Sage Journals: Discover world-class research

Abstract

We examined the psychometric properties of the Borderline Personality Features Scale for Children (BPFSC-11) in a sample of n = 235 adolescents. This involved an exploratory factor analysis (EFA) and a confirmatory factor analysis (CFA). We meticulously assessed internal consistency, convergent, and concurrent validity by analyzing the relationship between BPFSC-11 scores and measurements of emotional dysregulation, self-efficacy, resilience, depression, anxiety sensitivity, and social anxiety. The EFA revealed a two-factor solution, a finding that the CFA robustly confirmed. The BPFSC-11 demonstrated excellent internal consistency with an alpha value of 0.87 and adequate validity. Measurement invariance was examined across five groups: sex assigned at birth, born in Norway, speaking primarily Norwegian at home, by treatment group, and longitudinally. These findings confirm the reliability and validity of the BPFSC-11, adding new and significant findings to provide a solid foundation for its use in the assessment and treatment of adolescents with borderline personality features. Limitations of the current study include the absence of a clinical interview to examine diagnostic validity, and participants’ responses may have been influenced by response desirability bias. Future research should examine the validity of the scale, both cross-sectionally and longitudinally, to enhance understanding of its application in different cultural contexts.

Keywords

BPSFC11 borderline traits psychometric properties adolescents emotion regulation

Introduction

The essential feature of borderline personality disorder (BPD) is a pervasive pattern of instability of interpersonal relationships, self-image, and affects, and marked impulsivity that is present in a variety of contexts (APA, 2013). This disorder is associated with significant functional impairments, frequent use of intensive treatments, and substantial societal costs (Leichsenring et al., 2024).

BPD and BPD symptoms in adolescence have a substantial negative impact on different areas of functioning and are recognized as a severe form of psychopathology (Haltigan & Vaillancourt, 2016). BDP is a unique disorder but has symptoms that overlap and have similar etiological roots to depression and anxiety. Like depression, the development of borderline traits seems closely linked to the experience of adverse events (Fonagy & Bateman, 2016). Moreover, impaired emotion regulation is a hallmark feature of major depressive episodes (APA, 2013) and the core feature of BPD (APA, 2013). Further, BPD has been associated with anxiety in general and with social anxiety disorder, specifically with increased self-focused negative emotions, along with more escape tendencies and less intent to engage in pleasant activities (Gutz et al., 2016). In addition, lower resilience (Guillén et al., 2021) and self-efficacy (Heiland & Veilleux, 2021) have been reported among patients with BPD. The present study evaluates the psychometric properties of a Scandinavian adaptation of the Brief Borderline Personality Features Scale for Children (BPFSC-11; Sharp et al., 2014). For comparative purposes, it is advantageous to incorporate assessment instruments utilized in prior validation studies. Furthermore, the inclusion of psychometrically robust measures assessing psychological constructs empirically and theoretically associated with borderline personality disorder (BPD) features is recommended to enhance the comprehensiveness of the evaluation. Historically, there has been debate over the appropriateness of diagnosing BPD in pre-adolescents, and there is no current consensus that BPD is both a valid and reliable diagnosis for this age group. Therefore, early identification of BPD features allows for prompt treatment for these young individuals, ultimately alleviating personal distress and reducing societal expenses. However, early identification of BDP features depends on a careful and accurate assessment of personality pathology in pre-adolescents and adolescents (Sharp et al., 2014). Over the past two decades, the field has witnessed significant evolution in recognizing and assessing borderline traits in youth. Early skepticism was rooted in concerns about the stability of personality in childhood and the potential for stigma (Miller et al., 2008). However, accumulating evidence has demonstrated that core features of BPD—such as affective instability, impulsivity, and interpersonal difficulties—can be reliably identified in adolescence and even late childhood and that these features are associated with significant functional impairment (Chanen & Kaess, 2012). This shift has prompted the development of age-appropriate assessment tools, including both self-report measures and structured interviews.

Recent reviews further underscore the importance of dimensional approaches to assessment, highlighting the continuity of borderline pathology across development and the utility of early screening for prevention and intervention (Leichsenring et al., 2024). These advances reflect a growing consensus that, while caution is warranted, assessing borderline traits in children and adolescents is feasible and clinically meaningful. As such, ongoing research continues to refine assessment strategies, improve early detection, and inform tailored interventions for at-risk youth (Laczkovics et al., 2025; Leichsenring et al., 2024).

To complement clinical assessment, reliable and valid self-report instruments are both important and valuable, yet they are also time- and cost-effective. Moreover, such tools should focus on dimensional personality models, particularly when assessing maladaptive personality functioning among adolescents. By evaluating symptoms across the entire latent trait of borderline personality, it becomes possible to identify not only those exhibiting clinically significant levels of symptoms but also individuals who may be at risk (Sharp et al., 2014). Such an approach also aligns with the DSM-5 (APA, 2013) alternative dimensional model for assessing BPD (Leichsenring et al., 2024).

The brief Borderline Personality Features Scale for Children (BPFSC-11; Sharp et al., 2014), using item response theory (IRT), was developed from the 24-item Borderline Personality Features Scale for Children (BPFSC; Crick et al., 2005). The BPFSC-11 is a dimensional measure developed by collecting data from a community sample of 964 adolescents (mean age = 15.1 years, SD = .79; 55.9% female; Sharp et al., 2014). Sharp et al. (2014) examined the BPFSC-11, and they reported adequate construct validity by comparing the interviews with the original BPFSC among adolescents who met BPD criteria and those who did not, using semi-structured clinical interviews. Correlation analyses were conducted using the Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) and the Deliberate Self-Harm Inventory (DSHI: Gratz, 2001) to assess emotional dysregulation and self-harm. A hypothesized four-factor structure was not supported. Instead, support for a unidimensional factor structure was reported (Sharp et al., 2014). Examining the BPFSC-11 within a sample of 805 (mean age = 16.42 years, SD = 1.47; 64% female) Italian adolescents, Fossati et al. (2019) did not support a one-factor solution of the BPFSC-11 items. A bi-factor model analysis indicated a general factor and two specific factors. Fossati et al. (2019) reported Cronbach’s α = .78, a moderate (r = .50) for 6-month test-retest correlation, and gender invariance across the BPFSC-11 items. Calvo et al. (2023) assessed a mixed sample of 145 Spanish children and adolescents (mean age = 15.51 years, SD = 2.21; 62.1% female), comprising 58 individuals diagnosed with BPD, 58 with ADHD, and 29 healthy volunteers as a control group. Applying an Explorative Factorial Analysis (EFA) demonstrated a two-factor solution, whereas six items (1, 4, 6, 7, 9, 11) were included in factor 1 (Emotional dysregulation), and three items (5, 8, 11) were included in factor 2 (impulsivity/recklessness). Two items (2, 3) exhibited nearly identical loads across both factors. The reliability of the total BPFSC-11 score based on the 11 items was α = 0.86 overall, and α = 0.89 and α = .59 for factors 1 and 2, respectively. Furthermore, the BPFSC-11 total scores on factor 1 appeared to discriminate between the BDP, ADHD, and healthy control groups, with the BDP group reporting significantly higher scores. In contrast, the ADHD group reported significantly higher scores on factor 2 than the other two groups. A mixed sample of 1600 adolescents was assessed in Pakistan, including outpatients seen in a BPD psychiatric ward program (n = 300) and healthy controls between 12 and 19 years old, 55% of whom were female. Bibi and Kazmi (2021) reported a two-factor structure of the BPFSC-11 applying an EFA and a CFA. Factor 1 was named Intra-/interpersonal preoccupation (items 2, 6, 9, 11), and factor 2 (Items 1, 4, 5), with Cronbach’s α = .79 and α = .84 for the two factors, respectively.

