Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Psychol Trauma . Author manuscript; available in PMC: 2026 Apr 15. Published before final editing as: Psychol Trauma. 2026 Apr 13:10.1037/tra0002175. doi: 10.1037/tra0002175 Search in PMC Search in PubMed View in NLM Catalog Add to search When Strengths May Become Vulnerabilities: Reconsidering the Role of Social Support and Psychological Strengths Among Sexual Assault Survivors Shaina A Kumar Shaina A Kumar 1 Department of Psychological Sciences, William & Mary Find articles by Shaina A Kumar 1 , Rebecca L Brock Rebecca L Brock 2 Department of Psychology, University of Nebraska-Lincoln Find articles by Rebecca L Brock 2 , David DiLillo David DiLillo 2 Department of Psychology, University of Nebraska-Lincoln Find articles by David DiLillo 2 Author information Copyright and License information 1 Department of Psychological Sciences, William & Mary 2 Department of Psychology, University of Nebraska-Lincoln ✉ Correspondence should be addressed to Shaina Kumar, [email protected] . PMC Copyright notice PMCID: PMC13078691 NIHMSID: NIHMS2154511 PMID: 41973782 The publisher's version of this article is available at Psychol Trauma Abstract Objective: Many women who experience child or adolescent sexual assault (CASA) are revictimized as adults. Although researchers have made considerable progress in understanding mechanisms that increase risk for adult revictimization, including CASA-related posttraumatic stress symptoms (PTSS), few have explored factors that protect against this association. Drawing from the Resilience Portfolio Model and prior research, this study examined perceived social support, gratitude, self-compassion, and optimism as potential strengths-based factors that may reduce revictimization risk. Method: Participants were 405 college women with a history of CASA who completed three assessments over one year. Results: As expected, PTSS stemming from CASA predicted greater adult revictimization across the yearlong study period. A multidimensional approach to these data revealed that a latent psychological strengths factor including gratitude, self-compassion, and optimism predicted lower odds of revictimization among survivors with low to moderate PTSS, although this protective association weakened as PTSS increased. In contrast, perceived social support was unrelated to revictimization. Conclusions: Findings suggest that psychological strengths may be protective under certain conditions, highlighting the importance of matching strengths-oriented recommendations to survivors’ symptom severity. Future researchers should examine whether tailored approaches that both address posttraumatic stress and build personal strengths reduce revictimization risk and foster resilience and wellbeing among women CASA survivors. Keywords: posttraumatic stress, resilience, revictimization, trauma, wellbeing Over half of women (54.3%) in the United States experience unwanted sexual contact in their lifetime ( Basile et al., 2022 ), with at least 18% of cases comprising childhood or adolescent sexual assault (CASA; Stoltenborgh et al., 2011 ). Sexual victimization is linked to greater risk for long-term difficulties, such as posttraumatic stress symptoms (PTSS), depression symptoms, substance misuse, and interpersonal challenges ( Collin-Vézina et al., 2013 ). Moreover, women with a history of CASA are more likely than nonvictimized women to experience adult sexual victimization (i.e., revictimization; Collin-Vézina et al., 2013 ), showing increased symptoms with each victimization ( Steine et al., 2017 ). Here, we focus on developmental revictimization (i.e., victimization occurring across different periods such as childhood to adulthood) rather than chronic or repeated victimization within a single developmental period. Prior work has shown earlier victimization often evolves into new adult risk ( Fereidooni et al., 2024 ), making this a critical stage at which targeted intervention may break the cycle of victimization. The high prevalence and suffering associated with revictimization has encouraged researchers to identify factors that may increase revictimization risk among women who have experienced CASA. A range of PTSS, which are experienced by a majority of CASA survivors ( Maikovich et al., 2009 ), are one such factor. Symptoms such as intrusive memories and flashbacks may diminish survivors’ situational awareness and impair threat perception ( Messman-Moore & Long, 2003 ). Survivors experiencing PTSS may also avoid thinking about their safety due to triggers related to a traumatic experience ( Arata, 2000 ). Hypervigilance, which might seem protective, could paradoxically lead to misjudging threats or overreliance on unsafe coping mechanisms ( Risser et al., 2006 ). In support of these assertions, studies of undergraduate women indicate self-reported intrusion, avoidance, and hyperarousal symptoms each contribute to the link between CASA and adult revictimization ( Arata, 2000 ; Risser et al., 2006 ). Together, these findings provide evidence that overall PTSS severity, encompassing multiple clusters, is one possible pathway connecting early and later revictimization by reducing survivors’ ability to navigate potentially threatening environments in ways that are outside their control. Such circumstances reflect the unjust conditions that contribute to revictimization risk. Indeed, it is important to acknowledge revictimization is not a reflection of survivors’ actions or character but results instead from the enduring impacts of survivors’ traumatic experiences and systemic inequalities. To address these complexities, we advocate for research that not only explores factors increasing risk of revictimization but also identifies pathways that foster resilience and support survivors in their recovery. Here, we focus on factors promoting resilience, defined as the process of adapting well (e.g., decreasing the odds of revictimization) following exposure to trauma and its sequelae (e.g., CASA and CASA-related PTSS). The Resilience Portfolio Model ( Grych et al., 2015 ) provides a comprehensive, multidimensional framework for understanding how survivors of violence may adapt following exposure to traumatic events. Specifically, the model conceptualizes resilience as the presence of compounding protective resources and assets that enable individuals to cope effectively with adversity. It highlights the role of external resources, such as social support, alongside internal assets, such as interpersonal and regulatory strengths, in shaping healthy functioning. Guided by this framework, the present study focuses on perceived social support as a key external resource and three internal assets, namely gratitude , self-compassion , and optimism , as potential psychological strengths that may protect against the risk of revictimization. Gratitude and self-compassion are conceptualized within the model as interpersonal strengths, reflecting capacities that facilitate positive relationships with others and with oneself. In contrast, optimism represents a regulatory strength, promoting adaptive emotion regulation and future-oriented coping. Together, these factors may each help buffer the link between CASA-related PTSS and revictimization by reinforcing survivors’ ability to maintain supportive connections, foster self-kindness, and sustain a hopeful outlook in spite of their traumatic experiences. We prioritized examining these factors above other possible resources and assets because each is theoretically and empirically tied to processes central to recovery from sexual violence. To begin, social support has consistently emerged as one of the most robust protective resources among survivors. Supportive and caring relationships provide safety and nurturance while fostering self-regulation, self-worth and efficacy, and healthy coping ( Grych et al., 2015 ). In this way, survivors might also perceive an event as less stressful if they believe others will provide support. Studies show perceived support from friends, family, and significant others is an important factor in reducing distress among trauma survivors (see Sippel et al., 2015 ), even when the support is generalized and unrelated to the trauma ( Evans et al., 2013 ). For example, greater perceived support appears to boost resilience to interpersonal and intrapersonal difficulties, including PTSS, among women CASA survivors ( Dumont et al., 2007 ). Together with perceived support, gratitude, self-compassion, and optimism represent complementary dimensions of resilience. Whereas support may function as an external buffer against the psychological sequalae of trauma, these three internal assets encourage adaptive coping processes that promote wellbeing and