Skip to main content
NCBI home page
Research in Psychotherapy : Psychopathology, Process, and Outcome logoLink to Research in Psychotherapy : Psychopathology, Process, and Outcome
. 2022 Jun 30;25(2):606. doi: 10.4081/ripppo.2022.606

Reliability and factorial validity of long and brief versions of the inventory of personality organization in a Latvian sample

Aija de Starceva-Apele 1, Malgozata Rascevska 1,✉
PMCID: PMC9422320  PMID: 35796597

Abstract

The aim of this study was to examine the construct validity and reliability of long and brief versions (57 and 16 items) of the inventory of personality organization (IPO) in a Latvian nonclinical sample. The actuality of this study is dictated by the number of persons suffering from personality disorders, and the excessive need for reliable constructs for the assessment of personality in the normality-abnormality range. The total number of participants of the study was 1118, recruited in several subsamples, which were employed for the investigation of the factor structure, psychometric properties, test-retest, and convergent validity of the IPO-57 and IPO-16 scales. The 3-factor structure, obtained by confirmatory factor analysis, was replicated in the Latvian-speaking sample for long and brief IPO versions. Both constructs achieved adequate model fit psychometric indices and sufficient internal consistency, and the test-retest reliability was also approved. The convergent validity of IPO with aggressivity, negative affectivity, and non-adaptive personality measures was established. In the current study, we adapted a very fundamental psychodynamic/psychoanalytic construct in the Latvian language and replicated the previously empirically established IPO three-dimensional model, whereas the novelty in IPO construct convergent validity research was the establishment of convergent validity with non-adaptive personality traits measured by the Personality Inventory for DSM-5 Brief Form (PID-5-BF). Both versions of IPO are recommended for further use in the research in Latvian language samples, for clinical purposes to assess the level of psychic functioning, and for treatment planning and evaluation purposes.

Key words: Personality organization, personality structure, personality disorders, non-adaptive personality traits, psychometric properties

Introduction

Origin and core premise of the personality organization construct

Kernberg (1984) and his team created the theoretical model of personality organization (PO) construct (Kernberg, 1984) in the 1990s within the framework of contemporary psychoanalytic object relations theory. However, the longer history of the PO model goes back not only to the mid-1960s, when the term borderline personality organization was used by Kernberg (1965), but even back at the end of the 19th century, when the term pathogenic organization was mentioned to describe the set of multiple consequences of experienced trauma (Freud, 1893/1962). The most influential, direct forerunners of the concepts of defensive splitting and other primitive emotional states were Klein (1935) and Fairbairn (1946). According to Kernberg (2012), Erikson’s theories, especially - ego identity and ego diffusion concepts (Erikson, 1956), were also significant for the PO construct.

The core premise of the PO construct states that emotional ill-being, identity disturbances, and problematic behavioural derivates, typical of personality disorders, are caused by an underlying psychic structure or organization, which is of nature and nurture1 etymology (Kernberg, 1984; Kernberg & Caligor, 2005; Lenzenweger et al., 2001). From a developmental perspective, the internalized emotional relations between self and significant others become the fundamental building blocks of the human mind, stored in affective memory, represented by the internalized models of behaviour, and capacity of cognitive framing or the modulation and the contextualization of affects in perceived relations between the self and others (Kernberg, 2015; Yeomans et al., 2015). PO measures the patterns of affective, cognitive, and behavioural responses to environmental stimulation (Kernberg, 2015) through dimensions of psychic functioning (Kernberg, 1984; Lenzenweger et al., 2001; Yeomans et al., 2015), recently also called aspects (Yeomans et al., 2015) or key domains of personality functioning (Clarkin et al., 2017), namely: primitive defences (PD), identity diffusion (ID), and reality testing (RT) (described further in text). Accordingly, the actual degree of disturbance (or optimal condition) in these three dimensions form the particular order (Doering & Horz, 2012) or PO, which fits somewhere in the continuum from normal to abnormal personality characteristics. PO construct is primarily aimed to assess the middle layer of the normality-abnormality continuum, which is characteristic first, for personality disorders, second, for other enduring psychopathologies (Lenzenweger et al., 2001). Thus, PO assesses enduring (chronic) characterological organization of personality (Kernberg, 1984), also known by the analogous term - psychological structure, defined as a stable and enduring pattern of mental functions (Yeomans et al., 2015).

The clinical utility of the PO construct is not limited to practitioners of various forms of psychoanalytic/psychodynamic treatment, it can be used as a supplementary assessment tool for psychiatric purposes (Doering & Hoerz, 2012). However, for the professionals of the psychoanalytic/ psychodynamic approach, the PO construct is a daily working tool - essential not only for the diagnosis and formulation of psychotherapeutic goals and treatment strategies - but, above all, PO dimensions are embodied as a clinical thinking mode, permanently present in clinical work. The PO construct is one of the most central constructs in contemporary psychoanalytic approaches (Koelen et al., 2012). The aim of the current study was to examine the construct validity, internal consistency, and test-retest reliability of the inventory of personality organization (IPO; Lenzenweger et al., 2001) in a 57-item version (hereafter IPO-57) and its brief form IPO-16 (Zimmerman et al., 2013) in a Latvian non-clinical sample.

IPO tool and its dimensions

The construct of PO from a theory, initially based on clinical observations, was first embodied in the form of a semi-structured interview - structured interview of personality organization (STIPO; Stern et al., 2010). The reliability and validity of STIPO and its revised version STIPO-R (Clarkin et al., 2017) were established in English- speaking samples (Clarkin et al., 2017; Stern et al., 2010). Parallelly to the development of the semi-structured interview, which is a very informative, though demanding clinical assessment tool in terms of time and professional resources, a self-report measure - the IPO (Lenzenweger et al., 2001) was validated (even before the structured interview), which made the PO concept available to a broader circle of clinicians and researchers.

Although, the IPO dimensions PD, ID and RT share the common affect-related basis, any dimension reveals a particular aspect of psychic functioning. PD indicates an individual’s capacity/incapacity to experience complex and well-modulated affects without the loss of impulse control (Yeomans et al., 2015, p. 7). The PD dimension contains statements on the main primitive psychic mechanisms applied unconsciously and consciously to deal with the affects raised by intrapsychic or extra psychic stimuli. PD determines the subjective experience of individuals who suffer from poor/insufficient ability to regulate or to understand and absorb the affect (Yeomans et al., 2015, p. 7), also called deficiencies of cognitive framing, mentioned before (Kernberg, 2015). According to the most recent neuro-psychoanalytical assumptions, repression plays an actuating role (Smith & Solms, 2018) before other, secondary order PD (listed further) are activated due to the prevalence of peak affects, which cannot be regulated in a satisfactory way and thus bypass thinking by means of premature automatization (Smith & Solms, 2018). The main PD are splitting, projective identification, idealization/devaluation, primitive denial, omnipotence, omnipotent control, and acting out (Yeomans et al., 2015; Kernberg, 1984). These defence mechanisms are responsible for the excessive mood swings and impulsivity, but the latter - launches the impulsive outbursts of verbal and/or physical aggression and risky behaviour.

Second IPO dimension - ID reveals a person’s eventual tendency to a feeble, fragmented self-concept or persistent negative identity, an insufficiently integrated concept of others, boundary confusion, and inconsistent goals (Kernberg & Caligor, 2005). ID dimension reveals difficulties to modulate affects regarding the perception of self and other persons. A central aspect of coherent identity is the capacity to see oneself (and another person) in the range of healthy ambivalence, specifically - psychic ability of integrating positive and negative affects and aspects of self (and others) in the perception of self (and others) under the influence of defensive splitting, which brings diffused identity symptomatology: the contradictory and unstable self-regard, or seemingly very convincing and stable, but unrealistic image of oneself (or other) (Kernberg et al, 2015; Yeomans et al., 2015). The healthy end of ID dimension refers to the coherent concept of self as well as behaviour that reflects self-coherence (Yeomans et al., 2015).