In their examination of the Borderline Personality Features Scale for Children (BPFS-C; Crick et al., 2005), Haltigan and Vaillancourt (2016) found that the initially proposed four-factor structure was not supported in their sample of Canadian adolescents aged 13–16 years (N = 566; 55.5% female). Instead, their follow-up EFA and CFA indicated a refined two-factor latent structure consisting of an intra- and interpersonal preoccupation factor and an impulsivity/reactivity factor based on eleven items of the BPFS-C. Furthermore, measurement invariance modeling demonstrated that this two-factor structure was stable over time, suggesting that it held a consistent empirical meaning for both sexes across different time points. In a subsequent study, Vanwoerden et al. (2019) assessed the invariance of the BPFSC-11 items in a diverse community sample of 755 adolescents (mean age = 15.83, SD = .67; 56% female) in Texas, USA. Their findings revealed partial measurement invariance between genders and full longitudinal invariance from adolescence into young adulthood. Assessing a combined community-based sample (n = 379) and clinical sample (n = 74) of Lithuanian adolescents aged 11–18 revealed scale invariance for gender and partial invariance for age (Allman et al., 2025).

Taken together, previous studies (Allman et al., 2025; Bibi & Kazmi, 2021; Calvo et al., 2023; Fossati et al., 2019; Sharp et al., 2014; Vanwoerden et al., 2019) have examined the psychometric properties of the BPFSC-11, indicating promising validity and reliability for the inventory. Nevertheless, existing studies have shown varying factor structures (unidimensional, bi-factor, and two-factor), and there is a need for further establishment of concurrent and discriminant validity. Moreover, while the introduction mentions studies from various countries (USA, Italy, Spain, Pakistan, Lithuania, and Canada), a more comprehensive cross-cultural validation is needed. Finally, although Vanwoerden et al. (2019) reported partial measurement invariance between genders and full longitudinal invariance, more extensive measurement invariance testing across different groups and time is needed. More specifically, testing invariance not only across gender but also across an intervention and a control group, as well as two cultural groups living in the same cultural settings, is beneficial. Moreover, recognizing the multifaceted nature of borderline pathology, including additional psychometrically sound instruments that capture related psychological constructs, such as self-efficacy, resilience, anxiety, depression and emotion regulation, allows for a more nuanced evaluation of the BPFSC-11’s validity but also aligns with contemporary recommendations for personality pathology research in youth (Sharp & Cervantes, 2023). Addressing these aspects would provide a more comprehensive understanding of BPFSC-11’s psychometric properties and applicability across diverse populations and settings.

To our knowledge, no studies have reported using an EFA and a CFA of the BPFSC-11 on the same sample with a 1-year interval between the two assessments. Thus, this study has several objectives. First, we report descriptive statistics for the BPFSC-11. Second, we report internal consistency. Third, we conduct an EFA and a CFA one year after the initial analysis. Fourth, we examine convergent and concurrent validity by comparing the BPFSC-11 with other established batteries of inventories that assess areas such as resilience, self-efficacy, depression, social anxiety, anxiety sensitivity, and emotional dysregulation. Finally, we test measure invariance to ensure the stability of measure structure and meaning across five groups: sex assigned at birth, born in Norway, speaking primarily Norwegian at home, by treatment group, and longitudinally.

Materials and Method

Participants

This sample included 235 eligible first— and second-year high school students recruited from three senior high schools in Vestfold County, Norway. One hundred ninety-six (83.4%) participants participated in both assessment points within a 12-month interval using the same assessment battery. This sample of adolescents participated in an intervention study, with 124 participants in the intervention group and 72 participants in the control group. The intervention group received the School Resilience Program (Aune & Næss, 2025; Wolmer et al., 2011), while the control group received “treatment as usual.” Table 1 shows the demographic characteristics.

Table 1.

Demographic Characteristics of Participants. (N = 234).

Characteristic	Mean	SD	n	%
Age	16.79	1.86
Gender
Female			153	65.1
Male			82	34.9
School grade
VG1 (first year at high school/11^th year at school)			159	67.7
VG2 (second year at high school/12^th year at school)			71	30.2
Not reported			5	2.1
Living with adults
With mother and father, who live together			113	52.6
With mother and father, who do not live together (50/50)			16	7.7
Mostly with mother			44	18.7
Mostly with father			9	2.8
With mother and her new partner			21	8.9
With father and his new partner			4	1.7
With partner or friend			4	1.7
With other adults			8	3.4
Living alone			9	3.8
In institution			6	2.6
Not reported			1	0.4
Place of birth
Born in Norway			165	70.2
Not born in Norway			67	28.5
Not reported			3	1.3
Language use
Speaks Norwegian at home all or most days			180	76.6
Does not speak Norwegian at home at all or most days			47	20.0
Not reported			8	3.4
Family’s economic situation during last two years
Good all the time			81	34.5
Good most of the time			75	31.9
Neither good nor bad			45	19.1
Bad most of the time			19	8.1
Bad all the time			6	2.6
Not reported			9	3.8

Measures

The BPFSC-11 (Sharp et al., 2014) is an 11-item inventory that assesses borderline traits. Items are rated on a Likert-type scale of 1 (never true) to 5 (always true) (total score 11–55). It captures behaviors that reflect core BPD features (e.g., affective instability, identity problems, and negative relationships; Sharp et al., 2014). The BPFSC-11 has shown good criterion validity, internal consistency, test–retest reliability, and satisfactory Cronbach’s alpha coefficients of .79–.85 (Fossati et al., 2019; Sharp et al., 2014). Permission was obtained from the developer of the BPFSC-11, Carla Sharp, to translate it into Norwegian. The first author and Kyrre Svara translated it, with input from four adolescents as language consultants. Geir Pedersen and Øyvind Urnes, experts in assessing BPD, contributed to the final phase of developing the official Norwegian version. A licensed interpreter translated BPFSC-11 back into the source language, and Carla Sharp approved the back translation. In our sample, the BPFSC-11 demonstrated Cronbach’s alpha of .87.

Anxiety Sensitivity Index (ASI-3; Taylor et al., 2007). ASI is an 18-item self-report measure, rated on a four-point Likert scale (0 = “not at all” to 4 = “very much”). The higher the score, the more severe the level of anxiety sensitivity. The ASI-3 has shown good convergent, discriminant, and criterion-related validity (Taylor et al., 2007), and it is affected by age, education level, and gender. Taylor et al. (2007) reported Cronbach’s Alpha of .90 for the ASI-3. The present study observed an excellent Cronbach’s Alpha for the Inventory (α = .93).

