ASSESSING PERSONALITY DISORDERS USING THE MMPI-2-RF

A thesis submitted to Kent State University in partial fulfillment of the requirements for the degree of Master of Arts

by Ashley M. Smith August, 2010

Thesis written by Ashley M. Smith B.A., Kent State University, 2006 M.A., Kent State University, 2010

Approved by

_________________________________, Advisor Yossef Ben-Porath _________________________________, Chair, Department of Psychology Maria S. Zaragoza _________________________________, Dean, College of Arts and Sciences John R. D. Stalvey

ii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. iv CHAPTER I

Page

INTRODUCION ......................................................................................................1 Criterion Overlap and Disorder Co-morbidity.........................................................5 Assessing Personality Disorders .............................................................................6 The MMPI-2-RF ....................................................................................................11 The Current Investigation .....................................................................................14

II

METHOD ..............................................................................................................16 Participants .............................................................................................................16 Measures ................................................................................................................17 Procedure ...............................................................................................................21

III

RESULTS ..............................................................................................................22

IV

DISCUSSION ........................................................................................................32 Limitations .............................................................................................................42 Future Directions ...................................................................................................43

REFERENCES ..................................................................................................................45 APPENDIX A THE MMPI-2-RF SCALES ....................................................................51

iii

LIST OF TABLES Table

Page

1

Hypotheses .............................................................................................................15

2

Correlations Between Higher-Order (H-O) Scales and Restructured Clinical (RC) Scales and NEO-PI-R and SCID-II .................................................23

3

Correlations Between Somatic/Cognitive and Internalizing Scales and NEO-PI-R and SCID..............................................................................................24

4

Correlations Between Externalizing, Interpersonal, and Interest Scales and NEO-PI-R and SCID-II ...................................................................................25

5

Correlations Between the Personality Psychopathology Five (PSY-5) Scales and NEO-PI-R and SCID-II........................................................................26

6

Hypotheses and Results ........................................................................................34

iv

CHAPTER I

INTRODUCTION A Personality Disorder is “an enduring pattern of inner experiences and behavior that deviates markedly from the expectations of the individual’s culture, is pervasive and inflexible, has an onset in adolescence or early adulthood, is stable over time, and leads to distress or impairment” (APA, 2000). Consistent with an organizational framework introduced with DSM-III, the current classification system, DSM-IV-TR (APA, 2000) categorizes personality disorders into three clusters, A, B, and C, with each including several personality disorders. The 10 personality disorders are categorized into the clusters based on descriptive similarities so that the disorders grouped into a particular Cluster share similarities in their presentations, symptomatology, personality traits, and behavioral observations (APA, 2000). Although not without its problems and limitations (c. f. Kraus, 1991; Klonsky, 2000; Mahrer, 2000; Livesley, 1991), this clustering system is currently the gold standard for diagnostic purposes. The first set of personality disorders, Cluster A, includes: Paranoid Personality Disorder, Schizoid Personality Disorder, and Schizotypal Personality Disorder. This cluster includes individuals who appear to be odd or eccentric when compared with others (APA, 2000). Specifically, Paranoid Personality Disorder involves a pattern of distrust and suspiciousness, such that

1

2 the motives of others are interpreted as malevolent (APA, 2000). The prevalence rate for this disorder is 0.5-2.5% in the general population. In addition, both Alcohol and other Substance Use Disorders commonly co-occur with this diagnosis (APA, 2000). For example, Chiao-Chicy and colleagues (1999) found that 15.9% of their heroin-addicted sample obtained a diagnosis of Paranoid Personality Disorder Schizoid Personality Disorder involves a pattern of detachment from social relationships and a restricted range of emotional expression by the individual (APA, 2000). Prevalence rates for this particular disorder are not stated specifically in the DSM; however, it is uncommon in clinical settings (APA, 2000). Individuals with Schizoid Personality Disorder often appear to have flattened affect. Lastly, Schizotypal Personality Disorder is defined by a pattern of discomfort in close relationships with others, cognitive or perceptual distortions, and eccentricities of behavior (APA, 2000). The prevalence rate for this disorder is about 3% in the general population (APA, 2000). Individuals diagnosed with Schizotypal Personality Disorder may come across to others as socially phobic or as having bizarre experiences. Overall, individuals with Cluster A disorders are typically disconnected from reality and may appear paranoid, suspicious, or emotionally detached from others (APA, 2000). The Cluster B Personality Disorders include Antisocial Personality Disorder, Borderline Personality Disorder, Histrionic Personality Disorder, and Narcissistic Personality Disorder (APA, 2000). Overall, individuals with Cluster B Personality Disorders appear to be dramatic, overly emotional, erratic/reckless, or cold and uncaring (APA, 2000). More specifically, Antisocial Personality Disorder is defined by a pattern

3 of disregard for others and the violation of their rights (APA, 2000). These are individuals who have typically had much interaction with the law and judicial systems as both adolescents and adults. This is one of the more frequently occurring personality disorders at 3% of community samples for men and 1% for women; however, in forensic, substance abuse treatment and outpatient settings, the prevalence rates for this disorder increase dramatically (APA, 2000). Further, high rates of comorbidity have been noted between Antisocial Personality Disorder and Substance Use Disorders (APA, 2000). More specifically, a study by Craig (2000) found prevalence rates of Antisocial Personality Disorder ranging from 3 to 62% in an inpatient drug-abusing (i.e.- heroin and cocaine) population. Borderline Personality Disorder is classified as a pattern of instability and problem behaviors in interpersonal relationships, disruptions or fluctuations of the individual’s self-image, and general impulsivity (APA, 2000). In the general population, individuals are diagnosed with Borderline Personality Disorder at a 2% rate and up to 10% in outpatient treatment settings (APA, 2000). Individuals with a diagnosis of Borderline Personality Disorder are also likely to have a co-occurring Substance Use Disorder (APA, 2000). Research has demonstrated that individuals in treatment for a Substance-related disorder have comorbid diagnoses of Borderline Personality Disorder with prevalence rates ranging from 22.4% to 28.5% (Morgenstern, et. al., 1997; Skodol, et. al., 1999). Histrionic Personality Disorder involves a pattern of excessive emotionality and attention seeking thoughts, behaviors, and motivations (APA, 2000). Individuals with this

4 disorder may threaten to commit suicide, with no true intentions of doing so, to prevent their partner from ending a relationship with them. They frequently aim to be the center of attention in social situations as well. The prevalence rate for this disorder ranges from 2-3% in the general population and up to 15% in outpatient settings (APA, 2000). Narcissistic Personality Disorder is characterized by a pattern of grandiosity, a specific need for admiration, and a lack of empathy for others. The prevalence of this disorder is fairly low at less than 1% of the population, but higher in clinical settings at 216% (APA, 2000). In addition, individuals with Narcissistic Personality Disorder are at an increased risk for a comorbid Substance Use Disorder, especially related to cocaine abuse or dependence (APA, 2000). Individual with Narcissistic Personality Disorder may often appear very self-centered and/or selfish. In addition, they may appear very unempathic and require excessive admiration and praise from others (APA, 2000). Cluster C Personality Disorders include: Avoidant Personality Disorder, Dependent Personality Disorder, and Obsessive-Compulsive Personality Disorder. Individuals with a Cluster C Personality Disorder typically appear to be very anxious or fearful in a variety of situations (APA, 2000). Avoidant Personality Disorder is characterized by a pattern of social inhibition, feelings of inadequacy, and hypersensitivity to negative evaluation by others (APA, 2000). The prevalence rates for this disorder are 0.5-1% in the general population and about 10% in clinical settings (APA, 2000). People diagnosed with this disorder frequently over-react to criticism of any kind and may seem very down on themselves, believing that they will never be good enough.

5 Dependent Personality Disorder is defined as a pattern of submissive and clinging behavior that is related to an excessive need to be taken care of by others (APA, 2000). Despite the fact that there is no specific rate listed, this disorder is one of the most frequently reported personality disorders in mental health settings (APA, 2000). Individuals with Dependent Personality Disorder may tolerate excessive negativity in interpersonal relationships in order to maintain their dependency on others (APA, 2000). Obsessive- Compulsive Personality Disorder is characterized by a pattern of preoccupation with orderliness, perfectionism, and control (APA, 2000). This Personality Disorder occurs in about 1% of the general population and in 3-10% of clinical populations (APA, 2000). Individuals with Obsessive- Compulsive Personality Disorder may not be able to keep a steady job because they take several hours to complete a simple task. Several flaws and limitations of the DSM-IV personality disorder classification system have been identified. They include issues of comorbidity between the personality disorders, related overlap of symptoms or diagnostic criteria, and consequent difficulty in differential diagnosis of these disorders (Kraus, 1991; Klonsky, 2000).

Criterion Overlap and Disorder Co-morbidity One of the difficulties with the current Axis II clusters is that some of the symptoms or criteria within the clusters overlap. Although they may not be described with identical language, several symptoms from each of the personality disorders are similar within that same Cluster. In Cluster A, for example, suspiciousness, difficulties maintaining interpersonal relationships, and inappropriate or constricted affect are

6 characteristics of more than one disorder (APA, 2000). Similarly, in Cluster B, impulsivity, anger, aggressiveness, intense interpersonal relationships, and affective intensity are present in some form across the four personality disorders (APA, 2000). A fear of rejection and/or criticism, inhibited interpersonal relationships, cognitive preoccupation and difficulties in decision-making characterize more than one of the Cluster C disorders (APA, 2000). As a result of the overlap both within and between the three clusters of personality disorders, it is not surprising that there is a significant amount of comorbidity between the various disorders. In fact, research has demonstrated that an individual diagnosed with one Personality Disorder is at an increased risk to have a second Personality Disorder diagnosis (Kraus, 1991). More specifically, individuals diagnosed with any of the Cluster A Personality Disorders are at an increased risk to develop comorbid Axis II conditions, including the other Cluster A Personality Disorders, Avoidant, and Borderline Personality Disorders (APA, 2000). Further, as stated in the DSM-IV-TR (APA, 2000), an individual with any Cluster B Personality Disorder is at an increased risk to have another comorbid Cluster B diagnosis. In terms of comorbidity and Cluster C Personality Disorders, individuals diagnosed with any Cluster C Personality Disorder are also frequently diagnosed with a comorbid Cluster A Personality Disorder or Borderline Personality Disorder (APA, 2000).

