Mother reports of maternal support following child sexual abuse: Preliminary psychometric data on the Maternal Self-report Support Questionnaire (MSSQ)

Abstract

Objective

Maternal support is an important factor in predicting outcomes following disclosure of child sexual abuse; however, definition of the construct has been unclear and existing measures of maternal support are utilized inconsistently and have limited psychometric data. The purpose of this study was to develop a reliable and valid mother-report measure for assessing maternal support following the disclosure of child sexual abuse.

Methods

Data from 2 very similar samples of mother-child pairs seeking forensic evaluation following the discovery of child sexual abuse were combined, resulting in a final sample of 246.

Results

Exploratory factor analysis resulted in two reliable 7-item factors labeled “Emotional Support” and “Blame/Doubt,” each of which had acceptable internal consistency. Analyses with a child-report measure of general maternal support the construct validity of the MSSQ. Concurrent validity analyses revealed unique relations with maternal ratings of child behavior problems and case characteristic data.

Conclusions

The study resulted in the development of a brief, easily scored self-report measure of maternal support with reasonable preliminary psychometric properties that could easily be utilized in other studies of sexually abused children.

Practice implications

Adoption of this promising measure in future research will reduce the lack of cross-study measurement comparability that has characterized the maternal support literature to date, increase the feasibility of expanding upon current literature on maternal support, and may produce important information leading to clinical and theoretical innovation.

Introduction

A person's level of social support (e.g., the quality of their social relationships) is an important predictor of psychopathology (Fokias & Tyler, 1995). For instance, several researchers have documented relations between social support and general psychological functioning (e.g., Brown, Andrews, Harris, Adler, & Bridge, 1986). Social support has also been suggested to moderate the impact of various types of stressful and traumatic events (e.g., Roesler, 1994). In the child sexual abuse (CSA) literature, one particular form of support, generally referred to as “maternal support,” has been widely cited as relevant to the functioning of CSA victims after disclosure of the abuse. Several frameworks have been proposed from which to conceptualize the relation between sexual abuse and victim psychopathology (for a review, see Freeman & Morris, 2001). However, most of these conceptual frameworks do not meaningfully address the construct of maternal support.

The one exception is the Transactional Model which treats the construct of maternal support as a fundamental component (Spaccarelli, 1994). In the Transactional Model, sexual abuse is seen as comprising a host of potentially stressful experiences for the child. Some of these stressors are intrinsic to the abuse experiences themselves, others revolve around the circumstances and consequences of disclosure, and still others result from the intervention processes employed by the protective service system (Spaccarelli, 1994). The mother's reaction to discovery and disclosure of CSA is believed to be critical in determining how the child adjusts to disclosure-related stressors and adapts to intervention-related challenges. For instance, if the mother does not support or protect the child, this may cause the child to utilize maladaptive coping strategies, such as avoidance and self-blame. In addition, non-supportive responses from the mother (e.g., blaming the child for abuse-related problems) may lead a child to experience emotional distress and negative views of the mother and others.

Consistent with the Transactional Model's predictions, the construct of “maternal support” has been documented as a predictor of childhood adjustment following disclosure of sexual abuse (e.g., Adams-Tucker, 1982, Conte and Schuerman, 1987, Esparza, 1993, Everson et al., 1989, Feiring et al., 1998, Lovett, 1995, Tremblay et al., 1999). In fact, several studies have also posited that maternal support may be a better predictor of psychological adjustment than abuse-related factors (Fromuth, 1986, Johnson and Kenkel, 1991, Spaccarelli and Kim, 1995). Specifically, Everson et al. (1989) examined the relation between maternal supportiveness and victim psychopathology and found that low levels of abuse-specific maternal support were associated with higher levels of childhood distress and behavior problems. Similarly, Adams-Tucker (1982) found that lack of abuse-specific support by a primary caregiver predicted higher rates of a range of behavioral problems that included withdrawn behavior, suicide attempts, running away from home, fire setting, and aggression. Similar difficulties have been found in children who perceive their mothers to be more rejecting (Lovett, 1995), while positive mother-child relationship quality has been associated with fewer externalizing problems (Esparza, 1993, Tremblay et al., 1999).

Mannarino and Cohen (1996) reported a rare deviation from this pattern of findings. These authors reported no significant relations between abuse-specific maternal non-support (“blaming”) and child symptoms. They offered two explanations for the lack of consistency between their findings and results of similar research. First, they argued that mothers in their study tended to report a high level of support, thus restricting the range of maternal supportiveness. Also, mothers rarely endorsed questionnaire response options that were socially inappropriate (Mannarino & Cohen, 1996). These observations suggested that social desirability factors may have played a role in the accuracy of reporting when mothers were directly queried about their level of support following abuse disclosure.

Taken together, there is considerable agreement across studies that maternal support is positively associated with a healthier prognosis for children following the discovery of sexual abuse (e.g., Fromuth, 1986, Johnson and Kenkel, 1991, Spaccarelli and Kim, 1995). However, a closer examination of the studies reviewed above reveals conceptual and methodological limitations that suggest the need for further research before firm conclusions may be drawn.