Self-efficacy Questionnaire for Children (SEQ-C; Muris, 2001) is a 24-item self-report instrument measuring individual self-efficacy among adolescents using a five-point Likert scale in three main areas of self-efficacy. Xie et al. (2023) showed excellent internal consistency, good construct validity, and acceptable convergent validity for the SCQ-C. A total self-efficacy score is obtained by summing the scores across the responses to all 24 items. Tan and Chellappan (2018) reported an internal consistency of 0.89 from a study of Malaysian adolescents. The present study also observed excellent Cronbach’s alpha for the SEQ-C (α = 0.91).

Resilience Scale for Adolescents (READ) is a 28-item self-report questionnaire designed to assess the five protective resources: personal competence, social competence, structured style, family cohesion, and social resources to understand stress adaptation (Hjemdal et al., 2006). It has a five-point Likert scale. Higher scores indicate a higher level of protective traits. Assessing a large population-based sample of adolescents, Askeland et al. (2020) demonstrated adequate construct validity for the READ. Cronbach’s alpha, ranging from .88 to .94, has been reported (Askeland & Reedtz, 2015). This study used a 14-item abbreviated version of READ, revealing Cronbach’s alpha of .91.

The Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) is a 36-item self-report measure to assess emotion regulation. Items are rated on a scale of 1–5. Higher scores indicate more difficulty in emotion regulation. Gratz and Roemer (2004) have shown that DERS has strong construct validity. In the present study, four factors of the original DERS were included in the questionnaire booklet. They made up the DERS total score: nonacceptance of emotional responses, difficulty engaging in goal-directed behavior, lack of emotional awareness, and limited access to emotion regulation strategies. A total score was obtained by summing the items from the four subscales (ranging from 25 to 125). Cronbach’s alpha for the DERS total is 0.87.

The Social Phobia and Anxiety Inventory-23 (SPAI-23) (Roberson-Nay et al., 2007) is an abbreviated version of the original SPAI questionnaire (Turner et al., 1989). The SPAI-23 contains 23 items and includes two subscales—agoraphobia and social phobia—rated on a five-point Likert scale ranging from 0 to 4 (never to always). Roberson-Nay et al. (2007) reported strong factor loadings for all items, high correlations between the subscales of SPAI-23 and the original SPAI, and a Cronbach’s alpha of .95 for the SPAI-23 Social Phobia subscale. The present study observed an excellent Cronbach’s alpha coefficient for the SPAI-23 Social Phobia subscale (α = .96).

The Depression Anxiety Stress Scale (DASS-21; Lovibond & Lovibond, 1995) is a 21-item self-report measure that assesses three facets of negative emotion: depression, anxiety, and stress/tension. Each scale contains seven items on a four-point Likert scale. Brown et al. (1997) reported excellent internal consistency, reliability, and concurrent validity. We used the DASS-21 depression scale, demonstrating high internal consistency (Cronbach’s α = .90).

Statistical Analysis

We conducted descriptive analyses using SPSS Statistics for Windows (v. 27.0; IBM SPSS, Armonk, NY, USA). We assess and report concurrent and convergent validity using correlation estimates (Pearson product-moment) across all the included measurements. We expected a high positive correlation between BPFSC-11 and DERS, indicating convergent validity, whereas positive moderate correlations were expected between the BDFSC-11 and ASI, DASS-21, and the Spai-23 measure, showing concurrent validity. In contrast, negative moderate correlations were expected between BPFSC-11 and the READ and Self-efficacy scales. The means and standard deviations for the individual items in BPFSC-11 were reported, as well as the internal consistency (Cronbach’s alpha) of the overall BPFSC-11, and the various extracted factors. Additionally, we performed an exploratory factor analysis (EFA) for the sample examined at assessment point I and a CFA conducted one year later. A normed fit index (χ^2/df) close to 2 is acceptable and assesses how well the model fits the data given its complexity. The acceptable values for GFI (goodness-of-fit index), AGFI (adjusted goodness-of-fit index), CFI (comparative fit index), and IFI (incremental fit index) are.90 or greater. For RMSEA (root mean square error of approximation), <.05 and <.08 are considered excellent and acceptable fits, respectively (Tabachnick & Fidell, 2007). Measurement invariance was analyzed to assess consistency across five groups (sex assigned at birth, whether participants were born in Norway, whether Norwegian is the language spoken most at home, intervention group, and longitudinally). According to Chen’s guidelines for determining invariance (Chen, 2007), metric non-invariance is indicated by a change of ≥.010 in RMSEA and a change of ≤ −.005 in CFI, or a change of ≥.025 in SRMR when comparing configural and metric models. For scalar invariance, as determined by comparing the scalar and metric models, metric non-invariance is indicated by a change of ≥.010 in RMSEA and a change of ≥ −.005 in CFI, or a change of ≥.005 in SRMR.

Ethics

This research was performed according to the World Medical Association’s code of ethics (Declaration of Helsinki) for experiments involving humans. The study was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (REK, 2017/1938). The Norwegian Centre for Research Data (#53960) also approved the study. Written informed consent was obtained after the participants were provided with a complete study description.

Results

BPFSC-11 Descriptive Statistics and Internal Consistency Analyses

Table 2 shows the two factors’ mean, standard deviation, commonalities, and factor loadings for the BPFSC-11 individual items. The total score for the BPFSC-11 at assessment point I was 29.06 (SD = 9.15), while the mean score for the items was 2.65, with a range score of 1.79 and 3.19, respectively. Cronbach’s alpha for the BPFSC-11 full scale was 0.87. A 12-month test-retest correlation between the two sets was highly significant r = .71, p < .001 was found.

Table 2.

Means, Standard Deviation (SD), Factor Loadings, and Communalities for the EFA of the Borderline Personality Feature Scale for Children BPFSC-11.

BPFSC-11 Items	Mean	SD	Factors		Communalities
1. I feel very lonely	2.62	1.20	.744	−.117	.481
2. I want to let people know they’ve hurt me	2.72	1.37	.660	−.197	.345
3. My feelings are very strong	3.19	1.30	.541	.111	.365
4. There is something important missing	2.58	1.36	.625	.111	.472
5. I’m careless with things that are important	2.34	1.08	.305	.364	.335
6. People who were close have let me down	2.95	1.41	.701	.026	.510
7. I go back and forth between feelings	2.87	1.34	.828	.014	.698
8. I get into trouble do things without thinking	2.33	1.30	−.033	.774	.576
9. I worry that people I care about will leave	2.89	1.44	.809	−.019	.640
10. How I feel about myself changes a lot	2.83	1.30	.664	.055	.480
11. My friends and I are mean to each other	1.79	1.05	−.113	.504	.210

Note. Factor loadings are in boldface. Extraction method: Principal axis factoring. Rotation method: Promax. Factor 1, Abandonment, and Factor 2, Conflictual behavior. Communalities, Extraction.

Exploratory Factor Analysis

The BPFSC-11 questionnaire was subjected to EFA using the Principal Axis Factoring (PAF) extraction method and the ProMax oblique rotation, as it was expected that the items would be related conceptually. Communalities ranged from .21 to .70, as shown in Table 2. The Kaiser-Meyer-Olkin (KMO) test showed an adequacy index of .88. After considering multiple methods such as Kaiser’s criterion, Cattell’s scree plot, theoretical considerations, the previously reported factor analysis, and item inspection, it was concluded that the adequate solution was a two-factor structure.

