• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

A comparison of MMPI code types using MMPI and MMPI-2 norms

ProQuest Dissertations and Theses, 2011
Dissertation
Author: Andrew Garrett
Abstract:
The Minnesota Multiphasic Personality Inventory (MMPI) is recognized as one of the most widely researched and utilized personality instruments in clinical and professional settings (Greene, 2010; Hathaway & McKinley, 1989; Megargee, 2006). Concerns over dated norms and terminology, poor ethnic representation, and other issues led to the creation of a second version, the MMPI-2, in which the perceived flaws of the MMPI were eliminated and the accuracy of the instrument and its administration procedures were improved. However, after the MMPI-2 became public, controversy emerged among users over whether the MMPI-2 was superior to the original MMPI or whether it integrated new errors into a different testing format. While the MMPI-2 is officially recognized by the publisher as the superior instrument of the two, debate emerged concerning whether or not decades of research used to interpret code types could still be accessed using the code types that emerged from the new norms in the revised MMPI-2. The present research sought to make a systematic quantitative analysis of T -scores between the two versions of the MMPI to compare elevations and code patterns. This analysis clarifies how the MMPI clinical scales and corresponding code types change when they are converted to MMPI-2 norms and, conversely, how MMPI-2 code types change when they are converted to MMPI norms. The frequency and significance of these changes are explored with respect to the impact on the use of the MMPI or the MMPI-2 in clinical or professional practice.

TABLE OF CONTENTS Dedication iii Acknowledgments v Abstract vii List of Tables xi Chapter Page 1. Introduction 1 Background 3 Statement of the Problem 6 Significance of the Problem 8 Definition of Terms 8 Summary 9 2. Literature Review 10 Creation of the MMPI 10 Challenges to the Validity of the MMPI 13 Restandardization and Challenges to the Validity of the MMPI-2 18 Initial Challenges to the MMPI-2 21 The Argument Posed by Humphrey and Dahlstrom 24 Response by Ben-Porath and Tellegen 28 Response to Ben-Porath and Tellegen by Dahlstrom and Humphrey 29 ix

Chapter Page Second Response to Dahlstrom and Humphrey by Ben-Porath and Tellegen 30 Repercussions from the Debate 31 Research Questions 36 Summary 37 3. Methodology 39 Participants 39 Measures 39 Procedure 40 4. Results 42 5. Discussion 48 Significance of Findings 54 Limitations and Recommendations 57 References 61 Appendix Page A: Publishable Article 65 B: Curriculum Vitae 93 x

LIST OF TABLES Table 1: Number of Profiles Showing Clinical Range Scores 42 Table 2: Number of Profiles with at Least One Elevated Scale by Gender 43 Table 3: Number of Profiles Meeting Criteria for "Well-Defined" 43 Table 4: Number of Profiles Meeting Criteria for "Well-Defined" by Gender 44 Table 5: Frequency Count of Changes in Code Types from MMPI-2 to MMPI Profiles 44 Table 6: Frequency Count of Changes in Code Type among "Well-Defined" Profiles by Gender 45 Table 7: Frequency of Scales Ranked First or Second Highest in Elevation: MMPI vs. MMPI-2 45 Table 8: Frequency of High Point Scale Elevations: MMPI vs. MMPI-2 Profiles 46 Table 9: MMPI 34/43 Code Type Converted to MMPI-2 Norms: Resulting MMPI-2 Code Types 46 Table 10: MMPI-2 13/31 Code Type Converted to MMPI Norms: Resulting MMPI Code Types 47 XI

