Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jun 30;82(10):1471–1481. doi: 10.1002/jclp.70173

Maladaptive Perfectionism and Adolescent Nonsuicidal Self‐Injury: A Longitudinal Moderated Mediation Model of Hopelessness and Self‐Concept Clarity

Honglei Gu 1, Yuqing Xiao 2, Xichen Ai 3, Meng Fan 4,✉, Tiansheng Xia 5
PMCID: PMC13544487  PMID: 42376892

ABSTRACT

Objective

The Benefits and Barriers Model of nonsuicidal self‐injury (NSSI) suggests that individuals with maladaptive perfectionism may be prone to negative self‐views when facing failure, which may be associated with lowered psychological barriers to NSSI. Based on this framework, the present three‐wave longitudinal study examined whether hopelessness mediated the association between maladaptive perfectionism and later NSSI, and whether this indirect association varied as a function of self‐concept clarity.

Methods

A sample of 786 Chinese adolescents (51.8% girls; M age = 13.27, SD = 0.74) participated in a three‐wave longitudinal study with 6‐month intervals. Maladaptive perfectionism and NSSI were assessed at baseline, hopelessness and self‐concept clarity at the 6‐month follow‐up, and NSSI again at the 12‐month follow‐up.

Results

After controlling for gender, age, baseline hopelessness, and baseline NSSI, baseline maladaptive perfectionism was indirectly associated with NSSI at the 12‐month follow‐up through hopelessness at the 6‐month follow‐up. The hopelessness–later NSSI association varied by self‐concept clarity and was weaker among adolescents with higher self‐concept clarity.

Conclusion

These findings suggest that hopelessness may be relevant to understanding later NSSI among adolescents, and that self‐concept clarity may help identify variability in the strength of the hopelessness–later NSSI association. Any prevention or intervention implications involving self‐concept clarity should be considered tentative until future longitudinal, experimental, or intervention research determines whether changes in self‐concept clarity precede changes in NSSI risk or produce downstream improvements in self‐injury‐related outcomes.

Keywords: hopelessness, maladaptive perfectionism, nonsuicidal self‐injury, self‐concept clarity

1. Introduction

Nonsuicidal self‐injury (NSSI) is the direct, deliberate, and repetitive destruction of one's own body tissue in the absence of clear suicidal intent (Nock and Favazza 2009) and represents a prevalent concern among adolescents worldwide. International estimates indicate that approximately 22.9% of adolescents report lifetime NSSI and 18.6% report NSSI in the past 12 months (Gillies et al. 2018). In China, the 6–24‐month prevalence among adolescents has been reported at 23.3% (Lang and Yao 2018). Recent evidence suggests that global rates of NSSI have continued to rise (Mannekote Thippaiah et al. 2020). NSSI is strongly associated with a range of psychopathology (e.g., depression, eating disorders) (Schatten et al. 2015) and confers an elevated risk for subsequent suicidal behaviors (Kiekens et al. 2018). As a result, NSSI was listed in the DSM‐5 as a “condition for further study” (American Psychiatric Association 2013). Given its dynamic developmental course and serious consequences, longitudinal investigations into the mechanisms underlying NSSI are essential for informing early identification and intervention efforts among adolescents.

1.1. Maladaptive Perfectionism and NSSI

Maladaptive perfectionism denotes an irrational adherence to unrealistically high performance standards driven by fear of failure and harsh self‐evaluation (Frost et al. 1990). According to the Benefits and Barriers Model of NSSI, which explains NSSI from a decision‐making perspective, positive self‐perception may function as a protective barrier against self‐injury by supporting self‐preservation and self‐care; conversely, negative self‐views may remove that barrier and increase vulnerability to NSSI (Hooley and Franklin 2018). Individuals high in maladaptive perfectionism typically hold pervasive negative self‐appraisals, which may erode protective self‐regard and thereby facilitate self‐injurious behavior. Moreover, maladaptive perfectionism can motivate NSSI as a form of self‐punishment or self‐directed blame, whereby individuals use physical harm to express or atone for perceived failures (Claes et al. 2012).

Extensive research indicates that maladaptive perfectionism is a significant predictor of NSSI in adolescents (Cheng et al. 2026; Chester et al. 2015; Luyckx et al. 2015). For instance, a systematic review of 15 studies identified maladaptive perfectionism as a consistent risk factor for adolescent NSSI (Gyori and Balazs 2021). Although much of the evidence linking maladaptive perfectionism to NSSI has come from community samples, findings from clinical samples are also important for evaluating the clinical relevance of this association. For example, Fujimori et al. (2011) found that, among 80 patients diagnosed with eating disorders, those who reported NSSI showed significantly higher levels of perfectionism than those without NSSI. This finding provides a clinical illustration of the broader association between perfectionism and NSSI and supports the relevance of examining maladaptive perfectionism as a risk‐related factor.

Despite converging evidence, prior research has notable limitations. First, many studies have used cross‐sectional designs, limiting causal inference and underscoring the need for longitudinal investigations. Second, samples have often been female‐dominant, leaving male NSSI underexamined and constraining the generalizability of findings across sexes (Gyori and Balazs 2021). Theoretical accounts suggest plausible pathways linking maladaptive perfectionism to NSSI. Perfectionistic individuals are prone to intense negative reactions to perceived failure, including hopelessness (Frost et al. 1990). Hopelessness refers to a cognitive‐affective state characterized by negative expectations about the future and perceived difficulty in improving one's circumstances (Beck et al. 1974). Because the reduction of negative affect is among the most commonly reported functions of NSSI (Taylor et al. 2018), hopelessness may increase adolescents' vulnerability to NSSI. Conversely, higher self‐concept clarity, defined as the extent to which individuals' self‐beliefs are clearly and confidently defined, internally consistent, and stable over time (Campbell et al. 1996), has been associated with healthier emotion regulation and more adaptive coping. Accordingly, self‐concept clarity may help identify variability in the association between hopelessness and later NSSI, such that this association may be weaker among adolescents with higher levels of self‐concept clarity. Accordingly, the present study used a longitudinal moderated mediation design to examine whether hopelessness mediated the prospective association between maladaptive perfectionism and NSSI, and whether self‐concept clarity moderated the association between hopelessness and NSSI. The aim was to clarify theoretically relevant pathways and boundary conditions associated with adolescent NSSI and to inform future prevention and intervention efforts.

1.2. The Mediating Role of Hopelessness

The dual‐process model of perfectionism distinguishes adaptive perfectionists—who set realistic and attainable standards, derive satisfaction from their achievements, and therefore remain optimistic about future outcomes—from maladaptive perfectionists, who set unattainable goals and, driven by fear of failure, seldom experience satisfaction or confidence in future success (Slade and Owens 1998). Maladaptive perfectionists tend to respond to mistakes with harsh self‐criticism and hold negative beliefs about goal attainment, which increases vulnerability to feelings of hopelessness (Ferrari et al. 2018). Consistent with the theoretical perspective that maladaptive perfectionism may be associated with later hopelessness, Rice et al. (2006) reported that maladaptive perfectionism prospectively predicted greater hopelessness 10 weeks later in a sample of US college students.

