POSTPRINT VERSION. The final version is published here : Gamache, D., Savard, C., Leclerc, P. et Côté, A. (2019).

Introducing a short self-report for the assessment of DSM–5 level of personality functioning for personality disorders:
The Self and Interpersonal Functioning Scale. Personality Disorders: Theory, Research, and Treatment (Vol. 10, p. 438-447): Educational Publishing Foundation. doi: http://dx.doi.org/10.1037/per0000335

Introducing a Short Self-Report for the Assessment of DSM-5 Level of Personality Functioning

for Personality Disorders: The Self and Interpersonal Functioning Scale

Dominick Gamache, Claudia Savard, Philippe Leclerc, and Alexandre Côté

Université du Québec à Trois-Rivières

Author note

Dominick Gamache, Université du Québec à Trois-Rivières, and CERVO Brain Research

Centre; Claudia Savard, Université Laval, and CERVO Brain Research Centre; Philippe Leclerc,

Université du Québec à Trois-Rivières; Alexandre Côté, Université du Québec à Trois-Rivières

Correspondence concerning this article should be addressed to Dominick Gamache,

Université du Québec à Trois-Rivières, Département de psychologie, C.P. 500, Trois-Rivières,

Qc, Canada, G9A 5H7. Phone: (819) 376-5011 # 3530. Fax: (819) 376-5195. E-mail:

dominick.gamache@uqtr.ca.

©American Psychological Association, 2019. This paper is not the copy of record and may not exactly replicate the authoritative document published in the APA journal. Please do not copy or cite without author's permission. The final
article is available, upon publication, at: http://dx.doi.org/10.1037/per0000335
SELF AND INTERPERSONAL FUNCTIONING SCALE 2

Submitted: September 8, 2018

Revision submitted: January 11, 2019

© 2019, American Psychological Association. This paper is not the copy of record and may

not exactly replicate the final, authoritative version of the article. Please do not copy or cite

without authors' permission. The final article will be available, upon publication, via its

DOI: 10.1037/per0000335
SELF AND INTERPERSONAL FUNCTIONING SCALE 3

Abstract

In the present study, we report on the development and validation of the Self and Interpersonal

Functioning Scale (SIFS), a 24-item self-report questionnaire designed to assess the four core

elements of personality pathology (Identity, Self-direction, Empathy, and Intimacy) from the

DSM-5 Level of Personality Functioning (LPF) for personality disorders (PDs). Participants

from a community sample (n = 280) and patients from a specialized treatment facility for PDs (n

= 106) were included in the validation sample. Overall, the SIFS showed sound psychometric

properties. A second-order factor solution, which consisted of the four LPF elements and an

overarching personality pathology factor, showed the best fit indices. The four SIFS elements

showed a well-differentiated and conceptually meaningful pattern of associations with related

constructs. In light of these results, the SIFS should be considered as a promising, concise

measure of Criterion A for clinical screening and research purposes. Its relative strengths and

limitations in contrast with other existing self-report measures of Criterion A are discussed.

Keywords: Alternative DSM-5 Model for Personality Disorders; personality disorder;

Criterion A; test development; self-report measure.

SELF AND INTERPERSONAL FUNCTIONING SCALE 4

Introducing a Short Self-Report for the Assessment of DSM-5 Level of Personality Functioning

for Personality Disorders: The Self and Interpersonal Functioning Scale

Even without an “official” status, the Alternative Model for Personality Disorders

(AMPD), presented in Section III (Emerging Measures and Models) of the DSM-5 (American

Psychiatric Association, 2013) has generated considerable clinical and research interest over the

past few years (Waugh et al., 2017). The AMPD covers core difficulties in personality

functioning (Criterion A) in conjunction with pathological personality traits (Criterion B); it also

retains six categorical PDs, which can be diagnosed on the basis of Criteria A and B.

Criterion A is operationalized by the Level of Personality Functioning Scale (LPFS;

Bender, Morey, & Skodol, 2011), a clinician-rated tool which assesses the level of personality

pathology on a five-point scale, based on core impairments in self- and interpersonal functioning.

Self-functioning includes two elements, Identity and Self-direction, while Interpersonal

functioning refers to Empathy and Intimacy. Criterion A bridges important theoretical PD

formulations from various paradigms (e.g., psychodynamic, interpersonal, and social-cognitive),

incorporating structural elements, developmental processes, and personality dynamics relevant in

contemporary PD formulations (Bender et al., 2011; Waugh et al., 2017).

For pathological personality traits assessment (Criterion B), the AMPD proposes 25 traits

hierarchically organized into five higher-order dimensions: Negative Affect, Detachment,

Antagonism, Disinhibition, and Psychoticism. The availability of a well-validated self-report

measure, the Personality Inventory for DSM-5 questionnaire (PID-5; Krueger, Derringer,

Markon, Watson, & Skodol, 2012), has bolstered research on Criterion B, which has generated a

large number of findings over the past few years (Al-Dajani, Gralnick, & Bagby, 2016; Waugh et

al., 2017). However, the corpus of research for Criterion A has been more modest; the lack of a
SELF AND INTERPERSONAL FUNCTIONING SCALE 5

comparably detailed and validated tool has most certainly been a factor to explain the relative

paucity of findings (Morey, 2017). Available evidence on Criterion A and its operationalization,

the LPFS, has been promising for the most part. Its criterion validity has been demonstrated in

multiple studies, as Criterion A significantly predicted the presence and number of PDs in

diverse samples (Few et al., 2013; Morey, Bender, & Skodol, 2013). Morey and colleagues

(2013) reported that an LPFS score of moderate or greater severity (≥ 2) demonstrated good

sensitivity and specificity (84.6% and 72.7%, respectively) for identifying patients with at least

one PD diagnosis; Criterion A also showed incremental predictive power on functional

impairment, prognosis, and treatment intensity over the 10 categorical DSM PDs.

Self-report instruments for assessing Criterion A are needed for large-scale studies and

for screening purposes. An efficient self-report may help clinicians to systematically pay

attention to potential personality pathology, which would allow them to identify patients more

likely to require a more thorough assessment and/or subsequent treatment; it may also help

patients in self-assessing potential personality pathology and prompt them to get treatment

(Hutsebaut et al., 2016). Furthermore, considering that PD patients are often subject to lengthy

assessment procedures, developing concise measures might be especially relevant.

Early research on the LPFS relied on measures which antedated the DSM-5 AMPD, but

were intended to measure very similar constructs. The General Assessment of Personality

Disorder (GAPD; Berghuis, Kamphuis, Verheul, Larstone, & Livesley, 2013) posits a two-factor

structure for personality, aligned with the AMPD Criterion A’s two higher-order dimensions,

Self-pathology and Interpersonal dysfunction. It does not explicitly mention the four Criterion A

elements, which are merely considered as subscales among others. The Severity Indices of

Personality Problems (SIPP-118; Verheul et al., 2008) has yielded a five-dimension solution,
SELF AND INTERPERSONAL FUNCTIONING SCALE 6

which includes Self-control, Identity integration, Relational capacities, Social concordance, and

Responsibility. While somewhat different from current AMPD conceptualization, these five

dimensions qualitatively and theoretically map onto the four Criterion A dimensions.

Since the publication of the DSM-5, a growing number of self-report measures for

Criterion A have been concurrently developed and validated by a number of research teams

worldwide. Morey (2017) reported on the Level of Personality Functioning Scale-Self Report

(LPFS-SR), an 80-item self-report questionnaire, with each item answered on a four-point scale.

