Rachna Varia and Richard
R. Abidin
The minimizing style: perceptions of psychological abuse and quality
of past and current relationships
Child Abuse & Neglect, 1999, 23, 11
Abstract
Objective: The goals of this current project were to expand and elaborate on the results of Varia, Abidin, and Dass' (1996) study to better understand what historical parenting factors may contribute in creating a "Minimizing" perception of childhood psychological abuse. A second objective included examining the association between perceptual styles and adult ratings of adult relationship satisfaction.
Method: Ninety individuals from a non-clinical adult sample completed self-report questionnaires regarding psychological maltreatment and adult relationship satisfaction. Three groups were created and compared using analysis of variance. The groups were (I) "Non-Abused," consistent reporters of no abuse; (II) "Acknowledgers," consistent reporters of abuse; and (III) "Minimizers," reporters of abuse, but do not label themselves as having been abused.
Results: Results indicated a continuum effect with the Non-Abused individuals reporting the highest level of maternal warmth and affection, the Acknowledgers reporting the lowest, with the Minimizers in the middle. The Non-Abused group reported the healthiest adjustment in terms of adult relationship satisfaction. While Minimizers and Acknowledgers reported similar levels of psychologically abusive experiences, Acknowledgers reported more difficulties with adult relationships.
Conclusions: This study highlights
the importance of studying Minimizers, a group which describes abusive
events but minimizes the meaning and scope of these experiences. The
relatively higher levels of maternal care which the Minimizers reported
is hypothesized to be associated with their perceptions that they were
not abused and acted as a protective factor associated with better quality
adult relationships. The varying degrees of nurturance and care in the
early family environment of abused individuals is speculated to be associated
with distinct perceptual styles.
RESEARCH IN THE area of maltreatment has predominantly functioned within a direct and linear design framework. Early abusive childhood experiences have been shown to affect adult adjustment negatively. However, few studies have branched out of this cause and effect structure to examine the role of mediating variables. Possible risk or resiliency factors may be significant in leading abuse survivors down markedly different psychological pathways. The current project was in response to a number of studies that call for abuse research to identify and understand the role of mediating factors in adult outcome (Briere and Coffey). This study was also in response to Garrison's (1987) push for research which targets and defines psychological maltreatment.
Studies have begun to more critically examine the methodological issues involved with differing professional abuse definitions and criteria (Gauthier; Mullen; Socolar and Straus). However, most research has not integrated the role of personal perception, or self-definition of oneself as abused in study designs. Swann (1984) warned that "the strategy of abstracting the person perception process" (p. 457) can have negative implications for methodology.
Individuals may have different understandings of what constitutes abuse than clinicians, researchers, the popular press, or by the culture. Hart and Brassard (1987), writing for the APA's forum on psychological maltreatment, stated that "operational definitions should also consider the subjective impact of maltreatment on the child victim because the degree of abuse (and the degree of harm experience) may be related to the subjective meaning that a particular action or pattern of interaction has for a victim" (p. 162). Berger, Knutson, Mehm, and Perkins (1988) point out that the range of personal definitions for what constitutes abuse can create a disparity between reporting acts of abuse and subsequently defining oneself as abused. This discrepancy can create great inconsistency in research and confound results which may be a reflection of only one subset of the abused population, those who acknowledge their abuse.
In this current study, attention was given to the important intervening variable of perception when examining the relationship precursors and sequelae of psychological abuse. Varia and colleagues (1996) identified a subgroup of abused individuals who reported levels of abuse above the sample mean, but failed to acknowledge themselves as abused. These individuals, defined as "Minimizers," consistently reported more psychological and interpersonal difficulties than the "Non-Abused" group. Furthermore, Minimizers also had more areas of detrimental personality adjustment than those who acknowledged their abuse, suggesting that the interpretation or perception of an event can have a significant effect on later adjustment. Varia and colleagues (1996) focused on current personality and relationship constructs. This study examines how a Minimizing style may be formed by examining perceptions of abuse in relation to other historical parent-child transactions and how a Minimizing style may in turn be associated with current adult marital and parenting relationships.
The current study is part of Phase II of the Relationship Research Project at the University of Virginia. All Phase II participants were part of a nonclinical sample of parents with children between the ages of 3 and 12 who were recruited from primary care pediatric settings, public swimming pools, day-care centers, and a University of Virginia general electronic news group in the Charlottesville, VA and Washington DC, areas. Participants needed to be in a relationship with someone the subject considered a "parenting partner" who was also willing to complete the questionnaires.
Participants were part of a community sample as opposed to a clinical sample, and it is assumed that the presence of abuse and clinical pathology reflects the base-rate for the population at large. The sample inclusion criteria that participants be part of a couple, that they both participate, and that they have a child most likely skew the sample toward healthy interpersonal functioning.
Fifty-nine of the Phase I couples who 2 years previously had expressed further interest in the overall project agreed to participate in Phase II. Of these 59 couples, 30 ultimately participated in Phase II (51%). Additional efforts to recruit participants resulted in another 36 participants participating in the study. The additional group was obtained in the same manner as the original Phase I sample. The response rate for individuals contacted was 43%. The final adult sample consisted of 96 people, representing 45 couples and 6 single parents.