Current definitions of maternal support lack clear, precise, and consistently utilized criteria. Although several studies have included assessment of variables that have been termed “maternal support,” few have offered clear operational definitions. In fact, the construct of maternal support has often been assessed without reference to the abuse itself, but rather has been reflected by overall level of support or general parent-child relationship quality (e.g., Conte and Schuerman, 1987, Esparza, 1993, Feiring et al., 1998). As such, research has yet to determine if “abuse-specific” support is distinct from the pre-existing mother-child relationship or a mother's general ability to provide her child with support.

In studies focusing on abuse-specific support, assessed variables have included one or more forms of protective action (e.g., maternal separation from perpetrator), verbal/emotional support (e.g., empathic responding) following disclosure, and/or belief of the child's report of abuse (e.g., Adams-Tucker, 1982, Everson et al., 1989, Heriot, 1996), making comparisons across studies difficult. Furthermore, Bolen and Lamb, 2004, Bolen and Lamb, 2007 highlighted the importance of ambivalence in maternal reactions following CSA. They defined ambivalence as a normative reaction that can occur when mothers feel a strong positive valence toward both the child and the alleged offender. Exploratory research by Bolen and Lamb (2007) suggests that ambivalence and support may be independent constructs with mothers able to feel ambivalence and still be supportive of their children at the same time.

Although it is clear that extant studies have not shared a clear operational definition of the maternal support construct, research suggests that “abuse-specific maternal support” is a multi-dimensional construct likely consisting of mother's belief in the disclosure of sexual abuse, protective action taken by mother to prevent further victimization and to initiate appropriate intervention, and recognition and support of the child's distress following abuse and disclosure. However, it remains unclear whether the benefits of maternal support are due primarily to the effects of one type of support (e.g., emotional support), the overall quality of the mother-child relationship, or some combination of several types of support.

Measurement has also been problematic, as researchers have used different strategies for assessing maternal support. For instance, methods of data collection have ranged from subjective chart review performed by non-researchers (Heriot, 1996, Pintello and Zuravin, 2001) to direct administration of interviews and/or self-report questionnaires (e.g., Everson et al., 1989, Mannarino and Cohen, 1996). Among studies that have used direct administration of assessment instruments, the identity of the respondent has varied. In most studies, mental health professionals or caseworkers have provided judgments of maternal support based upon reviews of charts or interviews with the mother (e.g., Bolen and Lamb, 2002, Bolen and Lamb, 2007, Heriot, 1996, Leifer et al., 2001, Pintello and Zuravin, 2001, Sirles and Franke, 1989). Although this perspective provides important information about the mother's behavior toward the child, there are clear limits to how well these professionals are aware of the mothers’ feelings and thoughts about the child and the allegations. In the interest of obtaining a more personal perspective, several studies have had non-offending mothers report their own levels of supportive behavior (e.g., Bolen and Lamb, 2004, Bolen and Lamb, 2007, Hsu and Smith, 2000, Mannarino and Cohen, 1996). Four studies have obtained children's perceptions of their relationship with their mothers or of general “social support” (Esparza, 1993, Feiring et al., 1998, Lovett, 1995, Tremblay et al., 1999) and only two studies obtained child-reported perceptions of abuse-specific maternal support (Cyr et al., 2003, Morisson and Clavenna-Valleroy, 1998).

Not only have respondents varied across studies; so have instruments. This is not surprising, given that many studies have established their own idiosyncratic definition of and method to assess maternal support. One relatively commonly used rating scale, the Parental Reaction to Incest Disclosure Scale (PRIDS; Everson et al., 1989), was modified to apply more broadly to both incestuous and non-incestuous cases of CSA (Cross, DeVos, & Whitcomb, 1994). While it has been frequently used as a means of measuring maternal support (e.g., Bolen and Lamb, 2002, Heriot, 1996, Leifer et al., 2001, Pintello and Zuravin, 2001, Sirles and Franke, 1989), little psychometric data on the measure have been reported. Bolen, Lamb, and Gradante (2002) developed the Needs-Based Assessment of Parental Support (NAPS-C), a 26-item, clinician-report rating scale, which demonstrated good internal consistency and strong correlation with the PRIDS. The NAPS-C requires training to administer and has not been used widely by investigators other than the developers. Mannarino and Cohen (1996) developed the Parental Support Questionnaire (PSQ) to measure parental perceptions associated with their own responses to their child's sexual abuse. The PSQ targets verbal/emotional support and “blame,” but not belief in disclosure or protective action. Similar to the PRIDS, this measure has been used with small samples to date (e.g., Cohen & Mannarino, 2000) and inadequate psychometric data are available to evaluate its utility.

Taken together, research suggests that maternal support is an important factor in predicting child outcomes following disclosure; however, definition of the construct of maternal support has been unclear and existing measures of maternal support are utilized inconsistently in the literature, target the construct of maternal support in different ways, and have limited psychometric data to support reliability and validity. The purpose of this study was to develop a simple, reliable and valid mother-report measure for assessing maternal support following the disclosure of CSA. This article reports data collected in two separate studies that assessed similar populations using several measures in common, including a newly developed, self-report measure of maternal support. Both samples include mothers of children receiving forensic CSA exams at child advocacy centers. Using comparable data from two studies permitted the creation of a sample adequate for a meaningful, if preliminary, evaluation of the maternal support scale's psychometric properties.