Research: Family Environment

Part of NewgonWiki's research project
Research flaws and false constructs   Methodological flaws/false constructs Minor-Adult sex   Prevalence of harm Association or causation? Secondary harm Family environment Effects of age on outcomes Minors   Commercial and online victimization Youth sexuality Sexual repression Cognitive ability Teen pregnancy Effects of pornography "Child Sex Offenders"   Characteristics of the offender Who offends and how often? Recidivism Minor attraction   Child pornography Cognitive distortion Abnormal psychology Pedophilia as an orientation Nonsexual aspects Prevalence Dangers of stigma A "cure" for pedophilia? Broader perspectives   Non-human relationships Historical relationships Nonwestern relationships Double-Taboo (Incest, Prostitution) Evolutionary Perspectives
Template: Research - This template

Web Archive (within series)

Research has consistently shown that having experienced 'child sexual abuse' is associated with having a negative family background. Since a poor family background may exert an independently negative effect, this raises the issue that, when the effects of 'child sexual abuse' (CSA) are studied, some or most of problems correlated with CSA may actually have been caused by the associated family background features.

The best and most well-known study to analyze the influence of family environment on effect size was A meta-analytic examination of assumed properties of child sexual abuse (CSA) using college samples, published in 1998 by Bruce Rind, Philip Tromovitch and Robert Bauserman in the Psychological Bulletin. They found that the presence of 'child sexual abuse' accounted for only 0.81% of adjustment varience -- compared to the 8.41% accounted for by negative family variables. The original text of this study is available at Ipce, and at webarchive and a layman-friendly explanation of the paper is available here

Note that this alone cannot be interpreted to prove that adult-child sex is generally harmless. Rather, it means that the long-term harm attributed to the construct of CSA, which is often defined to include exhibitionism, sexual hugging, consensual sex with teenagers, etc., is greatly exaggerated by studies that fail to control for family environment.

Additionally, the significance of family environment as a confounding variable is less prominent when studies exclude subjects whose experience of 'CSA' was not unwanted. The studies cited by Dallam and the Leadership Council to contradict Rind all examined unwanted CSA only.

Evidence that family environment is more predictive of harm than CSA

The findings of Rind and his colleagues are not unique. Other studies which reached the same conclusion as Rind et al. on family environment include:

• Nash, M.R., Hulsey, T.L., Sexton, M.C., et al. (1993). "Long-term sequelae of childhood sexual abuse: Perceived family environment, psychopathology, and dissociation," Journal of Consulting and Clinical Psychology, 61:1–8.

  "Our first hypothesis, that sexual abuse itself is associated with broad-spectrum, general psychological impairment, independent of the effects of perceived family environment, was not supported by the findings. Perceived family environment appears to be an important mediating variable in determining general level of adult psychological distress, so important that we found no significant residual effect for abuse per se on the extent of general psychological impairment. For some victims, sexual abuse may be a signal variable that the home environment is profoundly and broadly pathogenic. Subsequent adult impairment may be an effect not only of abuse but of the context in which it was embedded."

• Fromuth, M.D. (1986). "The relationship of childhood sexual abuse with later psychological and sexual adjustment in a sample of college women," Child Abuse & Neglect, 10:5-15

  "To further explore the nature of the significant relationships found between the SCL-90 variables and sexual abuse, multiple correlations were performed predicting the SCL-90 variables from the presence of childhood sexual abuse and the Parental Support Scale. As can be seen by examining Table 1, the Parental Support Scale was a better predictor of the SCL-90 variables than was the history of childhood sexual abuse. Indeed, semipartial correlations revealed that, except for the Phobic Anxiety Scale (see Table 2), a history of sexual abuse did not significantly increase the prediction of the SCL-90 variables over and above that predicted by the Parental Support Scale alone. This suggests that for all but the Phobic Anxiety Scale, the significant relationships found between childhood sexual abuse and the SCL-90 variables were due to the confounding of the sexual abuse with the family background."