CHAPTER 1 INTRODUCTION In the study of personality and psychopathology, the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is generally recognized as one of the most useful - if not the most useful - metric instruments for the assessment of personality (Butcher, 2004; Greene, 2010; Megargee, 2006; Nichols, 2001, 2006). It is also the most widely used personality inventory tool, having been cited in more than 12,000 peer-reviewed scholarly articles (Groth-Marnat, 2003, 2009). However, as the field of general psychiatric research and specialized psychometric research expanded, the applicability of the MMPI-2 also extended beyond the original parameters of what this personality assessment tool was initially designed to cover. The MMPI-2 and its predecessor, the MMPI, have been applied to the study of personality and were subsequently adapted to other areas of research and counseling, including health and employment research, criminal profiling, and the social sciences (Megargee, 2006). The process of extending the original MMPI beyond its initial narrow scope as an instrument measuring psychopathology resulted in disputes as to the MMPI's validity as well as challenges to its merit as a suitable instrument in these new areas of application. Additionally, as the number of questions concerning validity and appropriate use of the MMPI increased, researchers began to query whether the MMPI could be improved by updating norms, item content, and ethnic representation and constructing a more effective personality test 1

that incorporated new findings in the field (Megargee, 2006; Nichols, 2006; Pope, Butcher, & Seelen, 2006). The MMPI-2 was designed to resolve methodological issues and improve the effectiveness of the original MMPI. The revised version of the MMPI uses updated terminology and was designed for use in multiple contexts in addition to the original domains of personality typology (Butcher, 1990; Lewak, 1990; Pope et al., 2006). While the MMPI-2 fitedt many of the criteria for reform requested by researchers and clinicians, there were proponents of the original MMPI who argued that the revised version did not adequately resolve the identified problems with methodology and application. In effect, they argued that the MMPI-2 was inferior to the original version. It was also noted that the new norms resulted in different profile configurations, thus resulting in instances where the patterning or code types changed (Caldwell, 1997). Code types emerge when one, two, or three of the highest-scoring scales are combined to form a specific personality profile. If code types change when the scores are converted from the MMPI to the MMPI-2, then a different interpretation of the personality profile of the subject results. As the personality of the subject is unlikely to have changed during the conversion process, the problem may be that one of the two MMPI versions is inaccurately assessing the examinee's personality. MMPI experts noted that changes in code type occurred during the conversion of raw scores to /"-scores as they were transposed from the MMPI to the MMPI-2 (Humphrey & Dahlstrom, 1995). These changes were particularly problematic among male norms, as scales were suppressed virtually throughout the entire MMPI-2 assessment protocol. Although a different pattern of change emerged with female 2

participants, multiple discrepancies also existed in their profiles, as raw scores were converted to T-scores from the MMPI to the MMPI-2. These psychometric inconsistencies were concerning, especially when the MMPI-2 is so widely used as a diagnostic tool in clinical practice. Background The original Minnesota Multiphasic Personality Inventory (MMPI) was co- created by Starke Hathaway and J. Charley McKinley in 1943 (Lewak, 1990). The purpose of the MMPI was to serve in routine diagnostic assessments of personality and to standardize the criteria used by psychologists and psychiatrists to assess their individual patients (Graham, 1999). The first eight scales developed were "hypochondriasis, depression, hysteria, psychopathic deviate, paranoia, psychasthenia, schizophrenia, and hypomania" (Graham, 1999, p. 2). Hathaway and McKinley cross-validated these subgroups through representational selection of persons enrolled in clinical and outpatient counseling, and diagnoses were assigned based upon the diagnoses of the patients' therapists and outcomes associated with the patients' behaviors. This resulted in the development of personality scales based on findings from clinical pathology, suggesting that the MMPI was useful as an instrument in clinical settings. Following the development of the original scales, the MMPI was occasionally revised or updated to incorporate new data that emerged from experiences in application or from the context in which it was applied. Differentiation among clinical groups was refined through ongoing testing, retesting, and standardization within the MMPI test and its scoring procedures (Nichols, 2001, 2006; Pope et al., 2006). However, some professionals using the MMPI questioned the social relevance, the accuracy, and the 3

