Elsevier

Psychiatry Research

Volume 258, December 2017, Pages 544-550
Psychiatry Research

The Schizotypal Personality Questionnaire-Brief lacks measurement invariance across three countries

https://doi.org/10.1016/j.psychres.2017.08.088Get rights and content

Highlights

  • The SPQ-B lacks measurement invariance across cultures.

  • The cross-cultural validity of SPQ-B is limited and its use for non-Western populations is not supported.

  • The poor psychometric properties could be due to cultural and translation issues.

Abstract

The Schizotypal Personality Questionnaire-Brief (SPQ-B) is a commonly-used tool for measuring schizotypal personality traits and due to its wide application, its cross-cultural validity is of interest. Previous studies suggest that the SPQ-B either has a three- or four-factor structure, but the majority of studies have been conducted in Western contexts and little is known about the psychometric properties of the scale in other populations. In this study factorial invariance testing across three cultural contexts–Australia, China and Chile was conducted. In total, 729 young adults (Mean age = 23.99 years, SD = 9.87 years) participated. Invariance testing did not support the four-factor model across three countries. Confirmatory Factor Analyses revealed that neither the four- nor three-factor model had strong fit in any of the settings. However, in comparison with other competing models, the four-factor model showed the best for the Australian sample, while the three-factor model was the most reasonable for both Chinese and Chilean samples. The reliability of the SPQ-B scores, estimated with Omega, ranged from 0.86 to 0.91. These findings suggest that the SPQ-B factors are not consistent across different cultural groups. We suggest that these differences could be attributed to potential confounding cultural and translation issues.

Introduction

Schizotypy is a heterogeneous construct that has been used by researchers and clinicians to describe schizotypal traits, psychosis-proneness, and psychotic-like experiences. Coined by Meehl (1962), it refers to a personality organisation that represents the vulnerability for developing psychotic-spectrum disorders. Expression of schizotypal traits (e.g. hallucinatory and delusional experiences), despite having a transient nature and possibly vanishing over time, can occur in the general population, without necessarily being associated with a mental health condition (Fonseca-Pedrero and Debbané, 2017; Linscott and Van Os, 2013; Van Os et al., 2009). For example, an international study conducted by McGrath et al. (2015) sampling 31261 adults in the community from 18 countries, found that the average life-time prevalence of psychotic experience was 5.8%. Further, evidence from a meta-analysis showed that the risk of conversion to a clinical psychotic outcome among people who report subthreshold psychotic experiences (0.56%) is 3.5 times higher than those without exposure to such experiences (0.16%), particularly the experiences were severe or persistent (Kaymaz et al., 2012). Other longitudinal studies (e.g., Poulton et al., 2000; Zammit et al., 2013) also suggest that adolescents and young adults who report such schizotypal experiences are at greater risk of developing psychosis and related disorders than those who do not. These studies suggest that the presence of schizotypal features may represent a liability to developing symptoms of schizophrenia or transitioning from subclinical impairments to a full-blown psychosis, and this is especially the case if genetic and environmental risk factors are present (e.g. degree of relatedness to family member with schizophrenia, cannabis use) (Linscott and Van Os, 2013; Debbané et al., 2015; Van Os et al., 2009). It is therefore important to identify and address schizotypal traits as early as possible as early detection and intervention may enable prevention of, or reduce the probability of progression in psychosis (Fonseca-Pedrero et al., 2016, Ortuno-Sierra et al., 2013).

The Schizotypal Personality Questionnaire-Brief version (SPQ-B; Raine and Benishay, 1995) is one of the most popular scales used to assess schizotypal traits (Reynolds et al., 2000). Derived from the original 74-item scale developed by Raine (1991), it consists of 22 items, responded to using a Yes/No format, that reflect the diagnostic criteria of schizotypal personality disorder first described in the DSM-III-R (American Psychiatric Association, 1987). These items cover nine distinct domains (i.e. ideas of reference, excessive social anxiety, odd beliefs or magical thinking, unusual perceptual experiences, odd or eccentric behaviour, no close friends, odd speech, constricted affect, and suspiciousness) that are grouped into three factors which relate in part to the three components of schizophrenia: cognitive-perceptual problems (positive symptoms), interpersonal problems (negative symptoms) and disorganization (Raine and Benishay, 1995).