may mitigate vulnerability to revictimization. The value of these psychological strengths can be understood through the broaden-and-build theory ( Fredrickson, 2001 ), which proposes positive emotions stemming from strengths may enhance cognitive flexibility, build enduring psychological resources, and counter negative emotion. Gratitude reflects an experience of thankfulness, which involves valuing and appreciating positive experiences in daily life ( Rashid & Seligman, 2018 ). Gratitude may lessen the effects of trauma exposure by helping survivors counter trauma triggers and avoidance behaviors, thus facilitating experiences of positive affect and increased engagement in meaningful activities ( Kashdan et al., 2006 ). In support of this possibility, Vernon et al. (2009) examined gratitude in the immediate days following a traumatic event and found that college women who reported experiencing higher levels of posttrauma gratitude endorsed fewer PTSS. Vieselmeyer et al. (2017) also found gratitude functioned as a protective factor following trauma exposure, where gratitude was negatively related to PTSS and positively linked to a sense of personal growth. Self-compassion reflects kind and nonjudgmental relating to oneself and one’s emotional experiences ( Neff, 2003 ). Three interrelated components within self-compassion determine reactions to challenging experiences: self-kindness, which describes an understanding behavior toward oneself in the face of suffering; common humanity, which describes the perception of one’s experiences as part of humankind; and mindfulness, which describes the balanced awareness of negative thoughts and feelings ( Neff, 2003 ). Collectively, these elements interact to generate a self-compassionate frame of mind and may protect against PTSS. Engaging in self-compassion is said to decrease avoidance of discomfort and facilitate desensitization, reducing rumination and suppression of unwanted thoughts ( Neff, 2003 ). In support of this assertion, researchers have found that university students high in self-compassion engaged in fewer avoidance symptoms following trauma exposure, allowing for a natural exposure and recovery process ( Thompson & Waltz, 2008 ). Further, among women survivors of childhood interpersonal violence, self-compassion was negatively related to PTSS ( Scoglio et al., 2015 ). Optimism refers to a positive way of approaching the world and favorable expectancies toward the future ( Scheier et al., 1994 ). Optimistic individuals tend to present with adaptive responses to traumatic events, redirecting their attention to effective problem-focused coping strategies to lessen PTSS ( Tedeschi & Calhoun, 1996 ). Notably, empirical findings show college students who endorsed greater optimism experienced lower PTSS following a traumatic event ( Gil & Weinberg, 2015 ). Of relevance to this study, Kumar et al. (2022) also found a negative association between optimism and PTSS among undergraduate women sexual assault survivors. Importantly, psychological strengths can be cultivated through intervention and reinforced through habitual action ( Niemiec, 2018 ; Rashid & Seligman, 2018 ). If the strengths examined here show promise in buffering the contributions of CASA-related PTSS to revictimization, they may serve as useful targets in intervention with CASA survivors. Survivors may be especially receptive to activities that bolster strengths and wellbeing, as such interventions may be more engaging, less stigmatizing, and better aligned with goal-setting and achievement than those solely focused on symptom-based outcomes ( Cloitre et al., 2016 ). The Present Study Replicating past work, we expected greater levels of CASA-related PTSS among women entering the study would predict greater risk of adult revictimization over a one-year period ( H1 ). Drawing from the Resilience Portfolio Model and prior research, we further predicted perceived support would moderate the association between CASA-related PTSS and adult revictimization, such that the odds of adult revictimization would be lessened by greater levels of support ( H2 ). We also hypothesized the strengths of gratitude, self-compassion, and optimism would each moderate the association between CASA-related PTSS and adult revictimization, such that the odds of adult revictimization would be lessened by higher levels of each strength ( H3 ). Notably, the prospective design of this study stands in contrast to the majority of work on resilience to trauma. Whereas many prior studies of resilience assess predictors and outcomes concurrently, here we examine how social and psychological strengths might protect against adult revictimization before it has occurred, a key methodological contribution to the literature. Method Participants and Procedures Undergraduate women were invited to participate in a larger prospective study about sexual violence from Fall 2020 through Fall 2021. Recruitment occurred through email using a database of all undergraduate women attending University of Nebraska-Lincoln (UNL). At Time 1 (T1), women who were interested in participating completed an initial online consent form followed by screening items to determine eligibility (see below). Those who met criteria completed the full online consent form for all procedures and remaining T1 measures via Qualtrics. After six months, those who completed T1 were recontacted by email to complete the Time 2 (T2) measures. After another six months (12 months total), participants were emailed an invite to complete the Time 3 (T3) measures. Participants were compensated $15 for completing the T1 and T2 measures and $20 for the T3 measures. The UNL IRB approved all procedures. Our sample included 405 women who completed the study at T1, 313 (77.28%) at T2, and 260 (64.20%) at T3. At enrollment, women ranged in age from 17 to 28 ( M age = 19.44, SD = 1.66). The majority were in their first (35.3%; n = 143) or second (26.7%; n = 108) year of study. Reflecting the student body from which they were recruited, they identified as White (86.9%; n = 352), Black or African American (6.4%; n = 26), Asian (5.7%; n = 23), American Indian or Alaskan Native (1.7%; n = 7), Native Hawaiian or Other Pacific Islander (0.5%; n = 2), or “other” (4.2%; n = 17). Of these options, 20 women (4.9%) selected more than one race. Approximately 13.1% ( n = 53) identified as Latina, Hispanic, or Spanish. Most (69.1%; n = 280) identified as heterosexual and reported being in a romantic relationship but not married (51.6%; n = 209) or single (47.2%; n = 191). As an approximation of socioeconomic status, participants indicated their average yearly income if they supported themselves or their parents’ average yearly income if their parents supported them. The modal average yearly income was over $100,000 (20.5%; n = 83) but almost half (45.2%; n = 183) reported income at or below $50,000. Measures CASA History (T1) and Adult Revictimization (T2, T3) Only women who reported a history of CASA were eligible to participate in the present study. In order to capture survivors’ initial experiences of revictimization, we also excluded women with a history of adult revictimization (i.e., age 18 to present) from our sample. This decision allowed us to examine adult revictimization as a distinct outcome rather than a continuation of prior adult revictimization experiences. Participants’ CASA experiences were assessed at T1 using 19 items from the Modified Sexual Experiences Survey (MSES; Messman-Moore et al., 2010 ). These yes/no items evaluate unwanted sexual contact encompassing sexual play (e.g., fondling; four items), oral-genital contact (four items), attempted sexual intercourse (two items), completed sexual intercourse (five items), and other sexual acts (e.g., penetration by fingers or other objects; four items) that occurred as a result of perpetrator tactics related to verbal coercion, misuse of a position of authority, physical force, or when the participant was incapable of consent due to alcohol or drugs (note that attempted intercourse was only assessed in relation to physical force or incapacitation). Questions were adapted to apply to sexual assault perpetrated by individuals of any gender identity. Participants who endorsed at least one of 19 items indicating victimization prior to the age of 18 were classified as having experienced CASA. One additional question inquired whether the participant had experienced any of the listed MSES prompts from age 18 to present to aid in exclusion criteria. At T2 and T3, participants reported on any new sexual victimization experiences since the prior assessment (i.e., in the past six months) using the MSES. If they endorsed at least one of the 19 items related to sexual assault experiences (e.g., “Have you given in to sex play [fondling, kissing, or petting, but not intercourse] when you didn’t want to because you were overwhelmed by that person’s continual arguments and pressure?”), they were considered having experienced adult revictimization at the respective time point (0 = no revictimization in past 6 months ; 1 = reported revictimization in past 6 months ). Internal consistency was not calculated for the MSES because categories of sexual assault experiences are not necessarily related. PTSS Linked to CASA (T1) PTSS linked to participants’ CASA experiences were assessed using the 20-item PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013 ) at T1. Participants responses were anchored with the following prompt: “Thinking about your worst unwanted sexual experience prior to the age of 18, please read each problem carefully and then indicate how much you have been bothered by that event in the past month.” Each item is rated from 0 ( not at all ) to 4 ( extremely ), e.g., “In the past month, how much were you bothered by repeated, disturbing, and unwanted memories of the stressful experience?” We calculated a mean score such that higher ratings indicate more severe PTSS (possible range: 0–4; α = .94). Note we averaged (versus summed) items to align the PCL-5 scale with other composite scores in our analytic models. Perceived Social Support and Psychological Strengths (T1) Participants completed the 12-item Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et al., 1988 ) to assess perceived social support from friends, family, and significant others at T1. Participants rated each item from 1 ( very strongly disagree ) to 7 ( very strongly agree ), e.g., “I can talk about my problems with my friends.” Items were averaged such that higher scores indicate greater perceived support (possible range: 1–7; α = .91). Gratitude was assessed with the six-item Gratitude Questionnaire (GQ-6; McCullough et al., 2002 ) at T1. Participants rated each item from 1 ( strongly disagree ) to 7 ( strongly agree ), e.g., “I am grateful to a wide variety of people.” Negatively-worded items were reverse-scored and items were averaged so higher scores indicate greater gratitude (possible range: 1–7; α = .85). Self-compassion was assessed with the 12-item Self-Compassion Scale-Short Form (SCS-SF; Raes et al., 2011 ) at T1. Participants rated each item from 1 ( almost never ) to 5 ( almost always ), e.g., “I try to be loving towards myself when I’m feeling emotional pain.” The SCS-SF consists of six subscales: self-kindness, self-judgment, common humanity, isolation, mindfulness, and over-identification. Items on the self-judgment, isolation, and over-identification subscales were reverse-scored, and a total score was calculated as the mean of the six subscales so higher scores indicate greater self-compassion (possible range: 1–5; α = .84). Optimism was assessed with the 10-item Life Orientation Test-Revised (LOT-R; Scheier et al., 1994 ) at T1. Participants rated each item from 0 ( strongly disagree ) to 4 ( strongly agree ), e.g., “In uncertain times, I usually expect the best.” Negatively-worded items were reverse-scored and the six optimism-related items were averaged such that higher scores indicate greater optimism (possible range: 0–4; α = .83). Data Analytic Plan First, we examined descriptives and correlations among study variables. To understand patterns of missingness, we also examined if demographic characteristics and study variables were related to a participant missing the follow-up assessments. Of all our demographic (i.e., age, year in school, race, ethnicity, sexual orientation, relationship status, income) and study variables (i.e., CASA-related PTSS, perceived support, gratitude, self-compassion, optimism), only self-compassion was associated with missingness at both the T2, t (403) = 2.12, p = .035, and T3, t (403) = 2.10, p = .037, assessments. We address this finding in the results section. Next, using Mplus 8 ( Muthén & Muthén, 1998–2017 ), we tested our original hypotheses focused on the direct effect of CASA-related PTSS on adult revictimization ( H1 ) as moderated by perceived support ( H2 ) and psychological strengths ( H3 ). Utilizing multilevel modeling, data were hierarchically organized at Level 1 (within-person level) such that the two repeated measures of revictimization collected at T2 and T3 were nested within participants (0 = no revictimization in past 6 months ; 1 = reported revictimization in past 6 months ). (Recall that reports at the baseline assessment were used to determine whether revictimization occurred across the study.) Given the binary nature of the outcome, each model was tested using a logit link function. At Level 2 (between-person level), CASA-related PTSS, perceived support, and psychological strengths were examined as predictors of the overall probability of revictimization across the two repeated measures (i.e., over the yearlong observation period). For each moderation model, we entered the predictor, moderator, and interaction term (e.g., CASA-related PTSS × gratitude) simultaneously as predictors of revictimization. Following, we conducted a regions of significance analysis, which computes simple slopes at all levels of the moderator and identifies at what levels of the moderator the predictor is related to the outcome. The proportion of data present ranged across variables. Data for Level 2 predictors were collected at T1 from survivors who agreed to informed consent and completed key measures assessing experiences of PTSS linked to CASA as well as social and psychological strengths ( N = 405). Level 1 outcome data were collected at T2 and T3, such that 313/405 (77.28%) participants provided data at T2 and 260/405 (64.20%) at T3, resulting in 573 of 810 person-waves (70.74%). Missing data were handled through full information maximum likelihood (FIML; Enders, 2010 ), leveraging all available data. CASA-related PTSS were entered into each model uncentered at Level 2 as zero was a meaningful score in our sample (scores ranged from 0.00–3.85). Moderators were mean-centered to facilitate interpretation of main effects as zero was not a meaningful score on scales of support (scores ranged from 1.25–7.00), gratitude (scores ranged from 1.67–7.00), self-compassion (scores ranged from 1.08–4.67), or optimism (scores ranged from 0.33–4.00). Although each moderator was centered, raw values are reported in the regions analyses for interpretation. Note that, following a series of unexpected findings, we also conducted post-hoc analyses to better understand the role of support and strengths in our sample of survivors. We first ran a parallel set of moderation analyses such that PTSS instead functioned as the moderator in our models. We probed the conditional effect of each strength on revictimization risk at −1 SD , the average, and +1 SD of PTSS and conducted regions of significance analyses. All terms were entered into the models as outlined above, and all centering decisions were identical. Finally, guided by the Resilience Portfolio Model’s emphasis on understanding an integrative model of both resources and assets, we sought to test the unique contributions of support and strengths after accounting for their shared variance as a theory-informed extension to our original hypotheses. We initially tested a model with PTSS, support, gratitude, self-compassion, and optimism entered simultaneously, along with all PTSS interaction terms, using similar procedures outlined above. Given the magnitude of between-person correlations among our variables, we determined the three measures of psychological strengths shared substantial variance, and therefore remodeled gratitude, self-compassion, and optimism as a latent factor (PSYSTR), retained PTSS and support as observed predictors, and included PTSS × PSYSTR and PTSS × support interactions in the final model. The latent strengths factor (PSYSTR) was well-defined by its three indicators, as described more below. All reported coefficients are unstandardized. Estimates are reported with robust standard errors (MLR estimator) to account for any violations of normality. All model parameters for our original and post-hoc analyses are presented in Table 1 . Table 1. Regression Model Parameters for Analyses Predicting Revictimization Model 1: PTSS Predicting Revictimization ( R 2 = .056) Term B SE p -value 95% CI PTSS 0.41 0.18 .021 [0.06, 0.76] Model 2: Perceived Social Support as a Moderator ( R 2 = .150) PTSS 0.43 0.18 .016 [0.08, 0.79] Perceived Social Support −0.63 0.20 .001 [−1.02, −0.24] PTSS * Perceived Social Support 0.46 0.14 .001 [0.18, 0.74] Model 3: Gratitude as a Moderator ( R 2 = .190) PTSS 0.30 0.19 .104 [−0.06, 0.67] Gratitude −1.04 0.27 < .001 [−1.57, −0.51] PTSS * Gratitude 0.55 0.16 .001 [0.23, 0.87] Model 4: Self-Compassion as a Moderator ( R 2 = .189) PTSS 0.44 0.20 .024 [0.06, 0.83] Self-Compassion −1.29 0.39 .001 [−2.05, −0.53] PTSS * Self-Compassion 0.77 0.29 .009 [0.20, 1.35] Model 5: Optimism as a Moderator ( R 2 = .119) PTSS 0.31 0.18 .091 [−0.05, 0.69] Optimism −0.88 0.36 .015 [−1.59, −0.17] PTSS * Optimism 0.38 0.24 .119 [−0.10, 0.86] Model 6: Multidimensional Model with Observed Variables ( R 2 = .532) PTSS 0.38 0.20 .050 [−0.001, 0.768] Perceived Social Support −0.30 0.21 .155 [−0.72, 0.11] Gratitude −0.64 0.33 .051 [−1.278, 0.002] Self-Compassion −0.84 0.44 .057 [−1.71, 0.03] Optimism −0.01 0.49 .992 [−0.96, 0.95] PTSS * Perceived Social Support 0.32 0.17 .057 [−0.01, 0.64] PTSS * Gratitude 0.29 0.20 .157 [−0.11, 0.68] PTSS * Self-Compassion 0.60 0.34 .081 [−0.07, 1.27] PTSS * Optimism − 0.14 0.31 .642 [−0.75, 0.46] Model 7: Multidimensional Model with Latent and Observed Variables ( R 2 = .322) PTSS 0.27 0.21 .201 [−0.14, 0.68] Perceived Social Support −0.12 0.24 .637 [−0.59, 0.36] PSYSTR −1.57 0.52 .002 [−2.59, −0.56] PTSS * Perceived Social Support 0.23 0.19 .215 [−0.13, 0.59] PTSS * PSYSTR 0.67 0.34 .049 [0.002, 1.344] Open in a new tab Note . Models 1–7 were estimated with N = 405 clusters. R 2 values represent the proportion of variance explained in revictimization at the between-person level. Perceived social support, gratitude, self-compassion, and optimism scores were mean-centered for analyses to facilitate interpretation of model parameters. The focal parameters of interest, representing interactions among CASA-related PTSS and support or psychological strengths predicting revictimization, are bolded. Regression coefficients represent change in the average log odds of revictimization for a one-unit increase in the predictor variable. Parallel moderation models that reversed the roles of PTSS and support or psychological strengths contained the same model terms, yielding identical effects (Models 1–5). These results are not presented separately. Results Descriptive and Bivariate Statistics Means, SD s, observed ranges, and correlations among variables at the between-person level are in Supplemental Table 1 . Regarding the nature of the child or adolescent sexual assault, women reported experiencing these unwanted events as a result of verbal coercion (85.7%; n = 347), physical force (34.3%; n = 139), lack of consent due to alcohol or drugs (26.7%; n = 108), or misuse of a position of authority (14.6%; n = 59; more than one option could be selected). With regard to endorsement of revictimization across the study (variables at the within-person level), 32.9% ( n = 103 out of 313) of women indicated experiencing unwanted sexual events six months from baseline (T2) and 24.6% ( n = 64 out of 260) indicated experiencing unwanted sexual events one year from baseline (T3). Further, 42.2% ( n = 132 of 313) indicated experiencing unwanted sexual events at either the six- or twelve-month assessment, and 11.2% ( n = 35 of 313) indicated experiencing unwanted sexual events at both assessments. CASA-Related PTSS and Adult Revictimization Replicating past work, we investigated whether greater PTSS linked to CASA (T1) were associated with average odds of adult revictimization across the two follow-up assessments (T2, T3). As expected, baseline PTSS increased risk for revictimization, B = 0.41, 95% CI [0.06, 0.76] ( H1 ). Next, we tested whether associations between CASA-related PTSS and revictimization varied as a function of perceived support ( H2 ) and psychological strengths ( H3 ). Because these single-strength models reflect our original analytic plan, we present them first before turning to the multidimensional analyses, which provide a more complete and theoretically aligned test of resilience. Perceived Social Support as a Moderator The association between PTSS at baseline and adult revictimization across the one-year study period was positively moderated by support, B = 0.46, 95% CI [0.18, 0.74]. Conditional effects indicated that, for individuals who endorsed relatively low levels of support (4.15, 1 SD below the mean), PTSS were not significantly associated with revictimization, B = −0.10, 95% CI [−0.53, 0.32]. However, for individuals who endorsed relatively moderate levels of support (5.31, the mean), PTSS were positively associated with revictimization, B = 0.43, 95% CI [0.08, 0.78]. For individuals who endorsed relatively high levels of support (6.48, 1 SD above the mean), PTSS were positively associated with revictimization, and the magnitude of this effect increased, B = 0.97, 95% CI [0.44, 1.51]. To the extent survivors endorsed higher support, the association between CASA-related PTSS and risk of revictimization became more positive ( H2 ). Although our primary focus was on the moderating role of perceived support in the association between PTSS and revictimization, we also examined the conditional effect of support which demonstrated that greater support was associated with lower odds of revictimization ( B = −0.63, 95% CI [−1.02, −0.24]) in the absence of PTSS (i.e., when PTSS equaled zero). Gratitude as a Moderator The association between PTSS at baseline and adult revictimization across the one-year study period was positively moderated by gratitude, B = 0.55, 95% CI [0.23, 0.87]. Conditional effects indicated that, for individuals who endorsed relatively low levels of gratitude (4.61, 1 SD below the mean), PTSS were not significantly associated with revictimization, B = −0.27, 95% CI [−0.73, 0.19]. For individuals who endorsed relatively moderate levels of gratitude (5.65, the mean), PTSS remained unrelated to revictimization, B = 0.30, 95% CI [−0.06, 0.67]. However, for individuals who endorsed relatively high levels of gratitude (6.69, 1 SD above the mean), PTSS were positively associated with revictimization, B = 0.88, 95% CI [0.35, 1.41]. To the extent survivors endorsed higher gratitude, the association between CASA-related PTSS and risk of revictimization became more positive ( H3 ). The effect of gratitude on revictimization when PTSS were 0.00 was negative, such that greater gratitude was associated with lower odds of revictimization ( B = −1.04, 95% CI [−1.57, −0.51]). Self-Compassion as a Moderator The association between PTSS at baseline and adult revictimization across the one-year study period was positively moderated by self-compassion, B = 0.77, 95% CI [0.20, 1.35]. Conditional effects indicated that, for individuals who endorsed relatively low levels of self-compassion (1.94, 1 SD below the mean), PTSS were not significantly associated with revictimization, B = −0.07, 95% CI [−0.51, 0.36]. However, for individuals who endorsed relatively moderate levels of self-compassion (2.61, the mean), PTSS were positively associated with revictimization, B = 0.44, 95% CI [0.06, 0.83]. For individuals who endorsed relatively high levels of self-compassion (3.28, 1 SD above the mean), PTSS were positively associated with revictimization, and the magnitude of this effect increased, B = 0.96, 95% CI [0.32, 1.59]. To the extent survivors endorsed higher self-compassion, the association between CASA-related PTSS and risk of revictimization became more positive ( H3 ). The effect of self-compassion on revictimization when PTSS were 0.00 was negative, such that greater self-compassion was associated with lower odds of revictimization ( B = −1.29, 95% CI [−2.05, −0.53]). Optimism as a Moderator There was a trending pattern for optimism as a moderator of the association between PTSS at baseline and adult revictimization across the one-year study period, B = 0.38, 95% CI [−0.10, 0.86]. Conditional effects indicated that, for individuals who endorsed relatively low levels of optimism (1.29, 1 SD below the mean), PTSS were not significantly associated with revictimization, B = 0.03, 95% CI [−0.45, 0.51]. For individuals who endorsed relatively moderate levels of optimism (2.03, the mean), PTSS remained unrelated to revictimization, B = 0.31, 95% CI [−0.05, 0.67]. However, for individuals who endorsed relatively high levels of optimism (2.77, 1 SD above the mean), PTSS were positively associated with revictimization, B = 0.59, 95% CI [0.06, 1.12]. To the extent survivors endorsed higher optimism, the association between CASA-related PTSS and risk of revictimization became more positive ( H3 ). The effect of optimism on revictimization when PTSS were 0.00 was negative, such that greater optimism was associated with lower odds of revictimization ( B = −0.88, 95% CI [−1.59, −0.17]). 