Assessing Personality Disorders A variety of methods have been developed to assess Axis II conditions. One assessment technique that can be used is a structured interview, such as the Structured

7 Clinical Interview for the DSM- Axis II (SCID II; First, Gibbon, Spitzer, Williams, & Benjamin, 1997). This instrument asks standardized questions related to Personality Disorder symptomatology. Further, the format of the interview corresponds with the DSM-IV-TR (APA, 2000) criteria for each of the personality disorders (First, Gibbon, Spitzer, Williams, & Benjamin, 1997). Two additional assessment methods used to examine Personality Disorder symptomatology are semi-structured and unstructured (e.g. – social history) interviews. These interviews include both open and closed-ended questions aimed at gathering information about symptomatology, emotions, and so on. However, these two types of interviews were not developed specifically for the assessment of personality disorders; rather, they are also used to assess a wide variety of mental disorders. Various attempts have been made to assess personality disorders with self-report inventories. The MMPI (Hathaway & McKinley, 1943) has played a major role in this area. The Pd (Psychopathic Deviate) scale was one of the early attempts to measure personality disorder symptomatology. One of the eight original Clinical Scales, Pd was designed to identify individuals with what would today be labeled Antisocial Personality Disorder. Scores on this scale are associated with substance use/abuse, familial discord, and antisocial behaviors (Graham, 2006), which are, as described earlier, associated features of this disorder. Building on the work of Hathaway & McKinley (1943), Morey, Waugh, and Blashfield (1985) developed the first comprehensive set of scales used to measure personality disorders with the MMPI. These researchers based their conceptualization on

8 the Personality Disorder criteria, as defined by the DSM-III (APA, 1980), and used items from the original MMPI (Hathaway & McKinley, 1943). According to Morey, Waugh, and Blashfield (1985), two methods were employed during the construction of the scales. First, the researchers selected four experienced clinical psychologists and asked them to identify MMPI items that they believed to represent the specific diagnostic criteria of the DSM-III (APA, 1980). They allowed some of the items to appear on more than one scale because the diagnostic criteria listed in the DSM-III overlapped as well. These items were then organized into 11 scales, based on the 11 personality disorders (Morey, Waugh, & Blashfield, 1985). Secondly, the researchers conducted a series of internal consistency analyses with the aim of making the items on their scales more homogeneous and removing items to increase discriminant validity. This scale set contained 251 items with varying numbers on each scale. Because item overlap among the scales could be problematic, Morey, Waugh, and Blashfield (1985) decided to create a second set of scales without overlapping items. Items were assigned a single scale with which they were most correlated (Morey, Waugh, & Blashfield, 1985). The non-overlapping scale set also contained 251 items (Morey, Waugh, & Blashfield, 1985). Hurt, Clarkin, and Morey (1990) demonstrated that the overlapping Morey et al. (1985) scale set had adequate stability over a three to four week period in a sample of individuals in treatment for substance abuse. The correlations between the first and second time periods ranged from .82 to .93, demonstrating good test-retest reliability. Overall, the empirical literature on these scales, reviewed by Widiger and Frances (1987)

9 indicates that they possess good to excellent convergent validity (Hurt, Clarkin & Morey, 1990; Jones, 2005), but have some limitations in their discriminant validity (Wise, 1996). Somwaru and Ben-Porath (1994) developed a set of Personality Disorder scales from the MMPI-2 item pool based on the DSM-IV criteria. . The construction of this scale set was similar to the development of the Morey et al. scales (1985). However, Somwaru and Ben-Porath (1994) placed more emphasis on decreasing the amount of item overlap between the scales (Ben-Porath, 1994), but did not develop a set of nonoverlapping scales. The final results included 266 items assigned to 10 different scales. Reliability of the Somwaru and Ben-Porath (1994) scales was examined by the authors. The scales obtained alpha values in the range of .68-.93, indicating that scores on the scale sets had adequate to very good internal consistency (Somwaru, 1994). The scale authors also examined test-retest reliability and reported test-retest values ranging from .76-.92 (Somwaru, 1994). Hicklin and Widiger (2000) examined the convergent validity of the Somwaru and Ben-Porath scales using various criterion measures. They also compared them with the Morey, et, al. (1985) scales. These authors concluded that the MMPI-2 Somwaru and Ben-Porath scales are generally as valid as the Morey et al. scales (1985) in terms of convergent validity, but they possessed increased convergent validity for the assessment of Schizoid, Antisocial, and Borderline Personality Disorders. In examining discriminant validity, there were no significant differences in the performance of the Somwaru and Ben-Porath (1994), with discriminant validity coefficients ranging from .15-.52, and the

10 Morey (1985) scales demonstrating values from .14-.52. Thus, both scale sets demonstrated low to adequate discriminant validity (Hicklin & Widiger, 2000). The MMPI-2 Psychopathology 5 (PSY-5) Scales are also designed to assess personality disorder features. The five scales are: Aggressiveness (AGGR), Psychoticism (PSYC), Disconstraint (DISC), Negative Emotionality/Neuroticism (NEGE), and Introversion/ Low Positive Emotionality (INTR), (Harkness, McNulty, & Ben-Porath, 1995). These scales represent a dimensional approach to assessing personality disorder symptoms, predicated on the notion that these phenomena are continuous, rather than taxonic (Graham, 2006). Using a method they called replicated rational selection, the first step in developing the PSY-5 Scales was to identify MMPI-2 items that were representative of those constructs identified by Harkness and McNulty (1994). This was accomplished by first training a group of college students to understand what the constructs represent. Then, the students selected MMPI-2 items that they judged to be related to or facets of those constructs. The items chosen by a majority of the students were then combined to form the preliminary set of scales. A second step in scale construction involved the use of expert raters, where the experts reviewed the items for each of the preliminary scales to ensure that the items contained within that scale were clearly related to and measures of that particular construct. Experts could delete, but not add items to the provisional scales. The third step of scale development involved a series of statistical analyses. More specifically, items from each scale were correlated with the other scales and if any item

11 was too highly correlated with another scale, it was deleted. In addition, the scale authors also made sure that there was no item overlap between the scales, so that each item was only scored on one scale. The PSY-5 scale authors recommend that they be used to aid in the diagnosis of personality disorders (Harkness, McNulty, & Ben-Porath, 1995). The dimensional nature of the constructs assessed by these scales is particularly useful as the field of psychology begins to move towards a more dimensional conceptualization of personality disorders, such as the Five Factor Model of Personality (Lynam & Widiger, 2001). Recent research by Bagby, Sellbom, Costa, and Widiger (2008) suggests that the PSY-5 Scales better predict symptoms of several personality disorders compared to the Five Factor Model. In particular, the PSY-5 Scales outperformed the NEO-PI-R Scales in the prediction of personality disorder symptom counts for Paranoid, Schizotypal, Narcissistic and Antisocial Personality Disorders (Bagby, Sellbom, Costa, & Widiger, 2008). Another study by Wygant, Sellbom, Graham, & Schenk (2006) also demonstrated the utility of the PSY-5 Scales in the assessment of personality disorders. These authors illustrated the incremental validity of the PSY-5 Scales to assess personality pathology above and beyond the MMPI-2 Clinical and Content Scales. Thus, these scales provide additional useful information relevant to the assessment of personality disorders not readily available from other MMPI-2 Scale sets.

The MMPI-2-RF The Minnesota Multiphasic Personality Inventory-2 Restructured Form (MMPI-2RF; Ben-Porath and Tellegen, 2008) is a 338-item revised version of the MMPI-2

12 (Butcher, et al., 2001), designed to “represent the clinically significant substance of the MMPI-2 item pool with a comprehensive set of psychometrically adequate measures” (Tellegen & Ben-Porath, 2008, p.1). The tests consists of a total of 50 scales including: eight Validity scales, three Higher-Order scales, nine Restructured Clinical (RC) scales, twenty-three Specific Problems (SP) scales, two Interest scales, and five revised Psychopathology Five (PSY-5) scales. Appendix A lists and provides a brief description of the 50 scales of the MMPI-2RF. A significant change to the MMPI-2 was the construction of the Restructured Clinical (RC) Scales (Tellegen, Ben-Porath, McNulty, Arbisi, & Graham, 2003). The RC Scales were derived from factor analyses of the original Clinical Scales to identify the major distinctive component of each scale. A large common factor among the clinical scales was placed into a new scale, Demoralization (RCd). Each of the remaining RC scales represents a major distinctive component of one of the eight original Clinical Scales. The authors of the RC Scales indicated that they were not intended to constitute a comprehensive MMPI-2 assessment of psychopathology and personality characteristics and that some of the facets of these scales warrant independent assessment. Thus, the MMPI-2-RF was developed to add substantive scales that assess constructs either not targeted by the RC scales or warranting more specific assessment (Ben-Porath & Tellegen, 2008). The methodology used to construct the various substantive scales of the MMPI-2-RF paralleled the development of the RC Scales to a large extent (Tellegen & Ben-Porath, 2008).

