The Justice Sensitivity Inventory: Factorial Validity, Location in the Personality Facet Space, Demographic Pattern, and Normative Data

This article investigates the psychometric properties of a self-report inventory for measuring individual differences in four components of justice sensitivity (JS): victim sensitivity, observer sensitivity, beneficiary sensitivity, and perpetrator sensitivity. A representative sample (N = 2510) was employed to (a) estimate the reliability of a newly developed perpetrator sensitivity scale, (b) test the factorial validity of this scale together with three previously developed scales (victim, observer, and beneficiary sensitivity), (c) estimate correlations between JS and demographic variables (gender, age, education, employment status, marital status, and residency in East versus West Germany), and (d) provide normative data for the computation of standard scores. A demographically heterogeneous convenience sample (N = 327) was used to locate the JS dimensions in the personality space of narrow facet factors. Results from confirmatory factor analyses demonstrated the factorial validity of the JS scales. Regression analyses with JS scales as criteria and personality facet scales as predictors suggested that JS cannot be reduced to combinations of personality facets. Demographic effects were small, explaining a maximum of 1.4% of JS variance. Women and East Germans were found to be more justice sensitive than men and West Germans, respectively. Victim sensitivity decreased with age; perpetrator sensitivity decreased with education. Taken together, our results corroborate the validity of the JS Inventory and contribute to a better psychological understanding of JS.     

Association of the MMPI-2 Restructured Form (MMPI-2-RF) Validity Scales with Structured Malingering Criteria

The current study examined the validity scales of the Minnesota Multiphasic Personality Inventory-2 Restructured Form (MMPI-2-RF; Ben-Porath and Tellegen 2008) in relation to the structured malingering criteria developed to assess malingered neurocognitive dysfunction and pain-related disability. These criteria examined a sample of 251 individuals undergoing compensation-seeking evaluations, who completed a battery of response bias measures. The MMPI-2-RF over-reporting scales yielded large effect sizes in contrasting those in the probable/definite malingering groups from the incentive only groups. The largest effects were found for the Infrequent Responses and Gervais et al. (Assessment, 14, 196–208, 2007) Response Bias Scale, an experimental scale that can be scored on the MMPI-2-RF. Classification analyses were also utilized to examine various cut scores for the individual validity scales, as well as their use in combination. These results suggest that the MMPI-2-RF validity scales can be used to screen for malingering, as they exhibited good sensitivity at lower cutoffs. The implications of these results and their use in forensic evaluations are discussed.     

The More You Say the Less It Means: Overreporting and Attenuated Criterion Validity in a Forensic Disability Sample

A recent publication in Psychological Bulletin by McGrath et al. (Psychological Bulletin 136:450–470, 2010) challenged the necessity of response bias indicators in applied psychological assessment on the grounds that there is insufficient empirical support that shows that they are able to moderate the prediction of substantive measures. The current study challenges their conclusions by examining the effects of response bias in a sample of 2,275 disability litigants. We utilized the validity scales of the Minnesota Multiphasic Personality Inventory-2 Restructured Form (MMPI-2-RF; Ben-Porath and Tellegen, 2008) in order to establish a group of litigants who did not exhibit evidence of overreporting and compared them to a group of overreporting litigants. The overreporting group scored significantly higher on every Restructured Clinical Scale. Moreover, we compared the criterion validity between various substantive scales on the MMPI-2-RF and extratest measures [e.g., Beck Depression Inventory-II (Beck, Steer, & Brown, 1996)] and found evidence of significantly attenuated criterion validity in the overreporting group. Implications of the study in light of the review by McGrath and colleagues are discussed.     