Loading Patterns

In Table 2, the initial analysis showed that two factors accounted for 56.3% of the total variance. Factor 1 represented 44.1%, and Factor 2 represented 12.1% of the explained variance. The eigenvalues for Factors 1 and 2 were 4.86 and 1.34, respectively.

Factor 1 (Abandonment)

Factor 1 contained eight items (items 1, 2, 3, 4, 6, 7, 9, and 10), which appeared to measure identity difficulties/affective instability and abandonment (e.g., “How I feel about myself changes a lot”) and was named Abandonment. The factor loading varied from .54 to .83. Internal consistency (α = .88) was revealed for the Abandonment factor.

Factor 2 (Conflictual Behavior)

Factor 2 contained three items (items 5, 8, and 11), which appeared to measure hostile relations, hostility, and carelessness (e.g., “My friends and I are mean to each other”). This was called Conflictual behavior, although the factor loadings varied from .36 to .77. A moderate internal consistency of α = .56 was revealed for this factor.

Factor Correlations

The two factors were moderately correlated (r = .49), confirming the appropriateness of a Promax-rotated solution (Fabrigar et al., 1999).

Confirmatory Factor Analysis

Figure 1 depicts the CFA, demonstrating that most items reveal standardized correlations of.64 or higher. The exception is item 11, which correlates of .43 with Factor 2. This item also revealed a relatively low factor loading in the EFA (Table 2) and a low commonality (.21). This item was deleted from the model, and the CFA was re-estimated.

Figure 1.

The two-factor CFA model of the BPFSC-11 inventory shows the correlation factor and standardized regression coefficients.

Table 3 shows the fit indices for the CFA after item 11 is excluded. These indices, ranging from acceptable (RMSEA) to excellent (CFI), confirm a strong data fit to the model. Notably, a robust correlation of r = .80 was observed between the two factors, underscoring the model’s stability and relevance to the BPFSC-11.

Table 3.

Model Fit Indices for BPFSC-11.

Models	χ^2/df	Goodness-of-Fit Indices
Models	χ^2/df	GFI	AGFI	CFI	RMSEA	IFI
BPFSC-11	2.01	.94	.90	.96	.07	0.96
(10 items, two factors)

Note. BPFSC-11, Borderline Personality Features Scale for Children; χ^2/df, normed fit index; GFI, goodness-of-fit index; AGFI, adjusted goodness-of-fit index; CFI, comparative fit index; RMSEA, root mean square error of approximation; IFI, incremental fit index.

Convergent and Concurrent Validity of the BPFSC-11

Table 4 shows the correlations between the measures.

Table 4.

Person’s Correlations (r) Estimate Mean Scores (M) and Standard Deviations (SD) Across Various Scales.

Scales	n	M	SD	BPFSC-11	ASI	SEQ-C	READ	DERS	SPAI 23	DASS Dep
BPFSC-11	231	29.06	9.15	1
ASI	230	24.69	14.11	.605**	1
SEQ-C	230	76.91	16.11	−.470**	−.380**	1
READ	231	3.91	.78	−.472**	−.345**	.479**	1
DERS	230	95.30	25.51	.738**	.536**	−.622**	−.498**	1
SPAI23	229	39.28	14.64	.449**	.498**	−.448**	−.409**	−.478**	1
DASS Dep	230	1.93	1.24	.654**	.498**	−.507**	−.518**	−.704**	.479**	1

Note. BPFSC-11, Borderline Personality Features Scale for Children; ASI, Anxiety Sensitivity Index; SEQ-C, Self-efficacy Questionnaire for Children; READ, Resilience Scale for Adolescents; DERS, Difficulties in Emotion Regulation Scale; SPAI23, Social Phobia and Anxiety Inventory; DASS dep, Depression Anxiety Stress Scale (depression subscale).

**Correlation is significant at .01 level (two-tailed).

*Correlation is significant at .05 level (two-tailed).

As expected, the BPFSC-11 showed moderate to strong correlations with the included measurements. We found a positive, strong association of .74 between the BPFSC-11 and the DERS, indicating concurrent validity, which is similar to what was reported by Sharp et al. (2014) for these two scales. Moderate correlations were shown between the BPFSC-11 and DASS-21, ASI, and SPAI-23, with the strongest association between borderline features and depressive symptoms. We also discovered significant negative correlations of.47 between the BPFSC-11 and both self-efficacy and resilience.

Measurement Invariance (BPFSC-11)

Invariance testing metrics are in Table 5 (sex assigned at birth), Table 6 (Born in Norway), Table 7 (Speak Norwegian at Home), and Table 8 (Treatment Group). A three-step approach was used to test for measurement invariance. First, the highest factor loading on each factor was set to 1 with factor means in each group set to zero (configural invariance). A model has configural invariance if model fit is adequate. Next, all factor loadings were constrained to be equal across the five groups (sex assigned at birth, born in Norway, speak Norwegian at home, treatment group and longitudinally) to test metric invariance; changes in fit indices were compared to the configural model. Lastly, factor loadings and intercepts were constrained to be equivalent across the five groups and scalar invariance was tested by comparing this model to the metric invariance model. Chen (2007) notes χ² tests for invariance are sensitive to sample size and violations of normality, and consequently small discrepancies may result in model rejection. Thus, in larger samples with unequal group size and/or non-normal data, Chen outlined guidelines for determining invariance relying on other model fit indices (Chen, 2007). Specifically, (1) a change of ≤ −.005 in CFI and a change of ≥.010 in RMSEA OR (2) a change of ≥.025 in SRMR, when comparing configural to metric models, would indicate metric non-invariance. For scalar invariance (comparing the scalar to metric model) (1) a change in CFI ≥ −.005, and a change in RMSEA of ≥.010 OR (2) a change in SRMR ≥.005 indicates scalar non-invariance. The configural invariance model examines if the factor structure is similar across the five groups.

Table 5.

Measurement Invariance by Sex Assigned at Birth.

Variable	RMSEA	CFI	TLI	SRMR
Configural invariance	.098	.893	.864	.066
Metric invariance	.090	.901	.885	.068
Δ from Configural	−.008	+.008	+.021	+.002
Scalar invariance	.094	.882	.875	.079
Δ from metric	+.004	−.019	−.010	+.011

Table 6.

Measurement Invariance by Born in Norway Status.

Variable	RMSEA	CFI	TLI	SRMR
Configural invariance	.089	.916	.892	.062
Metric invariance	.091	.903	.888	.086
Δ from Configural	+.002	−.013	−.004	+.024
Scalar invariance	.098	.877	.870	.089
Δ from metric	+.007	−.026	−.018	+.003

Table 7.

Measurement Invariance by Speaking Norwegian at Home.

Variable	RMSEA	CFI	TLI	SRMR
Configural invariance	.087	.919	.896	.069
Metric invariance	.094	.897	.881	.092
Δ from Configural	+.007	−.022	−.015	+.023
Scalar invariance	.096	.881	.874	.095
Δ from metric	+.002	−.016	−.007	+.003

Table 8.

