Mothers' influence on when their children initiate sexual intercourse appears to be greater for daughters than for sons, according to an analysis of longitudinal data from matched pairs of adolescents and their mothers.1 Five maternal variables affected adolescent females' risk of becoming sexually active: Mothers' impression that their daughters had already kissed was associated with an elevated risk of early sexual initiation, while mothers' education, disapproval of their daughters' having sex, satisfaction with the mother-daughter relationship and frequency of conversations with the parents of their daughters' friends were all associated with a reduced risk. By contrast, the only maternal variable that mattered for young men was the mother's impression that her son had kissed.

The data come from the National Longitudinal Study of Adolescent Health (Add Health), a nationally representative survey of in-school youth in grades 7-12. Investigators used Add Health information from 2,006 matched pairs of sexually inexperienced 14- and 15-year-olds and their mothers. The adolescents and their mothers were interviewed separately at home during the 1994-1995 school year (Wave 1); follow-up data were collected one year later from the adolescents only (Wave 2).

Using Cox proportional hazards models, the researchers assessed how adolescents' and mothers' characteristics affected young people's probability of initiating sexual activity in the year separating the surveys. Among the factors controlled for were social and demographic characteristics and both the adolescents' and their mothers' impressions of how far the adolescent had progressed in a romantic relationship. The models also included maternal communication variables about sex, as well as variables measuring the mother's values and her satisfaction with the mother-child relationship.

Overall, 74% of the respondents were white, 12% Hispanic, 10% black and 4% of other races and ethnicities. Sixty-three percent of the adolescents lived with both married biological parents, and one-quarter of the mothers had completed college. A significantly higher proportion of female than of male adolescents had initiated sexual activity by Wave 2 (16% vs. 11%); this result is consistent with the differences by gender in the proportions who said at Wave 1 that they were currently in a romantic relationship (24% of females vs. 18% of males) or had ever been in one (42% of females vs. 35% of males).

Mothers of daughters reported discussing sexual issues more often than did mothers of sons (i.e., 30% of mothers of daughters discussed sex a "great deal" with their child vs. 24% of mothers of sons), although women who had a daughter also felt more uncomfortable discussing sex with their child (i.e., 43% of mothers of daughters strongly agreed that they felt uncomfortable vs. 33% of mothers of sons). On the other hand, mothers of sons were significantly more likely than mothers of daughters to recommend a contraceptive method to their child (i.e., 35% of mothers of sons agreed or strongly agreed that they had recommended a specific method to their child vs. 22% of mothers of daughters). The remaining maternal variables did not differ significantly by the adolescent's gender.

For adolescent males, three variables independently affected the probability of having initiated intercourse by Wave 2, and only one was a maternal variable (i.e., the mother's views on whether her son was in a romantic relationship). Being black (hazards ratio, 3.4), having been in a romantic relationship in the past 18 months (2.0) and having a mother who believed that her son had kissed (2.0) were associated with an increased likelihood of early sexual debut.

Although this last variable--a mother's impression that her child had kissed--also proved significant for females (hazards ratio, 2.0), four additional maternal factors affected the likelihood of initiating sexual activity between the surveys among adolescent women, and all of these significantly reduced that likelihood. Adolescent females had a reduced risk of sexual initiation if their mothers had graduated from college (0.6), strongly disapproved of their having sex (0.6), were satisfied with the mother-daughter relationship (0.6) or talked with more parents of their friends (a proxy for paternal monitoring, 0.9).

The investigators caution that the short observation period prevents them from drawing conclusions about causality and that their data are limited by the exclusion of youth who are no longer in school (and who would be at a high risk of early sexual debut). Further, they warn that the lack of maternal influence on young men's early sexual activity might stem from sample selection, higher attrition among males than females and males' lower rates of sexual initiation over the study period.

Nonetheless, the investigators observe that mothers appear to directly influence their daughters' sexual behavior by communicating a strong disapproval of sex, by maintaining a close mother-daughter relationship and by staying in touch with the parents of their daughters' friends. The reasons why mothers' influence seems to matter more for daughters than for sons are unclear, but the researchers suggest that the messages (verbal and nonverbal) that mothers send to sons about sexual activity may be less clear than the ones they transmit to daughters. The authors assert that "parents need to be clear about their values and then clearly articulate them to their children." --L. Remez


1. McNeely C et al., Mothers' influence on the timing of first sex among 14- and 15-year-olds, Journal of Adolescent Medicine, 2002, 31(3):256-265.