Next Article in Journal
Mental Health in Europe After COVID-19: A Systematic Review of Depression, Anxiety, and Stress Among Adult Primary Health Care Users
Previous Article in Journal
An Altered Gut Microbiota–Brain Axis in Fragile X Syndrome May Explain Autistic Traits in Some Patients
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Psychiatry International
Volume 6
Issue 3
10.3390/psychiatryint6030108
psychiatryint-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Validity and Reliability of the Chinese Version of the Screening Instrument for Borderline Personality Disorder

by
Hui Zhou
1,
Yu Chang
1,
Chaiyun Sakulsriprasert
1,2,*,
Tinakon Wongpakaran
1,3,
Nahathai Wongpakaran
1,3,*,
Chawisa Suradom
1,3,
Ronald O’Donnell
1,4 and
Nan Jia
5
1
Mental Health Program, Multidisciplinary and Interdisciplinary School, Chiang Mai University, Chiang Mai 50200, Thailand
2
Department of Psychology, Faculty of Humanities, Chiang Mai University, Chiang Mai 50200, Thailand
3
Department of Psychiatry, Faculty of Medicine, Chiang Mai University, Chiang Mai 50200, Thailand
4
College of Health Solutions, Arizona State University, 500 N. 3rd St., Phoenix, AZ 85004, USA
5
Yew Chung International School of Beijing, Honglingjin Park, No. 5 Houbalizhuang, Chaoyang District, Beijing 100025, China
*
Authors to whom correspondence should be addressed.
Psychiatry Int. 2025, 6(3), 108; https://doi.org/10.3390/psychiatryint6030108
Submission received: 23 June 2025 / Revised: 4 August 2025 / Accepted: 2 September 2025 / Published: 5 September 2025
Versions Notes

Abstract

Background: Borderline personality disorder (BPD), a significant personality trait frequently observed in young adults, is associated with challenges in mental health and academic performance. Screening for BPD symptoms is essential. The Screening Instrument for Borderline Personality Disorder (SI-Bord) is widely used to assess general BPD symptoms. However, despite being translated and culturally adapted, the psychometric properties of the Chinese version of the SI-Bord have not been thoroughly investigated in a Chinese population. Objectives: The aim of the study was to evaluate the psychometric properties of the Chinese version of the Screening Instrument for Borderline Personality Disorder (SI-Bord) among university students using confirmatory factor analysis (CFA). Methods: Participants completed the SI-Bord along with the Perceived Stress Scale (PSS), the Meaning in Life Questionnaire (MLQ), the Experiences in Close Relationships–Revised (ECR-R), and the Rosenberg Self-Esteem Scale (RSES). Results: A total of 715 Chinese university students (mean age = 20.33 years; age range = 18–25), including 385 males (54.2%) and 325 females (45.5%), participated in this study. The unidimensional model demonstrated adequate fit indices. The SI-Bord showed significant correlations with the PSS and ECR-R (attachment anxiety), alongside smaller correlations with the MLQ, supporting its convergent and discriminant validity. The Chinese version of the SI-Bord exhibited good reliability. Invariance testing confirmed at least metric invariance across various groups. Conclusions: The Chinese version of the SI-Bord demonstrates strong validity and reliability as a tool for screening for core BPD symptoms among Chinese university students. Further studies are encouraged to evaluate the validity of the SI-Bord across diverse groups, including age, socioeconomic status, and geographic regions. Applying it in clinical BPD samples will further enhance its utility across Chinese populations.

1. Introduction

Borderline personality disorder (BPD) is a multifaceted and severe mental health condition. According to Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), it is characterized by pervasive instability in mood, self-image, interpersonal relationships, and impulse control [1]. Individuals diagnosed with BPD frequently experience intense episodes of emotional dysregulation, chronic feelings of emptiness, impulsivity, recurrent self-harming behaviors, and profound fears of abandonment.
Attachment theory explains how early caregiver relationships shape social functioning throughout life [2,3]. Insecure attachment, particularly attachment anxiety and avoidance, is significantly associated with BPD characteristics; meta-analytic evidence shows attachment anxiety is most strongly correlated with BPD, while attachment avoidance also contributes [4,5] These two dimensions translate into distinct adult attachment styles: preoccupied (high anxiety, low avoidance), dismissing (low anxiety, high avoidance), and fearful (high anxiety, high avoidance). In addition, research highlights the role of disorganized attachment, characterized by a lack of coherent relational strategies and marked by both high anxiety and high avoidance, often accompanied by confusion and fear. Disorganized attachment—commonly observed in individuals with traumatic or inconsistent caregiving—has been strongly linked to the emotional instability and relational difficulties found in BPD [6].
The consequences of BPD extend beyond daily emotional suffering, often resulting in substantial impairment in academic, occupational, and social domains. Moreover, the elevated risk of self-harm, substance misuse, and suicide attempts among those with BPD underscores its public health significance. Epidemiological data reveal that individuals with BPD have a suicide rate nearly fifty times higher than the general population, with a mortality rate estimated at approximately 10% [1,7].
In university populations, the prevalence of BPD is a growing concern. Recent studies in China indicate that approximately 15% of undergraduate students screen positive for BPD using self-report instruments [3]. A broader survey of college freshmen found remarkable variation in prevalence estimates (ranging from 0.67% to 17.7%), likely due to differences in sample characteristics and screening methodologies [4]. These figures are alarming, not only because of the high prevalence but also because this period represents a critical developmental window for the emergence and detection of personality pathology. Early recognition and intervention for BPD in university students could play a pivotal role in mitigating its negative long-term outcomes.
Given the clinical and social burden of BPD, the necessity for reliable and efficient screening tools in both general and clinical populations remains urgent [5]. Various instruments are available for assessing BPD, each with its unique advantages and limitations. For instance, the Personality Diagnostic Questionnaire (PDQ-4+) is comprehensive but not BPD-specific and can be time-consuming [8]; the Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD), though clinically sensitive, requires trained raters [9]. Measures such as the Brief Borderline Symptom List (BSL-23), Borderline Personality Disorder Severity Index (BPDSI), Clinical Global Impression Scale for BPD (CGI-BPD), and Borderline Evaluation of Severity Over Time (BEST) primarily focus on symptom severity rather than direct screening [10,11,12]. These factors restrict the practicality of existing tools for rapid, large-scale screening, particularly among non-clinical populations such as university students.
To address these limitations, the Screening Instrument for Borderline Personality Disorder (SI-Bord) was developed as a concise and efficient self-report screening tool targeting core DSM-5 diagnostic criteria for BPD [13,14]. The SI-Bord builds upon its predecessor, the Short-Bord, and has demonstrated favorable psychometric properties in different cultural contexts. Validation studies in Thailand have reported strong reliability (e.g., Cronbach’s alpha = 0.80; area under the Receiver Operating Characteristic (ROC) curve (AUC) = 0.95 for the Short-Bord; AUC = 0.83 for the SI-Bord) and significant correlations with related constructs such as depression and perceived stress [13,14]. Likewise, positive results have been reported in Irish and Polish adaptations [15,16]. The item brevity and unidimensional structure of the SI-Bord facilitate its use as both a screener and a symptom measure, making it highly relevant for research and clinical application.
Despite the demonstrated efficacy of SI-Bord in other populations, there is a notable lack of data regarding its validity and reliability among Chinese university students, despite a recent Chinese translation becoming available [17]. The process of psychometric validation in new cultural contexts is not trivial. Tools initially developed in different sociocultural environments may not perform equivalently when translated and applied to new populations due to cultural differences in symptom expression, stigma, and language nuances. Ensuring the factorial structure, internal consistency, and measurement invariance of the SI-Bord within the Chinese context is thus a necessary step toward its responsible implementation.
One of the gold-standard approaches for establishing the adequacy of a new or adapted instrument is to evaluate its construct validity. Construct validity comprises two essential components: convergent validity and discriminant validity. Convergent validity is demonstrated when an instrument correlates highly with theoretically and empirically related constructs. For BPD, strong associations with constructs like attachment insecurity and perceived stress are expected, consistent with attachment theory and the established impact of psychosocial stressors on borderline pathology [2,18,19]. Discriminant validity, in contrast, is demonstrated by low correlations with unrelated or weakly related constructs, such as meaning in life or self-esteem [18,20]. This ensures that the instrument does not inadvertently measure other aspects of personality or well-being. The evaluation of both convergent and discriminant validity is a foundational requirement in modern psychometric research. It is crucial to validate the utility of the SI-Bord in new cultural contexts [17].
Furthermore, it is crucial to ensure that psychological screening instruments are valid and reliable across diverse demographic groups. Assessing measurement invariance (for example, across gender) is crucial to ensure that the SI-Bord provides unbiased assessments and enables meaningful comparisons between groups. Without such testing, the tool’s results could be misinterpreted or confounded by demographic factors.
This report was developed in response to the growing need for efficient and culturally appropriate BPD screening methods for Chinese university students and is based on data collected from a cross-sectional survey. Given the lack of comprehensive validation for the Chinese version of the SI-Bord, our study sought to fill this gap by thoroughly investigating its psychometric properties in this population.
In summary, we aimed to provide a comprehensive psychometric evaluation of the Chinese version of the SI-Bord among university students. Specifically, we assessed its factorial structure, internal consistency, convergent and discriminant validity (using validated measures of attachment, perceived stress, meaning in life, and self-esteem), and measurement invariance. By establishing the reliability and validity of the SI-Bord in a Chinese context, our findings would support the tool’s application for early identification and intervention for BPD features, contributing to both clinical and research advancements.