• Higgins, D. J., & McCabe, M. P. (2003). "Maltreatment and Family Dysfunction in Childhood and the Subsequent Adjustment of Children and Adults," Journal of Family Violence, 18(2)

  "Although child maltreatment scores predicted psychopathology, childhood family variables were better predictors of adjustment. [..] Standard multiple regression analyses were used to assess the contribution of scores on the five maltreatment scales to the adjustment measures of TSC-40 Total score and Self-derogation (see Table IV). Scores on these five maltreatment scales significantly predicted adults' reports of trauma symptoms, F(5, 132) = 15,36, p < .001, with psychological maltreatment being the only statistically significant unique predictor (sr² = .05). Maltreatment scores also significantly predicted self-depreciation, F(5, 132) = 7.14, p < .001, with neglect being the only statistically significant unique predictor (sr² = .03). Standard multiple regression analyses were used to assess the contribution of gender and scores on family background variables (closeness and quality of childhood relationships, physical and verbal affection, parental divorce, family adaptability, family cohesion, parental sexual punitiveness, quality of interparental relationship, traditionality of mother, traditionality of father) to the adjustment measures of TSC-40 Total score and Self-derogation (see Table IV). The family background variables significantly predicted trauma symptomology F(10, 127) = 4.44, p < .001; and self-depreciation, F(10, 127) = 3:87; p < .001. Family adaptability and quality of childhood relationships were unique predictors of both trauma symptomatology (sr² = .05 and .05 respectively) and self-depreciation (sr² = .04 and .03 respectively)."

• Friedrich, W., Beilke, R., and Urquiza, A. (1987). "Children from sexually abusive families: A behavioral comparison," Journal of Interpersonal Violence , 2, 391-402.

  "Friedrich et al. (1987) used mulitvariate analysis to explore the effects of abuse and family variables simultaneously on symptomatology. This study found that abuse severity was not a significant predictor of internalizing symptoms when controlling for family variables; instead, internalizing symptoms were predicted by family conflict, family cohesion, and time since disclosure. On the other hand, abuse severity and family variables both contributed significant variance in predicting externalizing symptoms and sexualized behaviors." (as cited by Spaccarelli and Fuchs, 1997, "Variability in Symptom Expression among Sexually Abused Girls: Developing Multivariate Models")

• Harter, S., Alexander, P.C., and Neimeyer, R.A. (1988). "Long-term effects of incestuous child abuse in college women: Social adjustment, social cognition, and family characteristics," Journal of Consulting and Clinical Psychology, 56:5–8.

  "Additional analyses suggested that family characteristics and increased perceptions of social isolation were more predictive of social maladjustment than abuse per se."

• Shen, F., Liu, Y. (2023). Perceived Parental Attachment and Psychological Distress Among Child Sexual Abuse Survivors: The Mediating Role of Coping Strategies. J Fam Viol. DOI:10.1007/s10896-023-00568-w

  “We found that parental attachment significantly predicted CSA survivors’ psychological distress. Additionally, both approach coping and avoidance coping mediated the effect of perceived parental attachment on psychological distress. CSA severity was found to be a nonsignificant predictor of psychological distress.”

• Dinwiddie, S., Heath, A. C., Dunne, M. P., Bucholz, K. K., Madden, P. A. F., Slutske, W. S., et al. (2000). Early sexual abuse and lifetimepsychopathology: A co-twin-control study. Psychological Medicine, 30, 41–52

"The association between CSA and psychopathology arises at least in part through the influence of shared familial factors on both risk of victimization and risk of psychopathology."

"[A] substantial part of the risk for psychopathology might be due to shared vulnerability factors rather than stemming solely from CSA. [...] [O]ur findings give further support to the conclusion that, rather than being a pure cause-and-effect relationship, the association between CSA and subsequent psychopathology in fact reflects a complex interplay of factors, very likely related to parenting."

Editor’s note: This study uses a narrow definition of CSA as a coerced activity, and yet (although not explicitly stated in the text of the article) Table 4 shows that almost all associations between CSA and psychopathology become statistically insignificant in pairs of twins discordant for CSA (that is, people with virtually identical environments).