13 The substantive scales of the MMPI-2-RF are organized into a three-tiered hierarchical structure. The Higher-Order Scales provide a broadband framework with which to organize information obtained from the test. The three dimensions measured by these scales, emotional, thought, and behavioral dysfunction, tap psychological factors relevant to the assessment of personality disorders. Several of the RC Scales can also be linked to personality disorder criteria as can the more narrowly-focused SP Scales. The revised PSY-5 scales were designed specifically to assess variables related to personality disorders. Several studies provide information on the link between MMPI-2-RF scales and personality disorder symptoms. Sellbom, Ben-Porath, and Stafford (2007) demonstrated that RC 4 (compared with the original Clinical Scale 4) was the best MMPI-2 based measure of Psychopathy, which is closely linked to Antisocial Personality Disorder. Kamphuis, Arbisi, Ben-Porath, & McNulty (2008) found that the RC scales outperformed the Clinical Scales in differential prediction of Axis II conditions. The MMPI-2-RF Manual for Administration, Scoring, and Interpretation (Ben-Porath & Tellegen, 2008) identifies certain personality disorders as diagnostic considerations (meaning that the interpreter should consider whether the test-taker meets the actual diagnostic criteria) for some personality disorder. More specifically, RC3 (Cynicism) may be linked to personality disorders involving both mistrust of and hostility towards others (e.g. – Paranoid and Antisocial Personality Disorders; Ben-Porath & Tellegen, 2008). RC8 (Aberrant Experiences) may identify individuals with personality disorders associated with unusual thoughts, perceptions, or experiences, such as Schizotypal Personality

14 Disorder (Ben-Porath & Tellegen, 2008). Diagnostic considerations related to personality disorders are also identified for some Specific Problems (SP) scales. For example, Juvenile Conduct Problems (JCP) is linked to Antisocial Personality Disorder (BenPorath & Tellegen, 2008), Interpersonal Passivity (IPP) to Dependent Personality Disorder (elevated scores > 65) and Narcissistic Personality Disorder (low scores < 38) (Ben-Porath & Tellegen, 2008). Lastly, because they assess the same constructs, the revised PSY-5 scales similarly provide a dimensional model of personality disorder symptoms. Ben-Porath & Tellegen (2008) link each of the PSY-5 scales with diagnostic considerations related to one of the personality disorder clusters. AGGR-r and DISC-r indicate possible Cluster B disorders, PSYC-r is linked with Cluster A disorders and both NEGE-r and INTR-r indicate possible Cluster C disorders.

The Current Investigation Personality disorder-related diagnostic considerations listed by Ben-Porath & Tellegen (2008) are, for the most part, inferential. The purpose of the current study is to examine the hypothesized link, empirically. Several hypotheses were developed prior to conducting statistical analyses. Individual scales from each of the MMPI-2-RF scale sets were hypothesized to be associated with the various personality disorders based on the DSM-IV criteria and the diagnostic considerations listed in the MMPI-2-RF manual. One Higher-Order scale and one PSY-5 scale, along with some RC or SP scales were hypothesized to be associated with each of the personality disorders. The two Interest Scales were not included in any hypotheses because they are not measures of psychopathology. A complete list of the hypotheses is available in Table 1.

15 Table 1. Hypotheses Cluster A Paranoid PD THD RC3 RC6 AGG PSYC-r Cluster B Antisocial PD BXD RC4 RC9 ANP JCP AGG AGGR-r DISC-r

Cluster C Avoidant PD EID RC2 RC7 SFD BRF SAV SHY INTR-r

Schizoid PD EID RC2 SAV DSF FML PSYC-r

Schizotypal PD THD RC2 RC6 RC8 SAV PSYC-r

Borderline PD EID BXD THD RC2 RC6 RC7 RC9 SUI SFD ANP AGG FML IPP (-) SHY (-) DISC-r NEGE-r

Histrionic PD EID BXD RC7 RC9 ACT SHY (-) DISC-r

Dependent PD EID RC7 HLP SFD NFC STW IPP NEGE-r

OCPD EID RC7 RC9 STW BRF NEGE-r

Narcissistic PD BXD RC4 RC9 IPP (-) AGGR-r

CHAPTER II

METHOD

Participants

Potential participants for this study included 1432 men enrolled in an addictions treatment program at a Midwestern Veteran’s Hospital. To be included in this study, the participants were required to complete all of the criterion measures in the study, including: the Minnesota Multiphasic Personality Inventory 2, Borderline Syndrome Index, NEO Personality Inventory Revised, and Structured Clinical Interview for the DSM, Axis II. If one or more of these measures was missing or incomplete, that individual was excluded from the final sample. After excluding individuals for missing data, the sample size decreased to 996 men. Individuals were also excluded from this study if they produced invalid MMPI-2RF protocols, based on the criteria in the MMPI-2-RF Technical Manual: Cannot Say [CNS] raw score ≥ 18, Variable Response Inconsistency [VRIN-r] and/or True Response Inconsistency [TRIN-r] T ≥ 80, Infrequent Responses [F-r] T = 120, or Infrequent Psychopathology Responses [Fp-r] T ≥ 100. A total of 244 individuals (24% of the sample) were excluded due to invalid protocols.

16

17 After removing invalid protocols and missing criterion measures, the final sample was made up of 752 men, with ages ranging from 23-75 (M = 44.4; SD = 8.7). The sample was 57% African American, 37% Caucasian, and 6% had other ethnicities, including American Indian, and Hispanic. The primary diagnoses in this sample, other than substance abuse, included: Post-Traumatic Stress Disorder (6% of the sample), Major Depressive Disorder (5% of the sample), and Pathological Gambling (4.5% of the sample).

Measures

Minnesota Multiphasic Personality Inventory- 2- Restructured Form (MMPI-2-RF) The MMPI-2-RF is a 338-item self-report inventory. The scales on the MMPI-2RF include: eight validity scales, three higher order scales, nine Restructured Clinical (RC) scales, and five revised Psychopathology Five (PSY-5) scales. The validity scales include 7 scales from the MMPI-2 that were revised and the addition of one new scale (Ben-Porath & Tellegen, 2008). The three higher order scales were developed to measure personality and psychopathology at their broadest levels. The next level of the hierarchy includes the Restructured Clinical (RC) scales, which are identical to the RC scales of the MMPI-2 (Ben-Porath & Tellegen, 2008). The PSY-5 scales are similar to those that appear on the MMPI-2, but were revised for the MMPI-2-RF. Harkness and McNulty (2007) used an iterative process consisting of both internal and external analyses. They removed 22 of the 96 items that transferred from the MMPI-2 to the MMPI-2-RF and added 30 new items. This resulted in five non-overlapping scales consisting of 104 items.

18 According to Harkness and McNulty (2007), the revised PSY-5 scales demonstrated lower intercorrelations and analogous external validity, compared to the original scales. However, they still assess the same dimensional models of personality pathology (BenPorath & Tellegen, 2008). The lowest level of the hierarchy includes the 23 Specific Problems (SP) Scales and two Interest Scales. These scales aim to measure certain somatic, internalizing facets, externalizing facets, interpersonal problems, and interests of individuals. Extensive empirical data regarding the psychometric characteristics of the MMPI-2-RF are provided in the Technical Manual (Tellegen & Ben-Porath, 2008).

NEO Personality Inventory- Revised (NEO-PI-R) The NEO-PI-R is a 240-item self-report questionnaire designed to assess five broad domains of personality, including: extraversion, agreeableness, neuroticism, conscientiousness, and openness. The measure also assesses six specific facets of each of these five broad domains, where each broad domain contains six facets or subscales. The facets for Extraversion are Warmth, Gregariousness, Assertiveness, Activity, Excitement Seeking, and Positive Emotion (Costa & McCrae 1992b). The facets for Agreeableness include: Trust, Straightforwardness, Altruism, Compliance, Modesty, and Tendermindedness. Further, the Neuroticism factor contains the facets of Anxiety, Hostility, Depression, Self-Consciousness, Impulsiveness, and Vulnerability to Stress. In addition, the Conscientiousness factor includes Competence, Order, Dutifulness, Achievement Striving, Self-Discipline, and Deliberation facets. Finally, the factor of Openness contains the facets of Fantasy, Aesthetics, Feelings, Actions, Ideas, and Values. Costa and McCrae (1992b) report internal consistency estimates for the broad domains

19 with alphas ranging from .86 to .92 for the broad domains and .56 to .81 for the thirty facet subscales. The authors also attribute the lower internal consistency values of the facets to the fact that each facet only contains eight items. Therefore, it is not unreasonable to expect lower estimates (Costa & McCrae 1992b). Examination of testretest correlations of six years demonstrates that the NEO-PI-R possess adequate to very good temporal stability estimates, with alphas ranging from .63 to .83. Lynam and Widiger (2001) developed a method for assessing the DSM-IV personality disorders using the NEO-PI-R. The authors asked DSM experts to rate prototype cases using all 30 facets of the Five Factor Model (FFM). Then, they combined the ratings to identify a pattern of NEO-PI-R prototype scores for each of the DSM-IV (APA, 2000) personality disorders. Lynam and Widiger (2001) reported good agreement (i.e., r = .48-.66) among the raters for those prototypes and stated that the prototypes map onto the DSM-IV personality disorder criteria well. In the present study, we compared the prototypes generated by Lynam and Widiger (2001) to each participant’s individual NEO-PI-R profile. This comparison yields a set of similarity scores. The similarity scores are generated into the same metric as the prototypes, allowing for the direct comparison of the individual’s NEO-PI-R scores and the Lynam and Widiger (2001) prototypes. An individual is more likely to have a particular personality disorder as their similarity scores becomes closer to the prototype for that disorder. Thus, our participants received a NEOPI-R prototype similarities score for each of the ten DSM-IV personality disorders.

20 The Structured Interview for the DSM, Axis II (SCID-II) The SCID-II (First, Gibbon, Spitzer, Williams, & Benjamin, 1997) is a highly structured clinical interview designed specifically to aid in diagnosis of Axis II disorders. Responses to the interview questions are rated as either present, absent, or sub-threshold. A study by Smith and colleagues (2003) examined the criterion validity the DSM-IV SCID-II, and demonstrated poor agreement between the DSM-IV SCID-II and the Wisconsin Personality Disorders Inventory-IV (WISPI-IV; Klein & Benjamin, 1996) with kappas at or below .40. While there are no studies examining the reliability of DSMIV SCID-II, previous studies have examined the reliability of the instrument with the DSM-III. For example, the authors of the instrument report interrater reliability with kappas ranging from .24 to .74 when the instrument was administered to patients (First, Gibbon, Spitzer, Williams, & Benjamin, 1997).