Construction of the scale was based on items generated for each information unit from Table 2 of

the DSM-5 AMPD. Each item is weighted according to its severity within the LPFS

conceptualization for scoring. Morey reported promising preliminary validity results for the

scale, including high degrees of internal consistency and very large intercorrelations for the four

LPFS elements, and large correlations with concurrent measures of personality pathology. Of

note, however, the four LPFS elements showed indiscriminate associations with various criterion

variables, raising doubts about the utility of distinct elements. Morey contends that these results

are coherent with the assumption underlying Criterion A development that the four LPFS

components are all considered to be indicators of a single, global, core dimension of personality

pathology. Hopwood, Good, and Morey (2018) have since reported additional data from three

large community samples in support of the instrument’s reliability and validity. The LPFS-SF’s

internal structure was once again best characterized by a single factor. It was highly reliable

across a brief retest interval, and showed conceptually meaningful and often large correlations

with external criteria (maladaptive personality traits, PD constructs, interpersonal problems).

Huprich and colleagues (2017) have recently developed and validated the DSM-5 Levels

of Personality Functioning Questionnaire (DLOPFQ), a 132-item self-report questionnaire which

SELF AND INTERPERSONAL FUNCTIONING SCALE 7

assesses the four LPF elements (Identity, Self-direction, Empathy, and Intimacy) across two life

settings (work/school and relationships). Items are scored on a six-point scale. Initial validation

data from a sample of 140 psychiatric and medical outpatients have been promising, including

good internal consistency, along with meaningful and expected correlations with a number of

external criteria. The scale also showed incremental predictive validity over DSM-5 trait

domains for interpersonal and general functioning. The DLOPFQ scales, however, had limited

discrimination with the external criterion variables.

While these two scales have shown promising results, they are both relatively lengthy

instruments. Hutsebaut and colleagues (2016) have developed and validated a very concise, 12-

item self-report questionnaire to assess the LPFS, the Level of Personality Functioning Scale-

Brief Form (LPFS-BF). Items, which are scored binary in a yes/no format, were generated by a

group of PD experts, who tried to capture the basic psychological aptitude implied by the

description of each of the 12 “facets” of the LPFS (which correspond to the 12 descriptive

statements—three for each LPF element—presented in Table 1, “Elements of personality

functioning”, of the DSM-5 AMPD; American Psychiatric Association, 2013, p.762). Hutsebaut

and colleagues (2016) reported a two-factor solution (Self and Interpersonal functioning) for

their original instrument. Because the scale showed some middling psychometric properties (e.g.,

poor association with PD pathology, low reliability, low explained variance of the factor

solution), a revised version was developed and validated: the LPFS-BF 2.0 (Bach & Hutsebaut,

2018; Weekers, Hutsebaut, & Kamphuis, 2018). Unlike the first version, the LPFS-BF 2.0 uses a

four-point Likert scale scoring format. This version represents an improvement over the original

scale, showing better internal consistency and stronger associations with PDs, along with a high

sensitivity to change. The two-factor structure of the revised instrument was demonstrated in
SELF AND INTERPERSONAL FUNCTIONING SCALE 8

both Exploratory and Confirmatory factor analysis (EFA and CFA), with some noteworthy

limitations (e.g., two items from their total sample did not load on their intended factor in EFA,

and an acceptable model fit could only be obtained after performing two post hoc tests in CFA).

The LPFS-BF 2.0 showed relevant associations with various external criteria, although there

were also indiscriminate patterns of associations between its two factors and most of the SIPP

Short Form domains.

Concurrently with these international efforts aiming to operationalize Criterion A, our

own research group was working on the development and validation of a new scale, the Self and

Interpersonal Functioning Scale (SIFS; Gamache & Savard, 2017). We wished to develop a scale

that: (a) was brief; (b) was user-friendly, with a simple scoring system; and (c) provided a

coverage of all key facets of personality functioning depicted in the LPFS. In a first step, the two

authors of the instrument, who have respectively 15 and 10 years of clinical experience with PD

patients, and both have significant experience in personality test development and validation,

generated items based on the DSM-5 AMPD definitions for the four LPFS elements (Identity,

Self-direction, Empathy, and Intimacy). This original pool of 36 items was then reviewed by a

panel of five PD experts, who share 62 years (range 3–19) of clinical experience with these

patients. They rated all items for their representativeness of the DSM-5 AMPD construct they

intended to depict, on a scale ranging from 0 (Not representative at all) to 5 (Totally

representative). After reviewing representativeness scores for all 36 items, 12 were deemed less

representative and dropped. All 24 remaining items were also rated by the same expert panel on

their clarity, on a scale ranging from 0 (Totally unclear) to 5 (Totally clear); items with a mean

score ≤ 4.5 were revised until a mean score > 4.5 was attained; this led to minor rewording for

six items. Thus, a total of 24 items, rated with a five-point Likert scale from 0 to 4, were
SELF AND INTERPERSONAL FUNCTIONING SCALE 9

included in the scale (with items 1, 6, 8, 12, 17, 19, and 24 as reversed items); higher scores are

indicative of pathological personality functioning. The four elements each originally had six

items; however, preliminary results (Leclerc, Gamache, & Savard, 2018) showed that item 7 (“I

often feel like my life has no meaning”), initially intended to capture the absence of meaningful

life goals (Self-direction), was unequivocally more strongly associated with the Identity element;

it was therefore moved to the latter.

The present study aims at exploring the psychometric properties of the newly developed

SIFS. Analyses include: (a) internal consistency and item properties based on classical test

theory (CTT); (b) differences between clinical and nonclinical participants, which are expected

to be significant and large; (c) test-retest after a two-week interval; (d) factor structure using

Confirmatory Factor Analysis (CFA). Making hypotheses about the optimal factor structure is

hazardous at this time, as a single-factor, a two- and a four-factor solution all appear defensible

in the light of DSM-5 AMPD formulation and previous empirical results on LPFS measures

(e.g., Hopwood, Good, et al., 2018; Hutsebaut et al., 2016; Zimmerman et al., 2015). However,

we contend that the scale should have clear, discernable factors in order to fulfill its intended

goal of being a useful tool for treatment planning, and for monitoring treatment course and

outcome; and (e) articulating the SIFS’ nomological network. At this point, there are relatively

few conclusive results on which to base definitive hypotheses regarding convergent and

discriminant validity, and the present work should contribute in expanding these findings. Most

notably, the relationship between Criteria A and B remains a thorny issue, as substantial overlap

and covariation have been noted in earlier studies (e.g., Few et al., 2013; Hentschel & Pukrop,

2014). Therefore, significant correlations between the two should be expected. As for the

specific nature of the associations between the SIFS’ elements and external criteria, including
SELF AND INTERPERSONAL FUNCTIONING SCALE 10

Criterion B, some hypotheses can be made, drawing on recent suggestions from Widiger et al.

(2018). Based on their extensive review, which included prior factor analytic work on Criteria A

and B (e.g., Zimmermann et al., 2015), the following pattern of associations between Criterion A

elements and external criteria (including Criterion B), should be expected: Identity with

measures of internalized pathology; Self-direction with measures of disinhibition-externalized

pathology; Empathy with measures of antagonism; and Intimacy with measures of detachment.

Method

Participants and procedure

A total sample of 386 French-speaking Canadian participants (273 women) aged 18 to 79

years old (Mage = 31.6; SD = 11.6) from two different subsamples were included in the study.

The first one is a community sample (n = 280, 208 women, Mage = 30.6, SD = 11.9) recruited

through social media, online message boards, and institutional e-mail from two universities in

the Province of Quebec, Canada; data were collected anonymously and computerized via an

online platform (SurveyMonkey). All community respondents were invited to participate in the

retest of the SIFS after a two-week interval; we had a return rate of 39.6% (n = 111, 80 women,

Mage = 30.7, SD = 12.0). The second subsample (n = 106, 65 women, Mage = 34.1, SD = 10.6) is a

clinical sample recruited during the intake procedure at a psychiatric outpatient clinic,

specialized in the treatment of PDs, in the Quebec City area. Provision of services is contingent

upon the presence of at least one DSM PD diagnosis established by the referring physician or

psychiatrist, and confirmed by a team of six licensed clinical psychologists. Main diagnoses

retrieved from patient files were as follows: narcissistic PD (28.4%), borderline-narcissistic PD

(22.7%), mixed or complex PD (i.e., three or more PDs; 18.1%), borderline PD (12.5%),

unspecified PD (8.0%), schizotypal PD (5.7%), syndrome disorder (3.4%), and histrionic PD

SELF AND INTERPERSONAL FUNCTIONING SCALE 11

(1.1%). The majority of participants from the community sample were full-time or part-time

students (58.4%), while the majority of the clinical group (59.8%) was unemployed (Cramer’s V

= .58, p < .001). Most participants from the community sample had a university degree (55.7%),

which was the case for only 18.7% of the clinical group (Cramer’s V = .48, p < .001).

