Abstract
Despite the widespread assumption that paternal investment is substantial in our species, previous studies have shown mixed results in relation to the impact of fathers on both offspring survival and reproductive outcomes. Using data from a large representative sample of British men, we tested whether father absence is associated with the timing of reproduction-related events among boys, while controlling for various cues denoting early childhood adversity. We further tested whether the loss of the father at different childhood stages matters, so as to assess whether early life is the most important period or if effects can be seen during later childhood. The results show that father absence before age seven is associated with early reproduction, while father absence between ages 11 and 16 only is associated with delayed voice-breaking (a proxy for puberty), even after adjusting for other factors denoting childhood adversity. We conclude that fathers do exert an influence on male reproductive outcomes, independently of other childhood adversities and that these effects are sensitive to the timing of father absence.
1. Introduction
Compared with other mammals, human fathers invest heavily in their offspring [1]. A tenacious theory in evolutionary anthropology is that fathers are essential providers of resources for their mates and highly dependent offspring and that this leads to stable nuclear families [2]. Empirical evidence, however, does not always support this view; fathers are often reported to have little if any positive effect on infant survival or health [3], and more recently, evolutionary anthropologists have suggested that women draw upon other individuals such as grandmothers and their older children for help in raising viable offspring [4,5]. It is possible, however, that fathers benefit older offspring by positively affecting their reproductive outcomes.
Research on the influence of fathers on reproductive outcomes has mainly focused on girls within a developed world context and generally shows that father absence accelerates life events: fatherless daughters tend to mature earlier [6–9], fare poorly academically [10] and reproduce earlier [11,12] than girls from two-parent families. This is usually attributed to the life-history theory tenet that sub-optimal conditions in early life are cues to a shortened life expectancy and so, assuming that father absence is indicative of a stressed childhood environment, early reproduction is the optimal strategy [13,14]. These studies often do not separate father absence from general childhood adversity and so do not necessarily show that father absence itself is significant for reproductive timing. A handful of studies have reported similar findings for males in Western populations showing father absence to be associated with younger ages at puberty [8] and parenthood [15], in contrast to studies among forager populations that generally find that father absence is more likely to result in the delaying of male reproduction [16,17] or to have no effect at all [18].
This study aims to disentangle these two alternative (although not necessarily mutually exclusive) explanations for father-absence effects by determining whether fathers exert an independent influence on offspring outcomes or if father absence may simply be an indicator of a sub-optimal environment. We do this by developing models that examine the association between father absence and the timing of reproduction-related outcomes, controlling for other factors representing cues to a deficient early environment. We also separate father absence into three age groups in order to isolate early childhood factors (argued to be crucial time influencing life-history strategy [7]) from later childhood stages.
2. Material and methods
We used data from the UK National Child Development Study (NCDS), to examine the association between father absence and timing of voice-breaking (a proxy for puberty), marriage, reproduction and intended ages for both of these (some evidence shows that childhood adversity is also related to female teenagers' intentions for timing of marriage and childbearing [12]).
The NCDS is an ongoing survey of all British children born in the first week of March 1958, providing an array of information on socio-economic, health and family circumstances over time [19]. We use data from five waves: birth, age seven, 11, 16 and 23, because information about early-life conditions as well as the timing of reproductive events can be captured by these ages (n = 9596). There is a substantial amount of missing data in the NCDS and so multiple imputations for missing data were performed [20]. Binomial logit models were fitted to estimate the odds ratios for the timing of each life event in relation to father absence at different times during childhood, adjusting for other early childhood factors. Father absence is defined as having no natural father in the household at age seven, or the father becoming missing from the household between ages seven and 11, or between ages 11 and 16; the reference category included children whose natural father resided in the household up to age 16. Table 1 shows the proportions of boys who had experienced these events.
Descriptions of key variables.
Variables denoting a sub-optimal early environment are: breastfed for less than one month, more than two household moves, separated from mother for more than one month and father's social class at birth. These were included simultaneously in the regression models in order to control for them. These variables have all been shown to have an accelerating effect on female reproductive outcomes in this cohort [14]. We also control for birth weight and body mass index (BMI) at age seven. Among girls, BMI is associated with early pubertal maturation [21,22] and birth weight with early reproduction [12]. Among boys, some evidence shows that BMI at age five is correlated with pubertal maturation [22,23], and that birth weight may also be important [24].