validity of the MMPI in specific circumstances; as time progressed, these arguments were made more frequently and targeted multiple aspects of the MMPI. Of note was the increasing distance between the MMPI as a diagnostic tool in psychiatric therapy and its applications in other venues of the health and human services (Nichols, 2001). While the MMPI was showing signs of wear and potential failure in its use with clinical populations, it was increasingly used as a personality measurement tool in multiple domains, including employment testing, mental health profiling for large cohorts, and as a means of assessing and predicting outcome based upon personality type in sociological research (Butcher, 1990; Graham, 1999). The MMPI-2 was proposed as a means of resolving such issues. Theoretically, improvements made in the restandardization of the MMPI would be used to reclaim the MMPFs utility as a tool in clinical therapeutic practice through a process that involved updating normative data, providing additional scales to reflect new findings in personality research, and removing dated wording that could potentially affect the test-taker's perceptions of specific items (Butcher, 1990). In addition to restructuring the MMPI so that its successor, the MMPI-2, could be more effectively used in clinical practice, it was proposed that the MMPI-2 could resolve many of the issues of social relevance that had plagued the original version. A major goal of the MMPI-2 was to achieve contemporary relevance such that the new test was designed to reflect new or revised information from the scientific community (Butcher & Megargee, 1989). It was decided that a broader national sampling of the nonclinical population was required to establish a baseline that reflected the general population (Butcher & Megargee, 1989; Groth-Marnet, 2003; 2009; Nichols, 2001). 4

Researchers and specialists with experience in administering the MMPI created a revised edition, and the MMPI-2 was released in 1989. The MMPI-2 was specifically designed for use with patients over the age of 18 years; but a secondary test, the MMPI- A, was specifically designed to be used with adolescents (Archer & Krishnamurthy, 2001). The newly-restructured MMPI-2 incorporated greater flexibility into the test and ostensibly addressed many criticisms associated with the MMPI. However, proponents of the original MMPI argued that its replacement had significant potential for procedural errors and ongoing validity-related challenges (Butcher & Megargee, 1989). Researchers and clinicians conducted a debate in the form of published journal articles to elaborate upon their positions and refute the arguments of their counterparts. The main arguments were offered by Humphrey and Dahlstrom (1995, 1996) and Ben-Porath and Tellegen (1995; Tellegen & Ben-Porath, 1993, 1996), in which Humphrey and Dahlstrom argued in favor of preserving particular features of the MMPI, while Ben-Porath and Tellegen argued that the MMPI-2 was a superior replacement, and no preservation was required. The debate was initiated by an article published by Dahlstrom (1992) in which he indicated that the raw scores between the MMPI and the MMPI-2 were comparable, but the patterns resulting from T-score conversions demonstrated inconsistencies between the two tests. Dahlstrom suggested that these observable inconsistencies could cause problems for researchers and others seeking to use the MMPI-2 in personality analysis, and, given that the original MMPI had a strong baseline of research data based on decades of use, he strongly recommended that the original version be used until consistency errors in the MMPI-2 could be resolved. Dahlstrom subsequently co- authored a paper with Humphrey (Humphrey & Dahlstrom, 1995) to clarify these ideas, 5

arguing that the MMPI-2 could not be used in the same context as the original MMPI because ongoing results of clinical testing demonstrated that the MMPI and the MMPI-2 garnered different results. Advocates of the MMPI-2, on the other hand, did not believe that the MMPI-2 should be dismissed because of perceived challenges to its validity. Ben-Porath and Tellegen (1995) stressed that the original hypothesis used by Humphrey and Dahlstrom (1995) and the subsequent conclusions derived from this hypothesis were unfounded and were based upon faulty research methods. Specifically, Ben-Porath and Tellegen believed that Humphrey and Dahlstrom had placed incorrect values on certain scores that, in turn, accounted for the distorted analysis. Several articles elucidating the points of these opposing authors were exchanged, each delineating and debating their respective positions. These were augmented by an opinion essay by Alex Caldwell (1997), a leading authority on the MMPI who concurred with the position of Humphrey and Dahlstrom (1995, 1996). At the current time, there is no substantive active debate over the validity of the MMPI compared to the MMPI-2; the MMPI-2 has been openly accepted as the successor to the MMPI (Butcher, 2004). Statement of the Problem The value of the MMPI-2 as a unique diagnostic tool in therapy and its value in other areas of social research has been firmly established (Duckworth, 1995; Megargee, 2006; Nichols, 2001, 2006). The MMPI has, for all intents and purposes, been replaced by the MMPI-2. However, when the MMPI-2 was initially introduced, significant controversy arose concerning its comparability to the original MMPI (Caldwell, 1997; Humphrey & Dahlstrom, 1995, 1996; Megargee, 2006; Nichols, 2006). A core question 6