The Borderline Syndrome Index The Borderline Syndrome Index (BSI; Conte, Plutchik, Karasu, & Jerrett, 1980) is a 52- item self-report inventory intended to assess features and characteristics of Borderline Personality Disorder utilizing a true/false response format. More specifically, the measure focuses on borderline functioning in the areas of poor impulse control, absence of a consistent self-identity, depression, anhedonia, impaired object relations, depersonalization, and a number of “neurotic” symptoms (Sansone, Fine, Seuferer, & Bovenzi, 1989). The test authors examined reliability by calculating internal consistency and they demonstrated high internal consistency, with a Chronbach’s alpha of .92 (Conte, Plutchik, Karasu, & Jerrett, 1980). Convergent validity was examined between the BSI

21 and the Diagnostic Interview for Borderlines (DIS; Gunderson, Kolb, & Austin, 1981). The results indicated that convergent validity between the two measures is high, with correlations between the instruments of .88 (D’Angelo, 1991).

Procedure The archival data set was collected at a Midwestern VA medical center in an addiction treatment unit. The data collection took place over a three-day period upon admission to the hospital’s inpatient addiction treatment program. The addictions that individuals were being treated for included alcohol, drug, illicit substances, and gambling. The measures administered during this time include: a computerized version of the MMPI-2, a demographic questionnaire, and a set of extra-test measures. MMPI-2-RF scales were scored from individual responses to the MMPI-2 items. Tellegen and BenPorath (2008) demonstrated that individuals completing the two versions of the test produce interchangeable scores on the MMPI-2-RF scales. Included in this set of extra test measures were the NEO-PI-R, the SCID-II, and the BSI. Lastly, in accordance with ethical considerations and patient confidentiality, all identifying personal information was removed from the data.

CHAPTER III

RESULTS

Zero-order correlations were calculated between the substantive MMPI-2-RF scales and NEO-PI-R prototype similarities, the SCID-II personality disorder symptom counts, and the Borderline Syndrome Index total scores. Tables 2-5 provide results for all of the correlational analyses that were conducted. The zero-order correlations had to meet two requirements for interpretation. First, the correlation had to reach statistical significance, at p ≤ .05. In addition, correlations between MMPI-2-RF Scales and the NEO-PI-R prototype similarities and Borderline Syndrome Index scores had to reach a magnitude of at least .4, or a medium effect size as defined by Cohen (1981), for interpretation. Correlations between MMPI-2-RF Scales and the SCID-II symptom counts had to reach a magnitude of at least .2, or a small effect size (Cohen, 1981). The magnitude of ≥.4 was selected for the NEO-PI-R because the strength of the correlations between the MMPI-2-RF scales and the NEO-PI-R were consistently higher than those with the SCID-II, likely as a result of shared method (self-report) variance. These effect size requirements were set based on procedures followed in previous empirical research (e.g., McNulty, Ben-Porath, & Graham, 1999) to narrow the focus of interpretation of the analyses, as almost every correlation reached statistical significance.

22

Table 2. Correlations Between Higher-Order (H-O) Scales and Restructured Clinical (RC) Scales and NEO-PI-R and SCID-II Higher Order (H-O) Scales

Restructured Clinical (RC) Scales

Criterion Measure

EID

THD

BXD

RCd

RC1

RC2

RC3

RC4

RC6

RC7

RC8

RC9

NEO-PI-R Paranoid PD SCID Paranoid PD NEO-PI-R Schizoid PD SCID Schizoid PD NEO-PI-R Schizotypal PD SCID Schizotypal PD

.53* .21* .46 .10 .72* .16

.22 .24 .00 .15 .19 .27

.30 .18 -.12 .00 .19 .04

.46* .21* .37 .10 .67* .18

.28 .19 .20 .14 .34 .13

.46* .10 .61* .08 .68 .10

.31 .18 .05 .08 .20 .08

.32 .19 .00 -.02 .31 .06

.29 .19 .06 .12 .24 .24

.48* .27* .26 .13 .56* .22*

.21 .24* .00 .13 .22 .26

.24 .24* -.23 .05 .11 .11

NEO-PI-R Antisocial PD SCID Antisocial PD NEO-PI-R Borderline PD SCID Borderline PD BSI Borderline PD NEO-PI-R Histrionic PD SCID Histrionic PD NEO-PI-R Narcissistic PD SCID Narcissistic PD

.19 .05 .66 .27 .75 .09 .09 -.03 .14

.11 .05 .26 .20 .41 .12 .09 .14 .19

.50 .15 .45 .15 .34 .39 .09 .44 .17

.23 .07 .68* .29* .77* .19 .11 .01 .18

.08 .04 .32 .15 .42* .03 .04 .02 .07

.08 .04 .47 .16 .52 -.07 -.01 -.14 .02

.21 -.01 .26 .10 .34 .06 .06 .20 .14

.43 .12 .49* .16 .42* .34 .05 .32 .16

.16 .09 .30 .19 .40 .11 .12 .17 .19

.22 .09 .58 .29 .68 .15 .15 .10 .23*

.13 .05 .31 .23 .46 .16 .09 .13 .21*

.41 .09 .40 .20 .38 .38 .17 .42 .26

NEO-PI-R Avoidant PD SCID Avoidant PD NEO-PI-R Dependent PD SCID Dependent PD NEO-PI-R Obsessive-Compulsive PD

.66 .29 .40 .17 -.17

.14 .13 .00 .14 -.07

-.02 .01 -.23 .06 -.30

.59* .27* .39 .20* -.25

.31 .12 .16 .08 -.04

.67 .24 .45* .10 -.06

.13 .04 -.07 .08 -.04

.13 .05 -.07 .06 -.29

.17 .13 -.01 .15 -.07

.48 .25 .24 .18 -.15

.16 .15 .03 .15 -.12

-.07 .03 -.23 .12 -.27

SCID Obsessive-Compulsive PD .13 .22* .07 .13 .12 .01 .14 .07 .17 .23* .19 .20 Note: N= 752; Bold & underlining = Supported hypothesis; * = Unanticipated, but significant result; NEO-PI-R= The NEO-PI-R Personality Inventory Revised; SCID= Structured Clinical Interview for the DSM- Axis II; BSI= Borderline Syndrome Index.

23

Table 3. Correlations Between Somatic/Cognitive and Internalizing Scales and NEO-PI-R and SCID Somatic/Cognitive Scales

Internalizing Scales

Criterion Measures

MLS

GIC

HPC

NUC

COG

SUI

HLP

SFD

NFC

STW

AXY

ANP

BRF

MSF

NEO-PI-R Paranoid PD

.32

.17

.22

.23

.34

.27

.33

.37

.35

.36

.34

.51*

.18

.02

SCID Paranoid PD

.16

.15

.13

.18

.21*

.17

.10

.15

.19

.20

.20

.23*

.11

.02

NEO-PI-R Schizoid PD

.35

.13

.13

.16

.25

.18

.32

.31

.33

.19

.15

.14

.09

-.02

SCID Schizoid PD

.10

.06

.15

.12

.09

.05

.13

.06

.15

.12

.11

.11

.06

-.04

NEO-PI-R Schizotypal PD

.48*

.21

.28

.24

.48*

.38

.40*

.59*

.57*

.50*

.38

.40*

.26

.06

SCID Schizotypal PD

.11

.05

.11

.13

.20*

.12

.11

.14

.14

.15

.21*

.14

.16

.02

NEO-PI-R Antisocial PD

.09

.06

.09

.04

.18

.10

.09

.18

.14

.18

.14

.38

.06

-.04

SCID Antisocial PD

.02

.04

.01

.03

.07

.03

.04

.03

.05

.06

.08

.11

-.02

-.05

NEO-PI-R Borderline PD

.39

.20

.29

.22

.47

.37

.33

.60

.52*

.54*

.42*

.55

.27

.06

SCID Borderline PD

.18

.11

.11

.13

.24

.25

.19

.24

.20*

.24*

.24*

.24

.16

.01

BSI Borderline PD

.47*

.25

.34

.31

.61

.54

.48*

.68

.57*

.59*

.51*

.51

.37

.07

NEO-PI-R Histrionic PD

.01

.03

.06

.00

.16

.11

-.01

.17

.14

.19

.13

.20

.11

.02

SCID Histrionic PD

.00

.04

.01

.01

.09

.06

.06

.07

.08

.12

.11

.15

.11

.02

NEO-PI-R Narcissistic PD

-.06

.01

.04

.02

.05

.02

-.02

.05

-.05

.04

.09

.31

.03

-.04

SCID Narcissistic PD

.06

.06

.02

.07

.18

.16

.12

.15

.13

.18

.19

.15

.14

.07

NEO-PI-R Avoidant PD

.44*

.19

.24

.24

.41*

.32

.39

.54

.53*

.42*

.33

.27

.24

.09

SCID Avoidant PD

.16

.06

.09

.11

.22*

.24*

.15

.25

.25*

.21*

.17

.13

.12

.02

NEO-PI-R Dependent PD

.29

.09

.12

.12

.25

.18

.24

.40

.41

.26

.15

-.04

.16

.12

SCID Dependent PD

.09

.02

.08

.09

.17

.12

.13

.16

.17

.19

.14

.12

.13

.03

NEO-PI-R Obsessive-Compulsive PD

-.10

.05

-.06

.01

-.17

-.13

-.01

-.24

-.19

-.19

-.10

-.12

-.09

.02

24

SCID Obsessive-Compulsive PD .07 .06 .10 .08 .16 .13 .13 .09 .12 .16 .12 .15 .15 .02 Note: N= 752; Bold & underlining = Supported hypothesis; * = Unanticipated, but significant result; NEO-PI-R= The NEO-PI-R Personality Inventory Revised; SCID= Structured Clinical Interview for the DSM- Axis II; BSI= Borderline Syndrome Index.