2. Materials and Methods

2.1. Study Design

This study was a secondary analysis of data from the project titled “Personality Disorder Symptoms and Their Clinical Correlates among Chinese University Students: A Cross-Sectional Study.” [11], which was conducted among Chinese university students from November 2021 to January 2022. The original research was approved by the Ethics Committee of the Faculty of Medicine, Chiang Mai University (Approval Code: No. 476/2021; Approval date: 4 November 2021). The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Faculty of Medicine, Chiang Mai University.

2.2. Participants

The participants were Chinese-speaking undergraduate students from universities located all around China. The original study employed a convenience sampling strategy to recruit Chinese undergraduate students aged 18–25 years through various social media platforms, including WeChat, QQ, TikTok, and the websites www.douban.com and www.sina.com. Inclusion criteria required participants to be university undergraduates within the specified age range, proficient in speaking, reading, and writing Chinese, and able to access an internet-connected electronic device (e.g., computer, smartphone, or tablet). Exclusion criteria comprised individuals diagnosed with schizophrenia, bipolar disorder, or substance dependency, as well as those unable to complete the online questionnaire via an internet-enabled device independently. Eligible participants completed a Microsoft Form questionnaire, which commenced with eligibility screening and informed consent. The survey link was deactivated once the target sample size was achieved. Informed consent had been obtained during the prior study, and this research received an ethics exemption. To maintain analytic rigor, cases with missing data or disqualifying conditions were systematically excluded. Of the 1237 individuals who initially participated, 715 met the inclusion criteria. Participants’ ages ranged from 18 to 25 years, with a mean age of 20.33 years; 54.2% of the final sample were male.

2.3. Measurements

2.3.1. Screening Instrument for Borderline Personality Disorder (SI-Bord)

The SI-Bord is a short self-report questionnaire. It consists of 5 questions representing the DSM-5 criteria of BPD: abandonment avoidance, interpersonal relationship instability, affective instability, suicidal and self-harm behaviors, and identity disruption [21]. The SI-Bord utilizes a 4-point Likert scale (0–3). The total score ranges from 0 to 15. Sample statements include, “When people with ties to me leave me, I can barely live.”, and “I threaten to hurt myself or attempt to hurt myself or have attempted suicide”. Higher scores reflect increased severity of symptoms. It was found that a cutoff score of more than seven would result in a sensitivity of 0.75 and a specificity of 0.73. The SI-Bord was translated, back-translated, and culturally modified into Chinese. A pilot test with 39 Chinese participants, aged 19 to 29 years, revealed that the SI-Bord Chinese version has a Cronbach’s alpha of 0.738 [11].

2.3.2. Perceived Stress Scale (PSS-10)

The PSS-10 is a popular self-report tool that assesses how stressful life is [22]. This updated version of the original PSS comprises a total of 10 items [23]. The measurement is conducted a 5-point Likert scale (0–4). A total score between 0 and 13 on the Perceived Stress Scale means that the stress level is low, 14 to 26 means that the stress level is average, and 27 to 40 means that the stress level is very high. A value of 0.91 was found for the Cronbach’s alpha of the Chinese adaptation of the PSS-10 [24].

2.3.3. Meaning in Life Questionnaire (MLQ)

The MLQ is the primary measurement for assessing the significance of life in specific demographics and culturally diverse societies [25]. The scale consists of 10 questions, divided into the Presence of Meaning and Search for Meaning subscales. This study utilized the Presence and Search subscales to evaluate an individual’s sense of meaning or purpose. Through the use of a 7-point Likert scale, the response ranges from 1, indicating “absolutely untrue,” to 7, indicating “absolutely true.” Each subscale yields a range of 5 to 35 when the components are added together. Elevated total scores indicate a heightened quest for purpose in one’s life [25]. The MLQ has a consistent factor structure, good test–retest dependability, and strong internal consistency. With a generalizability score of 0.86 and a dependability index of 0.85, the Chinese adaptation was very effective in measuring its intended constructs. Specifically, the generalizability coefficients for the Existence of Meaning and the Search for Meaning were 0.76 and 0.85, respectively [26].