The brevity of the SPQ-B while still capturing the three factors reflected by the larger scale, makes it an attractive screening measure and it has been used both in non-clinical and psychiatric populations (Axelrod et al., 2001, Compton et al., 2007, Compton et al., 2009, Mata et al., 2005). In addition, the SPQ-B has been translated into different languages and validated in a variety of culture contexts, including China (Ma et al., 2010, Ma et al., 2015), Spain (Fonseca-Pedrero et al., 2009, Fonseca-Pedrero et al., 2011b, Fonseca-Pedrero et al., 2011a, Fonseca-Pedrero et al., 2014, Mata et al., 2005), Japan (Ito et al., 2008, Ito et al., 2010), and Turkey (Aycicegi et al., 2005).

Despite these advantages and its widespread use, Stefanis and colleagues (2004) argued that the 'positive' schizotypy factor in Raine’s (1991) three-factor model for the SPQ, which includes items of cognitive-perceptual and paranoid dimensions, should in fact be two independent factors. They suggested a four-factor model that consists of interpersonal, cognitive-perceptual, disorganized, and paranoid factors. While this four-factor structure has been well replicated in both the SPQ and SPQ-B (Cicero, 2015, Cohen et al., 2010, Fonseca-Pedrero et al., 2009, Fonseca-Pedrero et al., 2014, Lahmar et al., 2014, Smallman et al., 2010), there is little evidence to suggest that it is valid across cultures. For example, the four-factor model in the SPQ-B has been replicated in Spanish community samples (Fonseca-Pedrero et al., 2009, Fonseca-Pedrero et al., 2011b), but the original three-factor model (cognitive-perceptual, interpersonal and disorganised) was supported in China (Ma et al., 2015), and a two-factor model was identified by Principal Component Analysis in Turkey and Italy (Aycicegi et al., 2005, Preti et al., 2015).

Only two previous studies have provided evidence of the four-factor model solution of the SPQ across different cultural groups in the same study (Cicero, 2015, Fonseca-Pedrero et al., 2014). Cicero (2015) replicated the four-factor structure within cultural subgroup samples in the United States (i.e. Asian, Caucasian, Pacific Islander, etc.) while Fonseca-Pedrero and colleagues (2014) established measurement invariance for the four-factor model across samples of Spanish and US college students. Although the four-factor structure of SPQ-B has not been replicated in Eastern contexts, past evidence of measurement invariance across diverse ethnic groups demonstrated in Cicero’s study infers that a four-factor model may apply in both Western and Eastern cultures.

In relation to culture, Cohen et al. (2015) suggested that it is important to assess schizotypy in different cultures as social and affective functioning has the potential to vary considerably across cultures. In fact, previous studies have demonstrated variation in schizotypal traits and psychotic-like experiences across countries (Cicero, 2015; Linscott and Van Os, 2013). For example, Caucasian respondents have been reported to have lower scores on the SPQ than other ethnic groups on the ‘interpersonal’ schizotypal factor while Chinese respondents tend to report significantly lower SPQ scores than their Caucasian counterparts (Chen et al., 1997, Cicero, 2015). These findings suggest that potential ethnic or cultural factors were at play, which is worth investigation.

The current study aims to examine the measurement invariance of the four-factor structure of the SPQ-B across samples from both Western (i.e., Australia and Chile) and Eastern cultures (i.e., China). In light of Cicero’s (2015) findings in the United States across multiple ethnic groups, we hypothesized that the four-factor structure in the SPQ-B would be invariant across countries. In addition, in line with reported cultural variations in SPQ scores, we further hypothesized that scores would differ between countries.

Section snippets

Participants

Data for the current study were drawn from a larger series of studies examining schizotypal traits and social cognition (e.g., see Prado et al., 2014). The participants were 729 (M = 23.99 years, SD = 9.87 years) non-clinical community sample and university students recruited in Australia, mainland China, and Chile. A total of 304 males (M = 23.58 years, SD = 7.50 years) and 421 females (M = 24.07 years, SD = 11.05 years) took part in the study. Refer to Table 1 for detailed gender and age

Results

The means, standard deviations and the range of scores on the SPQ-B total scale as well as each of the specified subscales across three countries are presented in Table 2.