All participants gave informed consent, and no compensation or incentive for

participation was offered; for the clinical group, the decision to participate or not in the study had

no impact on service provision. This study was approved by three ethics committees from the

Université du Québec à Trois-Rivières, Laval University, and the Capitale-Nationale Integrated

University Health and Social Services Centre. Data were thoroughly examined by two of the

authors to rule out indiscriminate responding (i.e., selection of the same response option for

every item on a scale), and no participant had to be excluded on such basis.

Measures

Participants from the two samples completed slightly different test batteries. In addition

to the SIFS and a short sociodemographic form, both samples completed the following

questionnaires:

The short form of the Personality Inventory for DSM-5 (PID-5-SF; Maples et al., 2015;

French validation by Roskam et al., 2015) is a 100-item self-report derived from the original

220-item PID-5 (Krueger et al., 2012) using item-response theory. Items are rated on a four-point

scale. It covers 25 pathological personality traits, which can be hierarchically organized into five

dimensions: Negative Affect (NAF; α = 0.93), Detachment (DET; α = 0.92), Antagonism (ANT;

α = 0.90), Disinhibition (DIS; α = 0.83), and Psychoticism (PSY; α = 0.88).

The Interpersonal Reactivity Index-French Version (IRI-F; Davis, 1980; French

validation by Gilet, Mella, Studer, Grühn, & Labouvie-Vief, 2013) is a 28-item self-report
SELF AND INTERPERSONAL FUNCTIONING SCALE 12

questionnaire, scored on a seven-point Likert scale, which measures empathy and its

components. Two subscales assess the cognitive component of empathy: Fantasy (the propensity

to get involved in fictitious situations; α = 0.80); and Perspective Taking (the ability to adopt

others’ point of view; α = 0.82). Two other subscales focus on the affective component of

empathy: Empathic Concern (the motivation to care about others; α = 0.77); and Personal

Distress (the tendency to feel discomfort in response to others’ emotional distress; α = 0.85).

Community sample

The Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965; French validation by

Vallières & Vallerand, 1990) is a unidimensional measure, scored on a four-point Likert scale,

which includes 10 items assessing global self-esteem (α = 0.90).

The brief 20-item version of the Inventory of Personality Organization (IPO) developed

by Verreault, Sabourin, Lussier, Normandin, and Clarkin (2013) includes three scales from the

original IPO (Kernberg & Clarkin, 1995): Identity diffusion (α = 0.66), Primitive defenses (α =

0.75), and Impaired reality testing (α = 0.77). Items are scored on a five-point rating scale and

assess a continuum of personality pathology based on Kernberg’s PD model (e.g., 1984).

The Satisfaction With Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985;

French validation by Blais, Vallerand, Pelletier, & Brière, 1989), is a five-item measure

answered on a seven-point Likert scale, which asks straightforward questions about life

satisfaction (α = 0.89).

Clinical sample

The Brief Version of the Pathological Narcissism Inventory (B-PNI; Schoenleber, Roche,

Wetzel, Pincus, & Roberts, 2009; French validation by Diguer et al., 2014) was used to measure

two dimensions of pathological narcissism: Grandiosity (e.g., inflated self-image, entitlement,

SELF AND INTERPERSONAL FUNCTIONING SCALE 13

exploitative behaviors, and fantasies of power and perfection; α = 0.84) and Vulnerability (e.g.,

depleted self-image, feelings of shame/anger, and interpersonal hypersensitivity; α = 0.87). The

28 items are scored on a seven-point Likert scale.

The short version of the Borderline Symptom List (BSL-23; Bohus et al., 2009; French

validation by Nicastro et al., 2016) is a 23-item self-rating instrument assessing borderline PD

symptomatology in accordance with DSM-5 formulation. Items are scored on a five-point scale.

The BSL-23 assesses BPD symptom severity on a dimensional continuum. We only used the

global score (α = 0.92) in the present study.

The 12-item version of the Buss-Perry Aggression Questionnaire (BPAQ; Buss & Perry,

1992; French validation by Genoud & Zimmerman, 2009) covers four forms of aggression:

Verbal (α = 0.66), Physical (α = 0.88), Anger (α = 0.83), and Hostility (α = 0.75). It also yields a

global trait aggression score (α = 0.89). Items are scored on a seven-point scale.

The 36-item Medical Outcomes Study Short-Form Health Survey (SF-36; Ware &

Sherbourne, 1992; French validation by Richard, 2000) is a self-report measure of health-related

quality of life. It includes 36 items with scores that are transformed into a 0–100 scale, and then

averaged into eight subscales: Physical Functioning (α = 0.88), Role limitations due to physical

problems (α = 0.82), Bodily Pain (α = 0.70), General Health (α = 0.83), Vitality (α = 0.71),

Social Functioning (α = 0.63), Role limitations due to emotional problems (α = 0.83), and Mental

Health (α = 0.78). Higher scores reflect a better health-related quality of life.

Statistical analyses

Cronbach alphas were used to assess internal consistency, while bivariate zero-order

correlations were computed to assess test-retest reliability of the SIFS scores after a two-week

interval. T-tests and bivariate zero-order correlations were computed to explore item properties
SELF AND INTERPERSONAL FUNCTIONING SCALE 14

based on CTT. T-tests were also used to examine differences between the clinical and the

nonclinical groups on SIFS scores. These analyses were carried out using the Statistical Package

for the Social Sciences (SPSS) 25.0 software.

In line with DSM-5 AMPD theoretical formulations and previous studies on LPFS

measures (e.g., Hutsebaut et al., 2016; Morey, 2017), various factor analytic models were tested.

Five models were computed using Confirmatory factor analysis (CFA): a basic one-factor model

(Model 1); a two-factor correlated solution with Self and Interpersonal functioning as factors

(Model 2); a four-factor correlated solution with the theoretical elements of Identity, Self-

direction, Empathy, and Intimacy as factors (Model 3); a second-order orthogonal solution with

the four factors loading on a general personality pathology factor (Model 4); and a bi-factor four-

factor model, with all items loading both on the four elements and on an overarching general

personality pathology factor, consistent with recent studies (e.g., Sharp et al., 2015) investigating

the presence of a general personality pathology factor (Model 5). These analyses were performed

using Mplus version 8.0 (L. K. Muthén & B. O. Muthén, 2017), with data treated as categorical.

As suggested by Beauducel and Herzberg (2006), the robust weighted least square estimator

(WLSMV) was used. Adequate model fit was determined using the χ2 goodness-of-fit index,

alongside with sample-size independent fit indexes (e.g., Hu & Bentler, 1999): the comparative

fit index (CFI; > .90), the Tucker-Lewis index (TLI; > .90), and the root mean square error of

approximation (RMSEA; < .08). Nested model comparisons of fit improvement were evaluated

using the Mplus DIFFTEST function (MDΔχ2; Asparouhov & Muthén, 2009).

The SIFS’ nomological network was assessed, first, through bivariate zero-order

correlations between SIFS scores and external criteria. Unique contribution of each element,

after partialing out shared variance with the other three elements, was also computed. In order to
SELF AND INTERPERSONAL FUNCTIONING SCALE 15

do so, we created residualized scores for each SIFS element by computing a regression for each

element, with the three other elements as predictors. These residualized scores, which represent

the unique, unshared variance of each element, were then correlated with external criteria.

Results

Item properties results based on CTT are presented in Table 1. For the Global scale,

internal consistency was excellent, and was good for the four personality functioning elements.