Parents were asked to participate in a voluntary research project investigating the relationship between early family environment and later social relationships. Parents who chose to participate were given a pre-addressed, stamped packet containing an informed consent card and a 12-page questionnaire booklet that took approximately 45 minutes to complete. Each member of the couple was asked not to consult their parenting partner when answering the questions. Parents were instructed to mail the packets in self addressed envelopes to the Relationship Research Project office. Informed consent cards containing identifying information would be sent to the Project Director at a different address. Each family was assigned an identification number that was used instead of the family name to ensure confidentiality.
The measures from the packet which were used in this study include a demographic information form, the Psychological Maltreatment Scale, the Adult Attachment Scale, the Parenting Alliance Inventory, the Parental Bonding Inventory, and the Quality of Relationships Inventory.
The Psychological Maltreatment Scales. The Psychological Maltreatment Scale (adapted from Berger and Berger and Knutson 1984) was created using a rational/intuitive approach and used a Likert response. Briere's Psychological Maltreatment Scale consists of seven statements to which the subject is asked to respond on a scale of 0 to 6 corresponding with the frequency of events from an average year before the age of 16. Statements included: (A) Yelled at you; (B) Insulted you; (C) Criticized you; (D) Tried to make you feel guilty; (E) Ridiculed or humiliated you; (F) Embarrassed you in front of others; (G) Made you feel like you were a bad person; and (H) Other. Individuals were asked to respond to the Psychological Maltreatment Scale for the average year, based on the assumption that this form of abuse is more pervasive and less event related, and thus might be harder to specify in terms of time. Participants were asked to respond for events that involved their mother and father on separate and identical scales. The Psychological Maltreatment Scale was designed to tap parental behaviors which are primarily verbal in nature, as opposed to those involving physical actions or threats of physical harm. Alpha reliabilities for the Psychological Maltreatment Scale were found to be .87 for both mothers' and fathers' responses. Dass (1994) reports the alpha reliability of this scale to be .89, when using it with Phase I participants of the Relationship Research Project.
Scale scores for items within the psychological maltreatment area were summed across an individual's responses involving their mothers and fathers. The Total Abuse score was determined through a linear weighting system which attached degrees of intensity to both frequency and type of abuse (Varia et al., 1996).
Abuse belief question. Immediately after completing each maltreatment scale, participants were asked to respond to a question which asked whether they considered the maltreatment they received and reported as being abusive.
The Parenting Bonding Inventory (PBI; Parker, Tupling, & Brown, 1979) was designed to define and provide information about influences related to bonding. Respondents to the inventory rate both their mother and father on 25 statements ranging from "very like" to "very unlike." The PBI consists of two dimensions that evaluate the quality of parenting received. The Care dimension of the PBI consists of 12 items accounting for 28% of the total variance of the total scale. High scores indicate emotional warmth, affection, empathy, and closeness, while low scores reflect emotional coldness, indifference, and rejection. The overprotection dimension accounts for 17% of the variance and is also bipolar. High scores are indicative of over protective behaviors such as parental over-control, intrusion, excessive contact, infantilization, and prevention of individual behavior.
Norms for the PBI were based on the responses of 410 adults in Sydney, Australia. Test-retest reliabilities range from .76 for 17 participants taking the questionnaire 3 weeks apart, to .87 for depressed patients tested 9 weeks apart. Parker reports that the increased reliability in the patient sample is a result of higher motivation for accurate reporting. Split-half reliability was established at .88. Extensive validity research has shown the PBI to correlate with ratings made by investigators who interviewed participants with respect to their relationship with their parents. In accordance with the suggestions made by the authors, the two dimensions of the PBI, care and protection, are used in this study to assess separate parenting characteristics.
The Quality of Relationships Inventory. The Quality of Relationships Inventory (QRI; Pierce, Sarason, & Sarason, 1991) consists of 25 items designed to measure the perceived availability of social support from specific relationships. Individuals are asked to respond to questions using a four-point scale ranging from not at all to very much. The first dimension (support) of the scale measures the extent to which the individual can rely on the other person for support. The second dimension (depth) assesses the extent to which the person believes that they and the other person are committed to the relationship and place a positive value on it. The third dimension (conflict) measures the extent to which the individual experiences angry or ambivalent feelings toward the other person. Pierce and colleagues report adequate internal consistency of terms for the three factors with alpha reliabilities of .85 for the support dimension, .91 for depth, and .84 for conflict when participants reported on the quality of a friendship. Dass (1991) in her relationship study from Phase I of the project, found alpha reliabilities of .85, .82, and .78 for support, depth, and conflict respectively. Similar reliabilities were found for respondents reports of their relationships with their mothers and fathers.
Adult Attachment Scale (Hazan & Shaver, 1987). This scale is a single trichotomous item designed to assess the adult's predominant attachment style. The adult selects one of three descriptions of interpersonal style. These description correspond to "secure," "anxious-ambivalent," and "avoidant" attachment styles. The components of each description are conceptual translations of Ainsworth, Blehar, Waters, and Wall's (1978) descriptions of infants' attachment to their caretakers.
Hazan and Shaver (1987) report that the participants in their sample were able to classify themselves meaningfully. The frequency of each style was similar to that reported in the infancy attachment literature (i.e., approximately 60% secure, 20% avoidant, and 20% anxious-ambivalent). As further validation of the differences between attachment styles, participants responded predictably in the way they experience romantic love. In addition, responses varied as expected on a measure designed to describe an individual's internal working model of self and relationships.
