Discussion


In line with the findings at Phase 1 of the study when the children were in their preschool or early school years, no differences were identified between the single mother families and the two-parent families in parenting or child adjustment when the children reached middle childhood. With respect to parenting, the family types were similar in terms of the quality of mother-child relationships as assessed by standardized interview, the quality of mother-child interaction as assessed by an observational measure, and parental acceptance as measured by questionnaire. In addition, there were no differences between the single and partnered mothers in anxiety, depression, or stress associated with parenting, and the mothers’ mean scores for anxiety and depression were below the cut-off points of 40 (Grant, McMahon, & Austin, 2008) and 13 (Matthey, Henshaw, Elliott, & Barnett, 2006), respectively, based on normative data. Thus, using multiple methods involving both representational and behavioral measures of parent–child relationships (Imrie, Jadva, Fishel, & Golombok, 2019) with both mothers and children, and standardized measures of mothers’ mental health, it appeared that families formed by single mothers by choice were functioning as well as families with two parents when their children were around nine years old.
There was a similar pattern of findings for child adjustment. The children of single mothers by choice did not differ from children with two parents in terms of their scores on the Strengths and Difficulties Questionnaire as completed by mothers. It should also be noted that the average scores were in line with the population norms of 8.4 and 6.6 for the parent and teacher SDQ, respectively (Meltzer, Gatward, Goodman, & Ford, 2000). Moreover, the number of children with scores above the cut-off point for psychiatric disorder did not differ by family type (n = 6 and n = 6, respectively). The teachers’ scores on the Strengths and Difficulties Questionnaire did differ between family types, with the children of single mothers obtaining significantly higher scores than those in two-parent families. However, the teachers’ scores were within the normative range, suggesting that the children in both family types were generally well adjusted. It should also be noted that the significant finding for teachers’ SDQ scores may have resulted from chance, given the large number of group comparisons. In terms of the number of children with teachers’ scores above the cut-off for psychiatric disorder, there was no difference between the children with one and two parents, and the proportion scoring above cut-off in both family types was small (n = 7 and n = 4, respectively), reflecting a smaller percentage in the problematic range compared to general population norms (Meltzer et al., 2000). In addition, the independent psychiatrists’ ratings suggested that the children were functioning well, with only one child in each family type displaying severely problematic behavior. It seems, therefore, that even when donor-conceived children of single mothers reach the age at which they understand the significance of not having a father, they are no more likely to show adjustment difficulties than children who grow up with a father.
Although there was little difference in parenting or child adjustment between family types, in line with a developmental systems conceptual framework (Aldwin, 2014; Overton, 2015), higher levels of parenting stress and higher levels of children’s prior adjustment difficulties were each associated with children’s adjustment difficulties in middle childhood irrespective of family type. The finding for parenting stress replicates that of the first phase of the study (Golombok et al., 2016) and supports the hypothesis that children’s emotional and behavioral problems would be associated with maternal mental health problems. Similarly, the association between children’s adjustment difficulties at Phase 1 and Phase 2 of the study is consistent with the hypothesis that adjustment difficulties in middle childhood would be associated with children’s preexisting emotional and behavioral problems. The lack of association between either maternal anxiety or depression and children’s adjustment difficulties appears to reflect the low levels of anxiety and depression in the sample. For example, only five women’s ratings of depression fell within the problematic range on the Edinburgh Depression Scale. Our findings highlight the value of considering multiple components of parents’ wellbeing, everyday stressors, rather than clinically relevant symptoms of anxiety and depression, contributed to children’s adjustment difficulties.
Interestingly, in contrast to the findings reported at Phase 1 (Golombok et al., 2016), financial difficulties did not appear to contribute to children’s adjustment problems at Phase 2. Time-related characteristics may help explain this null effect, specifically the lack of stability in financial difficulties and the differences in working patterns from Phases 1 to 2, with 16 mothers entering the workforce during this period. The small minority of families who experienced definite financial difficulties at Phase 2 is consistent with previous research showing that single mothers by choice are generally financially secure (Bock, 2000; Graham & Braverman, 2012; Jadva et al., 2009), and may explain why financial difficulties were not associated with children’s adjustment problems.
Studying the adjustment of donor-conceived children born to single mothers by choice is not only of interest in its own right as little is known about the development and wellbeing of children in this new family form, but also is of theoretical interest as it enables the effects of growing up in a single mother family to be investigated without the potentially confounding effects of parental conflict, financial difficulties and maternal mental health problems. The similarities in parenting and child adjustment between children in one-parent and two-parent families in the present study are in direct contrast to the findings of studies of families headed by divorced or unmarried single mothers, which found higher levels of children’s emotional and behavioral problems compared to children in two-parent families. This discrepancy may be attributable to the differing social circumstances of single mothers by choice, who made an active decision to parent alone and planned their lives accordingly, and divorced and unmarried single mothers, who found themselves in this situation unintentionally. The findings of the present study thus add weight to the view that the raised levels of adjustment problems shown by children of divorced and unmarried single mothers result from the adverse circumstances that often accompany single motherhood, rather than single motherhood, in itself.
A limitation of the study is the modest sample size, which may have resulted in differences between the single mother and two-parent families not being detected. The relatively low intraclass correlation coefficients for mother-to-child warmth and child-to-mother warmth are likely to have resulted from the lack of information on nonverbal aspects of warmth, such as facial expressions, that were taken into account in the ratings made by the interviewer, but were unavailable to the second rater who was coding from audio recordings. A further limitation is that participants were originally recruited through a private fertility clinic, and thus the findings may not reflect the experiences of parents and children in families formed by single mothers by choice using other routes, such as sexual intercourse or online connection sites (Jadva, Freeman, Tranfield, & Golombok, 2018), about whose socioeconomic circumstances little is yet known.
An important advantage of the study was that the children in the comparison group of two-parent families had all been conceived by donor insemination thus controlling for the use of donor insemination by the single mothers by choice. In addition, a high retention rate was obtained at Phase 2, and the use of similar measures at both phases of the study enabled the Phase 1 scores to be controlled for in the Phase 2 analysis. A further advantage was that the families were recruited when the children were young, thus avoiding sample bias resulting from children who were distressed about their family declining to join the study.
With the exception of the investigation by Chan, Raboy, and Patterson (1998), which focused primarily on single lesbian mothers and produced similar findings to the present investigation, this is the only study of parenting and child adjustment in families formed by single mothers by choice when the children reach middle childhood. Although, by this age, children have acquired a more sophisticated understanding of what it means to be conceived by donor insemination to a single mother and not know the identity of their biological father, they continued to show positive relationships with their mothers and high levels of psychological adjustment. This suggests that the presence of two parents—or of a male parent—is not essential for children to flourish, thus adding to the growing body of evidence (Golombok, 2015; Lamb, 2012; Patterson, 2009) that family structure is less influential in children’s adjustment than the quality of family relationships.