willing to keep troops in Iraq until the “situation” has stabilized or “civil order has been restored” than to keep them there until there is a “stable government” or until “peace” has been achieved. Questions that cite no particular goal for our presence also tend to find larger numbers favoring a speedier departure.

Time frame is also salient. The longer the stay, the less support there is for it. CNN/USA Today/Gallup found 44 percent willing to keep troops in Iraq for less than a year and 31 percent willing to stay for one to two years. But just 21 percent expressed a willingness to keep U.S. troops in Iraq for more than three years.

Widely varying responses on any particular topic sometimes reflects public ambivalence or uncertainty. But it may instead reflect situationally specific views. If the mission seems reasonable (stability, civil order) and the time frame relatively brief, Americans are willing to keep our soldiers on station in Iraq. If the goal is overbroad (peace or a stable government) or the time frame too long, voters want to bring most troops home. Taken together, these polls tell us a great deal.

However, any one of these questions alone is fundamentally misleading. Asking only one question on a given topic is a lot less expensive but potentially much less accurate as well. No one question is a perfect indicator of underlying attitudes. Often, the true contours of public opinion can only be revealed by examining responses to multiple questions.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982, including Sen. John Kerry (D-Mass.) last year.