FOR the thousands of economists attending the annual meetings here of the American Economic Association, and for the millions of people in the world undergoing rapid and often chaotic change, there is a crucial question: How good is the state of economics?

The answer from some leaders of the profession: Not so good. Herbert A. Simon of Carnegie-Mellon University, a winner of the Nobel Memorial Prize in Economic Science, contends that the differences and sharp conflicts among economists demonstrate that the state of economic knowledge is insecure.

''The public is not mistaken in its perception that economists disagree frequently and vociferously,'' Mr. Simon said. ''We have Keynesian economics, monetarism, supply-side economics, rational expectations and budget balancing, not to mention free trade, protection for infant industries, and proponents of the income tax, the single tax, the sales tax and almost no taxes at all.''

Michael J. Boskin, chairman of the President's Council of Economic Advisers, sought to resolve these clashes by taking what he called a ''pragmatic, eclectic'' position. He said the public could not expect to find a ''pure monetarist, pure Keynesian or pure rational expectationist'' on his staff. ''We are not sticking slavishly to any one of them,'' he said, adding, ''This view is shared by most people in the Administration and fits in with the way President Bush thinks.'' He counseled much greater humility for economists.

Despite differences among economists on both analyses and policies, some critics insist that economists share a common and incorrect approach to understanding human behavior - that they exaggerate human rationality, alleging that people act on the basis of calculations and forecasts of how they can get the most income, profits or well-being.

Kenneth J. Arrow of Stanford University, who won the Nobel Memorial Prize in 1972, observed in an essay in a new book, ''The State of Economic Science,'' that ''the impossibility of carrying out such calculations is manifest from everyday observation and confirmed by the inability of economists using our theory and our computing power to make good forecasts - even good contingent forecasts.'' A contingent forecast is one based on certain assumptions.

Mr. Arrow noted that in experiments, cognitive psychologists had found systematic kinds of bias in human behavior, strongly contradicting the economists' assumption of rational behavior. Both Mr. Arrow and Mr. Simon urged the economists to learn from these psychological studies.

The inability of economists to forecast reliably remains a persistent criticism of the profession. There is increasing question, after the development of chaos theory in mathematics, about whether economic systems, given their complexity and the possibility that small changes in some variables will produce enormous changes in the overall system, is inherently chaotic and does not lend itself to dependable forecasts.

Even some of the most mathematical economists now complain that the field has become overly wedded to abstract theories and too careless and casual in amassing the empirical data on which understanding must be based.

While noting that no serious economist opposes applying mathematical techniques to economics, Henry Woo of the Hong Kong Institute of Economic Science said, ''What causes concern over present-day economics is the dominance of mathematics almost to the exclusion of other methods of inquiry and the imposition of a mathematical finality on the subject.''

Despite this bill of indictment, economists remain a cocky and feisty group. Because they are consulted routinely by public, press and Presidents, they believe, perhaps correctly, that they are the only wheel in town. Economics is certainly too important to be left to non-economists, with their own heavy political or personal interests.

But there is more to be said in behalf of the best of the profession. They have over the years developed concepts of the interdependence of elements within a system that are not obvious to lay people looking at one event at a time. They have tackled problems, increasingly interesting to other sciences, of decision-making in the face of uncertainty. And they have sometimes shown an imagination and inventiveness in problem-solving that society urgently needs.

The danger is that economists, with their infatuation with abstruse theory, neglect the hard empirical work on which public policies must depend.

In brief, as some outstanding economists see it, the field needs to be both richer in its intellectual and empirical content and more of a science in the true sense. As Mr. Simon put it: ''What distinguishes science from every other form of human intellectual activity is that it disciplines speculation with facts. Theory and data are the two blades of the scissors. But the metaphor is not quite right, for the blades are not symmetrical. When theories and facts are in conflict, the theories must yield. Economics has strayed from that simple principle, and it must return to it.''

One way of getting there is for more economists of this generation to get as involved in economic problem-solving as their predecessors during the Depression, World War II and the Marshall Plan.