Items 6 and 16 appeared problematic based on CTT; the first one was associated with low item-

scale correlations (ISC) figures (< .30), while the second had low variance and did not

discriminate between the clinical and the nonclinical groupsb.

Test-retest figures after a two-week interval (n = 111) were as follows: Global scale: r =

.89; Identity (SIFS-ID): r = .91; Self-direction (SIFS-SD): r = .63; Empathy (SIFS-EMP): r =

.78; and Intimacy (SIFS-INT): r = .92; all ps < .001.

For all CFA models, modification indices were consulted to determine whether

correlations between item residuals would improve model fits; only theoretically defensible

modifications for item pairs within a same element were considered. Three were eventually

implemented (item pairs 13–15, 19–24, and 22–23). Examination of fit indices (see Table 2)

revealed that the CFA two-factor (Model 2; MD∆χ2 = 65.241; df = 1; p < .001; ∆CFI = .018,

∆TLI = .019, ∆RMSEA = .011) and four-factor (Model 3; MD∆χ2 = 149.507; df = 6; p < .001;

∆CFI = .025, ∆TLI = .026, ∆RMSEA = .009 and nonoverlapping RMSEA 90% confidence

interval) models had significantly improved fit coefficients compared to the single-factor model

(Model 1), with the four-factor model fitting the data better than Model 2 (MD∆χ2 = 59.396; df =

5; p < .001; ∆CFI = .007, ∆TLI = .007, ∆RMSEA = .005). The second-order CFA model, nested

within the correlated four-factor model, did not significantly decrease model fit, and thus can be
SELF AND INTERPERSONAL FUNCTIONING SCALE 16

accepted as the more parsimonious alternative, especially considering its close alignment with

the Criterion A AMPD conceptualization, which posits that the four LPFS elements are all

indicators of an overarching global dimension of personality pathology (Morey, 2017). The bi-

factor solution (Model 5) had practically identical fits in comparison with Model 4; given that

the comparison of bi-factor and higher-order models is known to be biased in favor of the former

when there is unmodelled complexity (Murray & Johnson, 2013), the higher-order model was

preferred. Item loadings of the second-order CFA model solution are presented in Figure 1.

Results pertaining to the SIFS’ nomological network are presented in Table 3. SIFS-ID

showed the strongest unique associations with poor self-esteem (RSES), identity diffusion (IPO),

negative emotions (PID-5), and with subjective impression of mental health impairment and

limitations due to emotional distress (SF-36); it also showed a positive association with IRI

Personal distress. SIFS-SD showed the strongest unique associations with disinhibition (PID-5),

vulnerable narcissism (B-PNI), and with anger and hostility (BPAQ). SIFS-EMP showed the

strongest unique associations with antagonism (PID-5), with impaired perspective-taking and

empathic concern (IRI), and with trait, verbal, and physical aggression (BPAQ). Finally, SIFS-

INT showed the strongest unique associations with detachment (PID-5).

Discussion

The main purpose of the study was to report on the development and psychometric

properties of a new self-report measure of DSM-5 AMPD Criterion A, the Self and Interpersonal

Functioning Scale (SIFS). Overall, the scale has shown promising reliability and validity,

providing important preliminary results in support of its validity as an operationalization of the

DSM-5 AMPD personality pathology conceptualization.

SELF AND INTERPERSONAL FUNCTIONING SCALE 17

Internal consistency was excellent for the Global scale, and was good for the four

elements. Based on CTT, two items seemed fairly problematic. Item 6 (“I recognize myself in

how others describe me”) was associated with low ISC figures (< .30), calling into question

whether it should be retained in future iterations of the SIFS. We opted to keep it for now, as it is

the SIFS-ID item which taps most directly onto the facet “accuracy of self-appraisal”. Item 16 (“I

have little interest for other people’s feelings or problems”) was the only SIFS item that did not

discriminate between the clinical and nonclinical groups. It was weakly endorsed in both

subsamples, possibly because of its blunt formulation. However, we do not recommend its

deletion at this point, because of its potential value in contexts (e.g., forensic) where

psychopathic traits are likely to be high. All other items showed significant differences for the

clinical and the nonclinical groups, with large effect sizes for the four elements and the global

score, providing a first indication of the scale’s criterion validity. Items 3 (inner emptiness) and 7

(meaninglessness), which are central to the notion of identity diffusion (e.g., Kernberg, 1984),

were the most discriminant. Test-retest figures were substantial for most elements, which is

consistent with the short-term stability expected for core personality elements. Results were

moderate, however, for SIFS-SD, suggesting that this element may tap onto more fluctuating and

context-dependent personality constructs (e.g., the ability to self-reflect productively). However,

the substantial test-retest coefficient (.88) for the SD element of the LPFS-SR reported by

Hopwood et al. (2018) casts doubt on this hypothesis.

Factor analysis results revealed that the best fit was obtained with a second-order four-

factor CFA model, with a general personality pathology factor as a second-order overarching

construct. The model obtained acceptable fits, with only one item (Item 6) having a questionable

loading (< .40) on its factor. The retained model is coherent with the DSM-5 AMPD
SELF AND INTERPERSONAL FUNCTIONING SCALE 18

conceptualization that the four LPFS elements are all indicators of a global personality pathology

dimension; at the same time, it also suggests that it would also be justified to consider elements

as distinct, core features of personality pathology. Our results are in contrast with those reported

for both versions of the LPFS-BF (Bach & Hutsebaut, 2018; Hutsebaut et al., 2016; Weekers et

al., 2018), for which a two-factor solution (reflecting the Self and Interpersonal elements) was

found. Zimmermann et al. (2015) had also found support for a two-factor solution for the LPFS,

using other-ratings by laypersons and therapists, and did not find support for the four elements.

Authors of the LPFS-SR (Hopwood et al., 2018; Morey, 2017) have argued in favor of a single-

factor model based on their results, which relied on intercorrelations and PCA; however, they did

not use CFA, and therefore did not test for a second-order or a bi-factor model. Huprich et al.

(2017) provided no factor-analytic information on the structure of the DLOPFQ.

The underlying structure of Criterion A remains a contentious issue. The LPFS has been

described by its developers as a single dimension that should be represented by a single score

(see, e.g., Morey, 2017), which could arguably represent the general personality disorder factor

(e.g., Sharp et al., 2015) uncovered in recent investigations. However, a strong case could be

made that a clear and distinct factor structure for Criterion A is warranted, based on clinical and

conceptual considerations. Indeed, reducing Criterion A to a single dimension or score would

obfuscate potentially meaningful clinical information on specific pathological profiles of

patients. It seems highly unlikely that a one-factor solution, or even a two-factor solution, will be

able to accomplish the four objectives stated by the Criterion A authors of “(a) identifying the

presence and extent of personality psychopathology, (b) planning treatment, (c) building the

therapeutic alliance, and (d) studying treatment course and outcome” (Bender et al., 2011, pp.

340–341). Having distinct and robust factors will also enable to yield more reliable diagnoses for
SELF AND INTERPERSONAL FUNCTIONING SCALE 19

the six specific DSM-5 AMPD PDs; their description, albeit indirectly, acknowledges the

usefulness of distinct scores for the four elements, as the first criterion states that the presence of

moderate or greater impairment in personality functioning, as manifested by difficulties in two or

more Criterion A elements, is required. In sum, we contend that having a clear factor structure

for Criterion A, and a measure which allows for a valid assessment of these factors, is important.

This makes a case for the SIFS, which has shown the potential to distinguish “significant” levels

of impairment empirically, which has yet to be done by any other available Criterion A tool, to

our best knowledge. A second-order factor solution, which is consistent with our data, might

arguably be the best way to reconcile recent findings about a general personality disorder factor

(e.g., Sharp et al., 2015) with Criterion A’s intended goals.

Associations with external criteria showed distinct patterns for the four SIFS elements.

