Teen-agers who drive or hold jobs are more likely to run into trouble with the law, according to a government-backed study released yesterday.

The six-year study, entitled "Assessing the Relationship of Adult Criminal Careers to Juvenile Careers" and released by the Justice Department, said it was nearly impossible to predict whether juvenile delinquents, except those with long histories of criminal activity, will become adult criminals.

But it offered several observations about why young people get in trouble with the law in the first place, and some of those conclusions fly in the face of commonly accepted explanations. The study found, for example, that broken homes have little to do with producing juvenile delinquents and that the age at which a young person leaves home does not seem to have any impact on criminal behavior.

Potentially its most controversial finding was that teen-agers who work are more likely to run afoul of the law.

According to Pamela Swain, project manager for the Justice Department on the study, young people with "after-school jobs, evening jobs, are more likely to get into trouble. These youngsters have more opportunity to make a wider range of contacts outside the school setting, more opportunity to get involved with the criminal element."

Coincidentally, the Reagan administration proposed just last month a relaxation of federal child labor regulations that would allow 14- and 15-year-olds to work longer after-school hours. The proposals were withdrawn, at least until after the November elections, partly because of protests from labor and child development organizations that the change would contribute to academic trouble and delinquency among teens.

Experts in the juvenile justice field who were questioned about the findings of the study, however, expressed skepticism.

Alvin Bronstein of the National Prison Project said it may be valid that cars contribute to juvenile delinquency, "but with jobs, it seems to me it would be just the opposite."

David Howard, a lawyer with the National Center for Youth Law, said the conclusions relating to jobs and broken families "run counter to a uniform body of literature dating back a decade.

"If a kid or a young adult criminal has access to a job or marketable skills and has a family that cares about him or her, it would indicate to me that he is less likely to become a hard-core criminal," said Howard, who has worked in the juvenile law field for more than a decade. The study found that the delinquency rate significantly increases for those who hold their first full-time job at age 17 or earlier.

Swain, who works in the Justice Department's National Institute on Juvenile Justice, added that the study found no difference in the delinquency rate for young people who come from two-parent families, single-parent families or foster homes. "Instead, it seemed to be the quality of the relationship that matters," she said.

The study found driving is a major factor in leading young people into trouble. "We saw how frequently the automobile got juveniles into trouble with the police and how other behaviors were related to it," it said.

"It was not an early driver's license 'per se' that resulted in police contacts, but simply having access to the automobile, just as early employment may have exposed some juveniles to greater risk and also given them funds to be spent in a trouble-producing way during the years of socialization," the report said.

The six-year study tracked the criminal careers of 6,127 people from 1974 to 1980 in Racine, Wis., which was called a "microcosm" of the United States.

Bronstein objected that it is not valid to call Racine, a city of 100,000, a microcosm. "It is not a big, urban area, nor is it a suburb, nor is it a small town," he said. Swain said the study is just one of many being done in various cities.

"Much of the concern about juvenile delinquency has been based on the premise that it leads to adult crime," said study director Lyle Shannon, head of the Iowa Urban Community Research Center at the University of Iowa.

But he concluded that although there is some relationship between juvenile delinquency and adult crime, many adult criminals had no police records as youths, and many juveniles with five or more police contacts became law-abiding citizens.

A study several years ago by the National Prison Project would appear to support the second part of that conclusion, but not the first. Bronstein said the study found that only 10 percent of juvenile offenders go on to be adult offenders, but it also found that "90 percent of the people in state prisons had some problem with the law as juveniles in which they were in some custodial arrangement -- reform school, an institution, a halfway house."

The new study said the juvenile justice system often fails to reform young offenders, and those who go through the system, especially boys, often react by by committing more crimes.

"What we found . . . was an increase in frequency and seriousness of misbehavior in the periods following those in which sanctions court-ordered treatment and correction programs were administered," the study said. "With few exceptions intervention by the agencies of social control does not play even a moderate role in decreasing the seriousness of adult contacts with police ."

Shannon recommended that in most cases counseling and release by the police officer "is probably a wise policy because fewer youth are brought into the justice system, a step for which we see little evidence of positive results."

The study showed delinquent behavior generally declined in seriousness and discontinued after the teen-age years. Law-abiding adults who had been juvenile delinquents said they changed their behavior after realizing that what they thought was fun as a juvenile "was no longer appropriate behavior as an adult" -- not out of fear of being arrested, the study said.

Shannon concluded that the criminal justice system should concentrate its resources on the "hard core" juvenile offenders.

The study was funded with grants from the Justice Department, the Fleischman Foundation and the University of Iowa. The government contributed $360,000 to the study