1 Post-Hoc Analyses Given the above pattern of findings (i.e., a positive interaction between PTSS and protective factors predicting revictimization) was unanticipated, we conducted post-hoc analyses to better understand the role of support and psychological strengths. Specifically, we examined the level of PTSS at which each factor was no longer protective in the context of revictimization (e.g., at subclinical or probable levels of PTSD). In other words, PTSS instead became the moderator in these models. We probed the conditional effect of each strength on revictimization risk at −1 SD , the average, and +1 SD of PTSS and conducted regions of significance analyses. Perceived Social Support at Conditional Levels of PTSS Conditional effects indicated that, for individuals who endorsed relatively low levels of PTSS (0.37, 1 SD below the mean), support was negatively associated with revictimization, B = −0.46, 95% CI [−0.77, −0.15]. However, for individuals who endorsed relatively moderate levels of PTSS (1.21, the mean), support was no longer associated with revictimization, B = −0.07, 95% CI [−0.31, 0.17]. For individuals who endorsed relatively high levels of PTSS (2.05, 1 SD above the mean), support remained unrelated to revictimization, B = 0.31, 95% CI [−0.04, 0.67]. A regions of significance analysis revealed support lost its propensity for resilience at PTSS scores of 0.84 and above, which is equivalent to summed scores of 16.80 and above on the PCL-5, and seemed to be counterproductive (i.e., increasing risk for revictimization) when PTSS were very high (2.22 and above, equivalent to summed scores of 44.40 and above on the PCL-5). Gratitude at Conditional Levels of PTSS Conditional effects indicated that, for individuals who endorsed relatively low levels of PTSS (0.37, 1 SD below the mean), gratitude was negatively associated with revictimization, B = −0.83, 95% CI [−1.27, −0.40]. For individuals who endorsed relatively moderate levels of PTSS (1.21, the mean), gratitude was negatively associated with revictimization, although the magnitude of this effect decreased, B = −0.36, 95% CI [−0.65, −0.09]. However, for individuals who endorsed relatively high levels of PTSS (2.05, 1 SD above the mean), gratitude was no longer associated with revictimization, B = 0.10, 95% CI [−0.24, 0.43]. A regions of significance analysis revealed gratitude lost its propensity for resilience at PTSS scores of 1.40 and above, which is equivalent to summed scores of 28.00 and above on the PCL-5, and seemed to be counterproductive (i.e., increasing risk for revictimization) when PTSS were very high (2.83 and above, equivalent to summed scores of 56.60 and above on the PCL-5). Self-Compassion at Conditional Levels of PTSS Conditional effects indicated that, for individuals who endorsed relatively low levels of PTSS (0.37, 1 SD below the mean), self-compassion was negatively associated with revictimization, B = −1.01, 95% CI [−1.60, −0.41]. However, for individuals who endorsed relatively moderate levels of PTSS (1.21, the mean), self-compassion was no longer associated with revictimization, B = −0.36, 95% CI [−0.80, 0.08]. For individuals who endorsed relatively high levels of PTSS (2.05, 1 SD above the mean), self-compassion remained unrelated to revictimization, B = 0.29, 95% CI [−0.42, 1.00]. A regions of significance analysis revealed self-compassion lost its propensity for resilience at PTSS scores of 1.11 and above, which is equivalent to summed scores of 22.20 and above on the PCL-5, and seemed to be counterproductive (i.e., increasing risk for revictimization) when PTSS were very high (3.76 and above, equivalent to summed scores of 75.20 and above on the PCL-5). Optimism at Conditional Levels of PTSS Conditional effects indicated that, for individuals who endorsed relatively low levels of PTSS (0.37, 1 SD below the mean), optimism was negatively associated with revictimization, B = −0.74, 95% CI [−1.31, −0.17]. For individuals who endorsed relatively moderate levels of PTSS (1.21, the mean), optimism remained negatively associated with revictimization, although the magnitude of this effect decreased, B = −0.42, 95% CI [−0.81, −0.03]. However, for individuals who endorsed relatively high levels of PTSS (2.05, 1 SD above the mean), optimism was no longer associated with revictimization, B = −0.10, 95% CI [−0.65, 0.45]. A regions of significance analysis revealed optimism lost its propensity for resilience at PTSS scores of 1.29 and above, which is equivalent to summed scores of 25.80 and above on the PCL-5, and was at no point counterproductive (i.e., increasing risk for revictimization) when PTSS were high. A Multidimensional Model Approach Finally, guided by the Resilience Portfolio Model’s emphasis on a more holistic understanding of processes that promote resilience, we estimated an integrative model of both resources and assets to test the unique contributions of support and psychological strengths after accounting for their shared variance. This analysis was intended to examine whether any of these factors uniquely protected against or exacerbated the association between PTSS and revictimization, further clarifying implications for intervention. In running a model with PTSS, support, gratitude, self-compassion, and optimism entered simultaneously, along with all PTSS interaction terms, no main effects or interactions emerged ( p s ≥ .05, all 95% CIs contained zero). Based on these findings, as well as the magnitude of between-person correlations among variables, we determined that the three measures of psychological strengths shared substantial variance, and this overlapping variance may interact with PTSS. We therefore remodeled gratitude, self-compassion, and optimism as a latent factor (PSYSTR), retained PTSS and support as observed predictors, and included PTSS × PSYSTR and PTSS × support interactions. The latent strengths factor (PSYSTR) was well-defined by its three indicators. Standardized loadings, obtained using the Bayes estimator (required for standardized results under a two-level random model), were significant and salient for gratitude (λ = 0.76, 95% CI [0.68, 0.82]), self-compassion (λ = 0.64, 95% CI [0.55, 0.71]), and optimism (λ = 0.68, 95% CI [0.61, 0.76]). The PTSS × PSYSTR interaction was significant and positive ( B = 0.67, 95% CI [0.002, 1.344]). Consistent with our previous post-hoc analyses, we probed the latent interaction by examining the conditional effects of PSYSTR on revictimization across all observed values of PTSS (0.00–3.85). Results indicated the conditional effect of PSYSTR on revictimization was negative and significant at lower to average levels of PTSS, ranging from B = −1.57, 95% CI [−2.59, −0.56] at PTSS = 0.00 to B = −0.62, 95% CI [−1.242, −0.002] at PTSS = 1.41. Beginning at PTSS = 1.42 (equivalent to summed scores of 28.40 and above on the PCL-5), the conditional effect of PSYSTR on revictimization was no longer significant, as the 95% CIs for the simple slopes contained zero. Thus, PSYSTR significantly predicted lower likelihood of revictimization only at lower to average levels of PTSS, and this predictive effect weakened as PTSS increased. In contrast, support ( B = −0.12, 95% CI [−0.59, 0.36]) and the PTSS × support interaction ( B = 0.23, 95% CI [−0.13, 0.59]) were unrelated to revictimization. See Figure 1 for an illustration of the conditional effects of PSYSTR on revictimization at observed levels of PTSS. Figure 1. Conditional Effects of Latent Psychological Strengths on Revictimization at Observed Levels of PTSS. Open in a new tab Note . The bold line represents conditional effects of latent psychological strengths on sexual revictimization at observed levels of PTSS (0.00–3.85). The 95% CI for the conditional effects is represented by the light gray bands surrounding the bold line. Conditional effects were negative and significant (i.e., the 95% CI did not contain zero) for PTSS scores ranging from 0.00 to 1.41 and non-significant (i.e., the 95% CI contained zero) for PTSS scores ranging from 1.42 to 3.85. Estimated sample percentages in each region are based on complete data for the PTSS measure ( N = 405). Discussion The current study’s major contribution to resilience science lies in its prospective, longitudinal evaluation of factors that may attenuate or amplify risk for adult sexual revictimization. By focusing on revictimization as a clinically consequential outcome, these findings extend prior resilience research that has relied largely on cross-sectional designs and symptom-based outcomes. Indeed, our findings have direct relevance for clinical intervention efforts among CASA survivors, helping clarify for whom and under what conditions strengths-based approaches may be most beneficial for wellbeing. Consistent with this contribution, we first observed that PTSS stemming from CASA predicted greater revictimization across the yearlong study. This result