These results, along with results from factor analyses, suggest that it is justifiable to consider

SIFS element scores separately. This is a considerable strength, as the capacity to obtain a

personalized profile for each patient is one of the main anticipated advantages of evaluating

personality on a dimensional basis (e.g., a specific score for each element can help to document

treatment prognosis and to identify priority clinical targets for intervention). In contrast, the

LPFS-SR and the DLOPFQ showed mostly indiscriminate associations with external measures

for the four elements (Huprich et al., 2017; Morey, 2017). Of note, however, the external criteria

variables chosen for the present study (e.g., specific measures of identity diffusion, self-esteem,

or empathy) may have mapped more specifically onto LPFS constructs than the general

personality pathology measures used by Morey (2017) and Huprich and colleagues (2017).

Associations with external criteria were remarkably in line with hypotheses drawn from

Widiger et al. (2018). SIFS-ID showed theoretically meaningful and unique associations with
SELF AND INTERPERSONAL FUNCTIONING SCALE 20

measures of negative affectivity, poor self-esteem, identity diffusion, and personal distress. Thus,

this element appears to be strongly related to internalized psychopathology, a result also

supported by the negative associations with antagonism and various forms of aggression. SIFS-

SD showed associations with indices of disinhibition and externalization, in line with LPF

formulation of poor planning, compromised goal-setting abilities, and lack of prosocial internal

standards. Its associations with vulnerable narcissism and anger-hostility are intriguing and

warrant further studies. It may reflect, for both construct, an inhibition of the capacity to reach

gratifying and satisfying personal goals; in vulnerable narcissism, this may be due to impaired

self-evaluation (which may be secretly grandiose, or severely debased), and may entail enraged

reactions in the face of this failure (e.g., Pincus et al., 2009). The SIFS-EMP element showed

expected correlations with impaired empathic capacities, as well as coherent patterns of

associations with antagonism and aggression. Finally, the SIFS-INT element showed unique

associations with interpersonal detachment, and was also associated (though not specifically)

with other indicators of interpersonal problems (antagonism, low empathy), in line with the

LPFS formulation, which depicts negative, detached, and self-serving relationships.

These patterns of associations do not solve the redundancy problem pertaining to the

overlaps between Criteria A and B (e.g., Zimmermann et al., 2015), also noted by Hopwood et

al. (2018) for the LPFS-SR. Correlations between the SIFS global score and PID-5 trait domains

were in the high-very high range, from .49 (Antagonism) to .81 (Detachment). The SIFS-INT

element mapped to a significant degree onto the pathological trait domain of detachment. SIFS-

ID, SD, and EMP appear to have somewhat broader nomological networks, but remain closely

tied with negative affectivity, disinhibition, and antagonism, respectively. Solving the overlap

issues between Criteria A and B remains an important objective in demonstrating the clinical and
SELF AND INTERPERSONAL FUNCTIONING SCALE 21

scientific utility of the AMPD, and research geared toward that goal needs solid, theory-driven,

and statistically sound tools, such as the SIFS, to do so.

The main limitation of the present study is that validation data, at this time, are limited to

a sample of French-speaking Canadians. Proper validation of the existing English translation, as

well as translation-adaptation to other languages and validation in diverse cultural groups, is

needed. Test-retest, which could only be assessed for the community sample and for which we

had only a modest level of participation (39.6%), was assessed at only one point in time, and at a

brief interval (two weeks). PD diagnoses in the clinical sample, even though they were revised

by a team of licensed clinical psychologists, were not confirmed by a formal assessment using

structured/semi-structured interviews or validated self-reports. Invariance of the retained factor

model between groups, as well as between women and men, could not be performed due to the

relatively small clinical sample size and number of male participants. Results for convergent-

discriminant validity cannot be readily compared with those reported for other Criterion A self-

report measures, as our choice of external criteria comparators was restricted to those that are

well validated in the French language. The present study provided little information on the

incremental validity of the SIFS elements over the PID-5-SF traits; correlational results using

residualized scores showed an overlap between SIFS elements and pathological traits, suggesting

that a fine-tuned discrimination between LPFS and pathological traits is likely to remain

challenging (Hopwood, Good, et al., 2018). The use of a self-report to study personality

pathology might be considered in itself as a limitation; a number of authors (e.g., Hopwood et

al., 2008; Oltmanns & Turkheimer, 2009) have suggested that the use of informant reports, in

addition to self-reports, can provide more accurate information and data in the context of

evaluating personality and PDs. Thus, the possibility of developing an informant report version
SELF AND INTERPERSONAL FUNCTIONING SCALE 22

of the SIFS should be strongly considered. Finally, we did not assess the disorder-specific

impairments expected for the six specific PDs proposed in DSM-5 Section III.

Despite some limitations, the present investigation suggests that the SIFS possesses

sound psychometric properties, and should be seen as a valid, concise alternative for assessment

of Criterion A DSM-5 AMPD conceptualization. In contrast with other existing self-report

measures of Criterion A, its main strength relative to other measures is the presence of well-

differentiated factors, which we believe is important in order for a Criterion A measure to fulfill

its intended objectives (e.g., treatment planning and monitoring). Its psychometric properties,

overall, seem generally more robust in comparison with the original LPFS-BF (Hutsebaut et al.,

2016); a direct comparison with the 2.0 iteration of the LPFS-BF, which has also shown mostly

solid psychometric properties, should be tested in future studies. The rigorous strategy behind the

development of the LPFS-SR (Morey, 2017), for which each information unit from Table 2 of

the DSM-5 AMPD was turned into an item (up to a total of 80), makes it more closely aligned

with DSM LPF formulation in contrast with the SIFS; the added number of items also provides a

broader coverage of all 12 personality pathology facets. The DLOPFQ is considerably longer

than the SIFS (132 items), but in addition to a broader facet coverage and a better demonstration

of its incremental validity over Criterion B pathological traits, it also has the unique advantage of

assessing personality pathology manifestations across two meaningful life contexts (work/school

and relationships). Future comparison studies of Criterion A measures could also include semi-

structured interview schedules assessing the LPFS such as the Semi-Structured Interview for

Personality Functioning DSM-5 (Hutsebaut, Kamphuis, Feenstra, Weekers, & De Saeger, 2017).

Notes
a
The SIFS was originally developed in French. It has been since translated into English. The

translation and instructions for scoring are available as supplemental material.

SELF AND INTERPERSONAL FUNCTIONING SCALE 23

b
Proposed cutoff scores and basic statistical classification results for the SIFS and its subscales

are available as supplemental material.

SELF AND INTERPERSONAL FUNCTIONING SCALE 24

References

Al-Dajani, N., Gralnick, T. M., & Bagby, R. M. (2016). A psychometric review of the

Personality Inventory for the DSM-5 (PID-5): Current status and future directions. Journal

of Personality Assessment, 98, 62–81. doi:10.1080/00223891.2015.1107572

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders

(5th ed.). Arlington, VA: Author.

Asparouhov, T., & Muthén, B. O. (2009). Exploratory structural equation modeling. Structural

Equation Modeling, 16, 397–438. doi:10.1080/10705510903008204

Bach, B., & Hutsebaut, J. (2018). Level of Personality Functioning Scale—Brief Form 2.0:

Utility in capturing personality problems in psychiatric outpatients and incarcerated addicts.