RT dimension assesses the capacity to differentiate self from non-self or intrapsychic stimuli from outer reality, as well as the psychic ability for empathy in ordinary social reality contexts. Difficulties of RT are connected with excessive affective arousal - permanence of peak affects or painfulness of affects (in the past and/or in the present), what influences realistic perception and learning about reality - capacity to differentiate animate and inanimate reality, the perception of the present environment and the identification of social reality (Kernberg, 2015). Very severe disturbances of RT function are typical for the psychotic organization of thoughts and behaviour or psychotic states (Kernberg, 1984; Yeomans et al., 2015).

Existing research on IPO-57 construct validity provides evidence for several possible factor structures (Baretto et al., 2017; Berghuis et al., 2009; Ellison & Levy, 2012; Lenzenweger et al., 2001; Preti et al., 2015). According to Lenzenweger and colleagues (2001), IPO- 57 2- and 3-factor structures, yielded in the original study, provide the best fit for Kernberg’s theoretical construct, where the aspects of PD and ID collude and are considered as generic borderline phenomena, whereas RT impairments form an additional, separate factor. Several short IPOs have been validated: Japanese (Igarashi et al., 2009), German (Zimmerman et al., 2013), French-Canadian (Verreault et al., 2013), and others.

This research was motivated by clinical and scientific requests for a Latvian language version of IPO. Accumulated data from epidemiological studies reveal that the prevalence of at least one personality disorder ranges from 5% and 15% (median=11.5%) of research samples (Morgan & Zimmerman, 2018, p. 69). Therefore, we intended to investigate the theoretically described PO dimensions and previously empirically established factor structure of IPO-57 and IPO-16 by means of the hypothetical-deductive testing of existing models using confirmatory factor analysis (CFA; Haig, 2005), since factor analysis plays a crucial role in the development of comprehensive models for the definition of personality traits (Livesley, 2018). In regard to the original PO model, which claims that PO’s underlying structure forms three factors (PD, ID, and RT), we decided to examine the construct validity of IPO-57 (Lenzenweger et al., 2001) and IPO-16 (Zimmermann et al., 2013) in the Latvian population by highlighting two research questions: i) whether the adapted IPO-57 and IPO-16 Latvian versions will confirm the 3-factor structure and achieve sufficient model fit psychometrics; and ii) what the test-retest reliability, internal reliability, and convergent validity of IPO-57 and IPO-16 are.

Positive and negative affects and aggression measures were included to perform convergent validity (in the current and also in previous IPO research) because all IPO dimensions are directly related to the psychic ability to deal with affective experiences. Specifically, affective disbalance, negative affectivity as a particular psychopathological symptom, impulsivity, and uncontrollable expression of aggression as a problem of impulse control are broadly acknowledged psychopathological symptoms for character pathologies (Jorgensen, 2018; Livesley, 2018). PD dimension reveals the tendency to use primitive defensive mechanisms, ID dimension explores the affective perception of self and others, whereas RT examines the capacity to deal with external reality and to differentiate inner processes: affects, thoughts, fantasies from real-world phenomena. There is ongoing, intensive research on the relationship of non-adaptive personality traits construct, as a proposed measure of the third section of Diagnostic and Statistical manual of mental disorders (5th ed.; DSM-5; American Psychiatric Association, 2013a) with other measures of personality traits, for example, see the study of Stover and colleagues (2019). The authors of this research considered as highly important the examination of the association of PO construct with non-adaptive personality traits, which is a novelty in the IPO research field, because: i) the dimensions/factors of both constructs measure the personality in the normalityabnormality continuum; ii) the PO construct was one of the dimensional constructs that, among others, influenced the development of the alternative classification model for personality disorders (Livesley, 2018) of the DSM-5 (APA, 2013a).

Materials and methods

Participants

The total number of participants whose data were enrolled in the data analysis for this study was 1118. In the total number of this study 7 subjects or 6% of participants were not included, the data of 4 participants were not included due to the high percentage of missing values, whereas the data of 3 respondents were considered as outliers (see in Data analysis). All 1118 or 100% of participants were native Latvian speakers, which was the baseline condition for participation in the research (other sociodemographic characteristics, see Table 1). The population of Latvia comprise approximately 1.5 million adult persons (above 18 years), and 0.95 million or 62% of them, are Latvian speaking persons (Central Statistical Bureau Republic of Latvia, 2021).

We utilized several sets of research samples (see Table 1) at two stages of data gathering: i) the data for the IPO- 57 sample (n=925) and three subsamples (n=269, n=622, n=60) were gathered in 2020; and ii) the data for the IPO- 16 sample (n=193) were collected in the late spring of 2021. The sizes of samples for CFA were planned according to the following approach accepted in psychometrics: at least 10 respondents per item for CFA and at least N=200-300 sample size. For the IPO-57 sample, 925 voluntary participants were recruited - 70% (652) in four universities and 30% (276) in three other environments - the governmental and municipality institutions, the public services sector, and the hobby-groups; 96% (888) filled in pen and paper questionnaires, whereas 4% (37) preferred to fill in the electronic questionnaire. Students from one university were offered participation in the IPO testretest procedure (n=60). The IPO-16 sample consisted of 62% (120) of respondents from two universities and one design college and 38% (73) from two other environments - the governmental and municipality institutions and the public services sector; 16% (31) of IPO-16 sample filled in the pen and paper version of IPO-16 and PID-5-BF surveys, whereas 84% (162) of respondents filled in the electronic survey using Google Forms.

Measures

The demographic data survey included questions about: i) age, where respondents were asked to reveal their age in an exact number of years; ii) gender, where three categories (male, female, and other) were offered; iii) their education level, where six categories (primary, secondary, secondary-professional, college, bachelor, and master’s or higher) were offered; and iv) nationality, where two categories (Latvian and other) were offered (see in Participants and Table 1).

Personality organization was measured with the 57 items version of the Inventory of Personality Organization (IPO-57, Lenzenweger et al., 2001). IPO-57 is a self-assessment questionnaire that measures three factors: PD (16 items), ID (21 items), and RT (20 items). Every statement is evaluated with a 5-point Likert scale ranging between 1 (never true) and 5 (always true). IPO-57 was adapted for the Latvian population in this research (for the reliability coefficients of Cronbach’s alpha and Mc- Donald’s omega, see Table 2).

The second measure of personality organization applied in the current study was the 16 items version of the IPO-16 (Zimmermann et al., 2013). IPO-16 is a short version of IPO-57, validated and standardized for the German language. The PD subscale consists of 5 items, the ID subscale of 6, and the RT scale of 5 (for the reliability coefficients of Cronbach’s alpha and McDonald’s omega, see Table 2).

Positive and negative affects were measured with the Latvian language version of the positive and negative affect scale (PANAS; Watson et al., 1988; Upmane, 2010), similarly to the original scale, which consists of positive affect (PA; 10 items) and negative affect (NA; 10 items) subscales, evaluated with 5-point Likert scales from 1 (very slightly or not at all) to 5 (extremely). The Latvianlanguage version of PANAS was used as a measure for IPO-57 convergent validity; thus, the researchers repeated the convergent validity measuring pattern from the original IPO research (Lenzenweger et al., 2001; for the reliability coefficients of Cronbach’s alpha and McDonald’s omega, see Table 2).