replicates findings from the literature showing CASA is associated with revictimization and aligns with prior work demonstrating a positive link between PTSS severity and revictimization ( Arata, 2000 ; Messman-Moore & Long, 2003 ; Risser et al., 2006 ). Our findings affirm CASA-related PTSS are a robust predictor of later revictimization and reinforce the importance of addressing PTSS within trauma-focused treatment for survivors. Importantly, this pattern of results also emphasizes the deeply unfair reality that the harms of early victimization can mark people as more vulnerable to later victimization. As an initial step toward understanding for whom and under what conditions this link may be most relevant, we next tested perceived support as a potential moderator of the association between PTSS and revictimization. Although there was a significant interaction between PTSS and support, results suggested the association between PTSS and revictimization actually strengthens as support increases. However, post-hoc analyses helped unravel this unexpected finding, revealing that the odds of revictimization lessened with greater support but only to the extent that PTSS were low. This pattern of results builds on the Resilience Portfolio Model’s emphasis on additive effects of strengths by demonstrating that support reduced risk for revictimization, but its protective effects did not hold in the context of elevated symptoms. While unexpected, these findings are consistent with other work indicating that social support has shown mixed promise in resilience research (see Hamby, 2025 ) and that its protective effects might decrease at higher levels of trauma exposure ( Evans et al., 2013 ). It is possible that support systems may unintentionally bolster avoidance, such that the presence of certain types of support might encourage the recipient to escape distressing thoughts and feelings ( Calkins & Brock, 2020 ). In such cases, though well-meaning, support providers may inadvertently limit survivors’ opportunities to consolidate new learning, potentially undermining their ability to engage in self-protective behaviors ( Risser et al., 2006 ). This does not mean survivors cannot benefit from support but rather calls into question assumptions that support is invariably adaptive and underscores the need to adopt trauma-sensitive conceptualizations of social support. Like support, we found the positive association between PTSS and revictimization was strengthened at higher levels of each psychological strength. Moreover, post-hoc analyses indicated the effects of strengths paralleled those for support such that each strength was generally associated with reduced risk, but their protective capacity diminished as symptoms increased. These findings diverge from work suggesting positive emotions associated with strengths might “undo” the negative effects of PTSS (see Fredrickson, 2001 ), although recent research suggests strengths such as compassion may not be sufficient on their own to promote wellbeing following experiences of victimization ( Brooks et al., 2024 ). Given their novelty, any explanations for these results are speculative, although one possibility is that the challenging nature of traumatic experiences may make it difficult for survivors to effectively balance the use of their strengths, manifesting in the overuse of positivity. Although the value of positive emotions has long been recognized, a failure to also appreciate the negative emotions that naturally occur in the context of processing traumatic experiences could be harmful for survivors. Researchers should continue to investigate this phenomenon, exploring the optimal boundaries of the broaden-and-build theory and psychological strengths among CASA survivors. Another way to understand this pattern of findings is that the four-strength portfolio examined here (i.e., perceived support, gratitude, self-compassion, and optimism) may not be sufficient to help survivors with elevated PTSS achieve resilience, operationalized in this study as avoiding revictimization. Survivors with higher levels of PTSS may need a more robust resilience portfolio that includes additional resources and assets beyond those modeled above. For example, survivors’ sense of meaning in life, as reflected in feelings of personal significance, purpose, and coherence, may be especially important for reducing revictimization risk. Prior resilience literature supports the value of considering meaning-making strengths in the context of trauma exposure, as well as other assets such as self-regulation, perseverance, and forgiveness ( Grych et al., 2015 ; Hamby, 2025 ). Future work in this area may benefit from examining a broader set of meaning-related strengths and assets and identifying which combinations are most protective for survivors with high PTSS. Although we initially examined support and each psychological strength separately, these single-strength models evaluate protective factors in isolation and may overstate their unique effects. We address this limitation by prioritizing our multidimensional findings, representing the study’s primary contribution. When all predictors and their interactions with PTSS were entered simultaneously, no single strength or perceived support showed unique associations with revictimization once shared variance was controlled. However, when gratitude, self-compassion, and optimism were modeled as a latent factor, the combined strengths accounted for variance beyond support. This finding suggests the underlying commonality among these strengths, rather than any one asset in isolation, may be particularly relevant for understanding revictimization risk. Had a single strength emerged as uniquely predictive, that would have highlighted it as especially critical. Instead, our findings point to their collective role as mutually reinforcing factors, offering protection primarily when PTSS are low to moderate and losing value as PTSS increase. Thus, our results suggest the additive portfolio of strengths proposed by the Resilience Portfolio Model may operate as theorized only when PTSS are not highly elevated, showing the protective value of strengths is conditional rather than uniform across levels of symptoms. In reality, these strengths rarely exist in isolation, and analyses focused on a single strength could oversimplify the role of resiliency among women CASA survivors. Multidimensional approaches better reflect the complexity of survivors’ resources and assets. Ultimately, the goal of resilience science, and conceptual frameworks such as the Resilience Portfolio Model, is to identify how best to support people experiencing adversity by examining how multiple positive resources and assets operate together rather than focusing on any single factor in isolation. Such an approach helps move the field beyond a deficit-based lens and instead toward strength-oriented frameworks that recognize trauma survivors’ range of adaptive capacities, while also carefully considering for whom and under what conditions strengths promote resilience and wellbeing. With this in mind, we encourage researchers to continue investigating portfolios of strengths to clarify which are most central to survivors’ resilience and recovery following trauma exposure, contributing toward a science of healing ( Hamby, 2025 ). Limitations and Future Directions Several limitations of this study suggest directions for future research. To begin, our sample lacked certain elements of diversity that might impact the generalizability of results. All survivors were undergraduate women, most of whom identified as White, non-Hispanic, and heterosexual. As one might expect from a college sample, survivors also endorsed relatively low levels of PTSS, whereas their report of support and strengths were somewhat high, on average. Although subthreshold PTSS can result in clinically significant distress, future work is needed to increase the generalizability of findings across settings, or to adults more broadly, including men and people with diverse racial, ethnic, gender, and sexual identities. Related to generalizability, type of trauma experienced may be important to consider in the context of the current study’s findings. Research has shown that interpersonal trauma such as sexual assault may result in unique trauma-related outcomes, particularly due to its impact on factors such as safety, trust, power and control, esteem, and intimacy ( Resick et al., 2017 ). Moreover, it is possible our study’s findings are shaped by qualitative differences in trauma characteristics, such as acknowledgment of the assault and ongoing contact with the perpetrators. However, we did not collect information on survivors’ acknowledgement of their