Parenting Alliance Inventory. The Parenting Alliance Inventory (PAI; Abidin & Brunner, 1995) consists of 20 statements that parents rate on a five-point scale from "strongly agree" to "strongly disagree." Parenting alliance is defined as the interactions between partners regarding child-rearing. The PAI emerged as a response to Weissman and Cohen (1985), who found four aspects of good parenting alliances: (a) parental investment in the child by both parents; (b) valuing the partner's involvement with the child; (c) respect for the other parent's judgment; and (d) desire for communication between parents. The parenting alliance is seen as an evolving series of interactions occurring in response to childrens' and adults' developmental changes and the reactivation of adults' object-relations issues. Further, it is viewed as a distinct component of the marital relationship or other relationship between adults committed to the care of a child. Normally the alliance begins prior to the birth of the child. During pregnancy, the father or significant other provides psychological support for the mother. With the birth of the child, the alliance consists of the interactions between spouses or partners concerning child care which communicate respect for parental roles and tasks. The quality of these interactions are affected by the parents' ability to communicate with each other, work together as a team, and respect each other as parents.
The PAI has been found to co-vary with other measures of marital relationship, including .75 with the marital support subscale of the QRI in one study (Dass, 1994). However, the PAI was developed as a narrow band family measure, conceptually distinct from marital support. Despite its high correlation with the QRI in particular, both Brunner (1991) and Dass (1994) found that the PAI accounted for a significant amount of unique variance in both parenting behavior and child outcomes when compared to a measure of marital quality. Factor analysis of the PAI (Abidin & Brunner, 1995) has showed high item intercorrelation, with one factor accounting for 70.3% of the variance for mothers. Alpha reliability of the scale in the norm sample (321 mothers) was .97, a result replicated by Dass (1994).
The participants were divided into three categories for the purpose of this current study. The participants in responding to the Psychological Maltreatment Scale were questioned about particular incidents of abuse and asked to rate the frequency of occurrence of each within a one year time frame. Following completion of the Psychological Maltreatment Scale, participants responded "Yes" or "No" to whether they "believed" that they had been psychologically abused. The participants for this study were organized into four groups based on a combination of two variables, the first being their reports of the frequency and severity of their experiencing abusive behavior, and second by their belief as to whether of not they were abused: The groups are: (I) "Non-Abused," consistent reporters of no abuse; (II) "Acknowledgers" consistent reporters of abuse; and "Minimizers," reporters of abuse who label themselves as not having been abused. The minimizing style was defined by participants whose total abuse score was above the total weighted sample mean of reported abuse; and who said they were not abused. An Exaggerating group (IV), individuals who report that they had been abused but fall below the weighted mean for abuse, was not considered in any of the analyses due to the small number of individuals falling in this group.
Consistent reporters, both abused and non-abused adults, were those whose self reports were congruent. The "Acknowledgers" were those participants who state that they had been abused and report levels of abuse above the mean. In the same vein, the "Non-abused" group were those respondents who reports levels of abuse below the mean and endorse a "No abuse" label. The "Minimizers" were identified as those who report that they had not been abused but fall above the weighted mean for abuse. The weighted mean of abuse is not necessarily a clinically significant score, but rather a relative marker to distinguish between groups.
The study design used analysis of variance (ANOVA) and chi-square analyses using a SPSS 4.1 statistical program to examine differences among the three subgroups of responders in terms of abuse variables, early childcare experiences, personality traits and the quality of social relationships.
Of the 96 participants in the sample, three were omitted because of missing data involving maltreatment items and three were excluded because they met the criteria for being labeled Exaggerators. Of the remaining 90 participants, 48 individuals fell in the "Non-Abused" group (M = 61.5) and 22 participants met the criteria for being labeled "Acknowledgers" (M = 267); 20 participants were categorized as "Minimizers" (M = 236.1). Groups were created based on the sample mean (M = 147) of psychological abuse. Both the Minimizers and Acknowledgers reported significantly higher levels of abuse than the Non-Abused group, F(3, 89) = 81.57, p < .001.
Overall, the average age of respondents for the entire sample was M = 35.0 years (SD = 9.5/Range = 20¯48 years). The average number of years married was M = 8.1 years (SD = 5.9/Range = 1¯25 years), while the number of years of being in a relationship with their spouse was greater with the average being M = 10.8 years (SD = 5.8/Range = 2¯26 years). Females comprised 52.7% of the sample. No significant differences were found in the distribution of these variables across abuse perception styles.
However, significant differences
were found in other areas of the demographic data between groups within
the different types of abuse categories. In terms of racial make-up,
90.2% of the individuals were Caucasian; significantly more non-White
participants in the Minimizing group and more White participants in
the Non-Abused group than would be expected, 
2
(2, N = 89) = 8.87, p < .01. Of the entire sample,
82.8% of the responders were employed. Significantly more unemployed
participants fell in the Acknowledging group and more employed individuals
in the Non-Abused group than would be expected 2
(2, N = 90) = 6.61, p < .05.
Of the total sample, 93.7% of the participants reported having at least some high school education, with 81.1% of these people reported having some level of education beyond high school. For the purposes of analysis, the levels of education were collapsed into three groups, including: no high school diploma; no college degree; and college degree/advanced education. Using a chi-square analysis, no level of educational effects were found.
