Newsweek magazine devoted a cover story to the question April 5 and I thought I had it settled. ''Who are the poor?'' it asked. Officially they are the 13 percent of Americans who live at or below the census bureau's 1980 ''poverty threshold'' of an annual income of $8,414 for a family of four.
There are 29.3 million of them. I thought I had that settled. But wait a bit. Milton Friedman, the noted economist from the University of Chicago who contributes a column to Newsweek, takes exception. In the April 19 issue of the same magazine he uses a full page article in rebuttal; in his 16-year association with Newsweek he says ''only one other story'' has disturbed him as much as the one that gave me my supposed information. He doesn't disclose the other article. The new one is called ''Reagan's America: and the poor get poorer.'' Mr. Friedman charges that ''the story gives a most misleading impression of the source and extent of poverty and of the likely effect of the tax and budget measures enacted in 1981.'' The matter is certainly significant. President Reagan is becoming increasingly restive under the charge that his administration is helping the rich at the expense of the poor.
Take this matter of how many poor there are. The Newsweek writers say flatly: ''The number of poor began to edge upward in 1978. In 1980 it went through the roof.'' Not at all, says Friedman: ''The number of official poor has risen along with prosperity and burgeoning welfare-state spending, especially rapidly as inflation escalated.'' What is questionable, he says, is the official poverty line itself and the government's own inefficient efforts at dealing with the basic problem.
In this statistical hassle the Census Bureau has decided to make a new effort to explain. The official figures of poverty, it says, do not include substantial items like food stamps, subsidized school lunches, public housing, medicaid for the poor and medicare for the elderly. The statistical ''poverty'' figure has the virtue of consistency but is based on cash income alone. For example, a family might get $2,000 in medicaid benefits and this would not be included in the ''poverty'' reckoning.
After reading the evidence I conclude that poverty is too high in America for a rich country but that it isn't so high as the figures initially indicate. The total of ''29.3 million'' poor is often compared with other democracies but is inaccurate unless other material is included. It comes to 13 percent of all Americans or about one in eight. The census bureau last week offered three alternative methods of measuring poverty, all of which tend to reduce the amount substantially.
Nevertheless things have improved. Herbert Hoover in his 1928 acceptance speech declared the US was near ''the final triumph over poverty.'' Instead of that came the crash and poverty grew vastly worse. Some 36 years later came Lyndon Johnson who told Congress in his State of the Union speech Jan. 8, 1964: ''This administration today, here and now, declares unconditional war on poverty in America, and I urge this Congress and all Americans to join with me in that effort.''
Many present anti-poverty programs stemmed from that, but in the US the war on poverty has been less systematic.
So what is the result? The Census Bureau offers alternative methods of measuring poverty. Under one the rate would drop from 13 percent of Americans to 11.1 percent, and under another it would be only 9.8 percent, or even 6.4 percent. The Office of Management and Budget (which makes the calculations for the Census Bureau) does not plan to change its official definitions, it explains. This would be done only after a full public airing.
In short, the official estimate of poverty varies with the formula used. President Reagan and some supporting economists argue that the estimates are too high: Martin Anderson for example, chief domestic policy adviser, said that poverty had been virtually eliminated. On the other hand, most city slums tell their own story; the social strain has been growing through the recession.