Measurement Invariance by Treatment Group (Intervention or Control).

Variable	RMSEA	CFI	TLI	SRMR
Configural invariance	.098	.906	.880	.075
Metric invariance	.092	.908	.894	.073
Δ from Configural	−.004	+.002	−.014	−.002
Scalar invariance	.088	.909	.904	.077
Δ from metric	−.004	+.001	+.010	+.004

For sex assigned at birth (Table 5) the model fit the data well with significant factor loadings for all items, χ2(86) = 181.418, p < .001, CFI = .893, RMSEA = .098 (90% CI = .078, .118), SRMR = .066. Constraining factor loadings to be equivalent across sex did not result in changes to fit, relative to the configural model, indicating metric invariance. Finally, there was no substantial decreases in model fit, relative to the metric model, after constraining factor loadings and item intercepts to be identical and allowing factor means to vary across sex, indicating no scalar variance.

For the variable “whether participants were born in Norway” (Table 6), the model fit the data well with significant factor loadings for all items, χ2(86) = 164.695, p < .001, CFI = .916, RMSEA = .089 (90% CI = .069, .110), SRMR = .062. Constraining factor loadings to be equivalent across status of being born in Norway did not result in changes to fit, relative to the configural model, indicating metric invariance. Finally, there was no substantial decreases in model fit, relative to the metric model, after constraining factor loadings and item intercepts to be identical and allowing factor means to vary across status of birth in Norway, indicating no scalar variance.

For whether participants speak Norwegian primarily at home (Table 7) the model fit the data well with significant factor loadings for all items, χ2(86) = 159.553, p < .001, CFI = .919, RMSEA = .087 (90% CI = .066, .108), SRMR = .069. Constraining factor loadings to be equivalent across speaking Norwegian at home did not result in changes to fit, relative to the configural model, indicating metric invariance. Finally, there was no substantial decreases in model fit, relative to the metric model, after constraining factor loadings and item intercepts to be identical and allowing factor means to vary across speaking Norwegian at home, indicating no scalar variance.

For participants who completed both time points (n = 196), an analysis of measurement invariance was conducted by treatment group (Table 8). The model fit the data well with significant factor loadings for all items, χ²(86) = 166.467, p < .001, CFI = .906, RMSEA = .098 (90% CI = .076, .120), SRMR = .075. Constraining factor loadings to be equivalent across treatment group did not result in changes to fit, relative to the configural model, indicating metric invariance. Finally, there was no substantial decrease in model fit, relative to the metric model, after constraining factor loadings and item intercepts to be identical and allowing factor means to vary across treatment groups, indicating no scalar variance.

For longitudinal invariance across time (Table 9), the model revealed variance across time, indicating instability of the measure across timepoints. The model fit the data χ2(197) = 342.777, p < .001, CFI = .867, RMSEA = .076 (90% CI = .063, .090), SRMR = .080. Constraining factor loadings to be equivalent across time resulted in changes to fit, relative to the configural model, indicating metric variance. Finally, there was a substantial decrease in model fit, relative to the metric model, after constraining factor loadings and item intercepts to be identical and allowing factor means to vary across time, indicating scalar variance.

Table 9.

Longitudinal Measurement Invariance by Time 1 and Time 2.

Variable	RMSEA	CFI	TLI	SRMR
Configural invariance	.076	.867	.844	.080
Metric invariance	.078	.853	.836	.094
Δ from Configural	−.002	+.014*	+.008	+.014*
Scalar invariance	.078	.850	.839	.094
Δ from metric	+.000*	−.003*	+.003	+.000*

* = variance observed.

Taken together, the Norwegian version of the BPFSC-11 demonstrated strong internal consistency in our adolescent sample, with a Cronbach’s alpha of .87. This reliability estimate closely parallels findings from previous studies in European adolescent populations, including Fossati et al. (2019) in Italy (α = .78) and Calvo et al. (2023) in Spain (α = .86). The 12-month test-retest correlation in our sample was r = .71, which is notably higher than the six-month test-retest correlation of r = .50 reported by Fossati et al. (2019). The mean BPFSC-11 score in our sample (M = 29.06, SD = 9.15) was comparable to that reported by Fossati et al. (2019; M = 28.08, SD = 6.56), while Calvo et al. (2023) observed mean scores of 35.8 (SD = 7.9) in a BPD group, 27.7 (SD = 6.9) in an ADHD group, and 22.5 (SD = 8.3) in a control group. Consistent with Fossati et al. (2019), the best-fitting model in our sample was the two-factor model, with factor loadings ranging from .36 to .83, compared to .14 to .70 in Fossati et al. (2019). The same three items (5, 8, and 11) loaded on the same factor (factor 2) in both studies. Calvo et al. (2023) also identified two factors (Factor 1 item; 1, 4, 6, 7, 9, and 10: Factor 2; items 5, 8, and 11) with higher factor loadings (.64–.84) than those reported by Fossati et al. (2019), and a similar pattern to our findings. However, in contrast to our results, items 2 and 3 in Calvo et al. (2023) loaded similarly and above .40 in both factors, and three items (4, 7, and 10) from factor 1 also loaded above .40 on factor 2. Additionally, as we revealed in our study, gender invariance was confirmed in the Fossati et al. (2019) study.

Discussion

EFA indicates that two factors appeared to underline the 11 items (1. Abandonment; 2. Conflictual behavior). The significance of this solution was confirmed using a CFA. The high correlation between borderline personality features (BPFSC-11) and emotional dysregulation (DERS) indicates adequate convergent validity. Furthermore, correlation analyses between BPFSC-11 and measures assessing depression, anxiety sensitivity, and social anxiety disorder symptoms showed concurrent validity for the BPFSC-11. Compared with established measures of self-efficacy and resilience, the BPFSC-11 showed a moderate and negative correlation.

Upon inspection of the items, it was found that the mean score varies from 1.79 (item 11) to 3.19 (item 3), with an average score of 2.65. Fossati et al. (2019) reported similar results. Vanwoerden et al. (2019) reported a mean score of 2.31 and item scores ranging from 1.98 (item 11) to 2.73 (item 3) in a study using a community sample of 755 American adolescents. The differences in mean scores may be due to sample characteristics, such as age and gender distribution, as well as cultural differences between the US and Europe. Fossati et al. (2019) and Calvo et al. (2023) reported a two-factor model solution similar to the results of this study. However, the Fossati et al. (2019) and Calvo et al. (2023) studies differed in the items loaded on Factor 1. In contrast, comparing the two factors in this study with those of Calvo et al. (2023) revealed almost identical item distributions. On the other hand, in our study, all items were accurately distributed across two factors. In contrast, in the Calvo et al. (2023) study, only nine items were included in the final model. Moreover, in that study, most items loaded heavily on both factors.