Table 1.

Sociodemographic and descriptive statistics characteristics of participants.

Sociodemographic Samples
and descriptive IPO-57 PANAS and BPAQ PID-5-BF Test-retest IPO-16
statistics sample sample sample sample for IPO-57 sample
characteristics (n=925) (n=269) (n=622) (n=60) (n=193)
Age 18-68 18-47 18-68 18-44 18-68
M (SD) 26.54 (9.55) 23.17 (4.18) 28.05 (10.80) 23.10 (5.54) 23.17 (8.55)
Gender
Females 71.6% 171 64.6% 468 75.4% 49 81.7% 106 54.9%
Males 28.0% 98 36.4% 150 24.1% 11 18.3% 83 43.0%
Did not indicate 0.4% - - 2 0.3% - - -
Other - - - 1 0.2% - - 4 2.1%
Education
Primary 2.2% 4 1.5% 15 2.4% - - 20 10.4%
Secondary 31.9% 158 58.7% 122 19.6% 38 63.3% 93 48.2%
Professional 8.7% 38 14.1% 41 6.6% 11 18.3% 33 17.1%
College 18.8% 16 5.9% 153 24.6% 4 6.7% 5 2.6%
Bachelor 27.7% 45 16.7% 207 33.3% 5 6.3% 24 12.4%
Master’s and higher 10.5% 8 3.0% 81 13.0% 2 3.3% 18 9.3%
Did not indicate 0.3% - - 3 0.5% - - - -
Open in a new tab

Aggression was assessed with the Buss-Perry aggression questionnaire (BPAQ) consists of four subscales: physical aggression (8 items), verbal aggression (5 items), anger (5 items), and hatred (7 items), evaluated with 5- point Likert scales from 1 (extremely uncharacteristic of me) to 5 (extremely characteristic of me) (Buss & Perry, 1992; Gaitniece-Putane, 2008). The verbal aggression variable was excluded from the research because this subscale did not achieve an acceptable internal consistency level (for the reliability coefficients of Cronbach’s alpha and McDonald’s omega, see Table 2).

For the examination of non-adaptive personality traits, the Latvian-language version of the personality inventory for DSM-5-Brief Form (PID-5-BF; APA, 2013b; de Starceva-Apele & Rascevska, 2022) is a self-assessment measure of 25 items across 5 factors (5 items in each). Every item is evaluated with 4-point Likert scales from 0 (very false or often false) to 3 (very true or often true). Higher scores in subscales indicate non-adaptive personality traits (for the reliability coefficients of Cronbach’s alpha and McDonald’s omega, see Table 2).

Procedure

First, our study was accepted by the Ethical Committee of the Department of Psychology, Faculty of Education, Psychology and Art, University of Latvia (Protocol No. 134/15.03.2016). Then, we obtained the authorizations for the adaptation of IPO-57 and the use of PID-5- BF from their authors/intellectual rights holders (J. F. Clarkin for IPO-57; APA for DSM-5-BF). Additionally, we contacted one of the authors of the IPO-16 Germanlanguage version (J. Zimmermann) and received permission to conduct the adaptation survey with the use of the 16-item model construct. According to the standards of the Guidelines for Translating and Adapting Tests (2nd ed.; International Test Commission, 2017), IPO-57’s original English-language version was translated into Latvian by two independent translators who were not previously familiar with IPO-57 - one was an English philologist, the other was recruited via a translation agency. The authors of this research compared the two translated versions with the original, discussed the cultural and linguistic nuances, and agreed on the most clearly understandable formulations. We undertook a pilot round with IPO-57 with a group of 10 respondents, received their feedback, and made some final corrections to the sentence constructions of some of the unclear statements. We asked another independent translator to provide the back-translation to English, which was sent to Prof. J. F. Clarkin for approval. Since IPO-16 is a brief version of IPO-57, it did not require additional translation procedures.

Table 2.

Descriptive statistics, normality test and reliability of research variables.

Variables, samples M (SD) Me K-S/P Cronbach’s ɑ/Mc Donald’s ɷ Test-retest reliability
IPO-57 (n=925)
Primitive defences 33.22 (9.82) 33.00 0.001 0.89/0.89 0.80
Identity diffusion 45.58 (14.17) 45.00 0.001 0.92/0.92 0.81
Reality testing 34.80 (10.70) 32.00 0.001 0.90/0.90 0.87
Total 113.60 (31.05) 112.00 0.001 0.96/0.96 0.86
IPO-16 (n=193)
Primitive defences 12.00 (3.15) 12.00 0.001 0.71/0.73 0.84
Identity diffusion 15.51 (4.16) 16.00 0.025 0.73/0.74 0.75
Reality testing 11.05 (3.60) 11.00 0.001 0.75/0.75 0.73
Total 38.61 (8.94) 38.00 0.076 0.85/0.85 0.95
PID-5-BF (n=622)
Negative affect 4.20 (3.07) 4.00 0.001 0.78/0.81
Detachment 3.27 (2.13) 3.00 0.001 0.55/0.55
Antagonism 3.45 (2.45) 3.00 0.001 0.67/0.67
Disinhibition 3.86 (2.89) 4.00 0.001 0.78/0.79
Psychoticism 2.62 (2.19) 2.00 0.001 0.66/0.67
Total 17.39 (9.60) 15.00 0.001 0.88/0.88
PID-5-BF (n=193)
Negative affect 6.80 (3.42) 7.00 0.001 0.74/0.76
Detachment 4.98 (2.98) 5.00 0.001 0.68/0.69
Antagonism 4.55 (3.30) 4.00 0.001 0.77/0.78
Disinhibition 6.06 (2.96) 6.00 0.001 0.73/0.72
Psychoticism 5.75 (3.08) 6.00 0.001 0.71/0.72
Total 28.19(11.73) 27.00 0.058 0.88/0.88
BPAQ (n=269)
Physical aggression 18.69 (4.16) 18.00 0.001 0.60/0.61
Anger 11.94 (3.20) 12.00 0.001 0.62/0.61
Hatred 15.86 (3.37) 16.00 0.001 0.53/0.54
Total 67.59 (12.13) 66.00 0.001 0.78/0.80
PANAS (n=269)
Positive affects 33.02 (5.98) 33.00 0.055 0.80/0.80
Negative affects 24.87 (6.86) 38.00 0.039 0.80/0.80
Total 58.07 (11.37) 56.00 0.001 0.84/0.84
Open in a new tab

We contacted the responsible bodies of each setting, and the parties agreed on ethical obligations: voluntary participation, confidentiality, anonymity and the use of data only for research purposes. The ethical duties of the researchers (confidentiality, anonymity, use of data, etc.) were explained to each group of students and every single participant from any other environment.

Data analysis

SPSS 25.0 software (IBM Corp., 2017) was used for data processing and statistical data analysis. Additionally we employed R software (R Core Team, 2020). Due to the non-normal data distribution (see Results), we applied only non-parametric methods in this study: Spearman correlations and Chi-squared tests. The Mahalanobis distance was used for outlier analysis (on the basis of the IPO-57, 3-factor model), and three outliers were accordingly removed from the IPO-57 sample. No outliers were discovered in the IPO-16 sample.

To determine whether the underlying structure of our data matched the hypothesized models well, we applied diagonally weighted least squares (DWLS) estimations as the most popular CFA method for non-normal distributed data, based on the polychoric correlation matrix (Savalei & Rhemtulla, 2013). For the comparative fit index (CFI) and Tucker-Lewis index (TLI), the mandatory cut-off value is required to be above 0.90 (Hu & Bentler, 1999). For absolute model fit indices - the root mean square error of approximation (RMSEA) and the standardized root mean square residual (SRMR) - we relied on recommended cut-off values under 0.08 (Hu & Bentler, 1999). To increase model fit, we used the Modification Indices function, which is defined as χ2 value with 1 degree of freedom; values larger than 10.83 indicate that the parameter would have a P-value <0.001.

