* Should marijuana be legalized? * Is extramarital sex wrong?
* Do you attend a religious service regularly?
* Should a Communist be allowed to teach in college?
* Would you vote for a black president?
Answers to such questions in a survey conducted by the National Opinion Research Center point to a potentially significant pattern of change in attitudes of young people in the United States: Today's 18- to 24-year-olds, especially those with college training, tend to be more conservative than their counterparts of a decade ago where ''traditional values'' are concerned. At the same time, they remain liberal on rights of blacks and women and other matters of individual choice.
In the early 1970s, for example, 68 percent of the college group 18-to-24 years of age thought extramarital sex was generally wrong; in the early 1980s that figure has risen to 85 percent. About 35 percent of the young college group of the early 1970s favored capital punishment for convicted murderers; in the 1980s a high 67 percent do.
A decade ago 68 percent said a woman should be able to obtain a legal abortion if she is married and wanted no more children; today only 49 percent approve that choice. Moreover, today 49 percent of the 18-to-24 college group believe a woman should be able to obtain a legal abortion for any reason - a much smaller percent than for the older-age college groups.
Yet, reflecting their liberalism in other areas, 80 percent of the 18- to 24 -year-olds say there should not be laws against interracial marriage (the same percentage as in the early 1970s); 89 percent would vote for a qualified black for president (as against 86 percent in the early 1970s); and 94 percent would vote for a qualified woman for president (86 percent in the '70s).
The data are gathered and evaluated in the current issue of Public Opinion, a bimonthly published by the American Enterprise Institute. Among other interesting findings: In the early 1980s the 18- to 24-year-olds are more inclined to think that marijuana should not be legalized (61 percent as against 56 percent in the 1970s). They are also less disposed to accept that an avowed homosexual should be allowed to make a speech in the community (although the vast majority believe he should be).
High school seniors, for their part, have adopted the liberal view that men and women should be paid the same for the same work and that having a job gives a wife more opportunity to develop her capacities. But they appear to be having doubts about whether a wife should work if there are one or more preschool children in the family. Some 50 percent said yes in 1975: only 36 percent said yes in 1982.
Social analysts say it is too early to draw solid conclusions from such figures. But the data are viewed as a possible bellwether of where US society is headed, for it is the young who are the proponents and drivers of change. The survey seems to indicate that the 18- to 24-year-olds are more conservative on issues affecting how society operates and ''liberal and liberated'' where an expansion of individualism is concerned.
Today's generation is seen going through a process of ''sorting out'' the tumultuous changes that have taken place in moral and cultural values in the past 30 years. Daniel Yankelovich, in an accompanying interview in the magazine, suggests that the 1960s and '70s were a time of reaction against the moral and social conformity of the '50s. Instead of duty to others, young people turned to self-fulfillment and satisfying personal desires. Instead of commitment, they swung to ''freedom from commitment.'' Instead of external success, they focused on the ''inner life.''
There has been no great break with the 1960s and '70s, says Mr. Yankelovich, but with the sorting-out process has come a sense of the ''real world'' and the feeling that perhaps it is not possible to have every conceivable experience. He detects a shift away from preoccupation with self to the world outside, a greater sense of pragmatism, a willingness to cut corners to achieve success, and a redefining of pleasure away from sex and drugs to the pursuit of physical well-being.
This shift to the ''world outside,'' says Mr. Yankelovich, accounts for the increased support among young people for the death penalty. Less a sign of a shift to the right, he believes it reflects a concern on college campuses about rape and crime. Similarly, a ''new realism'' has led to a change of attitude about abortion.
''The attitude is a bit more hard-boiled,'' the sociologist says. ''Not soft, not idealistic, but anti-permissive and pro-discipline. It goes with the reaction to the view that anything goes.''
Today success is being redefined in a way that seems to synthesize both the material drives of the 1950s and the search for self-fulfillment in the 1960s and '70s: ''Now we emphasize both inner and external expressions of success,'' the pollster says. ''There is a great deal more flexibility. You don't have to make big bucks to be successful. You don't have to reach the pinnacle. There isn't the compulsive, driving, or rigid hierarchial meaning that characterized success in the 1950s. It's more supple, more varied, more pluralistic.''
Continuing changes are observed in family life as well. Disappointed with their experiments in self-fulfillment in the 1960s and '70s, people are now seeking closer bonds, Yankelovich says. There is greater emphasis on couples, on family, and on community. In line with this, the Public Opinion roundup shows that 34 percent of college students and graduates aged 18 to 24 thinks an ideal family should have three children - up from 20 percent in the early 1970s. Most still think two children is ideal but fewer favor just one.
Yankelovich, however, does not see a return to the times when everyone was supposed to get and stay married.
''There is no great pressure for everyone to get married,'' he says. ''There is no great pressure to have children. The idea of getting married and divorced and marrying again is much more acceptable than it was. I doubt we will see a return to the animosity toward divorce.''
As they evaluate the newly gathered data about values, the analysts caution against trying to make precise judgments about ''liberalism'' and ''conservatism.'' ''Liberalism and conservatism redefine themselves in every era ,'' says Yankelovich, ''and that is why the terms are confusing. . . . You could have a liberalism of the 1980s that was extremely pluralistic on life's choices for the individual. But where the lines have been drawn, there are harsh penalties if you cross them. The liberalism and conservatism lines are blurred. . . .''
According to the opinion roundup, 44 percent of the 18-to-24 college group and 36 percent of the 25-to-29 college group, and 50 percent of the 30-to-34 group consider themselves liberal. Those with only a high school or some college education tend to be even less liberal.