From the perspective of Escape Theory, which explains why individuals may engage in NSSI to escape or reduce aversive internal states, distressing negative states—such as hopelessness arising from perceived failure—may impair problem‐solving capacities and increase the likelihood of using NSSI as a maladaptive strategy to alleviate aversive emotions (Hoff and Muehlenkamp 2009). A substantial body of empirical evidence supports the link between hopelessness and NSSI (Dhingra et al. 2016; Gray et al. 2021; Gu et al. 2021). For example, a meta‐analysis of 20 prospective studies found that hopelessness is a strong predictor of NSSI (odds ratio = 3.08; Fox et al. 2015). Similarly, Gray et al. (2021) reported that elevated hopelessness—whether assessed explicitly or implicitly—was associated with both lifetime and recent NSSI. Taken together, theoretical frameworks and empirical evidence suggest that hopelessness may mediate the relationship between maladaptive perfectionism and subsequent NSSI.

1.3. The Moderating Role of Self‐Concept Clarity

Several studies have identified self‐concept clarity (SCC) as an important protective factor against NSSI (Goldberg and Israelashvili 2017; Lear and Pepper 2016; Scala et al. 2018). For example, in a cross‐sectional survey study of 147 college students, SCC was significantly and negatively associated with NSSI engagement, frequency, and versatility, even after controlling for negative affect and age (Lear and Pepper 2016). Likewise, ecological momentary assessment findings suggest that the association between intense negative emotions and self‐harm impulses may vary by SCC, with stronger associations observed among individuals with lower SCC and weaker associations among those with higher SCC (Scala et al. 2018).

Bandura's (2001) theory of self‐regulation posits that clearer and more coherent self‐knowledge may support individuals' capacity to regulate mood and adapt to environmental demands. Higher SCC may help individuals reinterpret their circumstances and experiences more positively (Y. Wu et al. 2020), thereby supporting more adaptive attributions and coping. Hopelessness—characterized by negative expectations for the future and diminished motivation—reflects pervasive pessimism about upcoming outcomes (Beck et al. 1974). Individuals with higher SCC tend to report greater self‐confidence and more constructive responses to failure and setbacks (J. Wu et al. 2010), which may be associated with more optimistic future expectations. Accordingly, SCC may moderate the association between hopelessness and later NSSI, such that hopelessness may be less strongly associated with later NSSI among adolescents with higher SCC.

1.4. The Present Study

The present study uses the Benefits and Barriers Model of NSSI (Hooley and Franklin 2018) as the primary organizing framework. This model conceptualizes NSSI as more likely to occur when the perceived benefits of self‐injury increase and/or when psychological barriers against self‐injury are weakened. Within this framework, related theories are integrated to clarify specific pathways in the proposed model. The Dual‐Process Model of Perfectionism supports the pathway from maladaptive perfectionism to hopelessness by suggesting that maladaptive perfectionistic tendencies may heighten negative self‐evaluation and vulnerability to hopelessness. Escape Theory further clarifies why hopelessness may be associated with later NSSI, as adolescents experiencing intense or persistent negative emotions may be more likely to use NSSI as a maladaptive strategy to reduce or escape aversive emotional states. Finally, Self‐Regulation Theory informs the proposed moderating role of self‐concept clarity. Adolescents with higher self‐concept clarity may be better able to regulate distress and maintain a coherent sense of self. These characteristics may offer a tentative explanation for why hopelessness was less strongly associated with subsequent NSSI among adolescents with higher self‐concept clarity.

Hence, the present study employed a longitudinal moderated mediation model (see Figure 1) to examine the associations among maladaptive perfectionism, hopelessness, self‐concept clarity, and adolescent NSSI. Specifically, the following hypotheses were proposed:

Figure 1.

Figure 1

Open in a new tab

A moderated mediation model of maladaptive perfectionism, hopelessness, self‐concept clarity, and NSSI.

Hypothesis H1

Hopelessness at T2 would account for the indirect association between maladaptive perfectionism at T1 and NSSI at T3.

Hypothesis H2

Self‐concept clarity at T2 would moderate the association between hopelessness at T2 and NSSI at T3, such that this association would be weaker among adolescents with higher self‐concept clarity.

2. Methods

2.1. Participants

This study employed a 1‐year longitudinal follow‐up of adolescents enrolled in a middle school in Guangdong Province, using cluster sampling. The first assessment (T1) was conducted in October 2021, the second (T2) in April 2022, and the third (T3) in October 2022. At T1 and T2, there were 878 and 813 valid participants, respectively, while T3 included 786 valid participants. A total of 92 students did not complete the last 2 assessments due to reasons such as school transfer or absence, yielding an attrition rate of 10.5%. Among the retained participants, 48.2% were male and 51.8% were female, with ages ranging from 12 to 16 years (M = 13.27, SD = 0.74). Analyses comparing retained and attrited participants revealed no significant differences in maladaptive perfectionism (t = 0.03, p > 0.05), hopelessness (t = 1.01, p > 0.05), self‐concept clarity (t = −0.51, p > 0.05), or NSSI (t = 0.14, p > 0.05).

2.2. Procedure

This study was approved by the research ethics committee of the corresponding author's university. The questionnaire survey was administered to seventh‐ and eighth‐grade middle school students at three time points, with a 6‐month interval between each wave. Measures of maladaptive perfectionism, hopelessness, NSSI behavior, and self‐concept clarity were administered according to the measurement schedule described below. Data were collected in classrooms after obtaining parental consent and adolescent assent.

Maladaptive perfectionism was assessed at T1 only and was conceptualized as a relatively stable dispositional vulnerability in the longitudinal model. Hopelessness was assessed at both T1 and T2, and T2 hopelessness was modeled as the mediator while controlling for T1 hopelessness. NSSI was assessed at both T1 and T3, and T3 NSSI was modeled as the outcome while controlling for T1 NSSI. This approach allowed us to examine whether T1 maladaptive perfectionism was indirectly associated with T3 NSSI through T2 hopelessness, after accounting for baseline levels of the mediator and outcome. Furthermore, self‐concept clarity was assessed at T2 only because the study aimed to examine whether adolescents' level of self‐concept clarity, measured at the same wave as hopelessness, moderated the association between T2 hopelessness and later T3 NSSI.

Each assessment followed a standardized procedure. A trained psychology graduate student served as the primary examiner, provided participants with an overview of the survey content and instructions, and informed them of their right to withdraw at any time. During administration, the examiner distributed the questionnaires and instructed participants to complete them independently. The questionnaires were collected immediately upon completion. Each session lasted approximately 15 min. After completing the assessment, all participants received a small gift as a token of appreciation.