All scales were assessed by two reliability indices for more precise internal consistency assessments. For the calculation of the internal consistency of McDonald’s omega coefficient, we used R software. For the test-retest reliability estimation, we considered correlation scores of r=0.70 and higher as acceptable (European Federation of Psychologists’ Associations, EFPA, 2013). For the calculation of convergent validity, Spearman’s two-tailed correlations were used. According to the test review model (TRM; EFPA, 2013), when instruments are less similar, lower values than r=0.60 are considered adequate, whereas for instruments measuring similar constructs, the required correlation values are above r=0.60.

Results

Factorial validity

The data distribution in the samples (see Table 2) only partly reached the criteria of normal distribution, according to the Kolmogorov-Smirnov test with Lilliefors criteria. This dictated the use of CFA for not-normally distributed data (see Data analysis). In the current research, we ran CFA for the 3-factor model of the full and brief versions of the Latvian language IPO (see Table 3 for all conducted CFA).

The original, unchanged IPO 3-factor solution of IPO- 57 (Lenzenweger et al., 2001) in the Latvian sample was very close to an adequate model fit level; only the cut-off value of the TLI index was insufficient. We analysed modification indices and discovered that in the IPO-57 sample, the modification indices for the path of the 1st item of the RT subscale (38th item of IP0-57) was above the recommended cut-off point (10.83). When we placed this item in the PD subscale (the content of the PD scale allowed us to do that), the slightly modified IPO-57 3-factor model reached an adequate level in all model fit indexes (TLI=0.907). However, in the following statistical data analysis for the current research, we used the original IPO-57 version. The goodness of fit indices in the original IPO-57 research for the 3-factor solution were TLI>0.90 (Lenzenweger et al., 2001).

Table 3.

Fit indices of retrieved IPO-57 and IPO-16 3-factor confirmatory models.

Model Items Factors χ2 df CFI TLI RMSEA SRMR
IPO-57 57 3 6968.428 1536 0.900 0.896 0.062 (0.060-0.063) 0.074
IPO-57* 57 3 6412.240 1536 0.910 0.907 0.059 (0.057-0.060) 0.070
IPO-16° 16 3 340.735 101 0.974 0.969 0.051 (0.045-0.047) 0.043
IPO-16# 16 3 230.814 101 0.920 0.905 0.082 (0.068-0.096) 0.078
Open in a new tab

*Item RT1 added to Primitive Defenses factor; °retrieved from main sample (n=925); #retrieved from independent sample (n=193).

When we conducted CFA on an IPO-16 3-factor model with IPO-57 sample data (by extracting 16 items from 57), the obtained cut-off values of the model fit indices were exceptional CFI=0.974, TLI=0.969 (see Table 3 for other indices). Lastly, we tested the IPO-16 model fit on the data of the independent IPO-16 sample, and model fit indices reached the adequate range of model fit cut-off values (see Table 3). The descriptive statistics of IPO-57 and IPO-16 items show that the majority of mean values of IPO-57 items are in the range of 1.54 to 2.39, while the mean values of IPO-16 items are from 1.94 to 3.16 (see Table 4). Corrected item-total correlations show results above 0.30 for the IPO-57 scale and above 0.28 for the IPO-16 scale, whereas standardized factor loadings are above 0.52 for IPO-57 and above 0.54 for IPO-16 (see Table 4).

Table 4.

Standardized factor loadings, descriptive statistics and corrected item-total correlations of IPO-57 and IPO-16 3-factor solution.

IPO-57 Factor loadings M (SD) Corrected item-total correlation IPO-16 item Factor loadings M(SD) Corrected item-total correlation
item PD ID RT PD ID RT
1 0.64 1.93 (0.82) 0.52
2 0.72 2.42 (1.01) 0.60
3 0.56 2.20 (0.94) 0.47
4 0.66 2.20 (1.07) 0.52 1 0.37 2.70 (0.83) 0.28
5 0.53 1.71 (0.83) 0.43
6 0.63 1.95 (0.93) 0.54 2 0.58 2.33 (0.98) 0.45
7 0.66 1.90 (0.92) 0.55
8 0.73 2.42 (1.06) 0.62
9 0.71 2.01 (0.84) 0.60 3 0.63 2.08 (0.90) 0.39
10 0.54 2.05 (1.09) 0.46
11 0.63 2.09 (1.04) 0.54
12 0.69 1.97 (0.95) 0.59 4 0.77 2.31 (0.91) 0.55
13 0.63 1.86 (0.95) 0.53
14 0.71 2.25 (1.01) 0.62 5 0.80 2.60 (0.95) 0.64
15 0.68 2.04 (0.94) 0.58
16 0.74 2.27 (1.13) 0.64
17 0.61 1.97 (0.96) 0.53
18 0.60 2.55 (1.26) 0.50
19 0.60 1.94 (0.93) 0.51 6 0.60 2.39 (0.94) 0.43
20 0.64 2.12 (1.02) 0.55
21 0.70 2.33 (1.13) 0.60
22 0.58 1.78 (0.91) 0.47
23 0.74 2.08 (1.09) 0.62
24 0.74 1.96 (0.91) 0.62
25 0.62 1.92 (0.95) 0.52 7 0.54 2.08 (0.91) 0.39
26 0.64 2.10 (0.56) 0.56
27 0.62 2.21 (1.02) 0.53
28 0.69 2.38 (1.14) 0.58 8 0.69 2.74 (1.11) 0.52
29 0.72 2.50 (1.23) 0.62 9 0.71 3.16 (1.26) 0.54
30 0.74 2.39 (1.09) 0.63
31 0.57 2.20 (1.15) 0.49 10 0.56 2.57 (1.21) 0.43
32 0.71 2.39 (1.12) 0.61
33 0.68 2.10 (1.12) 0.58
34 0.75 2.19 (1.08) 0.59
35 0.73 2.21 (1.11) 0.58
36 0.69 2.27 (0.99) 0.58 11 0.53 2.61 (0.88) 0.42
37 0.69 2.08 (0.94) 0.60
38 0.59 2.55 (1.15) 0.49
39 0.58 1.66 (0.82) 0.44 12 0.80 1.94 (0.96) 0.61
40 0.74 2.05 (1.01) 0.61
41 0.58 1.54 (0.78) 0.43
42 0.68 1.74 (0.85) 0.53 13 0.60 2.06 (0.96) 0.43
43 0.68 1.97 (1.03) 0.55
44 0.67 1.59 (0.89) 0.45
45 0.63 1.45 (0.81) 0.39
46 0.72 1.64 (0.79) 0.57
47 0.63 1.64 (0.86) 0.48
48 0.66 1.66 (0.82) 0.50 14 0.72 2.20 (1.04) 0.52
49 0.69 1.81 (0.93) 0.53
50 0.59 1.63 (0.87) 0.43
51 0.60 1.57 (0.81) 0.42 15 0.46 2.63 (1.07) 0.34
52 0.61 1.71 (0.91) 0.45
53 0.61 1.55 (0.88) 0.42
54 0.52 1.56 (0.87) 0.37
55 0.73 1.60 (0.85) 0.53 16 0.71 2.22 (1.01) 0.54
56 0.68 1.90 (0.93) 0.53
57 0.63 2.10 (1.04) 0.51
Open in a new tab

Reliability and convergent validity

Cronbach’s alpha and McDonald’s omega coefficients yielded psychometrically adequate, good and excellent results for the internal consistency of IPO-57 and IPO-16 scales (see Table 2). All acquired correlations of IPO-57 summary scale with PD, ID and RT subscales reached r=0.91, r=0.92 and r=0.84 respectively, whereas correlations on the subscale level showed the following values: r=0.81 for PD and ID, r=0.64 for PD and RT, and r=0.62 for ID and RT, all r values being significant at the level of P<0.001. The IPO-16 summary scale correlation with PD, ID and RT subscales reached r=0.81, r=0.84, and r=0.79, whereas IPO-16 subscale correlations were r=0.55 for PD and ID, r=0.52 for PD and RT, and r=0.47 for ID and RT, all at the significance level of P<0.001.