Discussion

The aim of the present study was to examine measurement invariance of the four-factor structure of the SPQ-B across Australian, Chinese, and Chilean young adult samples. Unexpectedly, the results of multiple CFA models revealed diverse underlying structures the samples (e.g. 4-factor for Australian and 3-factor for both Chinese and Chilean sample). Further, measurement invariance could not be established. These findings help to improve our understanding of expression of schizotypy with regard

Acknowledgment

This research did not receive funding from agencies in the public, commercial, or not-for-profit sectors. All authors made a substantial contribution to this research and the drafting of the manuscript. The first author would like to acknowledge Lanxi Huang and Gabrielle Wu for their assistance in verifying translations.

References (58)

  • W.F. Ma et al.

    Sensitivity and specificity of the Chinese version of the Schizotypal Personality Questionnaire-Brief for identifying undergraduate students susceptible to psychosis

    Int. J. Nurs. Stud.

    (2010)
  • I. Mata et al.

    Schizotypal Personality Questionnaire-Brief: factor structure and influence of sex and age in a nonclinical population

    Personal. Individ. Differ.

    (2005)
  • A. Preti et al.

    Bifactor structure of the Schizotypal Personality Questionnaire (SPQ)

    Psychiatry Res.

    (2015)
  • F.D. Schönbrodt et al.

    At what sample size do correlations stabilize?

    J. Res. Personal.

    (2013)
  • L.J. Siever et al.

    The search for a schizotypal personality: historical origins and current status

    Compr. Psychiatry

    (1983)
  • R.P. Smallman et al.

    Invariance testing of the 4-factor solution of schizotypal personality questionnaire across age, sex and ethnicity

    Schizophr. Res.

    (2010)
  • F. Van de Vijver et al.

    Bias and equivalence in cross-cultural assessment: an overview

    Rev. Eur. De. Psychol. Appliqué

    (2004)
  • V. Wuthrich et al.

    Reliability and validity of two Likert versions of the Schizotypal Personality Questionnaire (SPQ)

    Personal. Individ. Differ.

    (2005)
  • American Psychiatric Association

    Diagnostic and Statistical Manual Of Mental Disorders: DSM-III-R. (3 ed., revised)

    (1987)
  • S.R. Axelrod et al.

    Schizotypal Personality Questionnaire-Brief: factor structure and convergent validity in inpatient adolescents

    J. Pers. Disord.

    (2001)
  • A. Aycicegi et al.

    Validation of Turkish and English versions of the Schizotypal Personality Questionnaire-B

    Eur. J. Psychol. Assess.

    (2005)
  • B.M. Byrne

    Testing for multigroup invariance using AMOS graphics: a road less traveled

    Struct. Equ. Model.

    (2004)
  • B.M. Byrne

    Structural equation modelling with Mplus: Basic concepts, applications, and programming

    (2012)
  • D. Callaway et al.

    Schizotypal Personality Questionnaire–Brief Reviewed: psychometric replication and extension

    Personal. Disord.: Theory, Res., Treat.

    (2014)
  • W.J. Chen et al.

    Schizotypy in community samples: the three-factor structure and correlation with sustained attention

    J. Abnorm. Psychol.

    (1997)
  • D.C. Cicero

    Measurement invariance of the Schizotypal Personality Questionnaire in Asian, Pacific Islander, White, and multiethnic populations

    Psychol. Assess.

    (2015)
  • A.S. Cohen et al.

    Toward a more psychometrically sound brief measure of schizotypal traits: introducing the SPQ-Brief revised

    J. Personal. Disord.

    (2010)
  • A.S. Cohen et al.

    Schizotypy as an organizing framework for social and affective sciences

    Schizophr. Bull.

    (2015)
  • M. Debbané et al.

    Developing psychosis and its risk states through the lens of schizotypy

    Schizophr. Bull.

    (2015)
  • Cited by (4)

    • Are personality measures valid for different populations? A systematic review of measurement invariance across cultures, gender, and age

      2020, Personality and Individual Differences
      Citation Excerpt :

      see details in Appendix A) Notably, the studies showing strict/scalar invariance do not indicate that these personality measures are generally invariant because the levels of invariance might depend on the sample choices or the ways of creating subgroups (e.g., age groups). For example, Fonseca-Pedrero et al. (2018) and Liu et al. (2017) explored the measurement invariance of the Schizotypal Personality Questionnaire-Brief (Adrian Raine & Benishay, 1995) with different samples. As a result, configural invariance was not held in Liu et al.’s study but was established in Fonseca-Pedrero et al.’s work, which indicates the necessity for personality psychologists to test MI on their own samples if they plan to use those measures that were previously identified as invariant to compare factor correlations or means among new populations.

    View full text