2.3. Measures

2.3.1. Maladaptive Perfectionism Scale

At T1, maladaptive perfectionism was assessed using the “concern over mistakes” (nine items; e.g., “If I make a mistake, I should feel frustrated”) and “doubt about action” (four items; e.g., “I often hesitate over everyday little things”) subscales of the Chinese version (Zi and Zhou 2006) of the Multidimensional Perfectionism Scale (MPS; Frost et al. 1990). Items were rated on a 5‐point Likert scale (1 = strongly disagree, 5 = strongly agree), and mean scores were computed, with higher scores indicating greater maladaptive perfectionism. The MPS has demonstrated good reliability and validity in Chinese adolescent samples (Cheng et al. 2026). In the present study, Cronbach's α for T1 maladaptive perfectionism was 0.82.

2.3.2. Self‐Concept Clarity Scale

At T2, self‐concept clarity was measured using the Chinese version (Niu et al. 2016) of the Self‐Concept Clarity Scale (SCS; Campbell et al. 1996). The SCS comprises 12 items (e.g., “In general, I have a clear sense of who I am and what I am”), each rated on a 5‐point Likert scale (1 = strongly disagree, 5 = strongly agree). Mean scores were computed, with higher scores reflecting greater self‐concept clarity. This measure has demonstrated good reliability and validity in Chinese adolescent samples (e.g., Gu et al. 2024). In the present study, Cronbach's α for T2 self‐concept clarity was 0.81.

2.3.3. Hopelessness Scale

At T1 and T2, hopelessness was assessed using the short version of the Beck Hopelessness Scale (BHS; Perczel‐Forintos et al. 2013). The scale consists of four items (e.g., “My future seems dark to me”) and is rated on a 5‐point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The BHS has demonstrated good reliability and validity in Chinese samples (e.g., Gu et al. 2021). In this study, confirmatory factor analyses indicated that the standardized factor loadings for the item “Things just won't work out the way I want them to” were low (0.15 at T1 and 0.23 at T2), whereas all other items had factor loadings above 0.50. Consequently, this item was removed, and the average score was calculated across the remaining three items, with higher scores indicating greater hopelessness. The resulting scale demonstrated acceptable internal consistency, with Cronbach's α values of 0.71 and 0.75 at T1 and T2, respectively.

2.3.4. Nonsuicidal Self‐Injury Scale

At T1 and T3, adolescents' NSSI behavior was assessed using the NSSI Scale (You et al. 2013). The scale comprises seven items representing different self‐injurious behaviors: cutting, burning, biting, punching, scratching the skin, inserting sharp objects into the nails or skin, and banging the head or other body parts against walls. Participants were asked, “In the past 6 months, have you engaged in the following behaviors to deliberately harm yourself without suicidal intent?” Each item was rated on a 6‐point scale (0 = never, 1 = once, 2 = twice, 3 = three times, 4 = four times, 5 = five times or more). The NSSI score was calculated as the average score across the seven self‐injury methods assessed, with higher scores indicating more frequent NSSI behavior. The scale has demonstrated good reliability and validity in Chinese adolescent samples (e.g., Gu et al. 2022). In the present study, Cronbach's α was 0.83 at T1 and 0.85 at T3.

2.4. Data Analysis

First, descriptive statistics and Pearson's correlation analyses were conducted using SPSS 22.0. Second, Mplus 8.6 was used to test whether T2 hopelessness mediated the association between T1 maladaptive perfectionism and T3 NSSI, and whether this indirect association varied as a function of T2 self‐concept clarity. NSSI was assessed at baseline and at the 12‐month follow‐up, but not at the 6‐month follow‐up. Accordingly, the primary longitudinal model included baseline NSSI as a covariate and T3 NSSI as the focal outcome, while hopelessness and self‐concept clarity assessed at the 6‐month follow‐up were specified as the mediator and moderator, respectively. Because NSSI was not assessed at the 6‐month follow‐up, models including T2 NSSI could not be estimated. The reported model therefore reflects the actual measurement schedule and the intended temporal ordering of baseline maladaptive perfectionism and NSSI, 6‐month hopelessness and self‐concept clarity, and 12‐month NSSI.

Prior research suggested that age and gender were associated with hopelessness and NSSI (Lester 2015; Tang et al. 2018), so we included them as covariates to reduce potential confounding in the estimation of the focal relationships in our model. Age was treated as a continuous variable, and gender was treated as a dichotomous variable (0 = male, 1 = female). All nondemographic variables were standardized. The 95% confidence interval for the mediating effect was estimated using the bias‐corrected percentile bootstrap method, with significance indicated if the interval did not include zero (Hayes 2013). Finally, to further illustrate the moderating role of self‐concept clarity, simple slope analyses were conducted and the interaction effects were plotted.

The primary longitudinal models were estimated among participants with valid T3 NSSI outcome data, yielding an analytic sample of n = 786. Within this analytic sample, missing data on other model variables were handled in Mplus using full information maximum likelihood estimation (Muthén and Muthén 2019). Thus, FIML was used to retain participants with partial missingness on predictors, mediator/moderator, or covariates included in the model, but the primary models were restricted to participants with observed T3 NSSI because T3 NSSI was the focal outcome. The analytic sample size was the same across all primary models, as indicated by the number of observations reported in the Mplus output.

3. Results

3.1. Descriptive Statistics

Among adolescents in this sample, 35.1% (n = 276) reported engaging in NSSI at T1 and 21.3% (n = 167) at T3. Conversely, 64.9% and 78.7% of participants reported no NSSI at T1 and T3, respectively, indicating a substantial concentration at the floor of the scale. Cutting was the most commonly reported method of NSSI, with prevalence rates of 23.9% at T1 and 13.8% at T3, whereas burning was the least commonly reported method, with prevalence rates of 5.0% at T1 and 3.7% at T3. These prevalence rates suggest that NSSI was endorsed by a meaningful proportion of the sample and was not extremely rare. Nevertheless, the high proportion of participants reporting no NSSI indicates a clear floor effect; therefore, findings involving NSSI should be interpreted with appropriate caution.

Table 1 presents the means, standard deviations, skewness, kurtosis, and Pearson's correlation coefficients for all study variables. Correlation analyses indicated that T1 maladaptive perfectionism, T1 and T2 hopelessness, and T1 and T3 NSSI were positively correlated with one another. In contrast, T2 self‐concept clarity was negatively correlated with T1 maladaptive perfectionism, T1 and T2 hopelessness, and T1 and T3 NSSI.

Table 1.

Means, standard deviations, and Pearson's correlations of the study variables.