We also included established psychometrically sound instruments to examine the concurrent validity of the BPFSC-11. According to the DSM-5 (APA, 2013), the essential feature of BPD is a pervasive pattern of affective instability. High and statistically significant correlations between borderline personality features (BPFSC-11) and emotional dysregulation (DERS) demonstrate adequate convergent validity. The results are almost identical to those reported by Sharp et al. (2014) (r = .71) but significantly higher than those shown by Somma et al. (2017) (r = .55), combining the same two instruments. Anxiety sensitivity is a stable individual difference characteristic that involves a propensity to develop beliefs that anxiety-related symptoms will have negative, destructive somatic, cognitive, and social consequences (Gratz & Roemer, 2008; Reiss, 1991). Gratz and Roemer (2008) and Doyle et al. (2022) have shown that anxiety sensitivity distinguishes outpatients with BPD from outpatients with no personality disorders. Doyle et al. (2022) showed that anxiety sensitivity predicted BPD over and beyond emotion dysregulation of impulsivity. Further, BPD and major depressive disorder commonly co-occur (Yoshimatsu & Palmer, 2014) and have high comorbidity with satisfaction with life and depressive symptoms in adolescence (Winograd et al., 2008). Thus, a correlation of .65, explaining 42% shared variance between borderline disorder features (BPFSC-11) and depressive symptoms (DASS), indicates a high association between these two scales. A moderate and positive correlation was obtained between borderline personality features and social anxiety disorder symptoms (SPAI-23). For instance, Gutz et al. (2016) have shown that individuals with BPD and SAD symptoms specifically show a higher increase in self-focused negative emotions and propensity toward avoidance.

Burghart et al. (2023) conducted a comprehensive analysis of the latent structure of the Emotion Regulation Questionnaire (ERQ) for adults, encompassing a diverse sample of 11,288 individuals from 29 countries. Their findings suggest that emotion regulation strategies may not converge consistently across cultural contexts. This observation is particularly significant given that impaired emotion regulation is considered a fundamental characteristic of BPD, as outlined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR; APA, 2022). The implications of these findings underscore the critical importance of rigorously assessing and evaluating the psychometric properties of BPD scales across diverse cross-cultural settings. In an increasingly multicultural global landscape, psychological inventories demonstrating measurement invariance across various groups are paramount, as they provide evidence of a measure’s consistency and reliability.

Given observed measurement invariance across the factors for key demographic and contextual groups, namely sex assigned at birth, nativity (born in Norway), primary language spoken at home, and treatment condition, this supports the methodological robustness and cross-group applicability of the Norwegian version of the scale, reinforcing its value for comparative assessments in diverse populations. However, while these findings affirm the scale’s structural validity across groups, a deeper examination of the stability of these factors over time is warranted. The observed longitudinal variance suggests that the factor structure may not remain entirely consistent across different time points. This fluctuation could be attributed to two primary influences: the developmental nature of adolescence, a period characterized by rapid emotional, cognitive, and social changes, and the potential effects of clinical interventions, which may alter the way symptoms manifest or are reported over time. While reliable for cross-sectional comparisons among different groups, caution is warranted when interpreting changes over time. Future research should aim to clarify whether the observed longitudinal differences reflect true changes in borderline personality features or shifts in the measurement properties of the scale.

Limitations and Future Directions

This study has both strengths and weaknesses. It did not include a clinical interview to examine the diagnostic validity of the BPFSC-11, and another self-report inventory also intended to examine borderline features. As with all self-report instruments, there is a risk that social desirability bias may influence participants’ responses. Adolescents may underreport symptoms due to stigma or a desire to conform to perceived social norms. This potential bias could affect the accuracy of symptom reporting and, consequently, the scale’s psychometric properties. Moreover, cultural factors may have influenced the interpretation of certain items. Cultural differences in understanding or accepting risk-taking behaviors across various cultural settings might exist. Such nuances underscore the importance of cultural context in developing, translating, and adapting psychological measures. Moreover, while we followed established guidelines for the translation and cross-cultural adaptation of the BPFSC-11, including forward and back translation, expert review, and consultation with adolescents, some limitations remain. Despite these efforts, there is a possibility that specific cultural nuances or idiomatic expressions relevant to Norwegian adolescents may not have been fully captured. Furthermore, while back-translation and approval by the original developer help ensure fidelity to the source content, subtle differences in how symptoms are expressed or understood across cultures may impact the scale’s cultural validity. Future studies should examine the scale’s measurement invariance and cultural appropriateness in diverse Norwegian populations. Future research should incorporate qualitative approaches, such as cognitive interviewing, to further explore how Norwegian adolescents interpret specific items and to refine culturally sensitive content. Nonetheless, examining a sample of more than 200 using an 11-item inventory is sufficient (Field, 2013). However, a larger sample and a more balanced female-to-male ratio would have been preferable, specifically when assessing measurement invariance across various groups and genders. Conducting a CFA one year later with a sample from the same population strengthens the theoretical concept of the BPFSC-11 and confirms the model revealed by the EFA. Additionally, comparing well-recognized and psychometrically sound inventories increases confidence in the results, similar to previous works (Diotaiuti, et al., 2020, 2021). The statistical property of the measurement invariance indicates that the same underlying construct is being measured across groups and time.

This study makes an innovative contribution by providing the first validated Norwegian version of the BPFSC-11, expanding the availability of Evidence-Based assessment tools for borderline personality features in Norwegian adolescents. The scale’s strong psychometric properties support its use in both clinical and research settings, facilitating early identification and intervention for borderline personality pathology in Scandinavian youth. Future research should focus on several key areas. Larger longitudinal studies are needed to evaluate the predictive validity and sensitivity to change of the BPFSC-11, especially in clinical populations. Additionally, further exploration of the scale’s discriminant validity in relation to other personality disorders is essential, which may include comparisons with structured clinical interviews or other dimensional measures. Finally, cross-cultural validation efforts in diverse Norwegian subpopulations would enhance the scale’s generalizability across different cultural and demographic groups.

Footnotes

Acknowledgements

We thank all the participants in this study, Vestfold County, for their cooperation and support throughout the study process, and Kyrre Svarva for his support throughout the completion of this research.

Author Contributions

Tore Aune and Sigrid Flatås Aune contributed throughout the manuscript’s production phases. Roselyn Peterson conducted the measurement invariance testing and contributed to the manuscript’s writing. All three authors have read and approved the submitted manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Considerations

This research was performed using the World Medical Association’s code of ethics (Declaration of Helsinki) for experiments involving humans. It was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (2017/1938). The Norwegian Centre for Research Data (#53960) also approved the study’s conduct.

Informed Consent

Written informed consent regarding participation and publishing was obtained after a complete study description was provided to the participants and their parents.

ORCID iDs

Tore Aune

Roselyn Peterson

Data Availability Statement

The data is available upon request from the editor, reviewers, or others in the editor’s office.*

Author Biographies

Tore Aune, PhD, is a clinical psychologist working at the Norwegian Directorate of Children, Youth and Family Affairs. Aune holds a position as Director at Steinkjer Psychologist Center. He completed his clinical psychology degree and PhD at the Norwegian University of Science and Technology (NTNU) in Trondheim. Tore has also worked at Nord University and NTNU, contributing to research and teaching in the field of psychology.In 2009, he received the Norwegian Psychological Association’s award for the best doctoral thesis. Between 2018 and 2019, Tore served as a visiting professor at the RESTORES trauma clinic at the University of Central Florida in the United States. Dr. Aune’s research is at the forefront of understanding and addressing mental health challenges among children and adolescents, with a particular emphasis on the prevention and treatment of social anxiety disorders. He is deeply committed to translating research into practice, regularly lecturing and supervising practitioners who work in specialized and acute foster care settings across Norway.