The results of the test-retest reliability examination reveal an adequate strength of correlations between the first and second measures for IPO-57 and IPO-16 (see Table 2). The first and second measures were organized within an interval of four weeks. The IPO-16 subscales achieved a good range of test-retest reliability (e.g., 0.70≤r<0.80), while the total IPO-57 and IPO-16 scores reached an excellent range (e.g., r≥0.80). The convergent validity of IPO-57 and IPO-16 constructs was examined by correlating it with aggression, positive and negative affect measures, and non-adaptive personality trait measures (see Table 5). In the current study, PD, ID and RT subscales and the IPO-57 total scale significantly correlated with physical aggression, hatred, and anger in a range from r=0.33 to r=0.46. Also, the association of PD, ID and RT with negative affects were supported by our research; correlation coefficients were between r=0.38 and r=0.47, and thus an adequate range of cut-off values for the convergent validity for related constructs is reached. The expected negative association of IPO-57 factors with positive affects was proved only for the ID subscale; neither of the other two subscales (PD, RT) nor the total scale proved this association.

All IPO-57 variables showed statistically significant correlations with all non-adaptive personality trait domains (see Table 5). The majority of IPO-16 scales also significantly correlated with PID-5-BF suggested subscales: PD, RT, and the IPO-16 total scale correlated with all non-adaptive personality traits, but the ID subscale did not show a significant correlation with the antagonism trait, although it correlated significantly with other subscales. The recommended cut-off value of above 0.60 (EFPA, 2013), confirming the convergent validity of instruments measuring similar constructs (IPO-57, IPO-16 and PID-5-BF), was achieved by several subscales and the total scales of both IPO versions.

Discussion

In the current study, we explored the 3-factor structure and psychometric properties of IPO-57 and IPO-16 in a Latvian non-clinical population. The 3-factor structure, established in the original research into IPO-57 (Lenzenweger et al., 2001) and IPO-16 (Zimmermann et al., 2013), fitted the data of the Latvian-speaking population of the current study, and both acquired models showed an adequate level of goodness of fit indices. The findings on the factorial validity of IPO-57 in the Latvian sample support the 3-factor model established in the IPO original study (Lenzenweger et al., 2001), even more, the majority of items were not migrating among other subscales, except the 1st item of the RT subscale (38th item of IP0-57) which was prone to migrate to PD factor. The 3-factor structure of IPO-57 was acquired also in the Italian (Preti et al., 2015) and Portuguese (Barreto, 2017) samples and the values of the goodness of fit indices of both studies were close to the corresponding indices of IPO-57 Latvian language version, however Baretto and colleagues (2017) report on the mitigation of items among IPO-57 subscales. The results of our study are inconsistent with other established IPO factor structures in different cultural and social environments where CFA of the original IPO-57 3-factor solution did not reach an adequate model fit level (Igarashi et al., 2009; Ellison & Levy, 2012; Baretto et al., 2017). The model fit indices CFI=0.913, TLI=0.963, and RMSEA=0.067 of the 3-factorial IPO-16 German-language version (Zimmerman et al., 2013) were in a similar range to the corresponding indices of the Latvian language IPO-16, as well as the IPO-16 items did not migrate across the scales of IPO-16 German and Latvian language models.

Factor loadings, which is the criterion of the one-dimensionality of the models, were in a higher range for the IPO-57 model and in a lower range for the IPO-16 model in the Latvian language samples of the current study. According to the theoretical IPO model, the underlying psychic structures (PD, ID and RT), which reveal the actual level of psychic organization, are mutually highly related factors, which is empirically substantiated in both previous research and this study, but simultaneously PD, ID and RT are separate, distinguishable personality dimensions/ traits. Inter-subscale correlations of Latvian language IPO-57 are similar to the results of the original IPO-57 research, where PD and ID correlations are above 0.80, whereas for PD and RT and ID and RT in the area of 0.60 (Lenzenweger et al., 2001), which fully support the theoretical assumptions of the PO construct, that PD and ID scales contain the items on mental phenomena, which partly overlap and are prone to collude in one factor. Previous research on IPO-16 (Zimmermann et al., 2013) reveals a slightly higher range of correlation coefficients between the total scale and subscales, namely 0.87 till 0.94, in comparison to correlation coefficients obtained in actual study, which ranged from 0.79 till 0.81. PD, ID and RT subscales of the Latvian language IPO-16 correlate at the level of 0.50, which is lower than the correlation between IPO-57 subscales, thus the IPO-16 tool distinguishes better the separate dimensions of PO. According to Sellbom and Tellegen (2019, p. 4), in psychological assessments of personality, researchers should not be too fixated on model fit indices as models are hardly ever entirely unidimensional; thus, theoretical expectations should be tempered. To sum up all the results of the current study on the factor structure of IPO-57 and IPO- 16, we can conclude that the factorial validity of the Latvian- language IPO long and brief versions was established, and Kernberg’s theoretical model of PO, which is based on three personality dimensions/aspects (PD, ID and RT), was supported by our data. The internal consistency levels of Latvian language version of IPO-57, achieved in the current study, were in a good or excellent range, whereas those of IPO-16 reached an acceptable level on subscales and the good level on the total scale. The findings of the current study are consistent with the previous research, where Cronbach’s alpha coefficient of IPO-57 subscales was in a range from 0.81 to 0.88 (Lenzenweger et al., 2001), and Chronbach’s alpha coefficient for the IPO-16 German-language total scale reached 0.85 (Zimmermann et al., 2013). Accordingly, the results of our study show, that the items of both Latvianlanguage IPO constructs are internally congruent on the level of subscales and total scales.

Table 5.

Spearman correlation coefficients of IPO-57 and IPO-16 with affects and non-adaptive personality traits for convergent validity.