Variables 1 2 3 4 5 6
1. T1 MP —
2. T1 HO 0.63*** —
3. T1 NSSI 0.34*** 0.36*** —
4. T2 HO 0.50*** 0.53*** 0.34*** —
5. T2 SCC −0.53*** −0.44*** −0.27*** −0.62*** —
6. T3 NSSI 0.23*** 0.24*** 0.46*** 0.31*** −0.24*** —
M 2.62 2.49 0.31 2.50 3.09 0.23
SD 0.70 0.86 0.65 0.85 0.70 0.62
Skewness 0.30 0.56 3.15 0.50 −0.13 3.68
Kurtosis −0.15 −0.10 12.19 −0.14 −0.11 15.49
Open in a new tab

Abbreviations: HO = hopelessness, MP = maladaptive perfectionism, SCC = self‐concept clarity.

***

p < 0.001.

The skewness and kurtosis values for maladaptive perfectionism, hopelessness, and self‐concept clarity were within the commonly accepted range for normality assumptions, with skewness values below |3| and kurtosis values below |10| (Kline 2023). In contrast, NSSI scores at T1 and T3 were markedly positively skewed. Specifically, the original T1 NSSI scores showed a skewness of 3.15 and a kurtosis of 12.19, and the original T3 NSSI scores showed a skewness of 3.68 and a kurtosis of 15.49. Consistent with prior research (e.g., Gu et al. 2023), a natural logarithmic transformation was applied to the NSSI scores to reduce skewness and improve distributional properties. Because the NSSI variables included zero values, the transformation was calculated as ln(x + 1), where x represents the original NSSI score. Adding 1 allowed participants reporting no NSSI to be retained in the analyses, while the log transformation reduced positive skewness and the influence of extreme values. The transformed T1 and T3 NSSI scores showed improved distributional characteristics, with T1 skewness = 1.99 and kurtosis = 3.53, and T3 skewness = 2.56 and kurtosis = 6.22. These transformed NSSI scores were used in all subsequent inferential analyses, whereas descriptive prevalence and method‐frequency statistics were reported using the original untransformed NSSI responses.

3.2. Testing for the Mediation Effect of Hopelessness

Mplus was used to test the mediating role of hopelessness in the association between maladaptive perfectionism and NSSI. The mediation model demonstrated acceptable fit to the data, χ 2 = 35.14, df = 8, CFI = 0.95, TLI = 0.94, RMSEA = 0.066, SRMR = 0.043. The standardized path coefficients are presented in Figure 2. After controlling for gender, age, and T1 hopelessness, T1 maladaptive perfectionism positively predicted T2 hopelessness (β = 0.18, p < 0.001). After controlling for gender, age, and T1 NSSI, T2 hopelessness positively predicted T3 NSSI (β = 0.21, p < 0.001), whereas T1 maladaptive perfectionism did not significantly predict T3 NSSI (β = −0.02, p > 0.05). Regarding the covariates, age was not significantly associated with T2 hopelessness (β = 0.05, p > 0.05) or T3 NSSI (β = −0.03, p > 0.05). Gender was significantly associated with both T2 hopelessness (β = 0.08, p < 0.05) and T3 NSSI (β = 0.12, p < 0.001).

Figure 2.

Figure 2

Open in a new tab

Standardized path coefficients for the mediation model linking maladaptive perfectionism to NSSI through hopelessness. Note. Dashed lines represent non‐significant paths. For simplicity, the effects of gender and age on T2 hopelessness and T3 NSSI are not displayed. n = 786. ***p < 0.001.

Bootstrap analyses indicated that the standardized indirect effect of T1 maladaptive perfectionism on T3 NSSI via T2 hopelessness was significant, β = 0.036, 95% bootstrap CI [0.018, 0.062]. Overall, the model explained 29.0% of the variance in T3 NSSI.

3.3. Testing for Moderated Mediation Effect

Mplus was used to test whether self‐concept clarity moderated the mediating role of hopelessness. The moderated mediation model showed good fit to the data, χ 2 = 22.53, df = 8, CFI = 0.98, TLI = 0.96, RMSEA = 0.048, SRMR = 0.022. The standardized path coefficients are presented in Figure 3. After controlling for gender, age, and T1 NSSI, T2 hopelessness positively predicted T3 NSSI (β = 0.17, p < 0.01), whereas the interaction between T2 hopelessness and T2 self‐concept clarity negatively predicted T3 NSSI (β = −0.10, p < 0.05). The moderated mediation model explained 31.4% of the variance in T3 NSSI.

Figure 3.

Figure 3

Open in a new tab

Standardized path coefficients for the moderated mediation model of maladaptive perfectionism, hopelessness, self‐concept clarity, and NSSI. Dashed lines represent nonsignificant paths. For simplicity, the effects of gender and age on T2 hopelessness and T3 NSSI are not displayed. n = 786. *p < 0.05, **p < 0.01, ***p < 0.001.

To more clearly illustrate the interaction between self‐concept clarity and hopelessness in relation to NSSI, self‐concept clarity was divided into high and low groups based on one standard deviation above and below the mean, and a plot of the moderating effect was generated (Figure 4). Simple slope analyses showed that for adolescents with low self‐concept clarity (M − SD), the effect of hopelessness on NSSI was significantly positive (b simple = 0.28, p < 0.001), whereas for adolescents with high self‐concept clarity (M + SD), the effect of hopelessness on NSSI was not significant (b simple = 0.08, p > 0.05).

Figure 4.

Figure 4

Open in a new tab

The moderating effect of self‐concept clarity on the relationship between hopelessness and NSSI. Hopelessness, self‐concept clarity, and NSSI were standardized scores.

Finally, the conditional indirect effects (Hayes 2013) at different levels of self‐concept clarity were estimated. When self‐concept clarity was low (M − SD), the indirect effect of maladaptive perfectionism on NSSI via hopelessness was 0.048 (SE = 0.014, 95% CI [0.025, 0.083]). When self‐concept clarity was high (M + SD), the corresponding indirect effect was 0.014 (SE = 0.012, 95% CI [−0.005, 0.044]). The index of moderated mediation was −0.017 (SE = 0.007, 95% CI [−0.035, −0.005]), which was statistically significant. These results indicate that the mediating effect of hopelessness decreases as self‐concept clarity increases; that is, self‐concept clarity buffers the indirect impact of maladaptive perfectionism on NSSI.

3.4. Sensitivity Analyses

To examine the robustness of the primary findings, we conducted sensitivity analyses using two alternative specifications of the NSSI outcome: dichotomous NSSI status and untransformed NSSI frequency. The pattern of results was generally consistent with the primary model using log‐transformed NSSI. In the dichotomous‐NSSI model, baseline maladaptive perfectionism was positively associated with T2 hopelessness (β = 0.16, p < 0.01), and T2 hopelessness was positively associated with T3 NSSI (β = 0.23, p < 0.01). The hopelessness × self‐concept clarity interaction was also in the expected direction and was significant (β = −0.10, p < 0.05).