In terms of total family income,
31.5% of the participants earned up to $40,000 annually, 68.5% earned
over $40,000, with 28.3% earning more than $60,000 a year. For the purposes
of analysis, the levels of income were collapsed into three groups,
including: less than $30,000; $30,000¯$50,000; and $50,000
and above. More individuals with a family income over $50,000 fell in
the Non-Abused group, with more participants in the middle range falling
in the Minimizing group than would be expected. In addition, more lower
income individuals fell in the Acknowledging group, 2
(4, N = 89) = 12.55, p < .01.
Of the total sample, only 15.1%
had parents who were divorced or separated. More individuals with divorced/separated
parents fell in the Acknowledging category and more married parents
fell in the Non-Abused group, than would be expected, 2
(2, N = 90) = 7.90, p < .05.
Maternal care. On the maternal care dimension of the PBI, both the Non-Abused (M = 29.4; SD = 5.3) and Minimizing (M = 22.3; SD = 9.4) groups scored significantly higher than the Acknowledging (M = 17.0; SD = 8.4) group, F(3, 92) = 15.40, p < .001. In addition, the Non-Abused group scored significantly higher than the Minimizing group (see Table 1).
Maternal protection (overcontrol). On the maternal protection dimension of the PBI, the Acknowledging (M = 21.6; SD = 11.3) group reported higher levels of overprotective behaviors than the Non-Abused group (M = 11.1; SD = 6.7), F(3, 92) = 10.34, p < .001 (see Table 1).
Paternal care. On the paternal care dimension of the PBI, the Non-Abused (M = 25.0; SD = 7.0) group reported higher scores than both the Acknowledging (M = 17.0; SD = 9.1) and Minimizing (M = 18.0; SD = 10.2) groups, F(3, 90) = 6.29, p < .001 (see Table 1).
Paternal protection (overcontrol). On the paternal protection dimension of the PBI, the Acknowledging (M = 16.3; SD = 9.1) group reported higher levels of overprotective behaviors compared to the Non-Abused (M = 10.0; SD = 5.5) group, F(3, 89) = 4.77, p < .01 (see Table 1).
Spousal support. Degree of spousal support was found to be significantly higher for both the Non-Abused (M = 23.6; SD = 3.5) and Minimizing (M = 24.7; SD = 2.6) groups compared to the Acknowledging (M = 21.3; SD = .97) group, F(3, 91) = 4.44, p < .01 (see Table 2).
Spousal depth. No significant differences were found between abuse groups.
Spousal conflict. The Acknowledgers (M = 29.4; SD = 7.6) group scored significantly higher than the Non-Abused (M = 23.9; SD = 4.6) group on level of conflict with spouse (see Table 2).
Total score. On the total score of the PAI, the Non-Abused (M = 86.1; SD = 10.1) group scored significantly higher than the Acknowledging (M = 75.0; SD = 16.4) group (see Table 3).
Respect other parent. On the respecting other parent factor of the PAI, the Non-Abused (M = 26.5; SD = 3.6) group scored significantly higher than the Acknowledging (M = 22.0; SD = 6.1) group, F(3, 89) = 5.46, p < .01 (see Table 3).
Teamwork. On the teamwork factor of the PAI, the Non-Abused (M = 33.9; SD = 4.1) group scored significantly higher than the Acknowledging (M = 26.6; SD = 7.7) group, F(3, 92) = 3.46, p < .05 (see Table 3).
Other parent respects you. On the other parent respecting you factor of the PAI, the Non-Abused (M = 25.6; SD = .23) and Minimizing (M = 25.6; SD = 3.4) groups scored significantly higher than the Acknowledging (M = 22.8; SD = 3.7) group, F(3, 92) = 3.72, p < .05 (see Table 3).
Significantly more Acknowledgers
fell in the insecure attachment groups than would be expected, 2
(4, N = 89) = 18.96, p < .001. Nine Acknowledgers defined
themselves as Anxious-Avoidant, while six labeled themselves as Anxious-Ambivalent.
Seven Acknowledgers, 11 Minimizers, and 39 Non-Abused individuals were
classified in the secure attachment category. While significantly more
Minimizers were placed in the secure attachment category than would
be expected, seven of the 19 Minimizers also labeled themselves as Anxious-Avoidant.
The current findings demonstrate a continuum effect between the three
groups, with 81% of Non-Abused individuals, 57.9% of Minimizers, and
only 31.8% of Acknowledgers reporting a secure adult attachment.
The results of this study add to the accumulation of research documenting the detrimental effects of child abuse on subsequent adult social adjustment. The findings consistently demonstrate that Non-Abused individuals reported significantly better adult relationship satisfaction than individuals who reported punitive/abusive behaviors in childhood. Consistent with Berger and colleagues' (1988) results, the findings of this study support the existence of a subgroup of maltreated individuals who reported relatively higher levels of abusive events but fail to label themselves as abused. Of the 42 individuals who reported levels of psychological maltreatment above the mean, only 52.4% labeled themselves as having been abused. The results of this investigation demonstrate that an important, and often neglected group of individuals, here labeled Minimizers, are a significant group to study (Varia et al., 1996).
Parenting care. Individuals in the Non-Abused group reported experiencing more emotional warmth, affection, and empathy from their mothers than individuals falling in either of the other two groups. These results suggest that individuals who either do or do not label themselves as abused, but who have abuse scores above the mean report experiencing more emotional coldness, indifference, and rejection by their maternal caregiver. Minimizers, however, also reported significantly higher levels of emotional warmth and care than the Acknowledgers, indicating a continuum effect with the Non-Abused individuals reporting the highest level of maternal warmth and affection, and Acknowledgers reporting the lowest, with the Minimizers in the middle. The early family environment of abused individuals involve varying degrees of nurturance, warmth, and care (Egeland, Sroufe, & Erickson, 1983), is speculated to be associated with distinct perceptual styles.