Convergent validity measures IPO-57 IPO-16
Primitive defences Identity diffusion Reality testing Total defences Primitive Identity diffusion Reality testing Total
PID-5-BF
Dissociation 0.66** 0.65** 0.53** 0.69** 0.51** 0.31** 0.41** 0.49**
Detachment 0.46** 0.42** 0.40** 0.48** 0.26** 0.15* 0.33** 0.30**
Psychotism 0.39** 0.42** 0.58** 0.51** 0.39** 0.32** 0.59** 0.52**
Negative affects 0.64** 0.69** 0.53** 0.70** 0.49** 0.61** 0.40** 0.62**
Antagonism 0.51** 0.49** 0.42** 0.53** 0.35** 0.13 0.25** 0.53**
PID-5-BF total 0.73** 0.73** 0.66** 0.79** 0.54** 0.42** 0.53** 0.60**
BPAQ
Physical aggression 0.42** 0.41** 0.35** 0.46** - - - -
Anger 0.35** 0.39** 0.33** 0.42** - - - -
Hatred 0.38** 0.37** 0.39** 0.45** - - - -
BPAQ total 0.40** 0.40** 0.36** 0.45** - - - -
PANAS
Positive affects –0.06 –0.15* –0.01 –0.09 - - - -
Negative affects 0.42** 0.39** 0.38** 0.47** - - - -
Open in a new tab

Both Latvian language IPO versions were proved as reliable measures in terms of consistency of results at a time interval of four weeks. The IPO-16 and IPO-57 subscales achieved a good and excellent range of correlations between two measures, according to the criteria of TRM (EFPA, 2013). Our results go in line with previous findings of the test-retest reliability of IPO measures, where subscale and total scale correlation coefficients between first time and second times measures ranged from 0.72 to 0.78 for IPO-57 tool (Lenzenweger et al., 2001), and - from 0.79 to 0.86 for IPO-16 subscales (Zimmermann et al., 2013).To summarize, the 3-factor structure of the Latvian- language versions of IPO-57 and IPO-16 was validated, and internal consistency and test-retest reliability psychometrics were successfully approved. Thus, we conclude that the expectations (derived on the basis of Kernberg’s PO model and previous research) were met by the data of the Latvian sample of the current study.

The convergent validity of both IPO constructs was investigated through association with aggressivity, affectivity, and non-adaptive personality traits. We anticipated and were proven correct that IPO-57 scales were associated with PO dimensions and all subtypes of aggression and negative affects. These findings are in line with the original IPO research, where irritability and assault correlated with all IPO subscales in a range from 0.25 to 0.32 (Lenzenweger et al., 2001), nevertheless our study revealed an even higher correlation of the PO construct with aggression, which was above the level of 0.40, thus convergent validity for related (but not similar) constructs was reached and corresponded to the range of the suggested values of TRM (EFPA, 2013).

We also expected the negative association of PD, ID and RT with positive affects, but this was the only one that was not proved to the expected degree - only ID showed a negative association with positive affects. These findings contradict the findings of the original IPO research, where a negative association of positive affects with RT and PD were found (Lenzenweger et al., 2001). However, these results indirectly support the findings of Sollberger and colleagues (2011) that ID distinguishes individuals with mild borderline personality disorder from individuals with severe borderline disorder - it was found that the group with high ID scores had statistically significantly higher rates of negative affect, stress symptoms, general psychiatric symptoms, anxiety, and depression. Nevertheless, due to the affective instability of the personality disorders patients, who suffer from a sense of incoherence of their personalities and therefore feel, behave, and see themselves differently at different times (Lenzenweger et al., 2001), the presence/absence of positive affects has to be estimated cautiously by taking into consideration the hypomanic or even manic tendencies that may take place.

The IPO-57, IPO-16 and PID-5-BF constructs are designed for the assessment of dimensions or traits characteristic of personality disturbances. Statistically significant correlations between non-adaptive personality traits and IPO-57 and IPO-16 dimensions were expected, and the acquired correlation coefficients revealed a strong mutual associations between all scales of both constructs, except for ID (IPO-16) with antagonism. If we follow test development and adaptations guidelines (EFPA, 2013) that suggest high levels of correlations (above r=0.6, see more in Materials and methods and Results) for convergent validity measures, which assess similar phenomena, then from the current research, we can derive conclusions about the strongest associations between the PO construct and PID-5-BF (on the level of subscales) of two pairs of associations: PD and dissociation and PD and negative affects, and ID and dissociation and ID and negative affects. Thus, the convergent validity of the PO construct with aggression, positive and negative affects, and non-adaptive personality traits was established, which is in line with the theoretical premises of the PO construct and object relations theory.

Kernberg assumes that unprocessed experiences of peak-affective states motivate individuals’ general irritability and hypersensitivity to frustration and pain: Positive and negative affective memories are built up separately in early internalization of these experiences and, later on, are actively split or dissociated from each other […]. This development evolves into two major, mutually split domains of early psychic experience, an idealized and a persecutory or paranoid one (Kernberg, 2012, p. 9). In addition, Jorgensen (2018, p. 113) states that individuals with strong affects and impulses are unable to experience themselves as authors of their own behaviour, and their sense of agency and autonomy is impaired, which is associated with identity diffusion and painful experiences of incoherence and inauthenticity. Thus, we assume that the two pairs of associations illustrate very central areas of disturbance, which is in line with possible endophenotypes of personality disorders - affective instability, impulsivity, and cognitive disorganization (Siever, 2005).

Limitations and future directions

There are several limitations to the current study. Although, according to the rule of thumb, the sample sizes were sufficient and the acquired IPO models achieved an adequate model fit, we assume that bigger sample sizes would provide additional evidence about IPO-57 and IPO-16 models. We also admit that our research sample did not have optimal gender and age distribution. The reliability coefficients acquired for the PID-5-BF Detachment subscale and BPAQ Hatred subscale did not reach an ideally convincing level, therefore the association of these two variables with IPO factors should be interpreted cautiously. Unfortunately, we could not include verbal aggression since it did not reach an adequate internal validity level in our research subsample. The clinical validity of IPOs and gender-specific aspects of IPO-57 and IPO-16 are the main tasks for future research on the PO construct in the Latvian population, as this would allow us to prepare the instrumental basis for further use in clinical and non-clinical populations and in psychotherapy research.

Clinical implications

We consider the PO construct - the full and brief versions studied in the current research - as a dimensional map of the psychic continuum for mental health professionals dealing with personality pathology. We reason that the use of a psychometrically sound construct of IPO-57 or IPO-16, used in the form of a pen and paper, electronic questionnaire, or assessment interview is clinically beneficial in many ways, we will name three of them. First, it enriches practitioners’ ability to identify very difficult, sometimes seemingly overlapping clinical phenomena of disordered personality, for this reason - IPO dimensions (PD, ID, RT) help to group the symptoms and understand the peculiarity of the disturbance of every patient. Accordingly, IPO-57 and IPO-16 can be used as a main or an additional assessment tool for disturbances of personality and other psychopathologies.

Second, thorough assessment of PD, ID and RT helps to formulate a psychotherapeutic treatment plan and form a prognosis of treatment response. In the mental health institutions where the team of professionals decide which counselling and/or psychotherapy approach would be more beneficial for the particular client, assessment information gathered through an IPO would serve additional predictive accuracy. Whereas, specifically for the practitioners of psychodynamic/psychoanalytic treatment(s) assessment data would be considered by taking the decision between interpretive or supportive psychotherapy (depending on the severity of disturbances in every PO dimension/ aspect), predicting the eventual amount of transference-interpretations (non-low-medium-high) (Sollberger et al., 2011), and observing or assessing the patient’s clinical progress. Additionally, in regard to the view of Doering and Herz (2012) about the benefits of the application of the IPO tool parallelly to other psychiatric assessment tools, we consider that IPOs could be highly informative for treatment planning purposes if combined with other psychodynamic/psychoanalytic grounded instruments, for example, the assessment of anaclitic-introjective personality configurations for a more precise understanding of patients disturbances, emotional needs and progress of treatment, examined by Werbart and colleagues (2017).

Third, since the IPO items are thoroughly formulated and contain the evidence of several decades of clinical observations and empirical research from multiple studies in different cultural environments, the IPO item formulations are of great clinical help, in distinguishing a patient’s resistances in the course of psychotherapeutic treatment. As an illustration, the 3rd item on the PD subscale of IPO- 57 reads as follows: I feel it has been a long time since anyone really taught or told me anything I did not already know. There are patients who use this type of statement regularly during their treatment since omnipotence, boredom, devaluation, and a tendency to simplify are common defensive patterns for the patients with the sustainable character pathology. So, when a patient makes this type of utterance, it might be helpful for the clinician to recall the IPO-57 construct in which this position was specified as a commonly used manifestation of premature psychic defences. Thus, if the PO construct is used for thorough assessment of the peculiarities of the character pathology, it can prepare the professional for the transference-countertransference difficulties - this understanding can be creatively used for a better understanding of the patient-therapist inter-subjective relationship, especially - perception of therapists interpretations, and the difficulties of the co-creational process in the course of treatment, see also Knight (2021).