The Detection of Feigned Psychiatric Disorders Using the MMPI-2-RF Overreporting Validity Scales: An Analog Investigation

Individuals who are motivated to feign psychological problems to achieve a desired outcome (e.g., insanity defense) may overreport symptoms of psychopathology, with type of pathology being dependent on the setting. In the current investigation, we examined the utility of the overreporting validity scales (infrequent responses [F-r], infrequent psychopathology responses [F_P-r], infrequent somatic responses [Fs], and symptom validity [FBS-r]) on the Minnesota Multiphasic Personality Inventory-2-Restructured Form (Ben-Porath and Tellegen 2008) to detect research participants instructed to simulate one of three mental disorders: major depressive disorder (MDD), schizophrenia (SCH), or post-traumatic stress disorder (PTSD). The restructured clinical (RC) and overreporting validity scale scores of bona fide psychiatric patients with a primary diagnosis of either MDD, SCH, or PTSD were compared to two groups of simulators—naïve (i.e., undergraduate students with no training in mental disorders) and sophisticated (i.e., individuals with advanced training in psychopathology or personal experience with the disorder asked to overreport symptoms). Examination of the RC Scale profiles revealed that the sophisticated simulators produced symptom profiles more similar to the profiles of the psychiatric patients than did the naïve simulators. For the overreporting validity scales, the sophisticated simulators were less likely to be detected as feigning compared to the naïve simulators; overall, the validity scales were able to distinguish patients from simulators and accurately classify most of the simulators regardless of their level of “symptom” sophistication. Examination and comparison of the validity scales revealed that across disorders and level of research participant symptom sophistication, the F_P-r scale best differentiated simulators from patients.     

Validity and reliability of the Self-Report Delinquency among a sample of at-risk youths

Self-Report Delinquency scales have formed the basis of much understanding of juvenile delinquency today, but further improvement and extensive measurement research on self-report measures of crime are needed. The present study examined the psychometric properties of the shorter general delinquency scale version of the classic commonly used Self-Report of Delinquency (SRD) measure created for the original National Youth Survey (NYS) using a sample (N = 412) of male (n = 200) and female (n = 212) at-risk Portuguese youths. The shorter version of the SRD demonstrated satisfactory validity and reliability, namely in terms of its latent one-factor structure, internal consistency, convergent validity, divergent validity, concurrent criterion validity and discriminant groups validity that overall justifies its use among this population. The shorter general delinquency scale version of the SRD is a valid and reliable measure from a modern psychometric perspective that can be used with at-risk youths.     

The Personality Assessment Inventory (PAI): Treatment Scales and Interpersonal Characteristics in a Sample of Men Charged with or Convicted of a Sexual Offense*

An individual's interpersonal features are pertinent to treatment within clinical populations. The Personality Assessment Inventory (PAI) contains two scales that assess the interpersonal features of warmth (WRM) and dominance (DOM), as well as two additional measures to assess to treatment prediction, process, and rejection (RXR; TPI). The current study examined associations between these PAI scales in a sample of 92 men who underwent comprehensive evaluations of sexual behavior after being charged with or convicted of a sexual offense. Analyses indicated that RXR was positively associated with WRM and DOM, TPI was negatively associated with WRM, and the two interpersonal scales of WRM and DOM were positively correlated with each other. A significant inverse relationship was found between the two treatment scales RXR and TPI indicating that motivation for treatment may have a limited relationship with the treatment process. WRM significantly predicted scores on the TPI, and both WRM and DOM predicted individual scores on RXR. Higher scores on positive impression management (PIM) were predictive of lower TPI and higher RXR, as individuals with higher stakes cases may score higher on PIM and underreport obstacles within treatment or be unwilling to accept the need for treatment. Overall, findings suggest that interpersonal characteristics identified by the PAI scales may be advantageous in approaching treatment within this population.     

MMPI-2 and MMPI-2-Restructured Form Validity Scales: Complementary Approaches to Evaluate Response Validity

Responsible forensic assessment is dependent on an appreciation for how validity scales differ from one another and function across different evaluation contexts. The primary goal of this review is to evaluate whether there is sufficient empirical support to provide expert testimony using MMPI-2 or MMPI-2-RF validity scales. Towards this goal, we briefly describe the development of MMPI-2 and MMPI-2-RF validity scales and review empirical studies that support interpretative strategies. We describe and illustrate with case examples how the respective sets of MMPI-2 and MMPI-2-RF validity scales utilize complementary strategies to detect over- and under-reporting of psychological, cognitive, and physical symptoms. Additionally, the review describes standards for expert psychological testimony and illustrates how MMPI-2 and MMPI-2-RF validity scales are particularly well suited to assess symptom validity and psychological status in secondary-gain settings.     