2.3.4. Experience in Close Relationships–Revised (ECR-R)

The ECR-R scale is a commonly employed tool designed to evaluate two main aspects: attachment anxiety and attachment avoidance [27]. The updated edition of the original ECR consists of 18 items, each assessed using a 7-point Likert scale that spans from 1 (disagree) to 7 (strongly agree). The ECR-R18 has undergone validation and exhibits exceptional psychometric characteristics and reliability (Cronbach’s alpha = 0.90), satisfactory test–retest reliability, and validity [28]. The Chinese version of the ECR-R has sufficient validity and reliability [29].

2.3.5. Rosenberg Self-Esteem Scale (RSES)

This scale is the most widely employed self-esteem assessment tool for adult populations [30]. Five of the ten items on the scale are formulated in negative terms. It employs a 4-point Likert scale, ranging from 1 = strongly agree to 4 = strongly disagree. The scale ranges from 0 to 30. Total scores below 15 indicate low self-esteem, whereas scores between 15 and 25 are considered normal. Research reported a 1-week test–retest reliability of 0.82 and internal consistency: a coefficient alpha of 0.88 [31]. Among college students, the Chinese adaptation demonstrated a reliability coefficient of 0.8–0.89, while the test–retest reliability coefficient was 0.76 [32].

2.4. Statistical Analysis

Confirmatory factor analysis (CFA) was used to assess the SI-Bord’s factorial validity and identity as the best fit for the data. If the standardized root mean square residuals (SRMRs) [33] and the values for the root mean square error of approximation (RMSEA) [34] are relatively smaller and the comparative fit index (CFI) [35] and Tucker–Lewis Index (TLI) [36] exceed the 0.90 threshold, it is expected that the model fit will be more optimal. More precisely, it was anticipated that the SRMR would be less than 0.08, and the RMSEA would not surpass 0.08. Moreover, the evaluation of the quality of model fit would involve the use of the χ2 statistic.
A multi-group confirmatory factor analysis model was employed to examine the structural consistency of the SI-Bord across diverse demographic categories, including gender, age, and family financial level. This study implemented a standardized invariance testing protocol, systematically examining the measurement model’s structural equivalence through phased validation procedures. First, configural invariance analysis was conducted, requiring cross-group maintenance of equivalent factor configurations where free/fixed parameter linkages between latent variables and observed indicators remained consistent. Subsequently, metric invariance testing was performed by constraining factor loading matrices to numerical equivalence, thereby ensuring regression coefficient equivalence across sub-groups for all latent variable–observed indicator associations. Finally, strict scalar invariance verification was executed through cross-group equality constraints on variable intercepts and factor loading slopes, guaranteeing comparability of scale means across populations [37]. When ΔCFI and ΔTLI values were less than or equal to 0.01 and ΔRMSEA was less than or equal to 0.015, it was considered evidence of invariance [38].
In terms of reliability, McDonald’s omega (omega total, ωt) coefficients were used to determine the internal consistency, and a score of >0.7 is considered acceptable for reliability. Convergent validity was assessed by Pearson’s correlation coefficients on the SI-Bord, PSS-10, MLQ, ECR-R, and RSES to compare the magnitude of the relationship between the constructs of the measuring instruments. Data analysis was performed using IBM SPSS version 29 and AMOS version 26 software programs.

3. Results

Data were collected from Chinese university students between November 2021 and January 2022. Secondary data analysis commenced in November 2024 following approval from the Research Ethics Committee at Chiang Mai University.

3.1. Descriptive Analysis

The study involved 715 participants, predominantly male, with a mean age of 20.33 years. Over half identified as middle class and regularly received pocket money. The sample was embedded in a collectivist culture (Table 1). Descriptive analysis of SI-Bord items showed mean scores ranging from 1.12 to 1.24, except for Item 4 (self-harm/suicidal behaviors), which had a notably lower mean of 0.45, indicating less frequent endorsement. Standard deviations (~0.7–0.8) reflected comparable item variability. Item 4 also displayed significant positive skewness (1.38) and positive kurtosis (0.83), suggesting a peaked distribution with most responses at the lower end. In contrast, Items 1–3 and 5 exhibited mild negative skewness and flatter distributions, all within acceptable statistical ranges (Table 2).

3.2. CFA Model

As shown in Table 3, we used data from the Amos program in the unidimensional model. Preliminary analyses showed that the model is a good fit for the data. More specifically, the RMSEA (0.076) and SRMR (0.0314) are within acceptable thresholds (<0.08), showing a reasonable and excellent fit, respectively. The TLI (0.937) and CFI (0.969) exceed the recommended threshold (>0.9), reflecting strong model performance. And the CMIN/DF ratio (4.253) is under the acceptable limit (<5), suggesting a good fit between the model and data.
Table 4 displays the correlations among the five SI-Bord items and between each item and the SI-Bord total score. All items are significantly correlated with each other (p < 0.01), with coefficients ranging from 0.242 (between “Fear of abandonment” and “Self-harm/suicidal behaviors”) to 0.499 (between “Identity disturbances” and “Mood instability”). This indicates a moderate positive relationship among the items, suggesting that higher endorsement of one borderline symptom is associated with higher endorsement of the others. Each item shows a strong and significant correlation with the SI-Bord total score, ranging from 0.621 (“Self-harm/suicidal behaviors”) to 0.745 (“Mood instability”). This demonstrates that each item makes a substantial contribution to the overall measure of borderline symptoms. Among all items, “Mood instability” has the highest correlation with the total score (0.745), indicating that it may be a particularly central feature within this scale for this sample.

3.3. Convergent and Discriminant Validity

The SI-Bord showed a moderate positive correlation with perceived stress (PSS-10, r = 0.359, p < 0.01), indicating that individuals with higher levels of borderline features also report greater perceived stress. This relationship supports convergent validity, as stress is known to be associated with borderline personality pathology. There was a moderate, significantly positive correlation with attachment anxiety (r = 0.481, p < 0.01), consistent with theory and past findings showing that borderline features are linked to anxious attachment styles. This further supports convergent validity. The correlation between the SI-Bord and attachment avoidance is small and negative (r = −0.068, p = 0.52). While borderline traits can sometimes be associated with both dimensions of insecure attachment, these results suggest that the SI-Bord is more strongly related to attachment anxiety than avoidance in this sample. The SI-Bord is weakly and negatively correlated with meaning in life (r = −0.137, p < 0.01), a relationship that aligns with expectations but is small enough to support discriminant validity. The correlation with self-esteem is weakly negative (r = −0.075, p < 0.05), which supports discriminant validity, as self-esteem and borderline symptoms are related but distinct constructs (Table 5).