experiences or relationships to the perpetrators in the present study, and thus we were unable to examine how these contextual factors might intersect with support and psychological strengths. Future researchers would benefit from incorporating the above dimensions in their work and replicating these findings in other trauma-exposed samples. Notably, because data collection began during the COVID-19 pandemic, this may have also elevated overall distress and altered access to social support (e.g., reduced in-person contact) and utilization of strengths. Our findings should be interpreted in light of this context and replication in post-pandemic cohorts is warranted as well. Additionally, it is likely that the same vulnerabilities originally contributing to survivors’ experiences of victimization are still present, such as neglectful parenting, lack of family support, and living in more dangerous neighborhoods. A lack of resources, such as access to mental health care, also likely contributes to challenges in healing from earlier victimization experiences. Indeed, this is a central tenet of the Resilience Portfolio Model and other multidimensional models of resilience. Future researchers wishing to expand upon our work should consider exploring other risk factors that may contribute to experiences of adult sexual revictimization to better inform points of intervention for survivors. Researchers may also consider extending our moderation results to a mediation framework. In the context of longer-term longitudinal designs, support and strengths may function as mechanisms through which PTSS increase revictimization risk, as experiences of trauma and PTSS may erode these resources over time. Another extension would be to examine chronic, within-period victimization alongside developmental revictimization. Although we are not aware of any studies that have directly examined differing mechanisms between victimization types, these different forms may arise through partly distinct processes (e.g., shifts in social context versus ongoing coercive dynamics), pointing to the need for research that explicitly contrasts these pathways in the context of risk and resilience. Related to assessment, we encourage researchers to consider construct measurement in future work. For example, widely used global scales of perceived support (e.g., the MSPSS from Zimet et al., 1988 ) primarily assess the availability and frequency of support but not the adequacy, timing, or specific forms of support. These measures typically capture trait-like perceptions and may not reflect whether support could be accessed in real time during periods of heightened vulnerability. Moreover, global support scales may capture maladaptive as well as adaptive forms of support, such as well-intended responses that minimize survivors’ experiences or discourage disclosure. Thus, it may be more informative to assess specific, safety-relevant supportive behaviors proximal to victimization, such as supporting values-based decisions or assisting with safety planning, as these targeted actions may matter more for preventing revictimization than basic availability or quantity of contact. In addition, “strengths” questionnaires may inadvertently embed distress into their item content. For example, the SCS-SF includes several items that begin with negative stems (e.g., “When I’m going through a very hard time,” “When I’m feeling down”), which may lead people who frequently experience distress to endorse items more highly simply because they encounter these negative states more regularly. As a result, scores may partially reflect frequency of distress, rather than the quality of self-compassionate responding, and may be more positively correlated with symptom measures. This suggests a need for revised measures that better separate the presence of distress from the quality of self-compassionate responding, for example, by balancing neutral and negative item stems. Clinical Implications and Conclusion To our knowledge, this is the first longitudinal, prospective study to show that perceived social support and psychological strengths may function as both protective and risk-enhancing factors for adult sexual revictimization among CASA survivors, depending on PTSS severity. Moreover, our multidimensional findings suggest that although there might be value in boosting gratitude, self-compassion, and optimism as ways to reduce revictimization risk (beyond support), the clinical utility of this approach should be considered carefully, particularly in the context of heightened PTSS. Because this study did not test an intervention, treatment implications should be considered tentative; however, findings suggest that strengths-based strategies may be most beneficial for survivors when their PTSS are low to moderate, whereas elevated PTSS may indicate a need for other supports. Clinicians working with CASA survivors might, for example, deliver evidence-based trauma-focused treatments such as cognitive processing therapy ( Resick et al., 2017 ) to lower PTSS, while thoughtfully incorporating psychological strengths-based interventions as clinically indicated (e.g., positive psychotherapy [ Rashid & Seligman, 2018 ], goal-focused positive psychotherapy [ Conoley & Scheel, 2018 ]). Rather than privileging any single strength in isolation, a whole-person, trauma-informed approach that integrates symptom-focused care with a strategic emphasis on psychological strengths-based interventions may offer one promising way to reduce revictimization risk and foster resilience and wellbeing among survivors. Such an approach acknowledges these assets are not uniformly protective and makes clear that aligning interventions with survivors’ symptoms may determine whether they serve as strengths or become vulnerabilities. Supplementary Material Supplemental Material NIHMS2154511-supplement-Supplemental_Material.docx (19.5KB, docx) Clinical Impact Statement. This study suggests the need to revisit assumptions that social support and psychological strengths are invariably adaptive and to adopt trauma-sensitive conceptualizations of these factors for women survivors of child or adolescent sexual assault (CASA). Researchers and clinicians may also consider tailoring recommendations of strengths-based interventions among women CASA survivors, encouraging balanced approaches that both address posttraumatic stress and boost strengths in service of fostering resilience and wellbeing. Data Transparency Statement: Data reported in this manuscript were collected as part of a larger data collection, and no studies from the larger dataset are currently under review. One study from the larger dataset is currently in press. MS 1 (in press) explores the concurrent associations between CASA-related PTSS and character strengths, as well as CASA-related posttraumatic growth and character strengths, through the lens of network analysis. MS 2 (the current manuscript) focuses on the longitudinal associations among CASA-related PTSS, perceived social support and psychological strengths, and adult revictimization. Funding Statement: This research was supported by a grant from the National Institute of Child Health and Human Development (F31HD101271; PI: Kumar) under the supervision of David DiLillo and a dissertation research award from the Melissa Institute for Violence Prevention and Treatment (PI: Kumar). The views expressed are those of the authors and do not necessarily represent the views or policy of the National Institute of Child Health and Human Development or Melissa Institute for Violence Prevention and Treatment. Footnotes Conflict of Interest Statement : The authors have no conflicts of interest to disclose. Ethical Standards Statement : This study was approved by the University of Nebraska-Lincoln (IRB Approval #: 20200320058EP), and all participants provided informed consent. 1 Because self-compassion was correlated with missingness at both the T2 and T3 assessments, we ran a series of sensitivity analyses consistent with a saturated correlates approach ( Enders, 2010 ). In the models that did not include self-compassion as a moderator (i.e., perceived support, gratitude, optimism), we included self-compassion as an auxiliary variable, and results were consistent in direction, strength, and significance with those reported above. Data Availability Statement: De-identified data specific to the current paper are available upon reasonable request. References Arata CM (2000). From child victim to adult victim: A model for predicting sexual revictimization. Child Maltreatment, 5(1), 28–38. 