Table 4. Correlations Between Externalizing, Interpersonal, and Interest Scales and NEO-PI-R and SCID-II

Criterion Measures NEO-PI-R Paranoid PD SCID Paranoid PD NEO-PI-R Schizoid PD SCID Schizoid PD NEO-PI-R Schizotypal PD SCID Schizotypal PD NEO-PI-R Antisocial PD SCID Antisocial PD NEO-PI-R Borderline PD SCID Borderline PD BSI Borderline PD NEO-PI-R Histrionic PD SCID Histrionic PD NEO-PI-R Narcissistic PD SCID Narcissistic PD

Externalizing Scales JCP SUB AGG .24 .16 .46 .14 .17 .14 -.02 .00 -.05 .04 .04 .08 .19 .18 .31 .02 .05 .13 .34 .13 .34 .07 .23 .26 .03 .28 .10

.20 .06 .27 .13 .29 .16 .00 .12 .12

.43 .19 .50 .24 .46 .25 .15 .39 .21*

ACT .06 .18 .22 .03 .01 .13

FML .33 .20* .09 .03 .36 .16

Interpersonal Scales IPP SAV SHY .01 .50* .38 .07 .03 .09 .42 .68 .46* .06 .09 .10 .34 .56 .55* .03 .11 .11

DSF .47* .10 .49 .16 .44* .12

.16 .04 .19 .15 .25 .24 .13 .20 .22*

.23 .10 .45 .24 .49 .20 .14 .16 .21*

-.20 .01 .08 .01 .12 -.15 .08 -.38 .06

.09 .05 .25 .09 .38 -.18 .03 .03 .05

-.04 .06 .24 .10 .25 -.32 -.03 -.14 -.02

-.01 .03 .36 .13 .46* -.11 .01 -.16 .03

Interest Scales AES MEC -.21* -.10 .00 .07 -.23* -.22* -.03 .01 -.13 -.18 .01 .01 -.06 -.08 -.09 -.10 -.04 .14 .02 .03 .02

NEO-PI-R Avoidant PD .06 .09 .14 -.08 .26 .43* .42* -.17 .59 .57 SCID Avoidant PD .01 .06 .11 .04 .15 .17 .14 .00 .23 .27 NEO-PI-R Dependent PD -.09 -.01 -.17 -.12 .08 .26 .39 .11 -.10 .50 SCID Dependent PD .02 .10 .12 .09 .13 .04 .00 .10 .02 .01 NEO-PI-R Obsessive-Compulsive PD -.19 -.17 -.18 -.17 -.18 -.01 .21 .00 .12 -.17 SCID Obsessive-Compulsive PD .01 .10 .15 .20 .16 .07 .00 .09 .02 .03 Note: N= 752; Bold & underlining = Supported hypothesis; * = Unanticipated, but significant result; NEO-PI-R= The NEO-PI-R Personality Inventory Revised; SCID= Structured Clinical Interview for the DSM- Axis II; BSI= Borderline Syndrome Index.

.08 .03 -.06 .00 -.07 .10 .05 .15 .06 -.24 -.05 -.23 .01 - .06 .06

25

Table 5. Correlations Between the Personality Psychopathology Five (PSY-5) Scales and NEO-PI-R and SCID-II Personality Psychopathology Five (PSY-5) Scales Criterion Measure

AGGR-r

PSYC-r

DISC-r

NEGE-r

INTR-r

.14

.19

.16

.45*

.46*

SCID Paranoid PD

.13

.22

.15

.24*

.00

NEO-PI-R Schizoid PD

-.35

.00

-.17

.20

.70*

SCID Schizoid PD

-.01

.12

-.04

.12

.11

NEO-PI-R Schizotypal PD

-.23

.20

.11

.55*

.59*

SCID Schizotypal PD

.01

.29

.01

.17

.09

NEO-PI-R Antisocial PD

.33

.09

.42

.24

-.02

SCID Antisocial PD

.09

.05

.11

.09

.06

NEO-PI-R Borderline PD

.06

.26

.35

.63

.28

SCID Borderline PD

.07

.20*

.09

.29

.09

BSI Borderline PD

-.01

.42*

.25

.66

.30

NEO-PI-R Histrionic PD

.19

.13

.38

.21

-.27

SCID Histrionic PD

.12

.11

.05

.15

-.03

NEO-PI-R Narcissistic PD

.48

.11

.37

.08

-.19

SCID Narcissistic PD

.13

.22*

.14

.19

-.06

NEO-PI-R Avoidant PD

-.37

.15

-.08

.46*

.63

SCID Avoidant PD

-.12

.13

-.02

.22*

.21

NEO-PI-R Dependent PD

-.55*

.05

-.20

.25

.36

NEO-PI-R Paranoid PD

SCID Dependent PD

.00

.14

.03

.20

.03

NEO-PI-R Obsessive-Compulsive PD

-.02

-.09

-.32

-.21

.14

SCID Obsessive-Compulsive PD

.08

.25

.05

.18

-.04

26

Note: N= 752; Bold & underlining = Supported hypothesis; * = Unanticipated, but significant result; NEO-PI-R= The NEO-PI-R Personality Inventory Revised; SCID= Structured Clinical Interview for the DSM- Axis II; BSI= Borderline Syndrome Index.

27 For Paranoid Personality Disorder, an association was hypothesized with the Higher-Order Thought Dysfunction (THD) scale. The SCID-II measure of Paranoid Personality Disorder was significantly associated with THD; however, the NEO-PI-R measure of this disorder was not significantly associated with this scale, as the correlation did not reach a medium effect size. Moving down the hierarchy to the Restructured Clinical (RC) Scales, there were no interpretable associations between either the NEO-PI-R or SCID-II measures of Paranoid Personality Disorder and RC3 or RC6, as the correlations did not meet the required effect size. For the Specific Problems (SP) Scales, Aggression (AGG) was more strongly associated with the NEO-PI-R measure of this disorder; whereas, the SCID-II correlation did not meet the effect size requirement. Examination of the PSY-5 Scales revealed that the SCID-II Paranoid Personality Disorder was significantly associated with the MMPI-2-RF Psychoticism (PSYC-r) Scale; whereas the correlation with the NEO-PIR did not reach the magnitude required for interpretation. There were several unexpected, significant associations between EID, RCd, RC7, ANP, and NEGE-r and both the NEOPI-R and SCID-II measures of Paranoid Personality Disorder. In examining the results for Schizoid Personality Disorder, a significant association was found between the NEO-PI-R measure of this disorder and the HigherOrder EID Scale. In addition, there was a significant association between RC2 and NEOPI-R measure. The Interpersonal scales Social Avoidance (SAV) and Disaffiliativeness (DSF) were also significantly associated with the NEO-PI-R measure of Schizoid Personality Disorder; however, the Family Problems (FML) scale was not significantly

28 associated with this measure as hypothesized. The PSYC-r Scale was not significantly associated with the NEO-PI-R as anticipated. Finally, there were no interpretable associations between any selected MMPI-2-RF scales and SCID-II symptom counts, as the correlations were statistically significant, but failed to meet the effect size requirement. For Schizotypal Personality Disorder symptomatology, THD, RC2, RC6, RC8, SAV, and PSYC-r, were significantly associated with the SCID-II symptom counts and the NEO-PI-R similarities. Finally, there was also a significant association between RC8 and both the NEO-PI-R similarities and the SCID-II; however, this association was not initially hypothesized. For Antisocial Personality Disorder symptoms, the associations between hypothesized MMPI-2-RF Scales and NEO-PI-R prototype similarities were much stronger than the correlations with the SCID-II measure of this disorder. None of the correlations between select MMPI-2-RF Scales and the SCID-II were interpretable, as they did not meet the effect size requirement. However, the NEO-PI-R measure of Antisocial Personality Disorder was significantly associated with BXD, RC4, RC9, AGG, and DISC-r. Whereas the findings just described were in line with our hypotheses, there were no interpretable associations between the NEO-PI-R or SCID-II measures of this disorder and the Specific Problems Scales of Anger Proneness (ANP) and Juvenile Conduct Problems (JCP), or the PSY-5 Aggressiveness Scale (AGGR-r) as hypothesized. The results for assessing Borderline Personality Disorder symptomatology indicate that associations were present across all levels of the MMPI-2-RF measurement

29 hierarchy. Beginning with the Higher-Order Scales, EID was significantly associated with the NEO-PI-R, SCID-II, and BSI measures of this disorder. However, unexpectedly THD and BXD were also significantly associated with these measures. The Higher-Order BXD scale was only significantly associated with the NEO-PI-R. In terms of the RC Scales, the strongest relations were between RC7 and the NEO-PI-R, SCID-II, and BSI. Significant associations were also demonstrated between RC2 and the NEO-PI-R and BSI, and RC8 and the SCID-II and BSI. RC9 was also significantly associated with both the NEO-PI-R and SCID-II. Further, the Specific Problems Scale SFD was most strongly associated with Borderline symptomatology, as measured by the three criterion measures. Strong associations were also present for the ANP, AGG, and FML. In addition, a significant association was demonstrated between the SHY and the BSI. Inconsistent with the hypotheses, significant relations were not observed between Interpersonal Passivity (IPP) and any of the three criterion measures. Finally, the Negative Emotionality/Neuroticism PSY-5 Scale was strongly associated with the NEO-PI-R, SCID-II, and BSI; however, no significant associations were present for DISC-r, as hypothesized. There were also unanticipated associations found between RCd, COG, NFC, STW, and AXY and the NEO-PI-R, SCID-II, and BSI. In examining Histrionic Personality Disorder symptomatology, the results demonstrate that there were statistically significant associations between several of the predicted MMPI-2-RF scales and the NEO-PI-R and SCID-II measures of this disorder. However, none of those associations reached the effect size requirement for interpretation.