3.4. Invariance Test

Table 6 presents the results of the measurement invariance tests across gender, age, and family income groups. The results show that the unidimensional model fits all three groups well. Across all groups, CFI values remain above the recommended threshold of 0.95, while RMSEA values are below 0.06, indicating a satisfactory model fit. For gender groups, full measurement invariance (configural, metric, scalar) is supported. For age groups, configural invariance and metric invariance are upheld, but scalar invariance is not. Scalar invariance is achieved with negligible changes in fit indices (∆CFI ≤ 0.01, ∆TLI ≤ 0.01, and ∆RMSEA < 0.015). Significant deterioration (∆x2 = 19.175, df = 5) is observed, with CFI/TLI dropping (−0.021/−0.014) and RMSEA worsening (+0.009), indicating poor scalar invariance. For family income groups, configural invariance and metric invariance hold, but scalar invariance is only partially supported. ∆x2 = 13.024 (∆df = 5) is likely significant. CFI declines (−0.012), but RMSEA stabilizes, suggesting partial scalar invariance.

3.5. Reliability

The Screening Instrument for Borderline Personality Disorder demonstrated acceptable internal consistency, with a total omega (ωt) of 0.728. In this study, the reliability of the Chinese version of the SI-Bord was acceptable.

4. Discussion

The purpose of this study was to evaluate the psychometric properties of the Chinese version of the SI-Bord among Chinese university students. Overall, the results supported our hypotheses regarding the scale’s validity and reliability. To our knowledge, this is the first investigation of the SI-Bord’s psychometric performance in a sample of Chinese university students. Our findings indicate that the five-item unidimensional model provides a good fit for the data, supporting the use of both total and cutoff scores. This study provides substantial evidence for the structural validity of the Chinese SI-Bord. Inter-item correlation analyses demonstrated good internal consistency, with moderate-to-strong correlations observed among the items and between each item and the total score, indicating that the scale effectively assesses symptoms of borderline personality disorder. These results are consistent with previous research in Thailand, which also found support for a one-factor solution through factor analysis [21].
The Chinese version of the SI-Bord demonstrated adequate convergent and discriminant validity, as indicated by Pearson correlation analyses. Regarding the relationship between perceived stress and BPD symptoms, Stern’s seminal work posited that the “borderline” syndrome often arises as a decompensation in the presence of severe stress [39]. Consistently with this perspective, a significant correlation was observed between the SI-Bord and the Perceived Stress Scale (PSS) in this study, indicating some overlap in the constructs assessed by these instruments.
A strong and significant positive correlation was observed between anxious attachment and SI-Bord scores, consistent with previous findings and reinforcing the established link between attachment insecurity and borderline pathology [5,40]. Regarding discriminant validity, the SI-Bord demonstrated a negative correlation with meaning in life, as measured by the MLQ, underscoring the conceptual distinction between BPD and constructs such as meaning in life and self-esteem, despite some interrelations existing.
BPD is primarily characterized by emotional instability, impulsiveness, and disturbances in identity, while meaning in life represents a stable sense of existential fulfillment [5]. Recent studies have emphasized the relevance of meaning in life to the understanding of BPD symptoms, consistently showing negative associations between meaning in life and both behavioral and emotional symptoms of BPD [20]. Although BPD symptoms may diminish an individual’s sense of meaning in life, existential distress itself is not a diagnostic criterion for BPD. This distinction may explain why the SI-Bord’s behavioral focus demonstrates only limited correlation with more stable psychological traits such as meaning in life.
Similarly, self-esteem among individuals with BPD tends to fluctuate in response to situational contexts rather than remaining uniformly low. This instability of self-esteem is a distinguishing feature of BPD, separating it from the more persistent, trait-like self-esteem commonly measured in the general population and now recognized as central to BPD pathology [41]. As a result, the SI-Bord’s focus on the behavioral symptoms of BPD shows only limited overlap with measures of these more enduring psychological constructs, such as meaning in life and self-esteem.
However, the non-significant correlation between the SI-Bord and attachment avoidance, which did not confirm convergent validity as hypothesized, may be explained by several factors. While the Western literature often links borderline personality disorder (BPD) with both attachment anxiety and avoidance, our null finding aligns with Thai studies that highlight the limited role of avoidance in the development of psychopathology in this context [42,43]. In collectivist cultures such as Thailand, attachment avoidance may reflect culturally sanctioned emotional restraint rather than maladaptive detachment [44]. This suggests that emotional reserve can be adaptive and even desirable, rather than a marker of psychological dysfunction. This cultural dynamic may reduce the observable association between attachment avoidance and borderline symptoms as assessed by the SI-Bord.
Furthermore, it is possible that the SI-Bord’s focus on overt, emotionally expressive BPD symptoms may underestimate avoidance-related pathology in this population. Cross-cultural studies from East and Southeast Asia (including China and Thailand), where collectivism and group harmony are highly valued, emphasize minimizing visible discomfort with intimacy to preserve social cohesion. As a result, overt displays of attachment avoidance may be suppressed or expressed differently [45].
Finally, the Experience in Close Relationships–Revised (ECR-R) measure may not fully capture the nuances of attachment avoidance as influenced by the culturally unique interpersonal norms found in Thailand and neighboring Asian societies [11,21]. We recommend further empirical research to explore whether adaptations of either the SI-Bord or attachment measures may enhance the detection of avoidance patterns in non-Western samples.
Regarding the measurement invariance of the SI-Bord across groups, this study found that the gender group met strict criteria at all stages of measurement invariance, indicating that the scale allows for cross-group comparisons between different genders. However, the age and income groups exhibited certain variations in the intercept invariance test, particularly with ΔCFI and ΔTLI exceeding the recommended thresholds, which may affect the direct comparability of latent variable means. Therefore, when comparing the SI-Bord scores among different age and income groups, one has to be cautious. This discrepancy may stem from differences in how individuals from varying age groups or income levels interpret the measurement items. Scalar invariance is critical in ensuring that observed scores, such as total SI-Bord scores, maintain consistent meaning across diverse groups, including variations in age and income levels. Without achieving scalar invariance, a score of “20” may signify different levels of borderline personality disorder (BPD) severity, influenced by demographic factors such as age and income. As a result, comparative analyses between groups may yield misleading conclusions if the measurement scale operates differently across these groups. The absence of scalar invariance often indicates the existence of differential item functioning (DIF) [46], wherein certain items of the SI-Bord may be interpreted variably across different demographics. For instance, older adults may affirm the statement “I feel empty” due to experiences of grief rather than BPD. In contrast, individuals from low-income backgrounds may report “impulsivity” more frequently, as financial stress can exacerbate impulsive behavior [47]. Consequently, it is imperative to approach cross-group comparisons cautiously; such comparisons should be considered provisional. The subsequent course of action involves testing for partial scalar invariance by identifying items that fail to meet invariance criteria, utilizing techniques such as modification indices or DIF analysis. Alternatively, researchers may opt to conduct comparisons using latent factor scores derived from measurement models, rather than relying solely on raw totals. Should non-invariance remain an issue, it may be necessary to revise problematic items or develop group-specific norms, accompanied by distinct scoring guidelines tailored for various age and income sub-groups. This process should also be supported by replication efforts involving larger sample sizes.
In addition to the DIF analysis, special attention was paid to Item 4, which assesses either suicidal behavior or self-injury. While the prevalence of non-suicidal self-injury (NSSI) among Chinese youth is generally high—reported at 24.7%, with college students showing a lifetime prevalence of 21.2%, which points to early onset trends [48]—endorsement of this item was notably lower in our sample. Specifically, the mean score for the self-harm/suicidal behavior item was 0.45, compared to means of 1.12–1.24 for other items, suggesting that fewer participants acknowledged such behaviors. This discrepancy may be partially explained by confusion around the distinction between suicidal behavior and NSSI, as the boundaries between these concepts can be unclear for respondents [49]. Such ambiguity can result in underreporting or misclassification of these behaviors in self-assessment tools.
Furthermore, the infrequent endorsement of this item likely contributed to its weaker correlations with the overall scale. We acknowledge this limitation and recommend that future research adopt frequency-anchored items or specifically target high-risk sub-groups to improve detection rates. Nevertheless, despite the lower mean and weaker correlations, this item still demonstrated sufficient internal consistency, aligning with findings from studies with Thai samples [21].