10.1177/1077559500005001004 [ DOI ] [ PubMed ] [ Google Scholar ] Basile KC, Smith SG, Kresnow M, Khatiwada S, & Leemis RW (2022). The National Intimate Partner and Sexual Violence Survey: 2016/2017 report on sexual violence. Centers for Disease Control and Prevention. [ Google Scholar ] Brooks M, Taylor E, & Hamby S (2024). Polyvictimization, polystrengths, and their contribution to subjective well-being and posttraumatic growth. Psychological Trauma: Theory, Research, Practice, and Policy, 16(3), 496–503. 10.1037/tra0001489 [ DOI ] [ PubMed ] [ Google Scholar ] Calkins FC, & Brock RL (2020). The dark side of helping behaviors: Partner support increases daily alcohol use in outpatients with a history of alcohol dependence. Journal of Social and Personal Relationships, 37(5), 1389–1404. 10.1177/0265407519898257 [ DOI ] [ Google Scholar ] Cloitre M, Jackson C, & Schmidt JA (2016). STAIR for strengthening social support and relationships among veterans with military sexual trauma and PTSD. Military Medicine, 181(2), 183–187. 10.7205/MILMED-D-15-00209 [ DOI ] [ PubMed ] [ Google Scholar ] Collin-Vézina D, Daigneault I, & Hébert M (2013). Lessons learned from child sexual abuse research: Prevalence, outcomes, and preventive strategies. Child and Adolescent Psychiatry and Mental Health, 7, 1–9. 10.1186/1753-2000-7-22 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Conoley CW, & Scheel MJ (2018). Goal focused positive psychotherapy. Oxford. [ Google Scholar ] Dumont KA, Widom CS, & Czaja SJ (2007). Predictors of resilience in abused and neglected children grown-up: The role of individual and neighborhood characteristics. Child Abuse & Neglect, 31(3), 255–274. 10.1016/j.chiabu.2005.11.015 [ DOI ] [ PubMed ] [ Google Scholar ] Enders CK (2010). Applied missing data analysis. Guilford. [ Google Scholar ] Evans SE, Steel AL, & DiLillo D (2013). Child maltreatment severity and adult trauma symptoms: Does perceived social support play a buffering role? Child Abuse & Neglect, 37(11), 934–943. 10.1016/j.chiabu.2013.03.005 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Fereidooni F, Daniels JK, & Lommen MJJ (2024). Childhood maltreatment and revictimization: A systematic literature review. Trauma, Violence, & Abuse, 25(1), 291–305. 10.1177/15248380221150475 [ DOI ] [ Google Scholar ] Fredrickson BL (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56(3), 218–226. 10.1037/0003-066X.56.3.218 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Gil S, & Weinberg M (2015). Coping strategies and internal resources of dispositional optimism and mastery as predictors of traumatic exposure and of PTSD symptoms: A prospective study. Psychological Trauma: Theory, Research, Practice, and Policy, 7(4), 405–411. 10.1037/tra0000032 [ DOI ] [ PubMed ] [ Google Scholar ] Grych J, Hamby S, & Banyard V (2015). The resilience portfolio model: Understanding healthy adaptation in victims of violence. Psychology of Violence, 5(4), 343–354. 10.1037/a0039671 [ DOI ] [ Google Scholar ] Hamby S (2025). The resilience portfolio concept: New insights into how sufficient strengths can overcome even high burdens of trauma. Review of General Psychology, 29(3), 311–324. 10.1177/10892680251363859 [ DOI ] [ Google Scholar ] Kashdan TB, Uswatte G, & Julian T (2006). Gratitude and hedonic and eudaimonic well-being in Vietnam war veterans. Behaviour Research and Therapy, 44(2), 177–199. 10.1016/j.brat.2005.01.005 [ DOI ] [ PubMed ] [ Google Scholar ] Kumar SA, Jaffe AE, Brock RL, & DiLillo D (2022). Resilience to suicidal ideation among college sexual assault survivors: The protective role of optimism and gratitude in the context of posttraumatic stress. Psychological Trauma: Theory, Research, Practice, and Policy, 14(S1), S91–S100. 10.1037/tra0001141 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Maikovich AK, Koenen KC, & Jaffee SR (2009). Posttraumatic stress symptoms and trajectories in child sexual abuse victims: An analysis of sex differences using the National Survey of Child and Adolescent Well-Being. Journal of Abnormal Child Psychology, 37(5), 727–737. 10.1007/s10802-009-9300-x [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] McCullough ME, Emmons RA, & Tsang J (2002). The grateful disposition: A conceptual and empirical topography. Journal of Personality and Social Psychology, 82(1), 112–127. 10.1037//0022-3514.82.1.112 [ DOI ] [ PubMed ] [ Google Scholar ] Messman-Moore TL, & Long PJ (2003). The role of childhood sexual abuse sequelae in the sexual revictimization of women: An empirical review and theoretical reformulation. Clinical Psychology Review, 23(4), 537–571. 10.1016/S0272-7358(02)00203-9 [ DOI ] [ PubMed ] [ Google Scholar ] Messman-Moore TL, Walsh K, & DiLillo D (2010). Emotion dysregulation and risky sexual behavior in revictimization. Child Abuse & Neglect, 34(12), 967–976. 10.1016/j.chiabu.2010.06.004 [ DOI ] [ PubMed ] [ Google Scholar ] Muthén LK & Muthén BO (1998–2017). Mplus user’s guide (8th ed.). Muthén & Muthén. [ Google Scholar ] Neff KD (2003). Self-compassion: An alternative conceptualization of a healthy attitude toward oneself. Self and Identity, 2(2), 85–101. 10.1080/15298860309032 [ DOI ] [ Google Scholar ] Niemiec RM (2018). Character strengths interventions: A field guide for practitioners. Hogrefe. [ Google Scholar ] Raes F, Pommier E, Neff KD, & Van Gucht D (2011). Construction and factorial validation of a short form of the Self-Compassion Scale. Clinical Psychology & Psychotherapy, 18(3), 250–255. 10.1002/cpp.702 [ DOI ] [ PubMed ] [ Google Scholar ] Rashid T & Seligman M (2018). Positive psychotherapy: Clinician manual. Oxford. [ Google Scholar ] Resick PA, Monson CM, & Chard KM (2017). Cognitive processing therapy for PTSD: A comprehensive manual. Guilford. [ Google Scholar ] Risser HJ, Hetzel-Riggin MD, Thomsen CJ, & McCanne TR (2006). PTSD as a mediator of sexual revictimization: The role of reexperiencing, avoidance, and arousal symptoms. Journal of Traumatic Stress, 19(5), 687–698. 10.1002/jts.20156 [ DOI ] [ PubMed ] [ Google Scholar ] Scheier MF, Carver CS, & Bridges MW (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A re-evaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67(6), 1063–1078. 10.1037/0022-3514.67.6.1063 [ DOI ] [ PubMed ] [ Google Scholar ] Scoglio A, Rudat DA, Garvert D, Jarmolowski M, Jackson C, & Herman JL (2015). Self-compassion and responses to trauma: The role of emotion regulation. Journal of Interpersonal Violence, 33(13), 2016–2036. 10.1177/0886260515622296 [ DOI ] [ PubMed ] [ Google Scholar ] Sippel LM, Pietrzak RH, Charney DS, Mayes LC, & Southwick SM (2015). How does social support enhance resilience in the trauma-exposed individual? Ecology and Society, 20(4), 1–10. 10.5751/ES-07832-200410 [ DOI ] [ Google Scholar ] Steine IM, Winje D, Krystal JH, Bjorvatn B, Milde AM, Grønli J, Nordhus IH, & Pallesen S (2017). Cumulative childhood maltreatment and its dose-response relation with adult symptomatology: Findings in a sample of adult survivors of sexual abuse. Child Abuse & Neglect, 65, 99–111. 10.1016/j.chiabu.2017.01.008 [ DOI ] [ PubMed ] [ Google Scholar ] Stoltenborgh M, van IJzendoorn MH, Euser EM, & Bakermans-Kranenburg MJ (2011). A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment, 16(2), 79–101. 10.1177/1077559511403920 [ DOI ] [ PubMed ] [ Google Scholar ] Tedeschi RG, & Calhoun LG (1996). The Posttraumatic Growth Inventory: Measuring the positive legacy of trauma. Journal of Traumatic Stress, 9(3), 455–471. 10.1002/jts.2490090305 [ DOI ] [ PubMed ] [ Google Scholar ] Thompson BL, & Waltz J (2008). Self-compassion and PTSD symptom severity. Journal of Traumatic Stress, 21(6), 556–558. 10.1002/jts.20374 [ DOI ] [ PubMed ] [ Google Scholar ] Vernon LL, Dillon JM, & Steiner ARW (2009). Proactive coping, gratitude, and posttraumatic stress disorder in college women. Anxiety, Stress & Coping, 22(1), 117–127. 10.1080/10615800802203751 [ DOI ] [ PubMed ] [ Google Scholar ] Vieselmeyer J, Holguin J, & Mezulis A (2017). The role of resilience and gratitude in posttraumatic stress and growth following a campus shooting. Psychological Trauma: Theory, Research, Practice, and Policy, 9(1), 62–69. 10.1037/tra0000149 [ DOI ] [ PubMed ] [ Google Scholar ] Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, & Schnurr PP (2013). The PTSD Checklist for DSM-5 (PCL-5). Scale available from www.ptsd.va.gov Zimet GD, Dahlem NW, Zimet SG, & Farley GK (1988). The Multidimensional Scale of Perceived Social Support. Journal of Personality Assessment, 52(1), 30–41. 10.1207/s15327752jpa5201_2 [ DOI ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplemental Material NIHMS2154511-supplement-Supplemental_Material.docx (19.5KB, docx) Data Availability Statement De-identified data specific to the current paper are available upon reasonable request. ACTIONS View on publisher site PDF (471.8 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top