Journal of Personality Assessment. Advance online publication.

doi:10.1080/00223891.2018.1428984

Beauducel, A., & Herzberg, P. Y. (2006). On the performance of maximum likelihood versus

means and variance adjusted weighted least squares estimation in CFA. Structural Equation

Modeling, 13, 186–203. doi:10.1207/s15328007sem1302_2

Bender, D. S., Morey, L. C., & Skodol, A. E. (2011). Toward a model for assessing level of

personality functioning in DSM-5, part I: A review of theory and methods. Journal of

Personality Assessment, 93, 332–346. doi:10.1080/00223891.2011.583808

Berghuis, H., Kamphuis, J. H., Verheul, R., Larstone, R., & Livesley, J. (2013). The General

Assessment of Personality Disorder (GAPD) as an instrument for assessing the core features

of personality disorders. Clinical Psychology & Psychotherapy, 20, 544–557.

doi:10.1002/cpp.1811
SELF AND INTERPERSONAL FUNCTIONING SCALE 25

Blais, M. R., Vallerand, R. J., Pelletier, L. G., & Brière, N. M. (1989). L'échelle de satisfaction

de vie: Validation canadienne-française du "Satisfaction with Life Scale." [The satisfaction

scale: French-Canadian validation of the Satisfaction with Life Scale]. Canadian Journal of

Behavioural Science/Revue canadienne des sciences du comportement, 21, 210-223.

doi:10.1037/h0079854

Bohus, M., Kleindienst, N. Limberger, M. F., Stieglitz, R. D., Domsalla, M., Chapman, A. L., …

Wolf, M. (2009). The short version of the Borderline Symptom List (BSL-23): Development

and initial data on psychometric properties. Psychopathology, 42, 32–39.

doi:10.1159/000173701

Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and

Social Psychology, 63, 452–459.

Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS

Catalog of Selected Documents in Psychology, 10, 85.

Diener, E. D., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Satisfaction with Life

Scale. Journal of Personality Assessment, 49, 71–75. doi:10.1207/s15327752jpa4901_13

Diguer, L., Turmel, V., Da Silva Luis, R., Mathieu, V., Marcoux, L.-A., & Lapointe, T. (2014).

Development and initial structure analysis of a French version of the Pathological

Narcissism Inventory. Journal of the European Psychiatric Association, 29, 1–4.

doi:10.1016/S0924-9338(14)77997-0

Few, L. R., Miller, J. D., Rothbaum, A. O., Meller, S., Maples, J., Terry, D. P., … MacKillop, J.

(2013). Examination of the Section III DSM-5 diagnostic system for personality disorders in

an outpatient clinical sample. Journal of Abnormal Psychology, 122, 1057–1069.

doi:10.1037/a0034878
SELF AND INTERPERSONAL FUNCTIONING SCALE 26

Gamache, D., & Savard, C. (2017). Self and Interpersonal Functioning Scale. Unpublished

manuscript, Université du Québec à Trois-Rivières, Qc, Canada.

Genoud, P. A., & Zimmermann, G. (2009, August). French version of the 12-item Aggression

Questionnaire: Preliminary psychometric properties. Poster presented at the 11th congress of

the Swiss Psychological Society, Neuchâtel, Switzerland.

Gilet, A-L., Mella, N., Studer, J., Grühn, D, & Labouvie-Vief, G. (2013). Assessing dispositional

empathy in adults: A French validation of the Interpersonal Reactivity Index (IRI).

Canadian Journal of Behavioral Science, 45, 42–48. doi:10.1037/a0030425

Hentschel, A. G., & Pukrop, R. (2014). The essential features of personality disorder in DSM-5:

The relationship between Criteria A and B. Journal of Nervous and Mental Disease, 202,

412–418. doi:10.1097/NMD.0000000000000129

Hopwood, C. J., Good, E. W., & Morey, L. C. (2018). Validity of the DSM-5 Levels of

Personality Functioning Scale-Self Report. Journal of Personality Assessment. Advance

online publication. doi:10.1080/00223891.2017.1420660

Hopwood, C. J., Morey, L. C., Edelen, M. O., Shea, M. T., Grilo, C. M., Sanislow, C. A., …

Markowitz, J. C. (2008). A comparison of interview and self-report methods for the

assessment of borderline personality disorder criteria. Psychological Assessment, 20, 81–85.

doi:10.1037/1040-3590.20.1.81

Huprich, S. K., Nelson, S. M., Meehan, K. B., Siefert, C. J., Haggerty, G., Sexton, J., … Baade,

L. (2018). Introduction of the DSM-5 Levels of Personality Functioning Questionnaire.

Personality Disorders: Theory, Research, and Treatment. 9, 553–563.

doi:10.1037/per0000264
SELF AND INTERPERSONAL FUNCTIONING SCALE 27

Hutsebaut, J., Feenstra, D. J., & Kamphuis, J. H. (2016). Development and preliminary

psychometric evaluation of a brief self-report questionnaire for the assessment of DSM-5

Level of Personality Functioning Scale: The LPFS Brief Form (LPFS-BF). Personality

Disorders: Theory, Research, and Treatment, 7, 192–197. doi:10.1037/per0000159

Hutsebaut, J., Kamphuis, J. H., Feenstra, D. J., Weekers, L. C., & De Saeger, H. (2017).

Assessing DSM-5-oriented Level of Personality Functioning: Development and

psychometric evaluation of the Semi-Structured Interview for Personality Functioning

DSM-5 (STiP-5.1). Personality Disorders: Theory, Research, and Treatment, 8, 94–101.

doi:10.1037/per0000197

Kernberg, O. F. (1984). Severe personality disorders: Psychotherapeutic strategies. New

Haven, CT: Yale University Press.

Kernberg, O. F., & Clarkin, J. F. (1995). The Inventory of Personality Organization. White

Plains, NY: The New York Hospital-Cornell Medical Center.

Krueger, R. F., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. E. (2012). Initial

construction of a maladaptive personality trait model and inventory for DSM-5.

Psychological Medicine, 42, 1879–1890. doi:10.1017/S0033291711002674

Leclerc, P., Gamache, D., & Savard, C. (2018, March). Développement et validation d'un outil

dimensionnel du fonctionnement de la personnalité [Development and validation of a

dimensional instrument of personality functioning]. Poster session presented at the meeting

of the Société québécoise pour la recherche en psychologie, Quebec City, Canada.

Morey, L. C. (2017). Development and initial evaluation of a self-report form of the DSM-5

Level of Personality Functioning Scale. Psychological Assessment, 29, 1302–1308.

doi:10.1037/pas0000450
SELF AND INTERPERSONAL FUNCTIONING SCALE 28

Morey, L. C., Bender, D. S., & Skodol, A. E. (2013). Validating the proposed Diagnostic and

Statistical Manual of Mental Disorders, 5th ed., severity indicator for personality disorder.

Journal of Nervous and Mental Disease, 201, 729–735.

doi:10.1097/NMD.0b013e3182a20ea8

Morey, L. C., Skodol, A. E., & Oldham, J. M. (2014). Clinician judgments of clinical utility: A

comparison of DSM-IV-TR personality disorders and the alternative model for DSM-5

personality disorders. Journal of Abnormal Psychology, 123, 398–405.

doi:10.1037/a0036481

Murray, A. L., & Johnson, W. (2013). The limitations of model fit in comparing the bi-factor

versus higher-order models of human cognitive ability structure. Intelligence, 41, 407–422.

doi:10.1016/j.intell.2013.06.004

Muthén L. K., & Muthén B. O. (2017). Mplus user’s guide (8th ed.). Los Angeles, CA: Muthén

& Muthén.

Nicastro, R., Prada, P., Kunk, A.-L., Salamin, V., Dayer, A., Aubry, J.-M., … Perroud, N.

(2016). Psychometric properties of the French borderline symptom list, short form (BSL-

23). Borderline Personality Disorder and Emotion Dysregulation, 3, 4. doi:10.1186/s40479-

016-0038-0

Oltmanns, T. F., & Turkheimer, E. (2009). Person perception and personality pathology. Current

Directions in Psychological Science, 18, 32–36. doi:10.1111/j.1467-8721.2009.01601.x

Pincus, A. L., Ansell, E. B., Pimentel, C. A., Cain, N. M., Wright, A. G. C., & Levy, K. N.

(2009). Initial construction and validation of the Pathological Narcissism Inventory.

Psychological Assessment, 21, 365–379. doi:10.1037/a0016530.

SELF AND INTERPERSONAL FUNCTIONING SCALE 29

Richard, J. L., Bouzourène, K., Gallant, S., Ricciardi, P., Sudre, P., Iten, A., & Burnand, B.