30 For Narcissistic Personality Disorder, an association was found between the NEO-PI-R measure and BXD; whereas the SCID-II results did not have a large enough effect size for interpretation. RC9 was significantly associated with both the NEO-PI-R and SCID-II measures of this disorder. The PSY-5 AGGR-r scale was significantly associated with the NEO-PI-R, but not the SCID-II. Hypotheses were not supported for RC4 and IPP, which were not significantly associated to either criterion measure as expected. For Avoidant Personality Disorder symptomatology, significant associations were present between all hypothesized MMPI-2-RF Scales (EID, RC2, RC7, SFD, SAV, SHY, and INTR-r) and the NEO-PI-R and SCID-II measures of this disorder. However, one Specific Problems Scale, Behavior-Restricting Fears (BRF) was not meaningfully associated with either criterion measure as hypothesized. There were also several associations present that were not initially hypothesized. RCd, COG, NFC, STW, and NEGE-r were all significantly associated with both the NEO-PI-R and SCID-II. The magnitude of the correlations was greater between the expected MMPI-2-RF Scales and the NEO-PI-R measure, compared with the SCID-II. The results for Dependent Personality Disorder demonstrated significant associations between EID, SFD, NFC, and IPP and the NEO-PI-R Dependent Personality Disorder similarity scores. Only one scale, NEGE-r, was meaningfully associated with the SCID-II, but not the NEO-PI-R. Inconsistent with the hypotheses, RC7 and some of its facets, such as Helplessness/Hopelessness (HLP) and Stress/Worry (STW) were not

31 significantly associated with Dependent Personality Disorder symptomatology, as assessed by the NEO-PI-R or SCID-II. Finally, examination of Obsessive-Compulsive Personality Disorder symptomatology demonstrated a lack of meaningful relations between hypothesized MMPI-2-RF Scales and the NEO-PI-R and SCID-II measures of this disorder. Only one association reached the effect size requirement for interpretation, and a significant association was observed between RC7 and the SCID-II measure. In general, the magnitude of the correlations was much weaker for OCPD than for any other personality disorder symptomatology.

CHAPTER IV

DISCUSSION

The primary objective of this study was to examine associations between MMPI2-RF scales and measures of personality disorder symptomatology. The MMPI-2-RF has several new scales that may be particularly useful in the assessment of personality disorders. This study examined the link between scores on select MMPI-2-RF scales and personality disorder symptoms to determine whether the personality disorder-related diagnostic considerations listed by Ben-Porath & Tellegen (2008) are supported empirically. In addition, anticipated associations were also derived from the DSM-IV diagnostic criteria for each personality disorder. Thus, both the DSM-IV and the personality disorder-related diagnostic considerations (Ben-Porath & Tellegen, 2008) served as guidelines for which MMPI-2-RF scales would be expected to be related to each of the criterion measures. A correlational design was utilized and all participants completed several criterion measures under standardized instructions. Across the personality disorder clusters the magnitude of the correlation patterns was much stronger for the NEO-PI-R personality disorder (PD) measures and the Borderline Syndrome Index scores, in comparison with the SCID-II symptom counts. The strongest pattern of correlations for the Cluster A personality disorders was found for Schizotypal Personality Disorder, with all specific hypothesis supported by the results. 32

33 For Cluster B, the strongest pattern of correlations was demonstrated for Borderline Personality Disorder, where all but two hypotheses were supported. The two hypotheses that were not supported included a predicted negative association between the Interpersonal Passivity (IPP) Specific Problems Scale and criterion measures of Borderline symptomatology, and a predicted association between the PSY-5 Disconstraint Scale and Borderline symptomatology, as measured by the NEO-PI-R, SCID-II, and BSI. Finally, Avoidant Personality Disorder exhibited the strongest correlational pattern among the Cluster C personality disorders, with all hypotheses supported, except for the Behavior-Restricting Fears (BRF) scale. Across the three personality disorder clusters, Borderline Personality Disorder measures had the strongest associations with the MMPI-2-RF. Table 6 provides a detailed summary of the support for the individual hypotheses for each personality disorder. As mentioned, the associations between select MMPI-2-RF scales and the NEOPI-R prototype similarities were generally stronger than those with the SCID-II. One potential explanation from the stronger findings with the NEO-PI-R has to do with the dimensionality of the constructs being assessed. The NEO-PI-R is designed to measure the Five Factor Model (Costa & McCrae, 1992); a dimensional model of personality. The scales of the MMPI-2-RF are dimensional in nature, as higher scores or elevations are indicative of greater psychopathology. In addition, the PSY-5 Scales of the MMPI-2-RF are closely linked to Five Factor Model (Bagby, Sellbom, Costa, & Widiger, 2008); therefore, there is a strong connection between several of the constructs being measured by both the MMPI-2-RF and the NEO-PI-R. Thus, one may expect that the MMPI-2-RF

34 Table 6. Hypotheses and Results Cluster A Paranoid PD EID † RCd † THD ** RC3 RC6 RC7 † ANP † AGG * PSYC-r ** NEGE-r † Cluster B Antisocial PD BXD * RC4 * RC9 * ANP JCP AGG* AGGR-r DISC-r *

Schizoid PD EID * RC2 * SAV * DSF * FML PSYC-r INTR-r*

Schizotypal PD THD ** RC2 * RC6 ** RC7 † RC8 ** COG † SAV * PSYC-r **

Borderline PD EID ***β THD* β BXD **β RCd † RC2 *β RC6 β RC7 ***β RC9 *** COG † SUI **β SFD ***β NFC † STW † AXY † ANP *** β AGG *** β FML***β IPP (-) SHY (-)

Histrionic PD EID BXD RC7 RC9 ACT SHY (-) DISC-r

Narcissistic PD BXD * RC4 RC9 *** IPP (-) AGGR-r *

DISC-r NEGE-r ***β Note: * = Hypothesis supported for NEO-PI-R only; ** = Hypothesis supported for SCID-II only; *** = Hypothesis supported for both NEO-PI-R and SCID-II; β = Hypothesis supported for BSI only; *β = Hypothesis supported for NEO-PI-R and BSI only; **β = Hypothesis supported for SCID-II and BSI only; † = Significant result that was not hypothesized.

35 Table 6, continued Cluster C Avoidant PD EID *** RCd † RC2 *** RC7 *** COG † SFD *** NFC † STW † BRF SAV *** SHY *** NEGE-r † INTR-r ***

Dependent PD EID * RC7 HLP SFD * NFC * STW IPP * NEGE-r **

OCPD EID RC7 ** RC9 STW BRF NEGE-r

Note: * = Hypothesis supported by NEO-PI-R only; ** = Hypothesis supported by SCIDII only; *** = Hypothesis supported by both NEO-PI-R and SCID-II; β = Hypothesis supported by BSI only; *β = Hypothesis supported by NEO-PI-R and BSI only; **β = Hypothesis supported by SCID-II and BSI only; † = Significant result that was not hypothesized.

36 and NEO-PI-R would be more strongly linked as both measures assess personality from a dimensional perspective. In contrast, the SCID-II is a more taxonic-type of assessment tool. The SCID-II interview questions are closely aligned with the DSM-IV criteria for each personality disorder, which are also taxonic. Despite the fact that SCID-II symptom counts were utilized in this study, the symptom counts themselves are taxonic in nature. The rater is asked to indicate the presence or absence of a given symptom, and they do not rate how much or how little (i.e. - severity) of that symptom is present. Since the SCID-II and MMPI-2-RF scales take different perspectives (i.e. - taxonic vs. dimensional) to assessment of symptomatology, it is, perhaps, not surprising that the associations between these two measures are generally of smaller magnitude. Another potential explanation for stronger associations between the MMPI-2-RF and NEO-PI-R prototype similarities has to do with common method variance. Both the MMPI-2-RF and NEO-PI-R are self-report instruments. Therefore, the correlations between the two measures may be somewhat artificially inflated. It is possible that some of the co-variance between the MMPI-2-RF and NEO-PI-R may be attributed to the measurement method, rather than the constructs of interest. Thus, the common method variance shared between the two measures may have increased the systematic measurement error in this study. To address the common method variance concern, findings were only interpreted if they reached the magnitude of .4 or greater, which is a medium effect size, as defined by Cohen (1988).

37 Two sets of personality disorder symptoms do not appear to be adequately assessed by the criterion measures utilized in this study. First, there were no significant associations between select MMPI-2-RF scales and the NEO-PI-R prototypes or SCID-II symptom counts for Histrionic Personality Disorder. In fact, none of the hypotheses associated with Histrionic Personality Disorder were supported in this study. One potential explanation for the lack of support may relate to the substantial symptom overlap shared between Histrionic and Borderline Personality Disorders. Blagov and Westen (2008) examined the relationships between Histrionic and Borderline symptomatology, as they were skeptical of the validity of diagnosing Histrionic Personality Disorder. They demonstrated that a large majority of the patients in their study shared symptoms that overlapped between Histrionic and Borderline Personality Disorder, as defined by the DSM-IV. For example, both disorders share symptoms associated with internalizing and externalizing symptoms, such as anxiety, stress, worry, and acting out behaviorally. In this study, the strongest pattern of correlations, across all clusters, was demonstrated for Borderline Personality Disorder. However, Histrionic Personality Disorder demonstrated the weakest pattern of associations, as none of the hypotheses were supported. Therefore, a potential explanation for the discrepancy in patterns may have to do with the overlap in shared symptomatology between the disorders. Since the patterns are so discrepant, it appears that some of the Histrionic symptomatology may have been misclassified. Furthermore, the pattern of results in this study is consistent with the findings of Blagov and Westen (2008). The second set of personality disorder symptomatology that was largely unsupported in this study was for

38 Obsessive-Compulsive Personality Disorder. With this disorder, only one significant association was demonstrated between RC7 and the SCID-II. There were no significant associations between any MMPI-2-RF scale and the NEO-PI-R. A potential explanation for the unsupported hypotheses relates to the DSM-IV diagnostic criteria. The unanticipated associations found between the SCID-II and particular MMPI-2-RF scales, such as THD, RC8, and PSYC-r, suggest that there is a much larger thought distortion factor associated with OCPD. Thus, the initial hypotheses conceptualized OCPD as more of an internalizing disorder; whereas, the results of this study suggest that OCPD may be more appropriately labeled as a “thought disorder.” A related potential explanation for the unsupported hypotheses may also be low base-rates. Most of the participants in this study did not report significant symptoms associated with OCPD. As noted in Table 6, the results of this study also demonstrated some unexpected associations between the MMPI-2-RF scales and the criterion measures for several of the personality disorders, including: Paranoid, Schizotypal, Borderline, and Avoidant Personality Disorder. Thus, these unexpected associations occur across all three clusters of personality disorders. The unanticipated associations (e.g., EID, RCd, RC7, ANP, & NEGE-r) exhibited for the Cluster A personality disorders all involve scales that assess symptoms associated with general distress or demoralization and negative emotionality. For both Clusters B and C, the unexpected associations (e.g., - RCd, COG, NFC, STW, AXY, and NEGE-r) are related to symptoms of distress, cognitive difficulties, such as confusion memory problems, and internalizing symptomatology, including anxiety and worry for example.