In contrast to maternal care, Minimizers did not report higher levels of paternal care (affection and warmth) than the Acknowledgers. Rather, the Acknowledgers and Minimizers were seen as experiencing similar levels of paternal care. Non-Abused individuals reported experiencing higher levels of paternal emotional warmth and care than both Minimizers and Acknowledgers. Thus paternal care did not appear to play a role in the Minimizers developing their unique perceptual style.
Parenting overcontrol. While Minimizers reported experiencing higher levels of maternal care than the Acknowledgers group, this same pattern was not demonstrated with maternal overcontrolling behaviors. Non-Abused individuals reported having more positive experiences with their maternal caregiver in terms of promoting independent behavior and appropriate levels of nurturance than both other groups. Hence, both Acknowledgers and Minimizers reported experiencing more controlling behaviors by their maternal caregiver including infantilization, parental intrusion, and other negative parental control.
Non-Abused individuals reported higher levels of positive paternal behaviors in terms of control, contact, and individuation than individuals who acknowledged that they were abused. The Minimizing group reported that their experiences with parenting was neither as negative as that of the Acknowledgers, nor as positive as that of the Non-Abused individuals.
Overall, perceptions of paternal parenting did not distinguish the three groups from each other as strongly or as often as did perceptions of maternal parenting. This result may be due to the fact that mothers are more likely to be seen as and in fact more often play a larger nurturant role than fathers. Thus, the lack of maternal warmth and affection may have a greater influence on the child's perception of self as abused and on subsequent adjustment than paternal care behaviors. Conversely, maternal rejecting behavior may be more detrimental than paternal rejection despite the fact that significant differences were not found between maternal and paternal abuse perpetration rates (Herzberger, Potts, & Dillon, 1981).
Differences in parenting. A number of differences may be gleaned from this pattern of results. Rather than a stark dichotomy of abused versus non-abused groups, a continuum of individual experiences was observed. It is speculated that the Acknowledging group represents the "worst" group in terms of having experienced a combination of abusive events, cold parenting, and overcontrolling or rejecting behavior. In contrast, the Minimizing group, while experiencing abusive events and authoritarian parenting, also received more maternal warmth and care which may buffer their response to abuse. Finally, the Non-Abused group were recipients of positive protective behaviors, higher levels of affection, and lack of abusive events. These differential patterns of experiences are hypothesized to be associated with the participants' perceptual styles, and with their subsequent adult functioning. Integrating into research designs the early environment factors such as caregiver warmth, and rejection is necessary if an understanding of why individuals develop different perceptual styles is to be achieved.
Herzberger, Potts, and Dillon in their 1981 study of the child's perspective on parental treatment postulate that "seeing abuse as an indicant of parental rejection may have more harmful effects than perceiving the abuse as being caused by the parent's externally imposed frustrations" (p. 89). Rohner and Rohner (1980) describe parental acceptance as synonymous with parental warmth and affection, which corresponds with the PBI factor of maternal and paternal warmth. Regardless of the presence of various forms of abuse, children who experience more rejection than warmth have future problems with psychological adjustment.
The results of this current study suggest that parental rejection and lack of warmth are associated with adult relationship problems. Further, it appears that a rejected child perceives a given behavior as more abusive, and an accepted child perceives the same type of behavior as less abusive. These differences in perceptions are likely to influence and color later perceptions of self as well interactions with others.
Pierce and colleagues (1991) hypothesize that early gestalt formation can guide later behavior to fit a similar mold of interaction. The different manner in which individuals perceived the same childhood experiences, rather than the experiences themselves, therefore may be affecting later psychological adjustment. Perceptions of maternal care appear to be translated into expectancies which guide interpersonal interactions with significant others. Lower levels of maternal care may be an antecedent for impaired relationships where ideas about commitment and security are not sound. In contrast, higher levels of maternal care may lead to feeling that one is loved, valued, and cared for despite the presence of negative behaviors.
Acknowledger relationship outcome. More Acknowledgers fell in the "insecurely attached groups" than both Non-Abused individuals and Minimizers, suggesting that Acknowledgers have the most difficulty with their adult attachments and romantic relationships. Indeed, more Acknowledgers as indicated by the Adult Attachment Scale, than would be expected reported feeling that they felt others were reluctant to get close to them and they often worry that their partner does not really love them or will not stay with them.
The relationship difficulties and insecure attachments which the Acknowledgers reported experiencing may be the result of an impairment to the attachment system caused by the combination of both having been abused and other deleterious parenting they received as children. The resultant internal working models of relationships that the Acknowledgers form may be maladaptive in the long term because they will not test alternate strategies to deal with the world.
The anger, resentment, and insecurity abused individuals feel toward their parents may then be transferred to future adult relationships, especially with a spouse/parenting partner. Consistent with other research findings, individuals in the Acknowledging group reported the most problems with their spouses in terms of not feeling supported and having high levels of conflict as opposed to the Non-Abused group (Alexander; Briere; Browne and Steele).