Validation of a Short Form of the Marlowe-Crowne for Use with Law Enforcement Personnel

This study cross-validates a 13-question version of the Marlowe-Crowne Social Desirability Scale proposed by Reynolds (1982), using law enforcement professionals (n?=?241). A rational is offered for the importance of scales of this nature being used to assess honesty, social desirability and self-deception. External validity indicators including concurrent measures of aggression and prejudice as well as prehire MMPI scales K and L on a subset of this sample (n?=?70) are reviewed. Correlations between the 13-question version of the Marlowe-Crowne and external measures demonstrate robust effect sizes and statistical significance. Comparisons are also examined between officer scores and students from another cross-validation study of the 13-question short form of the Marlowe-Crowne (Zook & Sipps, 1984). Conclusions are made in consideration of the importance of honesty in law enforcement testing as well as the need for better means of accounting for variance due to social desirability in testing. Future directions for research are suggested.     

A Cross-cultural Validation of the Brief Child Abuse Potential Inventory (BCAP)

The aim of this study was to provide preliminary findings on the cross-cultural validation and reliability of the Brief Child Abuse Potential Inventory (BCAP) developed by Ondersma et al. (Journal of Clinical Child and Adolescent Psychology, 2005). A community sample of 324 parents was recruited through schools from a range of socioeconomic areas in a large UK city. The BCAP appears to be a reliable instrument that is easy to use and shows promise as a brief screening tool for parental child abuse potential within the UK. Modifications are suggested to the validity scales and the conceptual structure based on this sample, however, detailed testing of the utility of this scale for clinical practice outside the US should be pursued.     

A Meta-Analytic Review of Minnesota Multiphasic Personality Inventory—2nd Edition (MMPI-2) Profile Elevations Following Traumatic Brain Injury

Psychologists often use the Minnesota Multiphasic Personality Inventory (MMPI) and, more recently, its successor, the MMPI-2, to assess personality and psychological disturbances following traumatic brain injury (TBI). The present meta-analysis examined the pattern of mean Hedges’ d values on MMPI-2 validity (L, F, K) and clinical (1–4, 6–0) scales in individuals with TBI. Database keyword searches yielded ten studies providing post-TBI MMPI-2 profiles. Studies were required to include a pure TBI sample, individuals who were ≥18 at injury, and means and standard deviations for most MMPI-2 clinical scales. Analyses showed large effects for MMPI-2 scales F, 1, 2, 3, 7, and 8. Using Q statistics, moderating effects were found for TBI severity on scale 7 and for compensation-seeking/litigation status on scales 1, 2, 3, and 7. No significant effects were found for time post-injury. The available information was insufficient to examine the effect of lesion location, pre-injury personality and psychopathology, or time post-injury for samples with differing injury severities on MMPI-2 profiles. Results suggest that individuals with TBI report significant levels of psychopathology that may be moderated by TBI severity and compensation-seeking/litigation status. Discussion includes a literature critique given the meta-analytic findings and implications for future study of personality following TBI.     

The Portuguese adaptation of the Gudjonsson Suggestibility Scale (GSS1) in a sample of inmates

This paper comprises two studies which address the validity of the Portuguese adaptation of the Gudjonsson Suggestibility Scale, GSS1. In study 1, the means and standard deviations for the suggestibility results of a sample of Portuguese inmates (N = 40, M_age = 37.5 years, SD = 8.1) were compared to those of a sample of Icelandic inmates (Gudjonsson, 1997; Gudjonsson & Sigurdsson, 1996). Portuguese inmates' results were in line with the original results. In study 2, the means and standard deviations for the suggestibility results of the sample of Portuguese inmates were compared to those of a general Portuguese population sample (N = 57, M_age = 36.1 years, SD = 12.7). The forensic sample obtained significantly higher scores in suggestibility measures than the general population sample. ANOVA confirmed that the increased suggestibility in the inmates sample was due to the limited memory capacity of this latter group. Given that the results of both studies 1 and 2 are in keeping with the author's original results (Gudjonsson, 1997), this may be regarded as a confirmation of the validity of the Portuguese GSS1.     