39 The results of this study suggest that Cluster A and Cluster C personality disorders are both related to internalizing symptomatology. Further, The Cluster A and Cluster C personality disorders demonstrate strong associations with measures of negative emotionality in this study. Thus, it appears that individuals with Cluster A or Cluster C personality disorders experience significant amounts of negative emotionality, such as anxiety, worry, and stress. In addition, Cluster B personality disorder symptoms were found to be linked more to externalizing symptomatology, which is consistent with the criteria listed in the DSM-IV-TR (APA, 2000). However, Borderline Personality Disorder appears to represent a mixture of both internalizing and externalizing symptoms, which is also consistent with the DSM criteria. Research by Krueger and colleagues (2001) demonstrated an association between externalizing symptomatology and the construct of Disconstraint. The results of this study suggest that individuals with Cluster B personality disorders have a tendency to be more impulsive, act out behaviorally, and are likely to have difficulty controlling their own behavior. Evidence in support of this notion is demonstrated by the strong associations noted between the MMPI-2-RF scales of BXD, RC4, RC9, and DISC-r and the NEO-PI-R, SCID-II, and BSI measures of Cluster B symptoms. Evaluation of the personality disorder-related diagnostic considerations recommended by Ben-Porath & Tellegen (2008) indicated that almost every consideration was supported in this study. Specifically, support was found for the consideration of Antisocial Personality Disorder if an elevation on RC 4 is present. In

40 addition, the recommendation to consider a personality disorder manifesting unusual thoughts or perceptions when RC8 is elevated was demonstrated to be accurate as well. Support was also demonstrated for the consideration that elevations on RC9 are indicative of Narcissistic Personality Disorder. Furthermore, Ben-Porath and Tellegen’s (2008) recommendation to evaluate for Dependent Personality Disorder when Interpersonal Passivity is elevated was also supported. Elevations on the Social Avoidance Specific Problems Scales were also associated with Avoidant Personality Disorder, as Ben-Porath and Tellegen (2008) suggested. Also, according to Ben-Porath and Tellegen (2008), elevations on the Disaffiliativeness Specific Problems Scales warrant a consideration of Schizoid Personality Disorder. This recommendation was upheld by the results of this study as well. Ben-Porath and Tellegen (2008) also provide personality disorder-related diagnostic considerations for each of the PSY-5 scales and each consideration was supported by the findings in this study. Thus, the results provided support for consideration of a Cluster B personality disorder when elevations were present on both AGGR-r and DISC-r. In addition, elevations on PSYC-r warrant a consideration of a Cluster A personality disorder. Finally, the results also suggest that Cluster C personality disorders should be considered when an elevation was present on NEGE-r or INTR-r. There were two personality disorder-related diagnostic considerations that this study failed to support. No significant associations were found between RC3, Cynicism, and personality disorders characterized by mistrust or hostility (i.e. – Paranoid Personality Disorder). In addition, there was no support for the association between JCP,

41 the Juvenile Conduct Problems Scale, and Antisocial Personality Disorder. Perhaps the association between JCP and Antisocial Personality Disorder may not be present due to the fact that the NEO-PI-R does not assess criminal behaviors. If replicated, the results of this study indicate that the personality disorder-related diagnostic considerations by Ben-Porath and Tellegen (2008) are clinically useful and empirically supported. While a majority of the considerations suggested are supported by the results of this study, there are several associations between the criterion measures that were not initially anticipated. More specifically, the Cluster A personality disorders appear to be characterized by more internalizing-type symptomatology than initially hypothesized, as significant associations were demonstrated between RC7 and Paranoid and Schizotypal Personality Disorders, for example. In general, it appears that individuals with personality disorder symptomatology are reporting more distress than is reflected in the DSM-IV and the diagnostic considerations stated by Ben-Porath and Tellegen (2008). Thus, it is likely that the unanticipated correlations found in this study reflect associated features, rather than the core components of a given personality disorder Therefore, a combination of the personality disorder-related diagnostic considerations (Ben-Porath & Tellegen, 2008) plus elevations on scales assessing distress and internalizing psychopathology is more likely to indicate the presence of a given personality disorder. By using this combination of scale elevations, the potential exists to make test interpretation more accurate and clinically useful.

42 Limitations The current study has several limitations. First, the DSM-IV-TR (APA, 2000) diagnostic criteria for each personality disorder were used to develop the set of hypotheses utilized in this study. Currently, the DSM-IV-TR (APA, 2000) is the “gold standard” used by psychologists to diagnose personality disorders. Thus, the hypotheses utilized in this study were developed based on the most prominent set of criteria available. However, the criteria provided by the DSM are not without their own limitations and flaws. More specifically, empirical research has suggested that there are alternative methods of defining personality disorder symptomatology (Lynam & Widiger, 2001), such as examining symptoms on a continuum of severity, rather than a dichotomy. Thus, there are alternative methods and criteria that may have been used to develop the hypotheses for this study. Another potential limitation of this study relates to the sample. This study utilized an all- male sample and women were not included due to their small number. Previous research has demonstrated differences between men and women with a variety of disorders diagnoses (e.g., depression) and on a variety of different psychological constructs (e.g., aggression). Personality disorders also appear to be consistent with this pattern of differences between men and women as well. For example, differences between men and women have also been noted in the DSM-IV-TR (APA, 2000) prevalence rates of personality disorders. Men are more likely than women to be diagnosed with Antisocial Personality Disorder (APA, 2000). It appears that the opposite effect also exists, where women may be diagnosed with a particular personality more

43 often than men. For example, research by Bornstein (1996) demonstrated that Dependent Personality Disorder was diagnosed more often for women than men. Thus, it is unclear as to whether the results of this study would generalize to women in the same setting. The final potential limitation of this study also relates to the sample and generalizability of findings. The sample utilized in this study was enrolled in treatment in an addictions unit at a VA medical center. Individuals participating in the addictions treatment had a variety of addictions, including alcohol, drug, and gambling addictions. All of the participants also had Axis I conditions as well. Thus, there were tremendous rates of comorbidity between psychological diagnoses and substance abuse and/or dependence diagnoses, which could impact the ability of these results to generalize across outpatient treatment settings where not all individuals receiving mental health treatment have primary co-morbid substance abuse difficulties. In addition, the participants were veterans of various military branches. The generalizability of the results may also be influenced by the nature of the veteran sample as well.

Future Directions Future research should be conducted to evaluate the link between elevated MMPI2-RF scales and personality disorder symptomatology using samples of women, other outpatient samples, and samples from inpatient settings. Thus, it would be important to replicate the findings of this study with women, as the sample in this study is all men. In addition, it is unclear whether the results of this study would generalize to other outpatient samples where individuals are not enrolled in addictions treatment; therefore, future research should examine this link in outpatient settings that do not include

44 addictions treatment. Replicating the results of this study in inpatient settings would also be important for future research, as individuals in inpatient settings are likely to experience different symptoms of psychopathology. In addition, these associations should also be explored in forensic and medical settings, as personality pathology may impact diagnosis and treatment planning in those particular settings. Finally, future research should also examine the use of other criterion measures to assess personality disorder symptomatology. The criterion measures utilized in this study are not the only self-report or interview measures of personality disorder symptomatology available. Research in this area could be strengthened by implementing the use other criterion measures to examine the link between elevations on MMPI-2-RF and personality disorder symptomatology. For example, the Dimensional Assessment of Personality Pathology- Brief Questionnaire (DAPP-BQ; Livesley & Jackson, in press) and the Schedule for Nonadaptive and Adaptive Personality (SNAP; Clark, 1992) are two self-report measures that may be particularly useful in assessing personality disorder symptomatology, as both measures were developed with a focus on Axis II symptoms. In addition, therapist ratings of personality disorder symptomatology may also offer another alternative method of assessment as well.

REFERENCES

American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders- Fourth Edition- Text Revision. Arlington, VA: American Psychiatric Association. American Psychiatric Association (1980). Diagnostic and Statistical Manual of Mental Disorders- Third Edition. Arlington, VA: American Psychiatric Association. Bagby, R. M., Sellbom, M., Costa, P. T., & Widiger, T. A. (2008). Predicting diagnostic and statistical manual of mental disorders-IV personality disorders with the FiveFactor Model of personality and the Personality Psychopathology Five. Personality and Mental Health, 2(2), 55-69. Ben-Porath, Y. S. & Tellegen, A. (2008). Minnesota Multiphasic Personality Inventory2- Restructured Form: Technical Manual. Minneapolis: University of Minnesota Press. Blagov, P. S., & Westen, D. (2008). Questioning the coherence of Histrionic Personality Disorder: Borderline and hysterical personality subtypes in adults and adolescents. Journal of Nervous and Mental Disease, 196(11), 785-797. Bornstein, R. F. (1996). Sex differences in Dependent Personality Disorder prevalence rates. Clinical Psychology: Science and Practice, 3(1), 1-12.

45

46 Butcher, J. N., Dahlstrom, W. G., Graham, J. R., Tellegen, A., & Kaemmer, B. (1989). Minnesota Multiphasic Personality Inventory-2: Manual for administration and scoring.Minneapolis: University of Minnesota Press. Chiao-Chicy, C., Shang-Ying, T., Lien-Wen, S., Tien-Wei, Y., Chang-Jer, T., & HaiGwo, H. (1999). Psychiatric co-morbidity among male heroin addicts: Differences between hospital and incarcerated subjects in Taiwan. Addiction, 94(6), 825-832. Clark, L. A. (1993). Manual for the Schedule for Nonadaptive and Adaptive Personality (SNAP). Minneapolis: University of Minnesota Press. Conte, H. R., Plutchik, R., Karasu, T. B., & Jerrett, I. (1980). A self-report Borderline Scale: Discriminative validity and preliminary norms. The Journal of Nervous and Mental Diseases, 168(1), 428-435. Costa P. T., & McCrae, R. R. (1992). NEO-PI-R Professional Manual. Odessa, FL: Psychological Assessment Resources. Craig, R. J. (2000). Prevalence of personality disorders among cocaine and heroin addicts. Substance Abuse, 21(2), 87-94. D’Angelo, E. J. (1991). Convergent and discriminant validity of the Borderline Syndrome Index. Psychological Reports, 69, 631-635. First, M. B., Spitzer, R. L., Williams, J. B. W., & Gibbon, M. (1997). Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II) user's guide and interview. Washington, DC: American Psychiatric Press. Graham, J. R., Ben-Porath, Y. S., & McNulty, J. L. (1999). MMPI-2 Correlates for outpatient mental health settings. Minneapolis: University of Minnesota Press.