(2000). Validation et normes du SF-36 dans la population du canton de Vaud [Validation

and norms of the SF-36 in the Canton of Vaud population]. Lausanne, Switzerland: Institut

universitaire de médecine sociale et préventive.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton

University Press.

Roskam, I., Galdiolo, S., Hansenne, M., Massoudi, K., Rossier, J., Gicquel, L., & Rolland, J.-P.

(2015). The psychometric properties of the French Version of the Personality Inventory for

DSM-5. PLoS One, 10(7), e0133413. doi:10.1371/journal.pone.0133413

Schoenleber, M., Roche, M. J., Wetzel, E., Pincus, A. L., & Roberts, B. W. (2015). Development

of a brief version of the Pathological Narcissism Inventory. Psychological Assessment, 27,

1520–1526. doi:10.1037/pas0000158

Sharp, C., Wright, A. G., Fowler, J. C., Frueh, B. C., Allen, J. G., Oldham, J., & Clark, L. A.

(2015). The structure of personality pathology: Both general (‘g’) and specific (‘s’) factors?

Journal of Abnormal Psychology, 124, 387–398. doi:10.1037/abn0000033

Vallières, E. F., & Vallerand, R. J. (1990). Traduction et validation canadienne‐française de

l’Échelle de l’estime de soi de Rosenberg [French-Canadian translation and validation of the

Rosenberg Self-Esteem Scale]. International Journal of Psychology, 25, 305-316.

doi:10.1080/00207599008247865

Verheul, R., Andrea, H., Berghout, C. C., Dolan, C., Busschbach, J. J. V., van der Kroft, P. J. A.,

… Fonagy, P. (2008). Severity Indices of Personality Problems (SIPP-118): Development,

factor structure, reliability, and validity. Psychological Assessment, 20, 23–34.

doi:10.1037/1040-3590.20.1.23
SELF AND INTERPERSONAL FUNCTIONING SCALE 30

Verreault, M., Sabourin, S., Lussier, Y., Normandin, L., & Clarkin, J. F. (2013). Assessment of

personality organization in couple relationships: Factorial structure of the inventory of

personality organization and incremental validity over neuroticism. Journal of Personality

Assessment, 95, 85–95. doi:10.1080/00223891.2012.713883.

Wade, J. E., & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36).

Medical Care, 30, 473–483.

Waugh, M. H., Hopwood, C. J., Krueger, R. F., Morey, L. C., Pincus, A. L., & Wright, A. G.

(2017). Psychological assessment with the DSM-5 alternative model for personality

disorders: Tradition and innovation. Professional Psychology: Research and Practice, 48,

79–89. doi:10.1037/pro0000071

Weekers, L. C., Hutsebaut, J., & Kamphuis, J. H. (2018). The Level of Personality Functioning

Scale-Brief Form 2.0: Update of a brief instrument for assessing level of personality

functioning. Personality and Mental Health. Advance online publication.

doi:10.1002/pmh.1434

Widiger, T. A., Bach, B., Chmielewski, M., Clark, L. A., DeYoung, C., Hopwood, C. J., …

Thomas, K. M. (2018). Criterion A of the AMPD in HiTOP. Journal of Personality

Assessment. Advance online publication. doi:10.1080/00223891.2018.1465431

Zimmermann, J., Böhnke, J. R., Eschstruth, R., Mathews, A., Wenzel, K., & Leising, D. (2015).

The latent structure of personality functioning: Investigating Criterion A from the alternative

model for personality disorders in DSM-5. Journal of Abnormal Psychology, 124, 532–548.

doi:10.1037/abn0000059
SELF AND INTERPERSONAL FUNCTIONING SCALE 31

Figure 1
Second-order four-factor confirmatory factor analysis of the Self and Interpersonal Functioning
Scale (SIFS)

Note. All coefficients are significant at p < .001. R2 of Identity, Self-direction, Empathy, and
Intimacy were respectively at .771, .830, .839, and .745. Item 16 was excluded from the analysis.
SELF AND INTERPERSONAL FUNCTIONING SCALE 32

Table 1

Descriptive Statistics (Global, Nonclinical, and Clinical Samples) and Inter-item Correlations for the Self and Interpersonal

Functioning Scale (SIFS; N = 386)

Classical test theory Inter-item Correlations (N = 386)

Nonclinical Sample Clinical Sample

Global sample Student’s t d
(n = 280) (n = 106)

M SD ISC M SD ISC M SD ISC

Identity 2 3 4 5 6 7

1 1.80 1.17 .61 1.43 .97 .55 2.71 1.15 .29 10.04*** 1.20 .50*** .55*** .34*** .51*** .25*** .50***

2 2.46 1.25 .63 2.11 1.15 .55 3.33 1.08 .40 9.49*** 1.09 .59*** .41*** .57*** .17** .52***

3 1.56 1.43 .83 .98 1.06 .72 2.95 1.23 .69 15.32*** 1.72 .45*** .74*** .28*** .82***

4 1.09 1.11 .49 .84 .94 .38 1.70 1.70 .37 7.12*** 0.63 .47*** .11* .41***

5 1.42 1.41 .78 .97 1.14 .71 2.52 1.39 .68 10.93*** 1.22 .27*** .73***

6 1.87 1.05 .28 1.73 .98 .24 2.22 1.14 .10 4.21*** 0.46 .24***

7 1.51 1.46 .77 .93 1.14 .64 2.92 1.20 .59 14.61*** 1.70

M 1.67 .94 1.29 .71 2.61 .75 15.61*** 1.81

α .86 .80 .73

9 10 11 12

Self-Direction
SELF AND INTERPERSONAL FUNCTIONING SCALE 33

8 1.65 1.11 .47 1.35 .91 .29 2.39 1.19 .42 8.72*** 0.98 .31*** .33*** .35*** .37***

9 1.57 1.23 .55 1.27 1.06 .45 2.29 1.32 .47 7.78*** 0.85 .46*** .42*** .36***

10 1.14 1.10 .53 .91 .89 .37 1.69 1.34 .54 6.50*** 0.69 .44*** .25***

11 1.32 1.21 .52 1.09 1.06 .39 1.90 1.36 .52 6.05*** 0.66 .25***

12 1.11 .98 .42 .90 .76 .26 1.60 1.23 .37 6.76*** 0.69

M 1.36 .78 1.11 .58 1.97 .88 11.06*** 1.15

α .73 .60 .71

14 15 16 17 18
Empathy

13 .91 1.12 .62 .64 .89 .55 1.53 1.36 .55 7.38*** 0.77 .42*** .63*** .20*** .28*** .36***

14 .53 .95 .46 .42 .76 .46 .80 1.25 .40 3.51** 0.37 .24*** .45*** .24*** .17**

15 1.22 1.23 .49 .94 1.04 .41 1.90 1.39 .43 7.16*** 0.78 .15** .24*** .30***

16 .60 1.00 .33 .56 .96 .32 .72 1.09 .35 1.53 0.16 .20*** .15**

17 1.01 .98 .38 .83 .80 .18 1.46 1.21 .42 5.97*** 0.61 .29***

18 1.02 1.10 .39 .83 .90 .14 1.48 1.38 .52 5.37*** 0.56

M .88 .68 .70 .52 1.31 .83 8.59*** 0.88

α .71 .60 .71

20 21 22 23 24
Intimacy

19 1.64 1.29 .65 1.26 1.07 .63 2.57 1.31 .46 10.03*** 1.10 .45*** .19*** .46*** .45*** .71***
SELF AND INTERPERSONAL FUNCTIONING SCALE 34

20 .87 1.16 .58 .55 .84 .55 1.66 1.45 .41 9.29*** 0.94 .33*** .40*** .42*** .47***

21 .93 1.04 .32 .84 .89 .23 1.16 1.32 .38 2.75** 0.28 .28*** .25*** .19***

22 .79 1.08 .60 .63 .96 .59 1.17 1.25 .55 4.44*** 0.49 .54*** .46***

23 1.24 1.24 .57 1.01 1.09 .54 1.82 1.38 .47 5.97*** 0.65 .40***

24 1.32 1.29 .64 1.00 1.11 .62 2.10 1.38 .47 8.13*** 0.88

M 1.25 .82 .99 .67 1.90 .79 10.33*** 1.24

α .80 .78 .72

Global scale

M 1.30 .69 1.03 .50 1.97 .64 15.16*** 1.64

α .92 .87 .88

Note. d = Cohen’s d; ISC = Item-scale correlations (corrected). A five-point Likert scale (0 = This does not describe me at all; 4 = This

describes me totally) was used.

