Intuitively believing that attitudes determine behavior, we instinctively infer the former from the latter.

So it is with the demise of Hillary Clinton’s presidential campaign. Analyses mix examination of Barack Obama’s appeal and descriptions of strategic blunders with assessments of Clinton’s failure to provide what voters wanted.

One pundit suggested Clinton was defeated because “she turned into the living representation of … [the Washington] establishment and its myopic vision.” Another argued Democrats rejected Clinton because “The country is looking for something new and hip and next-generational.” Or perhaps it was because voters preferred “inspiration over restoration.”

Analysis of this kind is hardly limited to this presidential campaign. Michael Shellenberger and Ted Nordhaus prematurely proclaimed “The Death of Environmentalism,” based importantly on Congress’s 2002 failure to pass stricter Corporate Average Fuel Economy standards. After 2000, commentators inferred the centrality of Bush’s “likability” from his (disputed) electoral vote victory.

The genius of our Founders as social scientists lay in their recognition that institutional arrangements — the rules of the game, not just attitudes — play a central role determining outcomes and shaping behaviors.

Madison argued in the Federalist Papers that the institutional arrangements he recommended would “cure the mischief of faction.” Maintaining that public decisions should not merely reflect “the superior force of the majority,” the Federalists elaborated rules and inserted institutions into our Constitution designed to prevent outcomes from reflecting only the will of the majority. To the extent their plans were well-laid, observers should be unable to infer public opinion from political outcomes.

Before returning to politics, examine this process of inference in another realm. Despite the fact that nearly 100,000 Americans are waiting for life-saving transplants, just 35 percent of us are registered organ donors, compared to 99.9 percent of French, Hungarians and Austrians.

Struggling to explain these differences, an analyst oblivious to the role of rules might posit that American values around organ donation differ from those of the French. Attention might turn to the role of religiosity in the U.S. as against European secularism, or to Europe’s communitarian values, contrasted with American individualism.

Yet, Gallup finds 95 percent of Americans support organ donation. Attitudes don’t differ. Why the difference in behavior?

It’s the rules. In France, Austria and Hungary everyone is an automatic organ donor, unless they choose to opt out. In the U.S., potential donors must opt in. Different rules, not different values, produce the differing results.

Rules matter in politics as well. If Democrats had the same winner-take-all delegate selection rules adopted by Republicans, Hillary Clinton would have won the nomination without assistance from Michigan or Florida.

While the calculations are more difficult and the uncertainties greater, it is not at all clear that John McCain would have emerged victorious had the GOP used Democrats’ proportional representation system.

No one can dispute the validity of either nomination: Without rules there is no fairness, no legitimate contest, and there is nothing unfair about one set of rules or the other. They are just different. Everyone knew the rules in advance and everybody was bound to play by them.

But the centrality of the rules in determining the outcome makes it difficult to interpret wins and losses as endorsements of some particular version of public attitudes.

If the state-by-state and exit poll results had been exactly the same but today Hilary Clinton were celebrating her nomination because the rules were different, would pundits have concluded voters back the Washington establishment; were looking for restoration, abjuring inspiration; for tried and true, rather than new and cool?

If so, they would be making the same error many are now committing by inferring an array of hostile public attitudes toward Hillary Clinton from the outcome of this nomination battle.

Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982, including Sen. John Kerry (D-Mass.) in 2004.