The Non-Abused individuals also reported the healthiest parenting alliance compared to those individuals who acknowledge that they have been abused. In contrast, Acknowledgers reported having more problems with their ability to communicate with their partners, work together as a team, and respect the other parent's judgment as related to child care responsibilities. These results are consistent with other findings which report that abused children have difficulty with trust, confidence, and intimacy in their adult parenting and partnership relationship (Braver and Oates). These findings support the theory that the negative and rejecting nature of the abusive parent-child relationship carries through to influence later relationships in a dysfunctional manner (Howes & Espinosa, 1985). They also support the importance of an accepting and positive early attachment in influencing later social relationships (Bowlby, 1969).
Minimizer relationship outcome. In contrast to Acknowledgers, Minimizers report less relationship problems which probably relates to them having more of a secure attachment. The representational models of Minimizers may be more positive and flexible due to the presence of relatively higher levels of maternal care, which provided a more positive attachment and conferred a degree of protection from the effects of abusive experiences. Perhaps the negative nature of the abusive parent-child relationship does not have quite as profound an influence on later adult relationships with this added buffer of care. Indeed, the early attachment of Minimizers, which preceded the psychological abuse, may have been more positive resulting in a deeper sense of feeling supported. The fact that the Minimizers' profile does not illustrate a constellation of relationship outcome as detrimental as that of Acknowledgers, would be consistent with the theorized buffer effect of a secure early attachment.
While Erickson and Egeland (1987) found that the effects of abuse are inherently tied to feeling unsupported, Minimizers reported levels of supportive behaviors, that is, PAI respect, and QRI spousal support that are commensurate with Non-Abused individuals. Minimizers may be considered resilient adults in that they have the capacity of overcoming risk factors such as abusive events and avoiding social maladjustment (Hauser, Vieyra, Jacobson, & Wertreib, 1985).
While the majority of Minimizers reported a secure attachment, a significant number also reported feeling that they were somewhat uncomfortable being close to others and find it difficult to trust completely. This finding is not surprising, given that Minimizers did indeed experience more negative parenting and events that could be labeled as abusive compared to the Non-Abused group. The Minimizers did report problems with parenting teamwork and higher levels of spousal conflict. These areas may be more difficult for Minimizers to achieve success in because in the higher levels of controlling and abusive behaviors they experienced in childhood and both require more trust of the other person. Minimizers may also have normalized the combination of punitive and loving behaviors they experienced in their own childhood, which may in turn be negatively influencing how they deal with conflict and parenting teamwork in their current lives.
The results clearly demonstrate that Minimizers are not in the same category as Non-Abused individuals, although they self identify, as being non-abused. Rather, they appear to be an important middle group with a distinct profile that deserves attention in both research and clinical work. Thus, the present study supports suggestions (Varia et al., 1996) that differing perceptions of abusive events should be considered to have discriminable consequences on adult social adjustment. Indeed, researchers and clinicians have found that examining the parent-child relationship is crucial in understanding the different patterns associated with maltreatment (Belsky and Cicchetti). Early parenting styles, specifically higher levels of maternal care, appear to play a significant role in adult relationship satisfaction, indicating the importance of the early parent-child relationship on subsequent psychological functioning (Patterson, 1982).
This current study was in part a response to researchers calls' to address frequently cited methodological shortcomings (Braver; Briere and Lamphear) which plague maltreatment literature and research by (a) using the subject's personal criteria for labeling themselves as abused; (b) analyzing psychological maltreatment; (c) defining a nonclinical group of abuse victims and comparing them to a control group of non-abused individuals from the same population.
Definitions of abuse. The maltreatment literature is replete with definitional and classification problems (Gauthier; Mullen; Socolar and Straus). Rather than relying on widely discrepant professional criteria, the profound importance of personal interpretation of childhood experiences and events was utilized. While Socolar and colleagues (1995) state that "all definitions of abuse are trying to capture what really happened" (p. 572), this study attempts to also include how definitions of abuse must also incorporate how past events were experienced, coded, and possibly buffered or exacerbated by other contextual factors.
The strength of this study is that the label of abuse is left to the participants, rather than a professional criterion which may exclude an important subgroup of individuals. One major problem in threshold driven abuse definitions is that an abuse/no abuse dichotomy may not take the large percentage of grey cases into account (Rosenberg, 1987). Hart and Brassard (1987) suggest that "operational definitions should also consider the subjective impact of maltreatment ...because the degree of abuse (and the degree of harm experienced) may be related to the subjective meaning that a particular action or pattern of interactions has for a victim" (p. 162).
Nonclinical/control group sample. Another strength of this study is that the sample included reports from a nonclinical group with relatively high levels of education and income. Socolar and colleagues (1995) call for studying samples which are not biased in the direction of dysfunction and incorporate a wider range of individuals (Briere and Browne). Given this make-up, the extent of abuse-related symptomatology found in the current data is all the more significant. It is likely that a study of abused individuals in a mental health setting would show greater discriminability with regard to psychological abuse histories, and would reveal more long term maladjustment outcome. The current project's utilization of comparison groups as a method for teasing out the impact of maltreatment, represented an advantage in obtaining a more sophisticated understanding of possible confounds, such as maternal care and interpretation of events (Lamphear, 1985).