The authors of the current research share the view that the disorder-centred approach should evolve to a personcentred approach (Luyten & Fonagy, 2021) in the conceptualization (and treatment) of personality disturbances. We think, the alternative personality disorders model, offered by the DSM-5 classification (APA, 2013a), besides the categorical model, is an important turning point towards the development of an accurate assessment of personality in the normality-abnormality continuum and strengthening the client-centred values in both - defining and treatment of personality disorders. Ultimately, beside the aim to establish the factorial validity and reliability of IPO-57 and IPO-16 in the Latvian population, this study was our response to the invitation (APA, 2013a, p. 809) to continue the research on the dimensional assessments of personality traits and personality disturbances.

Conclusions

The current study proved the factorial validity and internal consistency of the PO construct in the sample of another cultural environment, which serves additional evidence that RT, ID and RT are scientifically grounded personality traits assessing normal and abnormal personality functioning among different cultures. The clinical utility of the construct of PO is in its capacity to reveal the tendency towards adverse or/and malignant character-structures, which was verified by proving the evidence of the convergent validity of Latvian language long and brief versions of the IPO with negative affects, aggression, and nonadaptive personality trait measures. From now on, two PO construct measurement versions in Latvian language - IPO- 57 and IPO-16 are available for clinicians for case conceptualization and clinical decision making (especially for personality disorders), as well for further research purposes.

Acknowledgements

We are grateful to Prof. J. F. Clarkin, Prof. J. Zimmermann, and Prof. S. Horz-Sagstetter for their support during the current study. This research received no specific grant from any funding agency in the public, commercial, or non-profit sectors.

Footnotes

1The co-influence of both neurobiological factors and emotional experience are responsible for the development of personality (Kernberg, 1984, Lenzenweger et al., 2001, Kernberg & Caligor, 2005), what is in line with the premise of the threshold liability model - a cornerstone for a dimensional approach, which states that the combination of genetic factors and exposure to emotionally stressful experiences predispose the development of personality disorders (Jang & Vernon, 2018). The empirical evidence verifies the continuity between childhood psychopathology and the onset of personality disorders in adulthood (Zarrella et al., 2017).