4.1. Implications

The validation of the Chinese version of the SI-Bord provides a trustworthy instrument for upcoming research on BPD symptoms. The theoretical methodologies employed in this study establish a strong basis for additional measurement and real-world applications. Early screening for students with high BPD risk may benefit educational authorities and institutions. Universities and teachers should target both high- and low-risk populations. Lastly, thorough evaluation facilitates the screening of BPD symptoms and offers a scientific foundation for clinical judgments, supporting the application of the SI-Bord in clinical practice.

4.2. Strengths and Limitations of This Study, and Future Research

One of this study’s strengths is that the SI-Bord scale performed well in the Chinese cultural setting and demonstrated strong consistency across diverse cultural backgrounds. Second, the study’s adequate sample size minimized the effect of measurement bias on the findings. Furthermore, the SI-Bord scale accurately measured the target psychological symptoms and showed satisfactory psychometric qualities, including good validity and reliability.
However, this study also has some limitations. This sample was collected during the COVID-19 pandemic, which may have impacted reactions due to China’s home isolation policy. Pandemic-related cognitive fog (reported by 81% of individuals with infections) [50] and media overexposure—both linked to heightened stress—could have particularly affected anxiously attached participants’ reactions [51].
The use of a secondary database limits the ability to implement additional, methodologically appropriate measures, particularly those necessary for establishing discriminant validity. Moreover, the convenience sampling approach may compromise the generalizability of the findings due to potential sampling bias. Reliance on self-report measures introduces the risk of biases, such as acquiescence and social desirability effects. To enhance the robustness of future research, replication studies should be conducted in non-pandemic contexts with larger and more diverse samples. It would also be advantageous to compare the SI-Bord with structured diagnostic interviews to delineate which specific facets of borderline personality disorder (BPD) are most accurately captured, as well as to examine its convergence with other established BPD assessment instruments to strengthen construct validity. Finally, further investigation into differential items functioning across demographic variables such as age and income is warranted to ensure the measure’s fairness and applicability across groups.

5. Conclusions

This study conducted a comprehensive analysis of the psychometric properties of the Chinese version of the SI-Bord and found that the scale demonstrates good validity and reliability in screening for core symptoms of borderline personality disorder. The unidimensional construct provides empirical support for the reasonableness and interpretability of the total score, ensuring its clear psychological significance. The scale showed meaningful correlations with perceived stress, attachment, meaning in life, and self-esteem, further supporting its ability to accurately capture the core symptoms of BPD in line with theoretical expectations. Although bias due to sex can be mitigated, age and income should be considered when SI-Bord scores are compared across samples. Acceptable reliability confirms the scale’s reproducibility. Further research is encouraged to improve the scale’s validity and reliability.

Author Contributions

Conceptualization and methodology, H.Z., C.S. (Chaiyun Sakulsriprasert), T.W., N.W., C.S. (Chawisa Suradom), Y.C., N.J. and R.O.; software, H.Z., C.S. (Chaiyun Sakulsriprasert), T.W. and Y.C.; formal analysis, H.Z., C.S. (Chaiyun Sakulsriprasert), T.W. and Y.C.; resources, H.Z., C.S. (Chaiyun Sakulsriprasert), C.S. (Chawisa Suradom), T.W., N.W., Y.C., R.O. and N.J.; data curation, H.Z., C.S. (Chaiyun Sakulsriprasert), T.W. and Y.C.; writing—original draft preparation, H.Z., C.S. (Chaiyun Sakulsriprasert), T.W. and Y.C.; writing—review and editing, H.Z., C.S. (Chaiyun Sakulsriprasert), C.S. (Chawisa Suradom), T.W., Y.C., R.O. and N.J.; supervision, C.S. (Chaiyun Sakulsriprasert), T.W., N.W. and Y.C.; project administration, C.S. (Chaiyun Sakulsriprasert) and N.W.; funding acquisition, H.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Faculty of Medicine, Chiang Mai University (Approval Code: No. 476/2021; Approval date: 4 November 2021).

