Demographic characteristics

There were no significant group differences on child age, child IQ, child ethnicity, number of parent/caregivers, parent/caregiver informant, birth order, total number of people living at home, and parental psychopathy. The groups differed on SES, with the TD group having significantly higher SES than the CP/HCU group, F (2,143) = 4.332, p = 0.015; no other group differences were found. The two CP groups were higher on ADHD, F (2,144) = 23.556, p < 0.0001, generalized anxiety, F (2,144) = 14.531, p < 0.0001, major depression, F (2,142) = 13.962, p < 0.0001, and child drug use, F (2,143) = 6.020, p = 0.003 use as compared to the TD group; no other significant group differences were found on these variables. The CP/LCU group reported higher alcohol use than the TD group, F (2,142) = 3.971, p = 0.021; no other group differences in alcohol use reached significance.

APQ parent report

There was a group difference on parent-reported poor monitoring and supervision, F (2,143) = 5.044, p = 0.008, (TD M = 9.12; CP/LCU M = 11.34; CP/HCU M = 12.99). Pairwise group comparisons revealed significant differences between TD and CP/HCU groups with a medium effect size (p = 0.005; d = 0.72). No other significant group differences were found.

There was also a group difference on parent-reported inconsistent discipline, F (2,143) = 6.783, p = 0.002, (TD M = 7.75; CP/LCU M = 10.26; CP/HCU M = 10.18). Pairwise group comparisons revealed significant differences between TD and CP/HCU groups with a medium effect size (p = 0.007; d = 0.67) and TD and CP/LCU groups with a medium effect size (p = 0.003; d = 0.68). The CP/HCU and CP/LCU groups did not differ significantly from each other.

No group differences emerged on parent-reported involvement, F (2,143) = 0.623, p = 0.538, positive parenting, F (2,143) = 0.915, p = 0.403, or corporal punishment, F (2,143) = 2.252, p = 0.109.

Covariate analysis

The effect of group on parent-reported poor monitoring and supervision was no longer significant after adjusting for AUDIT, F (2,125) = 1.405, p = 0.249.

The effect of group on parent-reported inconsistent discipline remained significant after adjusting for SES, ADHD, GAD, MDE, AUDIT and DUDIT, F (2,125) = 4.806, p = 0.010 (TD / CP/LCU d = 0.605; TD / CP/HCU d = 0.650).

APQ child report

There was a group difference on the child-reported involvement with father subscale, F (2,143) = 3.473, p = 0.034, (TD M = 23.24; CP/LCU M = 19.60; CP/HCU M = 16.95). Pairwise group comparisons revealed significant differences between TD and CP/HCU groups with a medium effect size (p = 0.026; d = 0.56). There was no significant difference between TD and CP/LCU groups or the two CP groups on the involvement with father subscale.

There was a trend-level difference in child-reported inconsistent discipline by group, F (2,143) = 2.904, p = 0.058 (two tailed), (TD M = 7.38; CP/LCU M = 8.91; CP/HCU M = 9.30). Because our predictions were one tailed, we conducted pairwise group comparisons which showed the difference between TD and CP/HCU groups had a large effect size (p = 0.066; d = 1.18). There was no significant difference between TD and CP/LCU groups or the two CP groups on the inconsistent discipline subscale.

The groups did not differ on child-reported involvement with mother, F (2,141) = 2.092, p = 0.127, positive parenting, F (2,141) = 0.055, p = 0.947, poor monitoring and supervision, F (2,143) = 1.168, p = 0.314, or corporal punishment, F (2,142) = 1.632, p = 0.199.

Covariate analysis

The effect of group on child-reported father involvement was no longer significant after adjusting for SES and MDE, F (2,125) = 0.625, p = 0.537.

Post hoc analyses on parent/caregiver–child agreement on APQ ratings

Because partially different patterns of findings emerged in parent and child APQ analyses, we ran post hoc intra-class correlation (ICC) analyses for all the APQ scales that were comparable between parents/caregivers and children (i.e., all except parental involvement, which was assessed separately with regard to mothers and fathers in the child APQ) in the CP groups. These analyses showed moderate agreement between parents/caregivers and children (ICC range = 0.44–0.68). These analyses suggest that although parent/caregiver and child assessments of parenting variables relate to each other meaningfully, they are not identical and likely explain why some differences emerge in the group analyses of parenting.

Qualitative findingsChallenges of parenting CP/HCU and CP/LCU children

Parents/caregivers of CP/HCU boys described grave concerns for their child’s safety, with one parent describing it as ‘my biggest fear’. Concerns over safety were described by parents/caregivers of CP/HCU boys both in terms of difficulties in monitoring their child’s whereabouts and ‘keeping him safe and off the streets’, as well as, problematic peer affiliations, which one parent described as, ‘worry over his safety and peer pressure to engage in unsociable behavior or illegal activity’.

Parents/caregivers of CP/HCU boys described challenges with extreme child behavior including, ‘violence’ and ‘aggression’, and, ‘the unpredictable outbursts which can escalate in seconds’. CP/HCU parents also described fatigue and stress from parenting their child. One CP/HCU parent described the last six years as being an ‘emotional and stressful time… this was very hard for the whole family, especially me’.

Parents of CP/LCU boys, on the other hand, described difficulties with parental influence, including challenges in motivating their child, ‘trying to persuade him to do something he doesn’t want to do’, as well as, difficulties with enforcing rules and boundaries, ‘will not confirm or follow a routine… cannot follow one instruction’, and, ‘instilling a stronger sense of discipline’.

CP/HCU and CP/LCU boys’ description of being parented

Both CP/HCU and CP/LCU boys described parental support, with CP/HCU boys describing parental willingness to support in the face of adversity, with one CP/HCU boy reporting ‘it didn’t matter how I treated her, she was always nice to me’. CP/LCU boys, on the other hand, described support in terms of parental understanding, ‘she understands me and now I realize how well she has raised me’, and guidance, ‘If I do something wrong, I am usually spoken to; If I do something right I am praised’.

A considerable number of CP/HCU (31%) and CP/LCU (22%) boys described the experience of being parented solely in terms of provision of basic needs with no mention of any emotional support or affection, ‘she gives me food, she dresses me, she pays for my house bills’. The absence of emotional descriptions occurred very infrequently in TD boys’ descriptions of caregiving (8%).

A full description of extracted themes and supporting quotations from parents/caregivers and children are included in Online Resource 4.