3. Results
Father absence was associated with pubertal maturation but only if the father became absent during the adolescent period (table 2 and figure 1). Father-absent boys were also less likely to have been married by age 23 but the only statistically significant association was for father loss between ages seven and 11. Father absence was not associated with stated intentions to either reproduce or marry; effect sizes were very small, non-significant, and the direction of the effects was inconsistent across all age groups. Father absence was, however, associated with early reproduction: boys with absent fathers were more likely to have had at least one child by age 23, regardless of the age at which they lost their father, although the only statistically significant effect was for those who lost their fathers before age seven, suggesting that this might be a particularly influential period (table 2 and figure 2). These results hold despite controlling for variables indicating a sub-optimal childhood environment. As expected, two early environment factors, social class and a short period of breastfeeding, are associated with early reproductive outcomes although not for voice-breaking.
Results from binary logit models showing odds ratios and 95% confidence intervals.
Father absence and predicted probabilities of voice breaking by age 13, n = 8638. Dark grey bars, father present; light grey bars, father absent.
Father absence and predicted probabilities of reproduction by age 23, n = 8640. Dark grey bars, father present; light grey bars, father absent.
4. Discussion
These findings suggest that father absence is associated with age at sexual maturity and reproduction in boys and that these effects are sensitive to the timing of father absence. Notably, the correlations between father absence and reproductive events are not consistent in their direction. For girls, father absence in developed societies is consistently associated with earlier maturation, earlier onset of sexual activity and earlier first childbirth [8,25]. Here, we show that for boys, father absence is associated with later puberty, later marriage, but earlier reproduction. Males are able to invest in either a mating (desertion) or parenting (commitment) strategy that is unavailable for females who are committed to intensive parental investment, at least among humans. Differences in the direction of the father-absence effects on marriage and reproduction may thus be attributable to the fact that investing in either mating or parenting are viable alternative strategies for males.
Voice-breaking, unlike early reproduction, is not associated with other indicators of early-childhood conditions, reinforcing the suggestion that it is specifically the absence of the father which influences reproductive events. It is possible that the stress of losing a father may have a proximate effect on adolescent hormonal responses and delay puberty [26]. Regarding reproduction, it may be that fathers are able to influence their son's educational attainment [10] thereby delaying starting a family in favour of investing in non-somatic capital. Another possibility is that sons may inherit (genetically or otherwise) their father's non-committal behaviour, as there is some evidence to show that daughters may genetically inherit a tendency for sexual precocity from their fathers [27].
It could also be that the presence of an unrelated male affects male reproductive outcomes and not father absence per se, as has been suggested in other similar studies [15]. We attempted to test for this by comparing single-mother families to those with unrelated male household heads and found no significant differences between these groups, although sample sizes were small and reported household residents may not necessarily reflect the presence or absence of an unrelated male in the child's life (results in the electronic supplementary material).
Overall, these findings suggest that father absence exerts an influence on male reproductive decisions, even after controlling for other indicators of early life adversity. This study also emphasizes the importance of analysing the role of fathers during both early and later childhood stages, as the effects of father absence on puberty are confined to boys who lost their fathers during adolescence. Although early reproduction appears to be particularly sensitive to father absence in early childhood, we argue that father absence is not simply one component of a sub-optimal childhood environment, because our results are robust to multiple controls denoting unfavourable early conditions, signifying that it is father absence itself that matters. In contrast to traditional societies, where father presence may facilitate marital and reproductive opportunities, father presence in developed world contexts may encourage the acquisition of human capital (such as education) in their sons, and therefore delay reproduction. Further research on the specific mechanisms through which father presence affects children's reproductive timing would be illuminating.
Acknowledgements
We thank The Centre for Longitudinal Studies, Piero Stanig, Jonathan Jackson, Jouni Kuha, David Lawson, Andy Wells and two anonymous reviewers for helpful comments. This study was funded by the Economic and Social Research Council.
- Received July 20, 2011.
- Accepted August 5, 2011.
- This journal is © 2011 The Royal Society
References