Informed Consent Statement

Informed consent was obtained from all the subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to ethical approval requirements.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
BPDBorderline Personality Disorder
SI-BordScreening Instrument for Borderline Personality Disorder
PSS-10Perceived Stress Scale
MLQMeaning in Life Questionnaire
ECR-RExperience in Close Relationships–Revised
RSESRosenberg Self-Esteem Scale

References

  1. APA. Diagnostic and Statistical Manual of Mental Disorder (DSM-5), 5th ed.; APA: Washington, DC, USA, 2013. [Google Scholar]
  2. Bretherton, I. The origins of attachment theory: John Bowlby and Mary Ainsworth. In Attachment Theory; Routledge: London, UK, 2013; pp. 45–84. [Google Scholar]
  3. Pearse, E.; Bucci, S.; Raphael, J.; Berry, K. The relationship between attachment and functioning for people with serious mental illness: A systematic review. Nord. J. Psychiatry 2020, 74, 545–557. [Google Scholar] [CrossRef]
  4. Levy, K.N.; Johnson, B.N.; Clouthier, T.L.; Scala, J.; Temes, C.M. An attachment theoretical framework for personality disorders. Can. Psychol./Psychol. Can. 2015, 56, 197. [Google Scholar] [CrossRef]
  5. Smith, M.; South, S. Romantic attachment style and borderline personality pathology: A meta-analysis. Clin. Psychol. Rev. 2020, 75, 101781. [Google Scholar] [CrossRef]
  6. Holmes, J. Disorganized attachment and borderline personality disorder: A clinical perspective. Attach. Hum. Dev. 2004, 6, 181–190. [Google Scholar] [CrossRef] [PubMed]
  7. Bourvis, N.; Aouidad, A.; Cabelguen, C.; Cohen, D.; Xavier, J. How do stress exposure and stress regulation relate to borderline personality disorder? Front. Psychol. 2017, 8, 2054. [Google Scholar] [CrossRef]
  8. Kellner, R. Psychosomatic Syndromes and Somatic Symptoms; American Psychiatric Publishing: Washington, DC, USA, 1992. [Google Scholar]
  9. Bourvis, N.; Aouidad, A.; Spodenkiewicz, M.; Palestra, G.; Aigrain, J.; Baptista, A.; Benoliel, J.J.; Chetouani, M.; Cohen, D. Adolescents with borderline personality disorder show a higher response to stress but a lack of self-perception: Evidence through affective computing. Prog. Neuropsychopharmacol. Biol. Psychiatry 2021, 111, 110095. [Google Scholar] [CrossRef]
  10. Xie, G.D.; Chang, J.J.; Yuan, M.Y.; Wang, G.F.; He, Y.; Chen, S.S.; Su, P.Y. Childhood abuse and borderline personality disorder features in Chinese undergraduates: The role of self-esteem and resilience. BMC Psychiatry 2021, 21, 326. [Google Scholar] [CrossRef]
  11. Jia, N.; Sakulsriprasert, C.; Wongpakaran, N.; Suradom, C.; Donnell, R.O. Borderline Personality Disorder Symptoms and Its Clinical Correlates among Chinese University Students: A Cross-Sectional Study. Healthcare 2022, 10, 1751. [Google Scholar] [CrossRef] [PubMed]
  12. Bozzatello, P.; Garbarini, C.; Rocca, P.; Bellino, S. Borderline Personality Disorder: Risk Factors and Early Detection. Diagnostics 2021, 11, 2142. [Google Scholar] [CrossRef]
  13. Zanarini, M.C.; Frankenburg, F.R.; Vujanovic, A.A. Inter-rater and test-retest reliability of the Revised Diagnostic Interview for Borderlines. J. Pers. Disord. 2002, 16, 270–276. [Google Scholar] [CrossRef] [PubMed]
  14. Bohus, M.; Kleindienst, N.; Limberger, M.F.; Stieglitz, R.D.; Domsalla, M.; Chapman, A.L.; Steil, R.; Philipsen, A.; Wolf, M. The short version of the Borderline Symptom List (BSL-23): Development and initial data on psychometric properties. Psychopathology 2009, 42, 32–39. [Google Scholar] [CrossRef]
  15. Arntz, A.; van den Hoorn, M.; Cornelis, J.; Verheul, R.; van den Bosch, W.M.; de Bie, A.J. Reliability and validity of the borderline personality disorder severity index. J. Personal. Disord. 2003, 17, 45–59. [Google Scholar] [CrossRef]
  16. Pfohl, B.; Blum, N.; St John, D.; McCormick, B.; Allen, J.; Black, D.W. Reliability and validity of the Borderline Evaluation of Severity Over Time (BEST): A self-rated scale to measure severity and change in persons with borderline personality disorder. J. Pers. Disord. 2009, 23, 281–293. [Google Scholar] [CrossRef]
  17. Zanarini, M.C.; Vujanovic, A.A.; Parachini, E.A.; Boulanger, J.L.; Frankenburg, F.R.; Hennen, J. Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD): A continuous measure of DSM-IV borderline psychopathology. J. Pers. Disord. 2003, 17, 233–242. [Google Scholar] [CrossRef]
  18. Marco, J.H.; Pérez, S.; García-Alandete, J.; Moliner, R. Meaning in life in people with borderline personality disorder. Clin. Psychol. Psychother. 2017, 24, 162–170. [Google Scholar] [CrossRef] [PubMed]
  19. Winter, D.; Steeb, L.; Herbert, C.; Sedikides, C.; Schmahl, C.; Bohus, M.; Lis, S. Lower self-positivity and its association with self-esteem in women with borderline personality disorder. Behav. Res. Ther. 2018, 109, 84–93. [Google Scholar] [CrossRef] [PubMed]
  20. Lorca, F.; Pérez, S.; Giner, F.; Marco, J.H. What dimension of meaning in life is the stronger predictor of borderline personality disorder symptom? J. Constr. Psychol. 2021, 34, 365–377. [Google Scholar] [CrossRef]
  21. Lohanan, T.; Leesawat, T.; Wongpakaran, T.; Wongpakaran, N.; Karawekpanyawong, N.; Oon-Arom, A.; Kuntawong, P. Development and validation of a screening instrument for borderline personality disorder (SI-Bord) for use among university students. BMC Psychiatry 2020, 20, 479. [Google Scholar] [CrossRef] [PubMed]
  22. Cohen, S.; Kamarck, T.; Mermelstein, R. A global measure of perceived stress. J. Health Soc. Behav. 1983, 24, 385–396. [Google Scholar] [CrossRef]