47 Graham, J. R. (2006). MMPI-2: Assessing personality and psychopathology. New York: Oxford University Press. Gunderson, J. D., Kolb, J. E., & Austin, V. (1981). The Diagnostic Interview for Borderline patients. American Journal of Psychiatry, 138, 896-903. Harkness, A. R., & McNulty, J. L. (1994), The Personality Psychopathology Five (PSY5): Issues from the pages of a diagnostic manual instead of a dictionary. In S. Strack & M. Lorr (Eds.), Differentiating Normal and Abnormal Personality (pp. 291-315). New York: Springer. Harkness, A. R., McNulty, J. L., & Ben-Porath, Y. S. (1995). The Personality Psychopathology Five (PSY-5): Constructs and MMPI-2 scales. Psychological Assessment, 7, 104-114. Harkness, A. R. & McNulty, J. L. (August, 2007). Restructured Versions of the MMPI-2 Personality Psychopathology Five (PSY-5) Scales. Paper presented at the meeting of the American Psychological Association, San Francisco, CA. Hathaway S. R., & McKinley, J. C. (1942). The Minnesota Multiphasic Personality Schedule. Minneapolis: University of Minnesota Press. Hicklin J. & Widiger, T. A. (2000). Convergent validity of alternative MMPI-2 Personality Disorder Scales. Journal of Personality Assessment, 75, 502-518. Hurt, S. W., Clarkin, J. F., & Morey, L. C. (1990). An examination of the stability of the MMPI personality disorder scales. Journal of Personality Assessment, 54(1&2), 16-23.

48 Jones, A. (2005). An examination of three sets of MMPI-2 Personality disorder scales. Journal of Personality Disorders, 19(4), 370-385. Kamphuis, J. H., Arbisi, P. A., Ben-Porath, Y. S., & McNulty, J. L. (2008). Detecting comorbid Axis-II status among inpatients using the MMPI-2 Restructured Clinical Scales. European Journal of Psychological Assessment, 24(3), 157-164. Klonsky, D. E. (2000). The DSM classification of personality disorder: Clinical wisdom or empirical truth? A response to Alvin R. Mahrer’s problem 11. Journal of Clinical Psychology, 56(12), 1615-1621. Kraus, A. (1991). Methodological problems with the classification of personality disorders: The significance of existential types. Journal of Personality Disorders, 5(1), 82-92. Krueger, R. F., McGue, M., & Iacono, W. G. (2001). The higher-order structure of common DSM mental disorders: Internalization, externalization, and their connections to personality. Personality and Individual Differences, 30, 12451259. Livesley, J. W. (1991). Classifying personality disorders: Ideal types, prototypes, or dimensions? Journal of Personality Disorders, 5(1), 52-59. Livesley, W. J., & Jackson, D. N. (in press). Manual for the dimensional assessment of personality pathology-basic questionnaire (DAPP-BQ). London, ON: Research Psychologists' Press.

49 Lynam, D. R., & Widiger, T. A. (2001). Using the Five Factor Model to represent the DSM-IV personality disorders: An expert consensus approach. Journal of Abnormal Psychology, 110(3), 401-412. Mahrer, A. R. (2000). So many researchers are sincerely scientific about factitious fictions: Some comments on the DSM classification of personality disorders. Journal of Clinical Psychology, 56(12), 1623-1627. Morey, L. C., Waugh, M. H., & Blashfield, R. K. (1985). MMPI scales for DSM-III personality disorders: Their derivation and correlates. Journal of Personality Assessment, 49, 245-251. Morgenstern, J., Langenbucher, J., Labouvie, E., & Miller, K., J. (1997). The comorbidity of alcoholism and personality disorders in a clinical population: Prevalence rates and relation to alcohol typology variables. Journal of Abnormal Psychology, 106(1), 74-84. Sellbom, M., Ben-Porath, Y. S., & Stafford, K. P. (2007). A comparison of measures of psychopathic deviance in a forensic setting. Psychological Assessment, 19, 430436. Skodol, A. E., Oldham, J., M., & Gallaher, P., E. (1999). Axis II comorbidity of substance use disorders among patients referred for treatment of personality disorders. American Journal of Psychiatry, 156(5), 733-738. Somwaru, D. P. & Ben-Porath, Y. S. (1995, March). Development and reliability of MMPI-2 based personality disorder scales. Paper presented at the 30th Annual

50 Workshop and Symposium on Recent Developments in the Use of the MMPI-2 and MMPI-A, St. Petersburg Beach, FL. Tellegen, A., Ben-Porath, Y. S., McNulty, J. L., Arbisi, P. A., & Graham, J. R., & Kaemmer, B. (2003). MMPI-2 Restructured Clinical (RC) Scales: Development, validation, and interpretation. Minneapolis: University of Minnesota Press. Tellegen, A. & Ben-Porath, Y. S. (2008). Minnesota Multiphasic Personality Inventory2- Restructured Form: Manual for Administration and Scoring (Revised Edition). Minneapolis: University of Minnesota Press. Widiger, T. A., & Frances, A. (1987). Interviews and inventories for the measurement of personality disorders. Clinical Psychology Review, 7(1), 49-75. Wise, E. A. (1996). Comparative validity of MMPI-2 and MCMI-II personality disorder classifications. Journal of Personality Assessment, 66, 569-582. Wygant, D. B., Sellbom, M., Graham, J. R., & Schenk, P. W. (2006). Incremental validity of the MMPI-2 PSY-5 scales in assessing self-reported personality disorder criteria. Assessment, 13(2), 178-18.

APPENDIX A

THE MMPI-2-RF SCALES

Validity Scales VRIN-r

Variable Response Inconsistency- Random responding

TRIN-r

True Response Inconsistency- Fixed responding

F-r

Infrequent Responses- Responses infrequent in the general population Infrequent Psychopathology Responses- Responses infrequent in psychiatric populations Infrequent Somatic Responses- Somatic complaints infrequent in medical populations Symptom Validity- Somatic and cognitive complaints associated at high levels of over-reporting

Fp-r Fs FBS-r L-r K-r Higher-Order (H-O) Scales EID THD BXD Restructured Clinical (RC) Scales

Uncommon Virtues- Rarely claimed moral attributes or activities Adjustment Validity- Avowals of good psychological adjustment associated at high levels with under-reporting

Emotional/Internalizing Dysfunction- Problems associated with mood and affect Thought Dysfunction- Problems associated with disordered thinking Behavioral/Externalizing Dysfunction- Problems associated with undercontrolled behavior

RCd RC1

Demoralization- General unhappiness and dissatisfaction Somatic Complaints- Diffuse physical health complaints

RC2

Low Positive Emotions- Lack of positive emotional responsiveness Cynicism- Non-self-referential beliefs expressing distrust and a generally low opinion of others Antisocial Behavior- Rule breaking and irresponsible behavior Ideas of Persecution- Self-referential beliefs that others pose a threat Dysfunctional Negative Emotions- Maladaptive anxiety, anger, irritability Aberrant Experiences- Unusual perceptions or thoughts Hypomanic Activation- Over-activation, aggression, impulsivity, and grandiosity

RC3 RC4 RC6 RC7 RC8 RC9

51

52 Specific Problems (SP) Scales Somatic Scales MLS GIC HPC

Malaise- Overall sense of physical debilitation, poor health Gastrointestinal Complaints- Nausea, recurring upset stomach, and poor appetite

NUC

Head Pain Complaints- Head and neck pain Neurological Complaints- Dizziness, weakness, paralysis, loss of balance, etc.

COG

Cognitive Complaints- Memory problems, difficulties concentrating

Internalizing Scales SUI

Suicidal/Death Ideation- Direct reports of suicidal ideation and recent suicide attempts

HLP

Helplessness/Hopelessness- Belief that goals cannot be reached or problems solved

SFD

Self-Doubt- Lack of confidence, feelings of uselessness

NFC STW

Inefficacy- Belief that one is indecisive and inefficacious Stress/Worry- Preoccupations with disappointments, difficulty with time pressure

AXY

Anxiety- Pervasive anxiety, frights, frequent nightmares

ANP

Anger Proneness- Becoming easily angered, impatient with others Behavior-Restricting Fears- Fears that significantly inhibit normal activities

BRF MSF Externalizing Scales

Multiple Specific Fears- Fears of blood, fire, thunder, etc.

JCP

Juvenile Conduct Problems- Difficulties at school and at home, stealing

SUB

Substance Abuse- Current and past misuse of alcohol and drugs

AGG

Aggression- Physically aggressive, violent behavior

ACT

Activation- Heightened excitation and energy level

Interpersonal Scales FML

Family Problems- Conflictual family relationships

IPP

Interpersonal Passivity- Being unassertive and submissive

SAV

Social Avoidance- Avoiding or not enjoying social events

SHY

Shyness- Bashful, prone to feel inhibited and anxious around others

DSF

Disaffiliativeness- Disliking people and being around them

53 Interest Scales AES MEC

Aesthetic-Literary Interests- Literature, music, the theater Mechanical-Physical Interests- Fixing and building things, the outdoors, sports

Personality Psychopathology Five (PSY-5) Scales AGGR-r

Aggressiveness-Revised- Instrumental, goal-directed aggression

PSYC-r

Psychoticism-Revised- Disconnection from reality

DISC-r NEGE-r

Disconstraint-Revised- Under-controlled behavior Negative Emotionality/Neuroticism-Revised- Anxiety, insecurity, worry, and fear

INTR-r

Introversion/Low Positive Emotionality-Revised- Social disengagement and anhedonia