Roselyn Peterson, PhD, is clinical faculty and the research director of the Clinical Psychology doctoral program at Bryant University. She completed her clinical psychology degree at the University of Central Florida, her clinical internship at the Warren Alpert Medical School of Brown University and went on to complete two years of postdoctoral training at the Center for Alcohol and Addiction Studies in the School of Public Health at Brown University. Her research focuses on the intersection of alcohol use and sexual violence, with a particular emphasis on protective behavioral strategies (PBS).

Sigrid Flatås Aune is a clinical psychology student at the Department of Psychology, Norwegian University of Science and Technology (NTNU). Her academic interests include research methodology, clinical and cognitive psychology. She has previously contributed to the research article “Underlying processes in the Norwegian universal preventive program for social anxiety”, reflecting her engagement in preventive mental health research.

References

Allman

Barkauskienė

Gervinskaitė-Paulaitienė

Sharp

(2025). Evaluation of the borderline personality features scale for children and life satisfaction in Lithuanian adolescents. European Journal of Psychological Assessment. Advance online publication. https://doi.org/10.1027/1015-5759/a000884

American Psychiatric Association . (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596

American Psychiatric Association . (2022). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425787

Askeland

Reedtz

(2015). Måleegenskaper ved den norske versjonen av Resilience Scale for Adolescents (READ). PsykTestBarn, 5(1), 1–12. https://doi.org/10.21337/0040

Askeland

K. G.

Hysing

Sivertsen

Breivik

(2020). Factor structure and psychometric properties of the resilience scale for adolescents (READ). Assessment, 27(7), 1575–1587. https://doi.org/10.1177/1073191119832659

Aune

Ness

(2025). Building resilience in teachers and students: A qualitative study. Tidsskrift for Norsk psykologforening. https://doi.org/10.52734/ZTTY6413

Bibi

Kazmi

S. F.

(2021). Urdu translation and validation of 11-item measure to assess borderline personality features in Pakistani adolescents. Sage Open, 11(1), Article 158244020986157. https://doi.org/10.1177/2158244020986157

Brown

T. A.

Chorpita

B. F.

Korotitsch

Barlow

D. H.

(1997). Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behaviour Research and Therapy, 35(1), 79–89. https://doi.org/10.1016/s0005-7967(96)00068-x

Burghart

Sahm

A. H. J.

Mier

(2023). Investigating measurement invariance of the Emotion Regulation Questionnaire-8 (ERQ-8) across 29 countries. Current Psychology, 1–7. Advance online publication. https://doi.org/10.1007/s12144-022-04220-6

10.

Calvo

Marin

J. L.

Vidal

Sharp

Duque

J. D.

Ramos-Quiroga

J. A.

Ferrer

(2023). Discrimination of Borderline Personality Disorder (BPD) and Attention-Deficit/Hyperactivity Disorder (ADHD) in adolescents: Spanish version of the Borderline Personality Features Scale for Children-11 Self-Report (BPFSC-11) preliminary results. Borderline Personality Disorder and Emotion Dysregulation, 10(1), 15. https://doi.org/10.1186/s40479-023-00223-2

11.

Chanen

A. M.

Kaess

(2012). Developmental pathways to borderline personality disorder. Current Psychiatry Reports, 14(1), 45–53. https://doi.org/10.1007/s11920-011-0242-y

12.

Chen

F. F.

(2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/10.1080/10705510701301834

13.

Crick

N. R.

Murray-Close

Woods

(2005). Borderline personality features in childhood: A short-term longitudinal study. Development and Psychopathology, 17(4), 1051–1070. https://doi.org/10.1017/0S0954579405050492

14.

Diotaiuti

Valente

Mancone

Grambone

(2020). Psychometric properties and a preliminary validation study of the Italian brief version of the Communication Styles Inventory (CSI-B/I). Frontiers in Psychology, 11, 1421. https://doi.org/10.3389/fpsyg.2020.01421

15.

Diotaiuti

Valente

Mancone

Grambone

Chirico

(2021). Metric goodness and measurement invariance of the Italian brief version of interpersonal reactivity index: A study with young adults. Frontiers in Psychology, 12, Article 773363. https://doi.org/10.3389/fpsyg.2021.773363

16.

Doyle

J. N.

Watt

M. C.

Cohen

J. N.

Couture

M. E.

Smith

M. M.

(2022). Relations between anxiety sensitivity and attachment in outpatients with borderline personality disorder. Journal of Personality Disorders, 36(5), 606–622. https://doi.org/10.1521/pedi.2022.36.5.606

17.

Fabrigar

L. R.

Wegener

D. T.

MacCallum

R. C.

Strahan

E. J.

(1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299. https://doi.org/10.1037/1082-989x.4.3.272

18.

Field

(2013). Discovering statistics using IBM SPSS statistics. Sage.

19.

Fonagy

Bateman

A. W.

(2016). Adversity, attachment, and mentalizing. Comprehensive Psychiatry, 64(1), 59–66. https://doi.org/10.1016/j.comppsych.2015.11.006

20.

Fossati

Sharp

Borroni

Somma

(2019). Psychometric properties of the Borderline Personality Features Scale for Children-11 (BPFSC-11) in a sample of community dwelling Italian adolescents. European Journal of Psychological Assessment, 35(1), 70–77. https://doi.org/10.1027/1015-5759/a000377

21.

Gratz

K. L.

(2001). Measurement of deliberate self-harm: Preliminary data on the deliberate self-harm inventory. Journal of Psychopathology and Behavioral Assessment, 23(4), 253–263. https://doi.org/10.1023/A:1012779403943

22.

Gratz

K. L.

Roemer

(2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. Journal of Psychopathology and Behavioral Assessment, 26(1), 41–54. https://doi.org/10.1023/B:JOBA.0000007455.08539.94

23.

Gratz

K. L.

Roemer

(2008). The relationship between emotion dysregulation and deliberate self-harm among female undergraduate students at an urban commuter university. Cognitive Behaviour Therapy, 37(1), 14–25. https://doi.org/10.1080/16506070701819524

24.

Guillén

Tormo

M. E.

Fonseca-Baeza

Botella

Baños

García-Palacios

Marco

J. H.

(2021). Resilience as a predictor of quality of life in participants with borderline personality disorder before and after treatment. BMC Psychiatry, 21(1), 305. https://doi.org/10.1186/s12888-021-03312-0

25.

Gutz

Roepke

Renneberg

(2016). Cognitive and affective processing of social exclusion in borderline personality disorder and social anxiety disorder. Behaviour Research and Therapy, 87, 70–75. https://doi.org/10.1016/j.brat.2016.08.020

26.

Haltigan

J. D.

Vaillancourt

(2016). Identifying trajectories of borderline personality features in adolescence: Antecedent and interactive risk factors. Canadian Journal of Psychiatry, 61(3), 166–175. https://doi.org/10.1177/0706743715625953

27.