Predictive Validity of the Neuropsychological Assessment Battery-Screening Module for Assessing Real-World Disability in Patients with Mild Traumatic Brain Injury

The present study set out to examine the ecological validity of the Neuropsychological Assessment Battery-Screening Module (NAB-S) and to examine more specifically, whether performance on this test battery is related to real-world disability status in patients who have sustained a mild traumatic brain injury (mTBI) as a consequence of a motor vehicle collision (MVC). Seventy-nine patients were recruited from a random sample of litigating patients who incurred an mTBI and were referred for a neuropsychological examination due to subjective complaints of cognitive function. Of the total sample, 54 patients were categorized into the non-disabled group and 25 patients were categorized into the disabled group using the 4th edition of the American Medical Association (AMA) Guides to the Evaluation of Permanent Impairment (AMA; 1993). A multivariate analysis of variance revealed that performance on the NAB-S significantly differed between the two groups, and that memory and attention domains provided a more discriminative index of disability than general neuropsychological functions. In addition, discriminant function analyses revealed that the NAB-S correctly classified 73% of patients according to their group membership. The current findings provide evidence for the ecological validity of the NAB-S as a reliable index of real-world disability status as defined by the AMA Guides.     

A Commentary on Ward and Stewart's Model of Human Needs

Abstract

It is often assumed that offenders employ socially desirable responding when completing self-report questionnaires, thereby invalidating such measures. The aim of this study is to examine the extent that sexual offenders employ socially desirable responding and the impact that socially desirable responding, as measured by the Balanced Inventory of Desirable Responding-6 (BIDR: 6, Paulhus, Journal of Personality and Social Psychology, 46, 598–609, 1984), has on self-report measures. The sample comprised 1730 adult males, all convicted and serving a sentence for a sexual offence. The entire sample had completed a set of self-report measures used to identify treatment needs, and examine progress in treatment. The results indicate that the extent of socially desirable responding is smaller than assumed, and its impact on a number of self-report measures is lower than expected. Furthermore, lower levels of risk of sexual re-offending were significantly associated with higher scores on the impression management subscale. We suggest that impression management should be viewed as more of a trait-like tendency rather than a validity check, and the use of this scale in determining whether or not to interpret other self-report instruments should be questioned. The present study also indicates that self-report questionnaires used with forensic populations may be accurate and valid. Implications for research and practice are discussed.     

Psychopathy and internalizing psychopathology

There is general consensus in clinical and research literature that the core feature of psychopathy consists of an affective deficit. However, previous studies tend to find weak and inconsistent associations between psychopathy and measures of internalizing psychopathology. In this study we test whether the predominant practice of using questionnaires to assess internalizing psychopathology has influenced the results of previous research. We argue that questionnaires measure general distress rather than specific symptoms of internalizing psychopathology, and that the validity of questionnaires might be impaired by psychopathic traits, such as impression management and lack of affective experience. Combining a questionnaire (Depression Anxiety Stress Scales-21; DASS-21) and a semi-structured interview (Structured Clinical Interview for DSM-IV-R Axis 1 Disorders; SCID-I) for internalizing psychopathology, we test the differential association of both measures with the Psychopathy Checklist-Revised (PCL-R) in a sample of 89 male detainees. In accordance with our prediction, we found moderate negative associations between the Interpersonal and Affective facets of the PCL-R and SCID-I, but no significant associations with the DASS-21. We found no evidence that psychopathic traits decrease the validity of the responses on a questionnaire. We conclude that the interpersonal and affective features of psychopathy are negatively related to specific symptoms of internalizing psychopathology, but not with general distress.     

Violence in relationships: A seventh grade inventory of knowledge and attitudes

The purpose of this research was to contribute to existing efforts of battered women's advocates, schools, and researchers by developing a valid and reliable inventory to evaluate educational interventions regarding violence in relationships. Content validity was established by national experts (N=32)solicited to complete an inventory rating scale and comment on the statements' ability to elicit early adolescents' knowledge and attitudes about violence in relation-ships. The revised inventory was tested with seventh grade health education students (N=99).The test-retest reliability criterion was .50. Spearman's rho correlations were .566 for the knowledge section and .669 for the attitude section. Knowledge and attitude sections were measured for internal con- sistency with Cronbach's alpha, with respective values of .3178 and .7207. Item analyses with Wilcoxon Rank Sum tests identified 11 knowledge statements and one attitude statement as significantly discriminating at both test and retest. Gender differences were noted and additional findings were reported. Recommendations for the further establishment of a valid and reliable inventory were made.     