Methodological limitations. Some interpretations made in this study must be tempered by the parameters of this study. The study was based on a sample of volunteers who were willing to report on psychological parenting behavior. How representative this sample is of the larger population who were psychologically abused as children is unclear. A further limitation of this study is that all analyses were conducted with adults and therefore all maltreatment data involved retrospective reports. Every effort was made to achieve the most accurate memories of childhood maltreatment as possible. All the questionnaires were anonymously completed so that the participants would not worry about revealing negative events to strangers. The possibility of over-reporting due to a wish to please or a search for a meaningful explanation of current difficulties was minimized by anonymity. Over-reporting is expected to be minimized in a nonclinical sample such as this in which no obvious advantage accrued to those reporting abuse. No effort was made to collect data from, or give special reward to, participants who had considered themselves abused. The results also indicate that there was no overall systematic response bias for Minimizers. In fact, they reported more abusive behavior and more negative parenting than the Non-Abused group, while still labeling themselves as non-abused. In addition, they reported more positive behaviors with their current relationships, suggesting a complex pattern of responding not skewed in a particular direction.
Another limitation of this study is that it involves all self-reported questionnaire data. The use of paper and pencil methods versus interviewing is seen by some researchers as a less accurate method of assessing abuse (Wyatt & Peters, 1986). Although, Stinson and Hendrick (1992), in their study of disclosure rates, found that both face-to-face inquiry and self-reported questionnaires obtained a greater disclosure than clinical records. They also report no differences were found between direct therapist inquiry and self-reported data.
Another methodological limitation of this study is the criteria used to form the different groups. The use of the mean of the entire sample to create the groups versus using the standard deviation as the threshold would have formed stronger boundaries between the subgroups. This more stringent cut-off may have identified the purest form of Minimizers. Nevertheless, the current design allowed for the full range of individual experiences to be studied.
In summary, the results of the present study describe long-term correlates of child maltreatment. Findings indicate that abusive histories and subsequent adult functioning need to be examined within the context of a matrix of overall childhood experience, especially in terms of the quality of parenting received and individual perceptions of events. The fact that differences between the three perceptual groups were found across a variety of relationship measures including parenting alliance, quality of the spousal relationship, parental bonding, and adult attachment, enhances the validity of these results.
While researchers are beginning to examine the nature of the circumstances in which abuse occurs, research is still lacking in examining what factors or circumstances may be buffering or exacerbating adult psychological and relationship adjustment. Coffey and colleagues (1996) call for research and clinical strategies which identify and address factors which may mediate the effects of abuse. The results of this study identified maternal care as a possible protective factor in the quality of adult relationships with those individuals who experienced relatively higher levels of abuse. These findings are consistent with Steele's (1986) hypothesis that the actual injury of abuse may not be the cause of negative psychological effects, but that the relational context of the injury and how it is inherently tied to the child's system of perceptions which are related to trust and love are the most prominent factors. Assuming these results are valid and replicable, assessing warmth and control in the family and not only the presence of abuse in isolation must be seen as important factors in understanding maltreatment sequelae.
The identification of a Minimizing group suggests the important role of perception. These are individuals who may not actively seek either personal or professional support, because they do not view themselves as victims. Nevertheless the current findings indicate that Minimizers do not have as healthy relationships as participants categorized in the Non-Abused group and they and their current partners are negatively affected by their abuse history. The authors speculate that Minimizers may be normalizing some of the punitive and negative parenting they received as children, because it was combined with a degree of positive behavior of warmth and love. Perhaps, Minimizers may be transmitting some of the negative parenting they received onto their own children, because they did not perceive it as abusive and they may also relate to their elevated level of conflict and problems with teamwork in their parenting relationship.
Future research should include
other environmental and early parenting variables which may exert an
influence on perceptions of abusive events and subsequent adult adjustment.
Both healthy and dysfunctional dynamics need to be examined to identify
possible protective and risk factors. The attenuating effect of positive
aspects of the family environment, despite the presence of abuse, needs
to be considered if a clear understanding of the effects and sequelae
of maltreatment is to be achieved. Furthermore, developing and using
measures which accurately tap into early environmental factors would
enhance methodological design. Personality measures which assess psychological
adjustment for a relatively healthy sample are also recommended. The
literature must integrate this level of complexity to better understand
indirect pathways of maltreatment outcome .Brassard
1993
Abidin, R.R. and Brunner, J., 1995. Development of a Parenting Alliance Inventory. Journal of Clinical Child Psychology 24, pp. 31¯40.
Ainsworth, M.D., Blehar, M.C., Waters, E. and Wall, S., 1978. . Patterns of attachment: A psychological study of the strange situation Erlbaum, Hillsdale, NJ.
Alexander, P.C., 1992. Introduction to the special section on adult survivors of childhood sexual abuse. Journal of Consulting and Clinical Psychology 60, pp. 165¯166.
Belsky, J., 1980. Child maltreatment: An ecological integration. American Psychologist 35, pp. 320¯335.
Berger, A. M., & Knutson, J. F. (1984). The Assessing Environments III: A questionnaire for assessing punitive abuse-related childhood histories. Unpublished manuscript..
Berger, A.M., Knutson, J.F., Mehm, J.G. and Perkins, K.A., 1988. The self-report of punitive childhood experiences of young adults and adolescents. Child Abuse & Neglect 12, pp. 252¯262.
Bowlby, J. (1969). Attachment and loss, Vol. 1. Attachment. New York: Basic Books..
Brassard, M.R., Hart, S.N. and Hardy, D.B., 1993. The psychological maltreatment rating scales. Child Abuse & Neglect 17, pp. 715¯729.
