This essay is a commentary on a Research Note by James Trussell, Barbara Vaughn and Joseph Stanford entitled "Are All Contraceptive Failures Unintended Pregnancies? Evidence from the 1995 National Survey of Family Growth."

During the last several decades, the concept of pregnancy intention has been used in many different research endeavors—to estimate "unmet need" for contraception, to make population projections, to examine couples' decision-making processes, to measure the extent to which couples successfully regulate their fertility, to explain women's contraceptive or sexual behaviors, to examine their pregnancy-related behaviors (and thus the outcomes for their infants), and even to explore the relative well-being of their intended and unintended children in later life. As different as these objectives may be, intention has been measured with a relatively small number of items over the years. The questions used evolved from items on ideal family size and from the simple "Do you want to have any (more) children?" to scaled and multiple measures of the strength of the fertility intention.

Because the most commonly used items can hardly be described as serious psychometric measures, it is hardly surprising that, even when a significant correlation is found between intention and behavior, there is generally a fairly large subset of women among whom that correlation is not observed. Moreover, the relationship between fertility intention and childbearing varies by stage of demographic transition and by societal context. Researchers probably use the term "unintended" rather than "unplanned" because intention is perceived as a more universal concept than consciously planned conception. But it is quite possible that, in the United States today, many respondents think "planning" when they are asked questions on intention. At the very least, such seeming contradictions as those reported by Trussell, Vaughan and Stanford suggest that there is not merely a range from unintended to intended, but a continuum from truly unintended, through unplanned, to intended and, finally, deliberately planned. Whereas even the concept of childbearing intentions is foreign to a traditional, more fatalistic society that does not practice fertility control, some Americans may describe a child as intended even when its conception was not consciously planned.

Demographers noted in the 1980s that measures of family size intentions were most useful in the context of stable marriage and infrequent nonmarital conception. In a society in which one-third of births—and an even greater proportion of pregnancies—occur out of wedlock, changes in the sexual dyad can have a profound effect on fertility desires, not only in terms of timing but also in terms of ideal family size. Because these dyads change—and because even marriage is often perceived as unstable—many old assumptions relative to fertility motivation are no longer relevant.

For example, the motivation to stop childbearing has commonly been assumed to be stronger than the motivation to postpone it, but the commitment of some young, single women to avoid premarital pregnancy may be more intense than the expressed inclination of some married women to stop at two. Similarly, young women who expect to marry at some time in the future may place greater importance on avoiding pregnancy with a casual partner than do women in similar alliances who do not foresee marriage at all. That scenario may explain why women in disadvantaged and unstable environments often see their pregnancies not as unintended (as the researcher expects), but as intended.

The relationship between fertility intention and childbearing—and the link between fertility intention and contraceptive use—are strongly affected by other, independent attitudes, such as the attitude toward contraception itself. Therefore, I would suggest that measures of intention have been used to predict what can only be predicted in the presence of cogent measures of contraceptive attitudes, and that the data available on the intention status of births in the United States today may not represent as serious a failure in contraceptive practice as is often supposed. Rather, they may tell us that timing intentions are not compelling. When childbearing is related more to social relationships than to economic necessity, as may be the case today, its timing within the limits of small family size may not be salient.

Knowing women's family size preferences is not enough; we need to understand their views, if any, on contraception. Thus, the difference between present and ideal family size does not truly describe unmet need; even if we believe that women who know nothing of contraception, or know of it and reject it, have a need for family planning, the measurement would more closely reflect demand if it included an assessment of women's attitudes on the subject. Research has demonstrated that, even when a young woman's intention to avoid pregnancy is unequivocal, her attitude toward contraception has to be absolutely positive if she is not to conceive.1 American women's attitudes toward and use of birth control methods clearly do not indicate such positive attitudes, and it is only when we care deeply about an outcome that we are willing to assume the costs of achieving it.

American women seem to be telling us that their timing intentions are often weak; perhaps we should believe them. We should believe them because, regardless of how ambivalent they feel about avoiding conception, and however they may feel about contraception, Americans are not having large families. They may not use contraceptive methods with the regularity or effectiveness with which such methods are used in other industrialized countries, but they stop having children after a relatively small number of births. Of course, it may be argued that, for the majority, the desire to stop childbearing does not lead to effective use of reversible contraceptives; rather, many men and women successfully limit family size by recourse to sterilization. Causality may indeed be expressed in reverse: Couples opt for sterilization because they eventually do have an unequivocal desire to avoid conception. And those who are not sterilized appear to handle contraception well enough to avoid having large families.

Thus, the concept of ambivalence toward both conception and contraception is of central importance in our understanding of contraceptive use. Ambivalence toward—often downright dislike of—birth control methods is no doubt implicated in many of the cases in which an "unintended" conception occurs. Moreover, the judgmental term failure is probably inappropriate in a large proportion of these cases. The evidence that American couples stop having children when they do, that they do not fail to control family size, should make us question whether "failure" really explains the lack of a strong connection between the expressed desire to avoid or postpone a pregnancy and its effective prevention. That lack may perhaps be better explained by the limited utility of the concept, on the one hand, and, on the other, by the weakness of couples' timing intentions.

If we follow the logic of this thinking, it may shed some light on the problem of adolescent childbearing, which has been of such public concern in this country. If, within the limits of a relatively small family, Americans are willing to allow some flexibility in both the ultimate number of children and the timing of their births, then there is probably not the powerful, almost universal relationship between intention and behavior, between the sex act and contraception, that one would hope for in a true "contraceptive society."

Perhaps the lack of such a relationship implies that, despite our high contraceptive prevalence rate, we do not value children enough to believe that unprotected sex should occur only when two persons share a positive and conscious desire for parenthood. Small wonder that this belief has not been transmitted to the young. Because most adults believe that the young should delay childbearing, we expect adolescents to demonstrate a level of cautious, effective contraceptive use well beyond that of large proportions of adult couples—a consummation devoutly to be wished but not always achieved. Perhaps only when childbearing is perceived by the entire society as sufficiently important to occur only in the presence of an unequivocal desire for parenthood can we expect that ideal to be adopted by the youngest couples among us.


1. Zabin LS, Astone NM and Emerson MR, Do adolescents want babies? the relationship between attitudes and behavior, Journal of Research on Adolescents, 1993, 3(1):67-86.