* p < .05. ** p < .01. *** p < .001.

SELF AND INTERPERSONAL FUNCTIONING SCALE 35

Table 2.

Goodness-of-fit statistics for the models estimated on the Self and Interpersonal Functioning

Scale (SIFS)

Models WLSMV χ2 (df) CFI TLI RMSEA [90% CI] p

1. CFA 1 factor 859.145* (227) .926 . 918 0.087 [.081 - .093] <.001

2. CFA 2 factors correlated 706.100* (226) .944 .937 0.076 [.069 - .082] <.001

3. CFA 4 factors correlated 637.270* (221) . 951 .944 0.071 [.065 - .078] <.001

4. Second-order CFA 663.354* (223) .949 .942 0.073 [.067 - .080] <.001

5. Bi-factor CFA 4 factors 668.565* (212) .947 .936 0.076 [.070 - .083] <.001

Note. WLSMV: Robust weighted least square estimator; χ2= WLSMV chi square; df = degrees of

freedom; CFI = comparative fit index; TLI = Tucker-Lewis index; RMSEA = root mean square

error of approximation; 90% CI = confidence interval. All models have been tested without item

16, and including correlations between error terms for items 19 and 24, 22 and 23, and 13 and

15.

* p < .001.
SELF AND INTERPERSONAL FUNCTIONING SCALE 36

Table 3

Nomological network of the Self and Interpersonal Functioning Scale (SIFS)

SIFS

Global SIFS Identity SIFS Self-direction SIFS Empathy SIFS Intimacy

Scale

r r Res r βc r Res r βc r Res r βc r Res r βc

SIFS (n = 369)

Global Scale .89*** .81*** .79*** .83***

Identity .67*** .53*** .61***

Self-direction .58*** .52***

Empathy .65***

Intimacy

RSES (n = 258)a -.69*** -.74*** -.44*** -.63*** -.48*** -.01 -.11* -.31*** .10 .03 -.42*** -.05 -.11

SWLS (n = 255)a -.68*** -.63*** -.27*** -.41*** -.47*** -.03 -.17** -.36*** .10 .05 -.54*** -.22*** -.32***

IPO (n = 257)a

Id. diffusion .56*** .57*** .33*** .45*** .41*** .03 .11 .33*** .05 .12 .29*** -.06 -.02

Pr. defenses .56*** .44*** .10 .19** .47*** .14* .26*** .41*** .09 .18* .37*** .05 .10

Reality testing .34*** .26*** .06 .14 .25*** .04 .10 .26*** .05 .13 .26*** .08 .08

PID-5 (n = 345)
SELF AND INTERPERSONAL FUNCTIONING SCALE 37

Negative affect .76*** .80*** .46*** .68*** .62*** .09 .13** .50*** .05 .08 .52*** -.01 -.02

Detachment .81*** .71*** .24*** .36*** .52*** -.05 -.09 .63*** .09 .14*** .81*** .38*** .54***

Antagonism .49*** .31*** -.11* -.17* .44*** .17** .24*** .54*** .26*** .37*** .44*** .11* .18**

Disinhibition .58*** .47*** .03 .04 .64*** .37*** .54*** .45*** .06 .09 .41*** .03 .05

Psychoticism .64*** .57*** .22*** .32*** .48*** .02 .02 .56*** .21*** .30*** .52*** .08 .11

IRI (n = 357)

Fantasy -.07 .06 .27** .37*** -.11* -.13* -.20** -.14* -.03 -.07 -.18** -.20*** -.22**

Pers. distress .37*** .41*** .29*** .42*** .32*** .05 .07 .25*** .08 .10 .17** -.14** -.17*

Pers.-taking -.51*** -.32*** .12* .17 -.46*** -.18** -.26*** -.59*** -.32*** -.46*** -.46*** -.10 -.11

Emp. concern -.17** .05 .31*** .49*** -.01 -.00 -.01 -.36*** -.28*** -.39*** -.30*** -.22*** -.37***

B-PNI (n = 106)b

Grandiose .50*** .36*** -.10 .15 .42*** .19 .21 .41*** .20* .19 .35*** .02 .08

Vulnerable .68*** .53*** -.12 .24** .62*** .34** .39*** .46*** .11 .05 .47*** .11 .17

BSL (n = 106)b .54*** .63*** .26** .58*** .38*** .02 .03 .40*** .22* .17 .28*** -.15 -.08

BPAQ (n = 106)b .63*** .37*** -.25* -.01 .56*** .29** .30** .61*** .35*** .41*** .44*** .01 .08

Verbal .40*** .20* -.19 -.05 .34*** .15 .15 .45*** .33*** .41** .26** -.05 -.01

Physical .47*** .20* -.26** -.12 .36*** .13 .10 .54*** .37*** .46*** .40*** .06 .14

Anger .52*** .32** -.25* -.04 .55*** .35*** .40*** .48*** .24* .26* .35*** -.01 .03

Hostility .61*** .48*** -.10 .19* .55*** .28** .32** .48*** .20* .18 .40*** .00 .07

SF-36 (n = 106)b
SELF AND INTERPERSONAL FUNCTIONING SCALE 38

Ph. Function. .14 .09 -.00 --- .05 -.05 --- .16 .13 --- .12 .01 ---

Role-Physical -.15 -.20* -.12 --- -.06 .05 --- -.08 -.02 --- -.11 -.01 ---

Bodily Pain -.01 -.11 -.16 --- .03 .08 --- .01 -.02 --- .05 .09 ---

General Health -.31** -.34*** -.09 -.28* -.25** -.08 -.09 -.19* -.05 .00 -.20* .01 -.03

Vitality -.29** -.30** -.07 -.27* -.15 .02 .05 -.14 .04 .11 -.31** -.18 -.27*

Soc. Function -.29** -.33** -.11 -.32** -.22* -.03 -.06 -.23* -.12 -.08 -.16 .07 .05

Role-Emotional -.16 -.31** -.24* -.43*** -.08 .05 .03 -.01 .07 .18 -.11 .01 -.00

Mental Health -.31** -.47*** -.30** -.54*** -.18 .08 .07 -.22* -.15 -.09 -.12 .17 .15

RSES = Rosenberg Self-Esteem Scale. SWLS = Satisfaction With Life Scale. IPO = brief 20-item version of the Inventory of Personality

Organization. Id. diffusion = Identity diffusion. Pr. Defenses = Primitive defenses. PID-5-SF = Short form (100 items) of the Personality Inventory

for DSM-5. IRI-F = Interpersonal Reactivity Index-French Version. Pers. Distress = Personal distress. Pers.-taking = Perspective-taking. Emp.

Concern = Empathic concern. B-PNI = Brief Version of the Pathological Narcissism Inventory. BSL = 23-item Borderline Symptoms List. BPAQ

= 12-item version of the Buss-Perry Aggression Questionnaire. SF-36 = 36-item Medical Outcomes Study Short-Form Health Survey. Ph.

Function. = Physical Functioning. Soc. Function = Social Functioning. All instruments in their validated French versions.

Note. a Data only available for the nonclinical subsample. b Data only available for the clinical subsample. c Standardized beta coefficients from

multiple regression analyses (controlling for age) using the four SIFS elements as statistical predictors and each external criterion as the predicted

variable. Coefficients not shown in the absence of a statistically significant regression result. Res r = bivariate correlations using the unique

contribution of each element, after partialing out shared variance with the other three elements.

* p < .05. ** p < .01. *** p < .001.