New polls come out every day. They measure the president’s approval ratings, they take the nation’s temperature, they measure the effect of the economy on the public psyche. They’re interesting in the way a Hershey’s Kiss is delicious: The pleasing effect lasts but a moment. They’re not fascinating in the way taco chips with cilantro-laced salsa are addictive; the sampling of one does not make a second tasting irresistible. Mostly the findings are meaningless mumbo-jumbo.

Mostly. But not always. And here is the exception, cloaked in about the most forbidding, unapproachable, opaque language possible: Top Five Values Related to Party Identification. But read on. In those seven words lie the key to understanding everything.

This Rosetta Stone of statistics was uncovered by the Pew Research Center and is buried on page 14 of a dutiful 23-page report that didn’t exactly dominate the American conversation the last few days. No matter. A look at these numbers and the whole world — well, maybe not the whole world, but surely the political world — suddenly makes sense.

Defiming parties

The numbers are derived from multiple regression analysis, which even my smarty-pants kids don’t fully understand. But they show which viewpoints prompt people to identify with the Republicans and which stir people to side with the Democrats. At the end of the 20th century, for example, if you knew someone’s views on whether government was almost always wasteful and inefficient, or whether government regulation of business were necessary, or whether homosexuality ought to be an accepted way of life, you could predict which party he would favor. Simple enough and, when you think about it, logical enough.

But the new poll, taken last month, shows that those questions, all bunched together as the leading hints to predicting party identification five years ago, no longer are the pass keys to political understanding. It also tells us that partisan gaps about government have actually narrowed, in part because Republicans, who now control the government, are less skeptical about government than they were only five years ago.

These changes tell us an enormous amount about our country and our times.

Right now your response to this assertion is by far the surest predictor of whether you are a Democrat or a Republican: The best way to ensure peace is through military strength. As an indicator of party preference, it is nearly 2 1/2 times more potent than any of the values (government’s waste, regulation’s importance or homosexuality’s acceptability) that held sway only a half-decade ago. The second-leading indicator (itself still more potent in 2004 than any of the other values in 1999) is your response to this notion: We should all be willing to fight for our country, whether it is right or wrong.

Security concerns

The message here is how national security concerns have raced to the top of America’s concerns and how these attitudes have been grafted onto the two parties’ characters. This reinforces rather than ameliorates the party divide; last November’s exit polls showed that 79 percent of those who supported President Bush said they believed the Iraq war had improved American security — while 88 percent of those who supported Sen. John F. Kerry of Massachusetts said they believed the war had not improved national security.

So everything you heard from the punditocracy about the 2004 election being driven by values was right. But, as usual, God is in the details, and in this case God (or, more precisely, people’s views about religious and cultural issues) had very little to do with it.

There are, to be sure, divisions in American life over religion, though they may well be less among various denominations than between those who are devout and those who are not. There are divisions, too, over homosexuality, divisions that may have been visible but not necessarily all that influential in states where gay-marriage ballot questions faced the public. But, then again, there are people in the nation who think the Steelers rule and others who favor the Eagles, just as there are people who don’t mind MSG in Chinese food and those who are devout in their revulsion over the very idea.

Those preferences persist — the convictions about the Steelers, the MSG, the role of government — but they don’t influence political behavior in our time nearly as much as do these national security questions.

Now to the difficult part. A facile reading of this would suggest that if the Democrats want to prevail in next year’s midterm congressional elections, they ought to have a midnight conversion to militarism as a guiding philosophy. Not so fast.

A tool of analysis

This is a tool of analysis, not necessarily a prescription for action. The president won the election, and he won it by an unambiguous margin, but that doesn’t mean that the peace-through-strength value is a sure winner in 2008, or next year, or even tomorrow.

Remember that the Democrats got a lot of votes in November. So many, in fact, that the following is true: Both the Republicans and Democrats got more of their market share than either Coca-Cola or Pepsi. A small change in the world — weariness over continuing losses in Iraq, for example, or some startling and unexpected event — could shift the public’s views.

These findings are a key to predicting voter behavior, but they are also a key to understanding the world in which voters act. They don’t tell us everything — they don’t tell a prospective congressional candidate what to say, for example, nor a member of the Senate how to vote — but they tell us about the kind of world in which we live. That’s the beginning of understanding, and now, at the beginning of the second Bush term, we can begin to understand.

— David Shribman is executive editor of the Pittsburgh Post-Gazette.