  23. Cohen, S.; Kamarck, T.; Mermelstein, R. Perceived stress scale. Meas. Stress A Guide Health Soc. Sci. 1994, 10, 1–2. [Google Scholar]
  24. Wang, Z.; Wang, Y.; Wu, Z.; Chen, D.; Chen, J.; Xiao, Z. Reliability and validity of the Chinese version of perceived stress scale. J. Shanghai Jiaotong Univ. (Med. Sci.) 2015, 35, 1448. [Google Scholar]
  25. Steger, M.F.; Frazier, P.; Oishi, S.; Kaler, M. The meaning in life questionnaire: Assessing the presence of and search for meaning in life. J. Couns. Psychol. 2006, 53, 80. [Google Scholar] [CrossRef]
  26. Chen, W.; Gao, R. A generalizability analysis of the meaning in life questionnaire for Chinese adolescents. Front. Psychol. 2021, 12, 687589. [Google Scholar] [CrossRef]
  27. Fraley, R.C.; Waller, N.G.; Brennan, K.A. An item response theory analysis of self-report measures of adult attachment. J. Personal. Soc. Psychol. 2000, 78, 350. [Google Scholar] [CrossRef]
  28. Wongpakaran, T.; Wongpakaran, N.; Wannarit, K. Validity and reliability of the Thai version of the Experiences of Close Relationships-Revised questionnaire. Singap. Med. J. 2011, 52, 100–106. [Google Scholar]
  29. Li-Tonggui, K.-K. Measuring adult attachment: Chinese adaptation of the ECR scale. Acta Psychol. Sin. 2006, 38, 399. [Google Scholar]
  30. Rosenberg, M. Society and the Adolescent Self-Image; Princeton University Press: Princeton, NJ, USA, 1965. [Google Scholar]
  31. Fleming, J.S.; Courtney, B.E. The dimensionality of self-esteem: II. Hierarchical facet model for revised measurement scales. J. Personal. Soc. Psychol. 1984, 46, 404. [Google Scholar] [CrossRef]
  32. Chen, F.; Bi, C.; Han, M. The reliability and validity of the Chinese version of the revised-positive version of Rosenberg self-esteem scale. Adv. Psychol. 2015, 5, 531–535. [Google Scholar] [CrossRef]
  33. Browne, M.W.; Cudeck, R. Alternative ways of assessing model fit. In Testing Structural Equation Models; Bollen, K.A., Long, J.S., Eds.; Sage: Newbury Park, CA, USA, 1993. [Google Scholar]
  34. Steiger, J.H. Structural model evaluation and modification: An interval estimation approach. Multivar. Behav. Res. 1990, 25, 173–180. [Google Scholar] [CrossRef]
  35. Bentler, P.M. Comparative fit indexes in structural models. Psychol. Bull. 1990, 107, 238. [Google Scholar] [CrossRef]
  36. Tucker, L.R.; Lewis, C. A reliability coefficient for maximum likelihood factor analysis. Psychometrika 1973, 38, 1–10. [Google Scholar] [CrossRef]
  37. Steenkamp, J.-B.E.; Baumgartner, H. Assessing measurement invariance in cross-national consumer research. J. Consum. Res. 1998, 25, 78–90. [Google Scholar] [CrossRef]
  38. Chen, F.F. Sensitivity of goodness of fit indexes to lack of measurement invariance. Struct. Equ. Model. A Multidiscip. J. 2007, 14, 464–504. [Google Scholar] [CrossRef]
  39. Allen, T.A.; Dombrovski, A.Y.; Soloff, P.H.; Hallquist, M.N. Borderline personality disorder: Stress reactivity or stress generation? A prospective dimensional study. Psychol. Med. 2022, 52, 1014–1021. [Google Scholar] [CrossRef]
  40. Ghiasi, H.; Mohammadi, A.; Zarrinfar, P. An Investigation into the Roles of Theory of Mind, Emotion Regulation, and Attachment Styles in Predicting the Traits of Borderline Personality Disorder. Iran. J. Psychiatry 2016, 11, 206–213. [Google Scholar]
  41. Kockler, T.D.; Santangelo, P.S.; Eid, M.; Kuehner, C.; Bohus, M.; Schmaedeke, S.; Ebner-Priemer, U.W. Self-esteem instability might be more characteristic of borderline personality disorder than affective instability: Findings from an e-diary study with clinical and healthy controls. J. Psychopathol. Clin. Sci. 2022, 131, 301–313. [Google Scholar] [CrossRef]
  42. Oon-arom, A.; Wongpakaran, T.; Kuntawong, P.; Wongpakaran, N. Attachment anxiety, depression, and perceived social support: A moderated mediation model of suicide ideation among the elderly. Int. Psychogeriatr. 2021, 33, 169–178. [Google Scholar] [CrossRef]
  43. Myint, K.M.; DeMaranville, J.; Wongpakaran, T.; Peisah, C.; Arunrasameesopa, S.; Wongpakaran, N. Meditation Moderates the Relationship between Insecure Attachment and Loneliness: A Study of Long-Term Care Residents in Thailand. Medicina 2024, 60, 622. [Google Scholar] [CrossRef] [PubMed]
  44. Lawrence, R.C. Collectivism at the individual level: A moderator of the relationship between emotion suppression and personal strain. Curr. Psychol. 2023, 42, 30364–30380. [Google Scholar] [CrossRef]
  45. Xu, C.; Yan, W. Negative parenting styles and social adjustment of university students: A moderated chain mediation model. Curr. Psychol. 2022, 42, 27719–27732. [Google Scholar] [CrossRef] [PubMed]
  46. Chen, W.-H.; Revicki, D. Differential item functioning (DIF). In Encyclopedia of Quality of Life and Well-Being Research; Springer: Berlin/Heidelberg, Germany, 2024; pp. 1783–1786. [Google Scholar]
  47. Zhu, J.; Chen, Y.; Zhang, W. Association between family economic hardship and university students’ risky alcohol use: Mediating and moderating roles of perceived discrimination and impulsivity. Personal. Individ. Differ. 2019, 141, 1–6. [Google Scholar] [CrossRef]
  48. Qu, D.; Wen, X.; Liu, B.; Zhang, X.; He, Y.; Chen, D.; Duan, X.; Yu, J.; Liu, D.; Zhang, X.; et al. Non-suicidal self-injury in Chinese population: A scoping review of prevalence, method, risk factors and preventive interventions. Lancet Reg. Health West. Pac. 2023, 37, 100794. [Google Scholar] [CrossRef] [PubMed]
  49. Poudel, A.; Lamichhane, A.; Magar, K.R.; Khanal, G.P. Non suicidal self injury and suicidal behavior among adolescents: Co-occurrence and associated risk factors. BMC Psychiatry 2022, 22, 96. [Google Scholar] [CrossRef] [PubMed]