In the untransformed‐NSSI model, baseline maladaptive perfectionism remained positively associated with T2 hopelessness (β = 0.18, p < 0.001), and T2 hopelessness remained positively associated with T3 NSSI (β = 0.17, p < 0.001). The hopelessness × self‐concept clarity interaction was again in the same direction and was significant (β = −0.10, p < 0.05). Overall, these sensitivity analyses suggest that the main conclusions were not dependent solely on the log‐transformed NSSI specification, although effect sizes and significance levels varied somewhat across outcome operationalizations.

4. Discussion

NSSI has become a major public health concern among adolescents. Consistent with the conceptual rationale presented in the Introduction, we interpret the findings within the Benefits and Barriers Model of NSSI rather than introducing separate theoretical accounts. From this perspective, our results showed that baseline maladaptive perfectionism was indirectly associated with NSSI at the 12‐month follow‐up through hopelessness at the 6‐month follow‐up. Specifically, higher levels of maladaptive perfectionism were associated with greater hopelessness at the 6‐month follow‐up, which in turn was associated with higher levels of subsequent NSSI. In addition, the association between hopelessness assessed at the 6‐month follow‐up and NSSI assessed at the 12‐month follow‐up varied as a function of self‐concept clarity assessed at the same 6‐month follow‐up. Thus, the strength of the hopelessness–NSSI association differed across levels of self‐concept clarity.

Importantly, however, this moderation finding should be interpreted as evidence of a conditional prospective association rather than as evidence that self‐concept clarity causally protects against or reduces the impact of hopelessness on subsequent NSSI. Several evidentiary features required for such an interpretation were not present in the current study. First, self‐concept clarity was not assessed prior to hopelessness, and therefore temporal precedence of the moderator cannot be established. Second, self‐concept clarity was measured at only one time point, which precludes examination of whether changes in self‐concept clarity prospectively alter risk trajectories. Third, because hopelessness and self‐concept clarity were assessed at the same occasion and using similar methods, shared method variance, concurrent psychological state, or unmeasured third variables may have contributed to the observed interaction. Accordingly, this moderation effect is best understood as identifying a subgroup pattern in which the association between hopelessness and later NSSI differed by level of self‐concept clarity, rather than as demonstrating self‐concept clarity as a general protective mechanism.

The observed indirect association is consistent with previous findings indicating that individuals with maladaptive perfectionism tend to approach life with an intense fear of failure, which is related to heightened hopelessness and depressive symptoms (Gnilka et al. 2013), and that hopelessness is associated with NSSI and has been shown to predict the frequency of such behaviors (Dhingra et al. 2016). From the perspective of the Benefits and Barriers Model of NSSI (Hooley and Franklin 2018), maladaptive perfectionism may be associated with pessimistic self‐perceptions and negative self‐associations. Such negative self‐related cognitions may intensify aversive affective states, such as hopelessness, and weaken positive self‐associations that typically serve as psychological barriers to self‐injury. In this context, NSSI may serve as a means of escaping aversive emotional states and expressing inexpressible thoughts or intolerable psychological pain through physical injury (Motz 2010).

With respect to self‐concept clarity, adolescents with higher self‐concept clarity may possess a clearer and more coherent understanding of themselves, ruminate less when experiencing negative emotions such as hopelessness, and may be less likely to become immersed in negative thoughts (Willis and Burnett Jr. 2016). These characteristics may offer a tentative explanation for why hopelessness was less strongly associated with later NSSI among adolescents with higher self‐concept clarity.

The effect sizes should be considered when interpreting the clinical importance of the findings. The mediation and moderated mediation models explained 29.0% and 31.4% of the variance in NSSI at the 12‐month follow‐up, respectively. Although this suggests that the models accounted for a nontrivial proportion of variance in later NSSI, several individual path coefficients were modest in magnitude and should not be interpreted as large clinical effects. Nevertheless, these effects may still have clinical relevance because NSSI is a clinically significant adolescent risk behavior, and even modest associations may be meaningful when they involve potentially modifiable psychological factors. The findings suggest that maladaptive perfectionism, hopelessness, and self‐concept clarity may contribute to cumulative risk assessment, particularly for identifying adolescents with elevated hopelessness and lower self‐concept clarity. However, these variables should not be viewed as sufficient standalone predictors of NSSI, and clinical assessment should also consider broader individual, family, peer, and contextual risk factors.

There are several limitations of this study that should be acknowledged and addressed in future research. First, all variables were measured via self‐report. Given the stigma surrounding NSSI (Ross and Heath 2002), self‐report data may be affected by issues such as social desirability and underreporting. Future studies should therefore incorporate multiple informants, including parents, teachers, and peers, and combine these reports with behavioral paradigms (e.g., the Cold Pressor Test and the Implicit Association Test; Ammerman et al. 2018) and ecological momentary assessments (Scala et al. 2018) to obtain more objective and ecologically valid data.

Second, the sample was drawn from a single middle school in one region of China, which limits the generalizability of the findings. Cultural and contextual factors may shape the associations observed in this study. For example, academic expectations, family socialization practices, attitudes toward failure, stigma surrounding mental health and self‐injury, and access to psychological support may influence the development of maladaptive perfectionism, the experience of hopelessness, and the likelihood that adolescents disclose or engage in NSSI (Chen et al. 2021). Therefore, the present findings should not be assumed to generalize to adolescents from other regions, school systems, socioeconomic backgrounds, or cultural contexts. Future studies should replicate the model in more diverse samples, including adolescents from different regions of China, other cultural contexts, rural and urban settings, and clinical or high‐risk populations.

Third, a measurement‐related limitation should also be noted. Although one hopelessness item was removed because of its low factor loading and the resulting scale showed acceptable psychometric properties, this decision was informed by the present data rather than specified a priori. Therefore, findings involving hopelessness could, in principle, be sensitive to this data‐driven measurement choice. Future studies should examine whether the present results replicate when hopelessness is assessed using the full original scale, preregistered scoring procedures, or alternative validated measures of hopelessness. Fourth, although the present study modeled NSSI as a continuous outcome using log‐transformed scores and conducted sensitivity analyses using dichotomous and untransformed NSSI outcomes, this represents one of several possible analytic approaches. Given the skewed distribution and substantial concentration of zero values in NSSI, future research could consider models specifically tailored for zero‐inflated or count‐like outcomes, such as zero‐inflated, hurdle, Poisson, or negative binomial models. Such approaches may provide additional insight into both the occurrence and frequency of NSSI.