References

  1. American Psychiatric Association (2013a). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association. Available from: 10.1176/appi.books.9780890425596 [DOI] [Google Scholar]
  2. American Psychiatric Association (2013b). Online assessment measures: The Personality Inventory for DSM-5-Brief Form (PID-5-BF) - Adult. Available from: https://www.psychiatry.org/File%20Library/Psychiatrists/Practice/DSM/APA_DSM5_The-Personality-Inventory-For-DSM-5-Brief-Form-Adult.pdf [Google Scholar]
  3. Barreto J. F., Matias M., Carvalho H. M., Matos P. M. (2017). Uncovering personality structure with the Inventory of Personality Organization: An exploration of factor structure with a Portuguese sample. European Review of Applied Psychology/ Revue Europeenne de Psychologie Appliquee, 67(5), 247-257. doi:10.1016/j.erap.2017.08.001. [Google Scholar]
  4. Berghuis H., Kamphuis J. H., Boedijn G., Verheul R. (2009). Psychometric properties and validity of the Dutch Inventory of Personality Organization (IPO-NL). Bulletin of the Menninger Clinic, 73, 44-60. doi:10.1521/bumc.2009.73.1.44. [DOI] [PubMed] [Google Scholar]
  5. Buss A. H., Perry M. (1992). The aggression questionnaire. Journal of Personality and Social Psychology, 63(3), 452-459. doi:10.1037/0022-3514.63.3.452. [DOI] [PubMed] [Google Scholar]
  6. Central Statistical Bureau Republic of Latvia. (2021). IRE020. Population, population change and key vital statistics by ethnicity 1980-2021. Official Statistics portal. Available from: https://data.stat.gov.lv/pxweb/en/OSP_PUB/START__POP__IR__IRE/IRE020/ Accessed: November 1, 2021 [Google Scholar]
  7. Clarkin F. J., Caligor E., Stern B., Kernberg O. F. (2017). Structured Interview of Personality Organization-Revised (STIPO-R). The Center for Transference-focused Psychotherapy. Available from: https://www.borderlinedisorders.com/assets/STIPORmanual.July2021.pdf [Google Scholar]
  8. de Starceva-Apele A., Rascevska M. (2022). Relationship of capacity to love with personality organization and non-adaptive personality traits [Manuscript in preparation]. Department of Psychology, Faculty of Education, Psychology and Art, University of Latvia. [Google Scholar]
  9. Doering S., Horz S. (2012). Handbuch der Strukturdiagnostik: Konzepte, Instrumente, Praxis. Schattauer GmbH. [Google Scholar]
  10. Ellison W. D., Levy K. N. (2012). Factor structure of the primary scales of the Inventory of Personality Organization in a nonclinical sample using exploratory structural equation modeling. Psychological Assessment, 24, 503-517. doi:10.1037/ a0026264. [DOI] [PubMed] [Google Scholar]
  11. Erikson E. H. (1956). The Problem of Ego Identity. Journal of the American Psychoanalytic Association, 4, 56-121. doi:10.1177/000306515600400104. [DOI] [PubMed] [Google Scholar]
  12. European Federation of Psychologists’ Associations. (2013). EFPA Review Model for the Description and Evaluation of Psychological and Educational Tests, version 4.2.6. Available from: http://assessment.efpa.eu/documents-/ [Google Scholar]
  13. Fairbairn W. D. (1946). Object-relationships and dynamic structure. International Journal of Psychoanalysis, 27(1-2), 30-37. [PubMed] [Google Scholar]
  14. Freud S. (1962). On the Psychical Mechanism of Hysterical Phenomena: A Lecture. In Strachey J. (Ed. and Trans.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 3, pp. 25-39). Hogarth Press. [Google Scholar]
  15. Gaitniece-Putane A. (2008). Relationship of aggression, emotional intelligence and stoicism for men and women in age of 20-25 and 30-35 (Publication No. 10094) [Doctoral dissertation, University of Latvia]. E-resource repository of the University of Latvia. Available from: https://dspace.lu.lv/dspace/handle/7/4530 [Google Scholar]
  16. Haig B. D. (2005). Exploratory factor analysis, theory generation, and scientific method. Multivariate behavioural research, 40(3), 303-329. doi:10.1207/s15327906mbr4003_2. [DOI] [PubMed] [Google Scholar]
  17. Hu L., Bentler P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55. doi:10.1080/10705519909540118. [Google Scholar]
  18. IBM Corp. (2017). IBM SPSS Statistics for Windows, Version 25.0. IBM Corp. [Google Scholar]
  19. Igarashi H., Kikuchi H., Kano R., Mitoma H., Shono M., Hasui C., Kitamura T. (2009). The inventory of personality organisation: its psychometric properties among student and clinical populations in Japan. Annals of General Psychiatry, 8, 9. doi:10.1186/1744-859X-8-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jang L. K., Vernon P. A. (2018). Genetics. In Livesley W. J., Larstone R. (Eds.), Handbook of personality disorders: Theory, research, and treatment (pp. 235-250). Guilford Press. [Google Scholar]
  21. Jorgensen C. R. (2018). Identity. In Livesley W. J., Larstone R. (Eds.), Handbook of personality disorders: Theory, research, and treatment (pp. 107-122). Guilford Press. [Google Scholar]
  22. Kernberg O. F. (1965). Notes on countertransference. Journal of the American Psychoanalytic Association, 13, 38-56. doi:10.1177/000306516501300102. [DOI] [PubMed] [Google Scholar]
  23. Kernberg O. F. (1984). Severe personality disorders. Yale University Press. [Google Scholar]
  24. Kernberg O. F. (2012). The inseparable nature of love and aggression: Clinical and theoretical perspectives. American Psychiatric Publishing. [Google Scholar]
  25. Kernberg O. F. (2015). Neurobiological correlates of object relations theory: The relationship between neurobiological and psychodynamic development. International Forum of Psychoanalysis, 24(1), 38-46. doi:10.1080/0803706X.2014. 912352. [Google Scholar]
  26. Kernberg O. F., Caligor E. (2005). A psychoanalytic theory of personality disorders. In Lenzenweger M. F., Clarkin J. F. (Eds.), Major theories of personality disorders (pp. 114-156). Guilford Press. [Google Scholar]
  27. Klein M. (1935). A contribution to the psychogenesis of manicdepressive states. International Journal of Psychoanalysis, 16, 145-174. [Google Scholar]
  28. Knight Z. G. (2021). A re-consideration of interpretation. A relational approach. Research in Psychotherapy: Psychopathology, Process and Outcome, 24(3). doi:10.4081/ripppo.2021 .555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Koelen J. A., Luyten P., Eurelings-Bontekoe L. H., Diguer L., Vermote R., Lowyck B., Buhring M. E. (2012). The impact of level of personality organization on treatment response: a systematic review. Psychiatry, 75(4), 355-374. doi:10.1521/psyc.2012.75.4.355. [DOI] [PubMed] [Google Scholar]
  30. Lenzenweger M. F., Clarkin J. F., Kernberg O. F., Foelsch P. A. (2001). The inventory of personality organization: Psychometric properties, factorial composition, and criterion relations with affect, aggressive dyscontrol, psychosis proneness, and self domains in a nonclinical sample. Psychological Assessment, 13, 577-591. doi:10.1037/1040-3590.13.4.577. [PubMed] [Google Scholar]
  31. Livesley W. J. (2018). Conceptual Issues. In Livesley W. J., Larstone R. (Eds.), Handbook of personality disorders: Theory, research, and treatment (pp. 3-24). Guilford Press. [Google Scholar]
  32. Luyten P., Fonagy P. (2021). Integrating and differentiating personality and psychopathology: A psychodynamic perspective. Journal of Personality, 1-14. doi:10.1111/jopy.12656. [DOI] [PubMed] [Google Scholar]
  33. Morgan T. A., Zimmerman N. (2018). Epidemiology of personality disorders. In Livesley W. J., Larstone R. (Eds.), Handbook of personality disorders: Theory, research, and treatment (pp. 173-196). Guilford Press. [Google Scholar]
  34. Preti E., Prunas A., De Panfilis C., Marchesi C., Madeddu F., Clarkin J. F. (2015). The facets of identity: personality pathology assessment through the inventory of personality organization. Personality Disorders: Theory, Research, and Treatment, 6, 129-140. doi:10.1037/per0000119. [DOI] [PubMed] [Google Scholar]
  35. R Core Team (2020). R: A language and environment for statistical computing. Available from: https://www.R-project.org/ [Google Scholar]
  36. Savalei V., Rhemtulla M. (2013). The performance of robust test statistics with categorical data. British Journal of Mathematical and Statistical Psychology, 66(2), 201-223. doi:10.1111/j.2044-8317.2012.02049.x. [DOI] [PubMed] [Google Scholar]
  37. Sellbom M., Tellegen A. (2019). Factor analysis in psychological assessment research: Common pitfalls and recommendations. Psychological Assessment, 31(12), 1428-1441. doi:10.1037/pas0000623. [DOI] [PubMed] [Google Scholar]
  38. Siever L. J. (2005). Endophenotypes in the personality disorders. Dialogues in clinical neuroscience, 7(2), 139-151. doi:10.31887/DCNS.2005.7.2/lsiever. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Smith R., Solms M. (2018). Examination of the hypothesis that repression is premature automatization: A psychoanalytic case report and discussion. Neuropsychoanalysis, 20(1), 47-61. doi:10.1080/15294145.2018.1473045. [Google Scholar]
  40. Sollberger D., Gremaud-Heitz D., Riemenschneider A., Kuchenhoff J., Dammann G., Walter M. (2011). Associations between identity diffusion, Axis II disorder, and psychopathology in inpatients with borderline personality disorder. Psychopathology, 45(1), 15-21. doi:10.1159/ 000325104. [DOI] [PubMed] [Google Scholar]
  41. Stern B. L., Caligor E., Clarkin J. F., Critchfield K. L., Horz S., MacCornack V., Kernberg O. F. (2010). Structured interview of personality organization (STIPO): preliminary psychometrics in a clinical sample. Journal of Personality Assessment, 92, 35-44. doi:10.1080/00223890903379308. [DOI] [PubMed] [Google Scholar]
  42. Stover J. B., Castro Solano A., Fernandez Liporace M. (2019). Dysfunctional personality traits: relationship with Five Factor Model, adaptation and symptomatology in a community sample from Buenos Aires. Research in Psychotherapy: Psychopathology, Process and Outcome, 22(2). doi:10.4081/ ripppo.2019.343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Upmane A. (2010). Factors predicting satisfaction with life: quantitative and qualitative analysis (Publication No. 34390) [Doctoral dissertation, University of Latvia]. E-resource repository of the University of Latvia. Available from: https://dspace.lu.lv/dspace/handle/7/4844 [Google Scholar]
  44. 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(1), 85-95. doi:10.1080/00223891.2012.713883. [DOI] [PubMed] [Google Scholar]
  45. Watson D., Clark L. A., Tellegen A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063-1070. doi:10.1037/0022-3514.54. 6.1063. [DOI] [PubMed] [Google Scholar]
  46. Werbart A., Alden S., Diedrichs A. (2017). Changes in the anaclitic-introjective personality configurations following psychoanalytic psychotherapy with young adults. Research in Psychotherapy: Psychopathology, Process and Outcome, 20(1). doi:10.4081/ripppo.2017.239.s [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yeomans F. E., Clarkin J. F., Kernberg O. F. (2015). Transference- focused psychotherapy for borderline personality disorder: A clinical guide. American Psychiatric Association. [Google Scholar]
  48. Zarrella I., Russolillo L. A., Caviglia G., Perrella R. (2017). Continuity and discontinuity between psychopathology of childhood and adulthood: a review on retrospective and prospective studies. Research in Psychotherapy: Psychopathology, Process and Outcome, 20(2). doi:10.4081/ripppo.2017.248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zimmermann J., Benecke C., Horz S., Rentrop M., Peham D., Bock A., Wallner T., Schauenburg H., Frommer J., Huber D., Clarkin J. F., Dammann G. (2013). Validierung einer deutschsprachigen 16-Item-Version des Inventarsder Personlichkeitsorganisation (IPO-16) [Validity of a German 16-item version of the Inventory of Personality Organization (IPO- 16)]. Diagnostica, 59, 3-16. doi:10.1026/0012-1924/a000076. [Google Scholar]

Articles from Research in Psychotherapy : Psychopathology, Process, and Outcome are provided here courtesy of PAGEPress

RESOURCES