Effect of life events

Not only family, but also the broader life context explains the correlation between CSA and adult psychological outcome

• Horwitz, A. V., Widom, C. S., McLaughlin, J., & White, H. R. (2001). The impact of childhood abuse and neglect on adult mental health: A prospective study. Journal of Health and Social Behavior, 42, 184–201. doi:10.2307/3090177

"These findings indicate that some impacts of early childhood victimization on adult mental health are not artifacts of retrospective recall or of the failure to use a matched comparison group. Adults who suffer sexual and physical abuse or severe neglect as young children are more likely to suffer from dysthymic symptoms and to act out, and adult women are more likely to have more symptoms of alcohol problems than persons who grew up in comparable environments but did not suffer from court-substantiated cases of childhood abuse and neglect. The results, however, also clearly indicate that there is no consistent or straightforward relationship between abuse and neglect in early childhood and subsequent mental health effects. They show that much of the perceived impact of childhood victimization that other studies without control groups report is likely to stem from a matrix of disadvantage that abused and neglected children suffer from, only one part of which consists of the abuse and neglect itself (Kruttschnitt, McLeod, and Dornfeld 1994). [...] In fact, whether respondents were in the abused or neglected or control group explains less than 2 percent of the variance on each mental health outcome. Childhood abuse and neglect do result in adverse subsequent mental health outcomes, but these outcomes are not independent of broader socio-economic contexts (Aneshensel 1992; Mullen et al. 1993; Pearlin 1989; Turner et al. 1995)."

"[S]tressful life events that occur later in the life course influence how much effect childhood victimization will have on subsequent outcomes. When childhood victims of abuse or neglect do not experience more stressors than controls, they do not have worse mental health outcomes (alcohol problems, dysthymia, or antisocial personality disorder) as adults."

("It includes the following 14 events: unemployment, fired from more than one job, held three or more jobs within a five-year period, money problems, homelessness, birth or fathering of a child, child care problems, divorce, separation, death of a parent, parent or sibling ever had a drug or alcohol problem, and parent or sibling ever arrested.")

• Turner HA, Finkelhor D, Ormrod R. (2005)The effect of lifetime victimization on the mental health of children and adolescents. Soc Sci Med. 2006 Jan;62(1):13-27. doi: 10.1016/j.socscimed.2005.05.030. Epub 2005 Jul 5. PMID: 16002198. (copy)

"Table 3 presents findings for depression and anger/aggression among 2–9 year olds.[...] All four types of victimization (sexual victimization, child maltreatment, witnessing family violence, and other major violence) had significant independent effects on depression, with child maltreatment having the strongest positive coefficient.[...] When non-victimization adversity was added to the model (Eq. (3)), it also showed a strong independent effect on depression. Although exposure to non-victimization stressors accounted for part of each victimization-depression association, all victimizations but sexual assault remained statistically significant.[...] The second half of Table 3 shows the same hierarchical regression analyses with anger/aggression as the dependent variable. [...] Finally, non-victimization adversity was added to the model, again showing a relatively strong independent effect. When this variable was controlled, the coefficients for sexual victimization and other major violence were no longer statistically significant."

Note 1: non-victimization adversity “included: (a) non-victimization traumas such as serious illnesses, accidents, parent imprisonment, and natural disasters; and (b) more chronic adversities, like substance abuse by family members, parental arguing, and chronic teasing about physical appearance.”

Note 2: in this sample of 1030 respondents, they use questions on forced sexual acts only. Nevertheless, if controlled for non-victimizing adversities it’s correlations with depression and anger become insignificant.

"Except for ‘witnessing family violence’ among the 10–17 year olds, all four types of victimization (sexual victimization, child maltreatment, witnessing family violence and other major violence) had significant effects on both depression and anger/aggression, with the other victimization domains controlled. [...] In most cases, child maltreatment (physical abuse or neglect by a caregiver) is the form of victimization that has the strongest independent association with depression and anger/aggression."

Note3: for this sample (n=1000), they used 8 questions, 7 of which were about forced sexual acts and 1 about wanted sexual act with somebody older than 18. There is no attempt to distinguish correlation with forced and mutual acts.

" We found that cumulative exposure to non-victimization adversity did indeed have an independent effect on depression and anger in both samples, and that the magnitude of these associations typically matched or exceeded the independent effects of child maltreatment."

See also

• Other methodological flaws of CSA studies see in the article Methodological flaws and syndrome construction