Fifth, the present findings should be interpreted in light of differences in how change was operationalized across study variables. Hopelessness and NSSI were assessed at multiple waves, allowing later levels of these variables to be modeled while accounting for baseline levels. Specifically, hopelessness at the 6‐month follow‐up was modeled while controlling for baseline hopelessness, and NSSI at the 12‐month follow‐up was modeled while controlling for baseline NSSI. Thus, the indirect effect involved residualized change components based on overlapping time intervals. Accordingly, the pathway should not be interpreted as definitive evidence of a strictly temporally ordered mediation process. Rather, it should be understood as an indirect association involving residualized change, which may partly reflect concurrent change in hopelessness and NSSI across the overlapping interval. In contrast, maladaptive perfectionism was assessed only at baseline, and self‐concept clarity was assessed only at the 6‐month follow‐up. Therefore, the present study could not examine changes in maladaptive perfectionism or self‐concept clarity, nor could it test whether self‐concept clarity assessed at an earlier wave would show the same moderating pattern. Because self‐concept clarity and hopelessness were both assessed at the 6‐month follow‐up, the moderation effect should not be interpreted as evidence of a general, stable, or causal moderating role of self‐concept clarity across time. Rather, the finding should be understood as a time‐specific conditional prospective association: the association between hopelessness at the 6‐month follow‐up and NSSI at the 12‐month follow‐up varied as a function of adolescents' self‐concept clarity measured at the same wave as hopelessness. Future studies should assess all key constructs at multiple waves to test these temporal and alternative explanatory accounts more rigorously.

Finally, it is important to note that the statistical model used in this study does not distinguish between‐person and within‐person effects. Therefore, the observed associations may reflect a combination of stable individual differences and within‐person changes over time. As such, the findings should be interpreted as reflecting general longitudinal associations rather than purely within‐person causal effects. Moreover, although our hypotheses were supported, only unidirectional paths among the variables were examined. For example, a longitudinal study of 785 Spanish adolescents reported that NSSI can significantly increase hopelessness 1 year later (Faura‐Garcia et al. 2024). Future research could employ random‐intercept cross‐lagged panel models to test potential reciprocal relationships between key variables (e.g., hopelessness and NSSI) and to disentangle between‐person and within‐person effects more effectively.

The findings of this study may have implications for adolescent NSSI prevention and intervention efforts. Given the observed associations among maladaptive perfectionism, hopelessness, and later NSSI, parents, teachers, and other educators may support adolescents by helping them set appropriate goals, develop more adaptive cognitive beliefs, and adopt healthier ways of interpreting and coping with failure. For example, educators and caregivers could prioritize process‐oriented praise that recognizes adolescents' effort, strategies, and progress rather than innate ability or perfect outcomes (Haimovitz and Dweck 2017). When adolescents experience setbacks, educators and caregivers could also model and teach attributional retraining by framing failures as temporary, situation‐specific, and modifiable through effort and strategy, rather than as evidence of fixed or inherent inadequacies (Haynes Stewart et al. 2011). Such practices may help reduce maladaptive perfectionistic self‐evaluations and hopelessness following perceived failure, while supporting more adaptive coping among adolescents who may be at elevated risk for NSSI.

At the same time, these prevention and intervention implications should be interpreted cautiously. Although previous research has documented significant associations between maladaptive perfectionism and NSSI across both clinical and nonclinical populations (Gyori and Balazs 2021), the present findings should not be assumed to generalize equally to youth with clinically significant NSSI or related clinical conditions. The current sample was community‐based and was not selected on the basis of clinical diagnosis, treatment‐seeking status, NSSI severity, or functional impairment. Thus, the findings may be relevant to clinical populations, but replication is needed among adolescents with more severe, recurrent, or clinically impairing NSSI.

Our results also indicate that higher self‐concept clarity was associated with a weaker positive association between hopelessness and later NSSI. This pattern suggests that self‐concept clarity may be relevant to future prevention and intervention research, particularly as a potential marker of variability in the hopelessness–NSSI association. Previous research has shown that mindfulness may help individuals revise their self‐concept, enhance context‐specific self‐concept clarity (Hanley and Garland 2017), and gain a clearer understanding of their own reactions and motivations (Dummel 2018). A clinical study further found that dialectical behavior therapy, which includes mindfulness training, was associated with significant improvements in self‐concept clarity among patients with borderline personality disorder (Roepke et al. 2011). On this basis, mindfulness‐based and related interventions may be worth exploring as possible approaches to supporting adolescents' self‐concept clarity. However, whether such approaches would reduce hopelessness or NSSI‐related risk among adolescents remains an important question for future research.

Although the present findings suggest that self‐concept clarity may have relevance for identifying adolescents for whom hopelessness is more strongly associated with subsequent NSSI, they do not establish self‐concept clarity as an intervention target. To more firmly support such a conclusion, future research would need to provide evidence that within‐person changes in self‐concept clarity temporally precede changes in NSSI risk, for example, by showing that increases in self‐concept clarity are followed by reductions in hopelessness or subsequent self‐injury. In addition, experimental or intervention studies are needed to determine whether programs designed to enhance self‐concept clarity produce downstream effects on hopelessness, emotion regulation, or NSSI outcomes. Such evidence would help clarify whether self‐concept clarity functions merely as a marker of risk heterogeneity or as a modifiable mechanism that can be targeted in prevention and intervention efforts.

This research was funded by grants from the Natural Science Foundation of Hunan Province (2025JJ50159) and Guangdong Provincial University Ideological and Political Education Project (2025GXSZ023).

Ethics Statement

The experimental protocol was approved by the Academic Ethics Committee of the institution of the first author.

Consent

All participants signed written informed consent forms.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgments

During the preparation of this manuscript, the authors used AI‐assisted tools to support language editing and improve clarity. The AI tools were not used to generate research hypotheses, conduct statistical analyses, create or modify data, interpret results independently, or determine the conclusions of the study. All AI‐assisted output was carefully reviewed, edited, and verified by the authors for accuracy, coherence, factual correctness, and consistency with the study data and analyses. The authors take full responsibility for the content of the final manuscript.