Heiland

A. M.

Veilleux

J. C.

(2021). Severity of personality dysfunction predicts affect and self-efficacy in daily life. Personality Disorders: Theory, Research, and Treatment, 12(6), 560–569. https://doi.org/10.1037/per0000470

28.

Hjemdal

Friborg

Stiles

T. C.

Martinussen

Rosenvinge

J. H.

(2006). A new scale for adolescent resilience: Grasping the central protective resources behind healthy development. Measurement and Evaluation in Counseling and Development, 39(2), 84–96. https://doi.org/10.1080/07481756.2006.11909791

29.

Laczkovics

Czernin

Bründlmayer

Zeiler

Bangerl

Prause

Plener

P. L.

Doering

Blüml

Carlitscheck

Bender

Krischer

(2025). Assessment of personality disorders in adolescents: A clinical validity and utility study of the structured interview of personality organization (STIPO). Child and Adolescent Psychiatry and Mental Health, 19(1), 49. https://doi.org/10.1186/s13034-025-00901-9

30.

Leichsenring

Fonagy

Heim

Kernberg

O. F.

Leweke

Luyten

Salzer

Spitzer

Steinert

(2024). Borderline personality disorder: A comprehensive review of diagnosis and clinical presentation, etiology, treatment, and current controversies. World Psychiatry: Official Journal of the World Psychiatric Association (WPA), 23(1), 4–25. https://doi.org/10.1002/wps.21156

31.

Lovibond

P. F.

Lovibond

S. H.

(1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck depression and anxiety inventories. Behaviour Research and Therapy, 33(3), 335–343. https://doi.org/10.1016/0005-7967(94)00075-u

32.

Miller

A. L.

Muehlenkamp

J. J.

Jacobson

C. M.

(2008). Fact or fiction: Diagnosing borderline personality disorder in adolescents. Clinical Psychology Review, 28(6), 969–981. https://doi.org/10.1016/j.cpr.2008.02.004

33.

Muris

(2001). A brief questionnaire for measuring self-efficacy in youths. Journal of Psychopathology and Behavioral Assessment, 23(3), 145–149. https://doi.org/10.1023/A:1010961119608

34.

Reiss

(1991). Expectancy model of fear, anxiety, and panic. Clinical Psychology Review, 11(2), 141–153. https://doi.org/10.1016/0272-7358(91)90092-9

35.

Roberson-Nay

Strong

D. R.

Nay

W. T.

Beidel

D. C.

Turner

S. M.

(2007). Development of an abbreviated Social Phobia and Anxiety Inventory (SPAI) using item response theory: The SPAI-23. Psychological Assessment, 19(1), 133–145. https://doi.org/10.1037/1040-3590.19.1.133

36.

Sharp

Cervantes

B. R.

(2023). Maladaptive self- and interpersonal functioning increments general psychiatric severity in the association with adolescent personality pathology. Children, 10(1), 120. https://doi.org/10.3390/children10010120

37.

Sharp

Steinberg

Temple

Newlin

(2014). An 11-item measure to assess borderline traits in adolescents: Refinement of the BPFSC using IRT. Personality disorders, 5(1), 70–78. https://doi.org/10.1037/per0000057

38.

Somma

Sharp

Borroni

Fossati

(2017). Borderline personality disorder features, emotion dysregulation and non-suicidal self-injury: Preliminary findings in a sample of community-dwelling Italian adolescents. Personality and Mental Health, 11(1), 23–32. https://doi.org/10.1002/pmh.1353

39.

Tabachnick

B. G.

Fidell

L. S.

(2007). Experimental designs using ANOVA (Vol. 724). Thomson/Brooks/Cole.

40.

Tan

S. K.

Chellappan

(2018). Assessing the validity and reliability of the Self-Efficacy Questionnaire for Children (SEQ–C) among Malaysian adolescents: Rasch model analysis. Measurement and Evaluation in Counseling and Development, 51(3), 179–192. https://doi.org/10.1080/07481756.2018.1435192

41.

Taylor

Zvolensky

M. J.

Cox

B. J.

Deacon

Heimberg

R. G.

Ledley

D. R.

Abramowitz

J. S.

Holaway

R. M.

Sandin

Stewart

S. H.

Coles

Eng

Daly

E. S.

Arrindell

W. A.

Bouvard

Cardenas

S. J.

(2007). Robust dimensions of anxiety sensitivity: Development and initial validation of the anxiety sensitivity index-3. Psychological Assessment, 19(2), 176–188. https://doi.org/10.1037/1040-3590.19.2.176

42.

Turner

S. M.

Beidel

D. C.

Dancu

C. V.

Stanley

M. A.

(1989a). An empirically derived inventory to measure social fears and anxiety: The Social Phobia and Anxiety Inventory. Psychological Assessment: Journal of Consulting and Clinical Psychology, 1(1), 35–40. https://doi.org/10.1037/1040-3590.1.1.35

43.

Turner

S. M.

Stanley

M. A.

Beidel

D. C.

Bond

(1989b). The social phobia and anxiety inventory: Construct validity. Journal of Psychopathology and Behavioral Assessment, 11(3), 221–234. https://doi.org/10.1007/BF00960494

44.

Vanwoerden

Garey

Ferguson

Temple

J. R.

Sharp

(2019). Borderline Personality Features Scale for Children-11: Measurement invariance over time and across gender in a community sample of adolescents. Psychological Assessment, 31(1), 114–119. https://doi.org/10.1037/pas0000640

45.

Winograd

Cohen

Chen

(2008). Adolescent borderline symptoms in the community: Prognosis for functioning over 20 years. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 49(9), 933–941. https://doi.org/10.1111/j.1469-7610.2008.01930.x

46.

Wolmer

Hamiel

Laor

(2011). Preventing children’s posttraumatic stress after disaster with teacher-based intervention: A controlled study. Journal of the American Academy of Child & Adolescent Psychiatry, 50(4), 340–348. https://doi.org/10.1016/j.jaac.2011.01.002

47.

Xie

Shlonsky

Harrigan

(2023). Psychometric evaluation of the Self-Efficacy Questionnaire for Children (SEQ-C): Validation among Chinese children and adolescents. Current Psychology, 42(21), 18405–18418. https://doi.org/10.1007/s12144-022-03046-6

48.

Yoshimatsu

Palmer

(2014). Depression in patients with borderline personality disorder. Harvard Review of Psychiatry, 22(5), 266–273. https://doi.org/10.1097/HRP.0000000000000045

Psychometric Properties and Measurement Invariance of the Borderline Personality Features Scale for Children (BPFSC-11)

Abstract

Keywords

Introduction

Materials and Method

Participants

Measures

Statistical Analysis

Ethics

Results

BPFSC-11 Descriptive Statistics and Internal Consistency Analyses

Exploratory Factor Analysis

Loading Patterns

Factor 1 (Abandonment)

Factor 2 (Conflictual Behavior)

Factor Correlations

Confirmatory Factor Analysis

Convergent and Concurrent Validity of the BPFSC-11

Measurement Invariance (BPFSC-11)

Discussion

Limitations and Future Directions

Footnotes

Acknowledgements

Author Contributions

Declaration of Conflicting Interests

Funding

Ethical Considerations

Informed Consent

ORCID iDs

Data Availability Statement

Author Biographies

References