Braver, M., Bumberry, J., Green, K. and Rawson, R., 1992. Childhood abuse and current psychological functioning in a university counseling center population. Journal of Counseling Psychology 39, pp. 252¯257.
Briere, J., 1992. Methodological issues in the study of sexual abuse effect. Journal of Consulting and Clinical Psychology 60, pp. 196¯203.
Briere, J. and Runtz, M., 1988. Multivariate correlates of childhood psychological and physical maltreatment among university women. Child Abuse & Neglect 12, pp. 331¯341. Abstract-EMBASE | $Order Document
Browne, A. and Finkelhor, D., 1986. Impact of child sexual abuse: A review of the research. Psychological Bulletin 99, pp. 66¯77.
Brunner, J. (1991). The development of a parenting alliance inventory. Unpublished manuscript. University of Virginia, Charlottesville..
Cicchetti, D. and Rizley, R., 1981. Developmental perspectives on the etiology, transmission, and sequelae of child maltreatment. New Directions for Child Development 11, pp. 31¯35.
Coffey, P., Leitenberg, H., Henning, K., Turner, T. and Bennett, R.T., 1996. Mediators of the long-term impact of child sexual abuse: Perceived stigma, betrayal, powerlessness, and self-blame. Child Abuse & Neglect 20, pp. 447¯455.
Dass, P. (1991). The effect of the parenting alliance on family environments: Implications for child rearing practices and child outcome. Unpublished Master's thesis, University of Virginia, Charlottesville..
Egeland, B., Sroufe, L.A. and Erickson, M., 1983. The developmental consequences of different patterns of maltreatment. Child Abuse & Neglect 7, pp. 459¯469.
Erickson, M.F. and Egeland, B., 1987. A developmental view of the psychological consequences of maltreatment. School Psychology Review 16, pp. 156¯168.
Garrison, E.G.A., 1987. Psychological maltreatment of children. American Psychologist 42, pp. 157¯159.
Gauthier, L., Stollack, G., Messe, L. and Aronoff, J., 1996. Recall of childhood neglect and physical abuse as differential predictors of current psychological functioning. Child Abuse & Neglect 20, pp. 549¯559.
Hart, S.N. and Brassard, M.R., 1987. A major threat to children's mental health: Psychological maltreatment. American Psychologist 42, pp. 160¯165.
Hauser, S.T., Vieyra, M.A., Jacobson, A.M. and Wertreib, D., 1985. Vulnerability and resilience in adolescence: Views from the family. Journal of Early Adolescence 5, pp. 81¯100.
Hazan, C. and Shaver, P., 1987. Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology 52, pp. 511¯524.
Herzberger, S.D., Potts, D.A. and Dillon, M., 1981. Abusive and nonabusive parental treatment from the child's perspective. Journal of Consulting and Clinical Psychology 49, pp. 81¯90.
Howes, C., & Espinosa, M. P. (1985). The consequences of child abuse for the formation of relationships with peers. Child Abuse & Neglect, 9, 497¯404..
Lamphear, V.S., 1985. The impact of maltreatment on children's psychosocial adjustment: A review of the research. Child Abuse & Neglect 9, pp. 251¯263.
Mullen, P.E., Martin, J.L., Anderson, J.C., Romans, S.E. and Herbison, G.P., 1996. The long-term impact of the physical, emotional, and sexual abuse of children: A community study. Child Abuse & Neglect 20, pp. 7¯21.
Oates, R.K., Forrest, D. and Peacock, A., 1985. Self-esteem of abused children. Child Abuse & Neglect 9, pp. 159¯163.
Parker, G., Tupling, H. and Brown, L.B., 1979. A parental bonding instrument. British Journal of Medical Psychiatry 52, pp. 1¯10.
Patterson, G.R., 1982. . Coercive family processes Castalia, Eugene, OR.
Pierce, G.R., Sarason, I.G. and Sarason, B.R., 1991. General and relationship-based perceptions of social support: Are two constructs better than one?. Journal of Personality and Social Psychology 61, pp. 1028¯1039.
Rohner, R.P. and Rohner, E.C., 1980. Antecedents and consequences of parental rejection: A theory of emotional abuse. Child Abuse & Neglect 4, pp. 189¯198.
Rosenberg, M.S., 1987. New directions for research on the psychological maltreatment of children. American Psychologist 42, pp. 166¯171.
Socolar, R.R., Runyan, D.K. and Amaya-Jackson, L., 1995. Methodological and ethical issues related to studying child maltreatment. Journal of Family Issues 16, pp. 565¯586.
Straus, M.A., 1980. Stress and physical child abuse. Child Abuse & Neglect 4, pp. 75¯88.
Steele, B.F., 1986. Notes on the lasting effect of early child abuse throughout the life cycle. Child Abuse & Neglect 10, pp. 228¯291.
Stinson, M.H. and Hendrick, S.S., 1992. Reported childhood sexual abuse in university counseling center clients. Journal of Counseling Psychology 39, pp. 370¯374.
Varia, R., Abidin, R.R. and Dass, P., 1996. Perceptions of abuse: Effects on adult psychological and social adjustment. Child Abuse & Neglect 20, pp. 511¯526
Weissman, S. and Cohen, R., 1985. The parenting alliance and adolescence. Adolescent Psychiatry 12, pp. 24¯45.
Wyatt, G. and Peters, S.D., 1986. Issues
in the definition of child sexual abuse in prevalence research. Child
Abuse & Neglect 10, pp. 231¯240.