Data Availability Statement

Data available on request from the authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. American Psychiatric Association. 2013. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Association.
  2. Ammerman, B. A. , Berman M. E., and McCloskey M. S.. 2018. “Assessing Non‐Suicidal Self‐Injury in the Laboratory.” Archives of Suicide Research 22, no. 2: 193–223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bandura, A. 2001. “Social Cognitive Theory: An Agentic Perspective.” Annual Review of Psychology 52, no. 1: 1–26. [DOI] [PubMed] [Google Scholar]
  4. Beck, A. T. , Weissman A., Lester D., and Trexler L.. 1974. “The Measurement of Pessimism: The Hopelessness Scale.” Journal of Consulting and Clinical Psychology 42, no. 6: 861–865. [DOI] [PubMed] [Google Scholar]
  5. Campbell, J. D. , Trapnell P. D., Heine S. J., Katz I. M., Lavallee L. F., and Lehman D. R.. 1996. “Self‐Concept Clarity: Measurement, Personality Correlates, and Cultural Boundaries.” Journal of Personality and Social Psychology 70, no. 1: 141–156. [Google Scholar]
  6. Chen, X. , Zhou Y., Li L., et al. 2021. “Influential Factors of Non‐Suicidal Self‐Injury in an Eastern Cultural Context: A Qualitative Study From the Perspective of School Mental Health Professionals.” Frontiers in Psychiatry 12: 681985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cheng, Y. , Han Y., and Gu H.. 2026. “Longitudinal Links Between Self‐Critical Perfectionism and Adolescent Non‐Suicidal Self‐Injury: The Roles of Basic Psychological Need Frustration and Intentional Self‐Regulation.” Journal of Youth and Adolescence 55, no. 3: 596–607. [DOI] [PubMed] [Google Scholar]
  8. Chester, D. S. , Merwin L. M., and DeWall C. N.. 2015. “Maladaptive Perfectionism's Link to Aggression and Self‐Harm: Emotion Regulation as a Mechanism.” Aggressive Behavior 41, no. 5: 443–454. [DOI] [PubMed] [Google Scholar]
  9. Claes, L. , Soenens B., Vansteenkiste M., and Vandereycken W.. 2012. “The Scars of the Inner Critic: Perfectionism and Nonsuicidal Self‐Injury in Eating Disorders.” European Eating Disorders Review 20, no. 3: 196–202. [DOI] [PubMed] [Google Scholar]
  10. Dhingra, K. , Boduszek D., and Klonsky E. D.. 2016. “Empirically Derived Subgroups of Self‐Injurious Thoughts and Behavior: Application of Latent Class Analysis.” Suicide and Life‐Threatening Behavior 46, no. 4: 486–499. [DOI] [PubMed] [Google Scholar]
  11. Dummel, S. 2018. “Relating Mindfulness to Attitudinal Ambivalence Through Self‐Concept Clarity.” Mindfulness 9, no. 5: 1486–1493. [Google Scholar]
  12. Faura‐Garcia, J. , Calvete E., and Orue I.. 2024. “Longitudinal Associations Between Nonsuicidal Self‐Injury, Depressive Symptoms, Hopelessness, and Emotional Dysregulation in Adolescents.” Archives of Suicide Research 28, no. 3: 800–814. [DOI] [PubMed] [Google Scholar]
  13. Ferrari, M. , Yap K., Scott N., Einstein D. A., and Ciarrochi J.. 2018. “Self‐Compassion Moderates the Perfectionism and Depression Link in Both Adolescence and Adulthood.” PLoS One 13, no. 2: e0192022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fox, K. R. , Franklin J. C., Ribeiro J. D., Kleiman E. M., Bentley K. H., and Nock M. K.. 2015. “Meta‐Analysis of Risk Factors for Nonsuicidal Self‐Injury.” Clinical Psychology Review 42: 156–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Frost, R. O. , Marten P., Lahart C., and Rosenblate R.. 1990. “The Dimensions of Perfectionism.” Cognitive Therapy and Research 14: 449–468. [Google Scholar]
  16. Fujimori, A. , Wada Y., Yamashita T., et al. 2011. “Parental Bonding in Patients With Eating Disorders and Self‐Injurious Behavior.” Psychiatry and Clinical Neurosciences 65, no. 3: 272–279. [DOI] [PubMed] [Google Scholar]
  17. Gillies, D. , Christou M. A., Dixon A. C., et al. 2018. “Prevalence and Characteristics of Self‐Harm in Adolescents: Meta‐Analyses of Community‐Based Studies 1990‐2015.” Journal of the American Academy of Child and Adolescent Psychiatry 57, no. 10: 733–741. [DOI] [PubMed] [Google Scholar]
  18. Gnilka, P. B. , Ashby J. S., and Noble C. M.. 2013. “Adaptive and Maladaptive Perfectionism as Mediators of Adult Attachment Styles and Depression, Hopelessness, and Life Satisfaction.” Journal of Counseling & Development 91, no. 1: 78–86. [Google Scholar]
  19. Goldberg, S. , and Israelashvili M.. 2017. “The Distinction Between Exclusivity and Comorbidity Within NSSI.” School Psychology Quarterly 32, no. 2: 156–172. [DOI] [PubMed] [Google Scholar]
  20. Gray, N. S. , Knowles J., George D., et al. 2021. “Explicit and Implicit Hopelessness and Self‐Injury.” Suicide and Life‐Threatening Behavior 51, no. 3: 606–615. [DOI] [PubMed] [Google Scholar]
  21. Gu, H. , Chen W., and Cheng Y.. 2024. “Longitudinal Relationship Between Harsh Parenting and Adolescent Non‐Suicidal Self‐Injury: The Roles of Basic Psychological Needs Frustration and Self‐Concept Clarity.” Child Abuse & Neglect 149: 106697. [DOI] [PubMed] [Google Scholar]
  22. Gu, H. , Hu C., and Wang L.. 2022. “Maladaptive Perfectionism and Adolescent NSSI: A Moderated Mediation Model of Psychological Distress and Mindfulness.” Journal of Clinical Psychology 78, no. 6: 1137–1150. [DOI] [PubMed] [Google Scholar]
  23. Gu, H. , Lu Y., and Cheng Y.. 2021. “Negative Life Events and Nonsuicidal Self‐Injury in Prisoners: The Mediating Role of Hopelessness and Moderating Role of Belief in a Just World.” Journal of Clinical Psychology 77, no. 1: 145–155. [DOI] [PubMed] [Google Scholar]
  24. Gu, H. , Yao Y., He F., and Cheng Y.. 2023. “Harsh Parenting and Adolescent Non‐Suicidal Selfinjury: A Moderated Mediation Model of Alienation and Cognitive Reappraisal.” Child Abuse & Neglect 141: 106188. [DOI] [PubMed] [Google Scholar]
  25. Gyori, D. , and Balazs J.. 2021. “Nonsuicidal Self‐Injury and Perfectionism: A Systematic Review.” Frontiers in Psychiatry 12: 691147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Haimovitz, K. , and Dweck C. S.. 2017. “The Origins of Children's Growth and Fixed Mindsets: New Research and a New Proposal.” Child Development 88, no. 6: 1849–1859. [DOI] [PubMed] [Google Scholar]
  27. Hanley, A. W. , and Garland E. L.. 2017. “Clarity of Mind: Structural Equation Modeling of Associations Between Dispositional Mindfulness, Self‐Concept Clarity and Psychological Well‐Being.” Personality and Individual Differences 106: 334–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Hayes, A. F. 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression‐Based Approach. Guilford Press.