  50. Gangemi, A.; Suriano, R.; Fabio, R.A. Longitudinal Exploration of Cortical Brain Activity in Cognitive Fog: An EEG Study in Patients with and without Anosmia. J. Integr. Neurosci. 2024, 23, 105. [Google Scholar] [CrossRef] [PubMed]
  51. Fabio, R.A.; Suriano, R. The Influence of Media Exposure on Anxiety and Working Memory during Lockdown Period in Italy. Int. J. Environ. Res. Public Health 2021, 18, 9279. [Google Scholar] [CrossRef]
Table 1. Sociodemographic characteristics and scale scores of the sample (n = 715).
Table 1. Sociodemographic characteristics and scale scores of the sample (n = 715).
VariablesFrequencyPercent
SexMale38554.2
Female32545.8
Perceived family wealthPoor212.9
Middle class41858.5
Wealthy27638.6
Personal incomePart-time job202.8
Daily pocket money from parents36751.3
Both32144.9
Others70.9
Scale scores (mean ± SD)
SI-Bord (0–14) 5.22 ± 2.58
Attachment anxiety (8–63) 28.92 ± 11.99
Attachment avoidance (9–63) 42.93 ± 9.68
Rosenberg self-esteem (11–45) 33.23 ± 6.31
Meaning in life (11–70) 49.72 ± 9.68
Perceived stress (0–32) 16.84 ± 5.23
Table 2. Descriptive statistics, skewness, and kurtosis of the SI-Bord items (n = 715).
Table 2. Descriptive statistics, skewness, and kurtosis of the SI-Bord items (n = 715).
SI-Bord ItemMeanSDVarianceSkewnessKurtosis
1. Fear of abandonment1.2300.7960.634−0.167−0.929
2. Unstable relationships1.1800.7340.538−0.107−0.767
3. Identity disturbances1.2400.7740.600−0.151−0.814
4. Self-harm/suicidal behaviors0.4500.7090.5021.3810.831
5. Mood instability1.1200.7240.5240.007−0.659
SD = standard deviation.
Table 3. Fit indices of the unidimensional model.
Table 3. Fit indices of the unidimensional model.
ModelCMINDFCMIN/DFRMSEASRMRTLICFI
Unidimensional model 25.51864.2530.0760.03140.9370.969
Note: CMIN: Chi-Square; DF: degrees of freedom; CMIN/DF: Chi-Square/degrees-of-freedom ratio; RMSEA: root mean square error of approximation; SRMR: standardized root mean square residual; TLI: Tucker–Lewis Index; CFI: comparative fit index.
Table 4. Correlation across items and between items and SI-Bord.
Table 4. Correlation across items and between items and SI-Bord.
SI- Bord Items12345SI-Bord Total
1. Fear of abandonment-
2. Unstable relationships0.379 **-
3. Identity disturbances0.268 **0.409 **-
4. Self-harm/suicidal behaviors0.242 **0.304 **0.269 **-
5. Mood instability0.308 **0.411 **0.499 **0.372 **-
SI-Bord total0.650 **0.723 **0.713 **0.621 **0.745 **-
SI-Bord: Screening Instrument for Borderline Personality Disorder; ** p < 0.01.
Table 5. Correlation results between SI-Bord, PSS-10, ECR-R, MLQ, and RSES.
Table 5. Correlation results between SI-Bord, PSS-10, ECR-R, MLQ, and RSES.
SI-BordPerceived StressAttachment
Anxiety		Attachment
Avoidance		Meaning in LifeSelf-Esteem
SI-Bord-
Perceived stress0.359 **-
Attachment anxiety0.481 **0.313 **-
Attachment avoidance−0.0680.152 **0.020-
Meaning in life−0.137 **0.101 **−0.095 *0.341 **-
Self-esteem−0.075 *−0.110 **−0.0430.0420.014-
Note: ** p < 0.01; * p < 0.05; SI-Bord: Screening Instrument for Borderline Personality Disorder; PSS-10: Perceived Stress Scale; ECR-R: Experience in Close Relationships–Revised; MLQ Meaning in Life Questionnaire; RSES: Rosenberg Self-Esteem Scale.
Table 6. Invariance test.
Table 6. Invariance test.
Modelχ2 (df)Δχ2 (Δdf)CFITLIRMSEAΔCFIΔTLIΔRMSEA
Sex group
Configural Invariance31.470 (11) 0.9680.9430.051
Metric Invariance36.742 (15)5.272 (4)0.9670.9550.045−0.0010.012−0.006
Scalar Invariance40.339 (20)3.579 (5)0.9690.9690.0380.0020.014−0.007
Age group
Configural Invariance26.137 (11) 0.9770.9580.044
Metric Invariance29.120 (15)2.983 (4)0.9780.9710.0360.0010.013−0.008
Scalar Invariance48.295 (20)19.175 (5)0.9570.9570.045−0.021−0.0140.009
Income group
Configural Invariance36.635 (11) 0.9610.9290.057
Metric Invariance38.914 (15)2.279 (4)0.9640.9520.0470.0030.023−0.010
Scalar Invariance51.938 (20)13.024 (5)0.9520.9520.047−0.0120.0000.000
Note: χ2 (df): Chi-Square value and degrees of freedom; TLI: Tucker–Lewis Index; CFI: comparative fit index; RMSEA: root mean square error of approximation.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhou, H.; Chang, Y.; Sakulsriprasert, C.; Wongpakaran, T.; Wongpakaran, N.; Suradom, C.; O’Donnell, R.; Jia, N. The Validity and Reliability of the Chinese Version of the Screening Instrument for Borderline Personality Disorder. Psychiatry Int. 2025, 6, 108. https://doi.org/10.3390/psychiatryint6030108

AMA Style

Zhou H, Chang Y, Sakulsriprasert C, Wongpakaran T, Wongpakaran N, Suradom C, O’Donnell R, Jia N. The Validity and Reliability of the Chinese Version of the Screening Instrument for Borderline Personality Disorder. Psychiatry International. 2025; 6(3):108. https://doi.org/10.3390/psychiatryint6030108

Chicago/Turabian Style

Zhou, Hui, Yu Chang, Chaiyun Sakulsriprasert, Tinakon Wongpakaran, Nahathai Wongpakaran, Chawisa Suradom, Ronald O’Donnell, and Nan Jia. 2025. "The Validity and Reliability of the Chinese Version of the Screening Instrument for Borderline Personality Disorder" Psychiatry International 6, no. 3: 108. https://doi.org/10.3390/psychiatryint6030108

APA Style

Zhou, H., Chang, Y., Sakulsriprasert, C., Wongpakaran, T., Wongpakaran, N., Suradom, C., O’Donnell, R., & Jia, N. (2025). The Validity and Reliability of the Chinese Version of the Screening Instrument for Borderline Personality Disorder. Psychiatry International, 6(3), 108. https://doi.org/10.3390/psychiatryint6030108

Article Metrics

Back to TopTop