The research, part of the larger UNC-based National Study of Youth and Religion, revealed a statistical association between religion and better behavior only among teens who went to religious services at least once a week, however, or who professed deeply held spiritual views, said study director Dr. Christian Smith.
Few associations were found among adolescents who attended irregularly or who said religion was only modestly important.
"We found that kids who go to church regularly or who say that religion is important in their lives are much less likely to be involved in various forms of substance abuse, get into trouble, commit crimes, are less involved in violence, have school problems and have difficulties with their parents," said Smith, professor of sociology at UNC. "They are more likely to behave safely, try to stay healthy and be involved in volunteering, sports and other community activities.
"Our findings are not radically surprising in that they support some earlier, smaller-scale work on this issue, which is something to which not much attention has been paid by most academics," he said. "For example, in the past, people who study adolescents have often neglected or completely ignored the religious factor in teen-agers' lives."
Some social science investigators have even assumed the religion had no effect or had a pernicious influence on teens, Smith said.
Conducted with doctoral student Robert Faris, the UNC study relied on data gathered through Monitoring the Future, a nationally representative University of Michigan survey of 2,478 high school seniors, he said. The new work, released in a report today (Sept. 18), is among the most comprehensive looks yet on the link between religion and positive and negative adolescent behavior.
"One of the most interesting observations is that the religious correlation doesn't seem to kick in until it reaches the level of the most religious kids," Smith said. "That suggests a threshold below which there's little or no association."
The sociologist said he wanted to be clear that the study revealed the constructive linkages without showing yet what caused what. He and others cannot separate the effects of what morals children were taught through religious traditions, for example, from possible effects of being part of social networks that include adult role models. A paper Smith will publish next year will offer nine different hypotheses, or possibilities, about how religion may influence adolescents.
"It could also be that kids who are initially religious and start getting into trouble drop out of religion because it feels uncomfortable for them," he said. "Then when someone takes a survey, those teens show up as being not very religious, and so there is an apparent association."
Among specific findings were that especially religious youths were less likely to smoke, drink and use drugs and more likely to start later and use less if they started at all, he said. They went to bars less often, received fewer traffic tickets, wore seat belts more, took fewer risks and fought less frequently. Shoplifting, other thefts, trespassing and arson also were rarer.
"Religious 12th--graders argued with parents less, skipped school less, exercised more, participated more in student government and faced fewer detentions, suspensions and expulsions," Smith said. "These findings were statistically significant even after we controlled for race, age, sex, region, education of parents, the number of brothers and sisters and other factors."
Lilly Endowment Inc. is funding the four-year UNC project, which began in 2001. Among goals are to identify effective practices in the religious, moral and social formation in young people's lives and to foster informed national discussions about the influence of religion on adolescents.
Note: To reach Smith or for copies of the report, call Roxann Miller, director of communications for the National Study of Youth and Religion at 919-966-1559. More information is available at www.youthandreligion.org.
AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.