  29. Haynes Stewart, T. L. , Clifton R. A., Daniels L. M., Perry R. P., Chipperfield J. G., and Ruthig J. C.. 2011. “Attributional Retraining: Reducing the Likelihood of Failure.” Social Psychology of Education 14, no. 1: 75–92. [Google Scholar]
  30. Hoff, E. R. , and Muehlenkamp J. J.. 2009. “Nonsuicidal Self‐Injury in College Students: The Role of Perfectionism and Rumination.” Suicide and Life‐Threatening Behavior 39, no. 6: 576–587. [DOI] [PubMed] [Google Scholar]
  31. Hooley, J. M. , and Franklin J. C.. 2018. “Why Do People Hurt Themselves? A New Conceptual Model of Non‐Suicidal Self‐Injury.” Clinical Psychological Science 6, no. 3: 428–451. [Google Scholar]
  32. Kiekens, G. , Hasking P., Boyes M., et al. 2018. “The Associations Between Non‐Suicidal Self‐Injury and First Onset Suicidal Thoughts and Behaviors.” Journal of Affective Disorders 239: 171–179. [DOI] [PubMed] [Google Scholar]
  33. Kline, R. B. 2023. Principles and Practice of Structural Equation Modeling. 5th ed. Guilford Press.
  34. Lang, J. , and Yao Y.. 2018. “Prevalence of Nonsuicidal Self‐Injury in Chinese Middle School and High School Students.” Medicine 97, no. 42: e12916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lear, M. K. , and Pepper C. M.. 2016. “Self‐Concept Clarity and Emotion Dysregulation in Nonsuicidal Self‐Injury.” Journal of Personality Disorders 30, no. 6: 813–827. [DOI] [PubMed] [Google Scholar]
  36. Lester, D. 2015. “Hopelessness in Adolescents.” Journal of Affective Disorders 173: 221–225. [DOI] [PubMed] [Google Scholar]
  37. Luyckx, K. , Gandhi A., Bijttebier P., and Claes L.. 2015. “Non‐Suicidal Self‐Injury in Female Adolescents and Psychiatric Patients: A Replication and Extension of the Role of Identity Formation.” Personality and Individual Differences 77: 91–96. [Google Scholar]
  38. Mannekote Thippaiah, S. , Shankarapura Nanjappa M., Gude J. G., et al. 2020. “Non‐Suicidal Self‐Injury in Developing Countries: A Review.” International Journal of Social Psychiatry 67, no. 5: 472–482. [DOI] [PubMed] [Google Scholar]
  39. Motz, A. 2010. “Self‐Harm as a Sign of Hope.” Psychoanalytic Psychotherapy 24, no. 2: 81–92. [Google Scholar]
  40. Muthén, L. K. , and Muthén B. O.. 2019. Mplus User's Guide. 8th ed. Muthén & Muthén.
  41. Niu, G. , Sun X., Zhou Z., Tian Y., Liu Q., and Lian S.. 2016. “The Impact of Teenagers' Use of Social Network Sites on Self‐Concept Clarity: The Mediation Effect of Social Comparison.” Psychological Science 39, no. 1: 97–102. [Google Scholar]
  42. Nock, M. K. , and Favazza A. R.. 2009. “Nonsuicidal Self‐Injury: Definition and Classification,” In Understanding Nonsuicidal Self‐Injury: Origins, Assessment, and Treatment, edited by M. K. Nock, 9–18. American Psychological Association.
  43. Perczel‐Forintos, D. , Rózsa S., Pilling J., and Kopp M.. 2013. “Proposal for a Short Version of the Beck Hopelessness Scale Based on a National Representative Survey in Hungary.” Community Mental Health Journal 49: 822–830. [DOI] [PubMed] [Google Scholar]
  44. Rice, K. G. , Leever B. A., Christopher J., and Porter J. D.. 2006. “Perfectionism, Stress, and Social (Dis)connection: A Short‐Term Study of Hopelessness, Depression, and Academic Adjustment Among Honors Students.” Journal of Counseling Psychology 53, no. 4: 524–534. [Google Scholar]
  45. Roepke, S. , Schröder‐Abé M., Schütz A., et al. 2011. “Dialectic Behavioural Therapy Has an Impact on Self‐Concept Clarity and Facets of Self‐Esteem in Women With Borderline Personality Disorder.” Clinical Psychology & Psychotherapy 18, no. 2: 148–158. [DOI] [PubMed] [Google Scholar]
  46. Ross, S. , and Heath N.. 2002. “A Study of the Frequency of Self‐Mutilation in a Community Sample of Adolescents.” Journal of Youth and Adolescence 31, no. 1: 67–77. [Google Scholar]
  47. Scala, J. W. , Levy K. N., Johnson B. N., et al. 2018. “The Role of Negative Affect and Self‐Concept Clarity in Predicting Self‐Injurious Urges in Borderline Personality Disorder Using Ecological Momentary Assessment.” Journal of Personality Disorders 32, no. Suppl: 36–57. [DOI] [PubMed] [Google Scholar]
  48. Schatten, H. T. , Andover M. S., and Armey M. F.. 2015. “The Roles of Social Stress and Decision‐Making in Non‐Suicidal Self‐Injury.” Psychiatry Research 229, no. 3: 983–991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Slade, P. D. , and Owens R. G.. 1998. “A Dual Process Model of Perfectionism Based on Reinforcement Theory.” Behavior Modification 22, no. 3: 372–390. [DOI] [PubMed] [Google Scholar]
  50. Tang, J. , Li G., Chen B., et al. 2018. “Prevalence of and Risk Factors for Non‐Suicidal Self‐Injury in Rural China: Results From a Nationwide Survey in China.” Journal of Affective Disorders 226: 188–195. [DOI] [PubMed] [Google Scholar]
  51. Taylor, P. J. , Jomar K., Dhingra K., Forrester R., Shahmalak U., and Dickson J. M.. 2018. “A Meta‐Analysis of the Prevalence of Different Functions of Non‐Suicidal Self‐Injury.” Journal of Affective Disorders 227: 759–769. [DOI] [PubMed] [Google Scholar]
  52. Willis, K. D. , and Burnett H. J. Jr. 2016. “The Power of Stress: Perceived Stress and Its Relationship With Rumination, Self‐Concept Clarity, and Resilience.” North American Journal of Psychology 18, no. 3: 483–498. [Google Scholar]
  53. Wu, J. , Watkins D., and Hattie J.. 2010. “Self‐Concept Clarity: A Longitudinal Study of Hong Kong Adolescents.” Personality and Individual Differences 48, no. 3: 277–282. [Google Scholar]
  54. Wu, Y. , Wu L., Niu G., Chen Z., and Liu L.. 2020. “The Effect of WeChat Moments Uses on College Students' Depression: The Effect of Negative Social Comparison and Self‐Concept Clarity.” Psychological Development and Education 36, no. 4: 486–493. [Google Scholar]
  55. You, J. , Lin M. P., Fu K., and Leung F.. 2013. “The Best Friend and Friendship Group Influence on Adolescent Nonsuicidal Self‐Injury.” Journal of Abnormal Child Psychology 41, no. 6: 993–1004. [DOI] [PubMed] [Google Scholar]
  56. Zi, F. , and Zhou X.. 2006. “Test on Reliability and Validity of Chinese Frost Multidimensional Perfectionism Questionnaire.” Chinese Journal of Clinical Psychology 14, no. 6: 560–563. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data available on request from the authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Journal of Clinical Psychology are provided here courtesy of Wiley

RESOURCES