The utility of the State-Trait Anger Expression Inventory with offenders

Self-report instruments can provide useful information as part of a thorough clinical assessment. However, their use in forensic settings can be problematic. The State-Trait Anger Expression Inventory (STAXI) has recently been proposed as an effective instrument for screening and outcome measurement in anger management programs. This study evaluated the effectiveness of this instrument in a sample of both voluntary and court-ordered anger-management clients, all of whom were determined through diagnostic interviews to have significant anger problems. Contrary to findings in nonforensic samples, the STAXI Trait Anger scale identified only about half of the participants as having anger-management problems severe enough to require intervention. Supplemental analysis with two additional scales did not significantly improve sensitivity. In addition to thorough diagnostic interviewing, forensic use of the STAXI (like similar assessment methods) may require additional validity scales to detect denial or socially desirable response patterns.     

The Self-Report Symptom Inventory (SRSI): a New Instrument for the Assessment of Distorted Symptom Endorsement

Self-report instruments to detect distorted symptom reporting play a crucial role in clinical and forensic psychology. Most of the instruments currently available for this purpose only list implausible symptoms, which makes them easily identifiable as symptom validity tests. We developed the Self-Report Symptom Inventory (SRSI), combining five self-report scales of genuine symptoms with five pseudosymptom scales to screen for distorted symptom reporting in various domains (e.g., depression, post-traumatic stress). With a preliminary questionnaire version, we collected data in a heterogeneous sample (N?=?239) and performed an item selection, resulting in the final 107-item version. This version was evaluated in civil forensic patients, inmates of a prison, and a population-based sample; N?=?387). Data show that (a) SRSI pseudosymptom scores correlate highly (≥.80) with other instruments tapping distorted symptom endorsement, notably the Structured Inventory of Malingered Symptomatology; (b) High SRSI pseudosymptom scores tend to correlate with underperformance; and (c) The psychometric features of the SRSI are satisfactory, with internal consistency for the total scales >.90 and retest reliability >.85. The instrument appears to be a promising tool for examining symptom exaggeration, but further work is required, in particular cross-validation with other samples and different methods.     

Measuring violence risk and outcomes among Mexican American adolescent females

Central to the development of culturally competent violence prevention programs for Hispanic youth is the development of psychometrically sound violence risk and outcome measures for this population. A study was conducted to determine the psychometric properties of two commonly used violence measures, in this case for Mexican American adolescent females. The Conflict Tactics Scales (CTS2) and the Past Feelings and Acts of Violence Scale (PFAV) were analyzed to examine their interitem reliability, criterion validity, and discriminant validity. A sample of 150 low-risk and 150 high-risk adolescent females was studied. Discriminant validity was indicated by the perpetrator negotiation scale and by the victim psychological aggression and sexual coercion scales of the CTS2 and the PFAV. Analysis indicates that the CTS2 scales and the PFAV demonstrate adequate reliability, whereas strong criterion validity was evidenced by eight of the CTS2 scales and the PFAV.     

Elaborating on the Construct Validity of the Levenson Self-Report Psychopathy Scale in Incarcerated and Non-Incarcerated Samples

The Levenson Self-Report Psychopathy Scale (LSRP) is a promising, efficient measure of the psychopathic personality. Previous examinations of its latent structure have yielded inconsistent results; therefore, the goal of the current study was to evaluate previously reported two- and three-factor models in three separate groups of male correctional inmates (n = 573), male college students (n = 202), and female college students (n = 200). Confirmatory factor analyses revealed that the best-fitting model across all the three groups was the three-factor model proposed by Brinkley et al. (Assessment 15:464–482, 2008). The convergent and discriminant validity of this three-factor structure as well as a revised 19-item total score were elaborated on in terms of predicting scores on conceptually relevant external criteria. The LSRP total and factor scales evidenced good convergent and discriminant validity with a few notable exceptions. Implications for theory and future research are discussed.