Virginity pledges, in which young people vow to abstain from sex until marriage, have little staying power among those who take them, a Harvard study has found.
More than half of the adolescents who make the signed public promises give up on their pledges within a year, according to the study released last week.
The findings have raised the ire of Concerned Women for America, a prominent conservative organization that advocates adolescent sexual abstinence.
"The Harvard report is wrong," said Janice Crouse, a fellow at a Concerned Women for America think tank.
"This study is in direct contradiction with trends we have been seeing in recent years," Crouse said. "Those who make virginity pledges have shown greater resolve to save sex for marriage."
Virginity pledges were introduced in the early 1990s as part of the Christian Sex Education Project. Their adult champions hail the promises, which rest solely on the individual's word, as being a major step toward reducing teen pregnancy and raising moral values.
By some estimates, at least 2.5 million adolescents around the world have publicly vowed to postpone sex until marriage. They include virgins, as well as those who have had sexual experiences but who swear to refrain from further activity.
Many wear rings or other jewelry to symbolize their pledge.
For the Harvard report, researcher Janet Rosenbaum analyzed data from the National Longitudinal Study of Adolescent Health, a survey conducted by the National Institute of Child Health and Human Development. It is the only government-sponsored study that asks about virginity pledges.
The 14,000 survey subjects were interviewed in 1995 and reinterviewed in 1996 and 2001. They ranged in age from 12 to 18 and came from across the country.
Rosenbaum found that 52% of those who said they had signed virginity pledges had had sex within a year. And of those who had sex after telling the first interviewers they had taken the pledge, 73% denied in the second interview having made the pledge.
"This may indicate that they are not that closely affiliated with the pledge," Rosenbaum said.
The adolescents also were unreliable in reporting their sexual experiences, Rosenbaum said. More than a quarter of nonvirgins in the first interview who later took a virginity pledge said in the next interview that they had never had sex.
"That puts a lot of error in these studies," Rosenbaum said. Virginity pledgers, she concluded, "are more likely to give bad information -- unreliable data -- about their sexual history."
Medical testing is a more reliable gauge of adolescent sexual activity than their own reporting, Rosenbaum said.