that initiated this debate has yet to be adequately resolved. Specifically, if MMPI-2 and MMPI profiles are using the same research base to interpret code types, then why are there differences in these code types, and what is the significance of these differences in the general research supporting the interpretation of code types? Although this question remains unresolved, interpretations of code types generated by the MMPI-2 may lead to different conclusions than interpretations of code types generated by the MMPI. As the application and use of the MMPI-2 is extensive, it is possible that differences in scale elevations and code types between the two versions of the MMPI may potentially cause serious interpretation problems in clinical and professional settings. As the original MMPI code types served as the research base for interpretation of the MMPI-2 code types, comparisons can and should be made between scale elevations and code types derived from these two tests to clarify the continuity and consistency of scale elevations and code types between the MMPI and the MMPI-2. Every day, practitioners continue to make code type interpretations using the MMPI-2, thereby sometimes unknowingly introducing different code type interpretations into their assessments. To better explore these issues, the following problem statement was used to guide the proposed research: Using a naturally occurring group of MMPI profiles found in a common clinical application, to what extent do scale elevations and code types from MMPI-2 profiles differ from comparative scale elevations and code types that would have resulted from using the same raw scores with original MMPI norms? In addition, when code types vary depending upon the norms used, what specific MMPI code types are associated with what specific MMPI-2 code types? 7

Significance of the Problem The heavy reliance on the MMPI-2 within the psychiatric community, coupled with its use in the workplace, emphasizes its critical role in personality testing. If there are inconsistencies in comparative scale elevations and code types between the MMPI and MMPI-2, then there will be a risk that an inaccurate representation of an individual's psychological functioning may be generated. This research attempted to clarify how MMPI scale elevations and code types change when they are converted to MMPI-2 norms and delineate the frequency of the types of changes that occur. Likewise, it elucidated some changes that occur when this process is reversed, and MMPI-2 code types are changed or converted into MMPI norms. If code types fail to remain stable when they are transformed from one version of the MMPI to another, then it becomes unclear how the code types should be interpreted. Definition of Terms In the context of this research, the following definitions are used: Clinical practice - Psychometrics conducted in a clinical setting, such as a psychologist's office, or used for clinical purposes, such as determining a personality diagnosis. Code type - A combination of the most highly elevated basic clinical scale scores. This summary pattern often corresponds with specific personality outcomes. Restandardization - The process of transforming the MMPI to the MMPI-2 in which new norms and other alterations were added to better reflect the sociocultural environment and broaden the sample population used as the baseline in the test. 8

T-score - also true score. In the study of psychometrics, the T-score is a test score that has been converted to an equivalent standardized score when there is a normal distribution with a mean of 50 and a standard deviation of 10. Summary The Minnesota Multiphasic Personality Inventory was used as an assessment of personality until it was restructured to improve accuracy of test scores. The second version, the MMPI-2, has been challenged in terms of validity and appropriateness of code type representation. A systematic quantitative comparative analysis of T-scores as represented in the MMPI and the MMPI-2 will demonstrate how scales compare in elevation when code types are converted to the norms used in each test. 9

CHAPTER 2 LITERATURE REVIEW For contemporary mental health practitioners, the debate over the validity of the Minnesota Multiphasic Personality Inventory (MMPI) as a diagnostic tool has been resolved from a practical perspective, as the publisher has officially replaced the original MMPI with the MMPI-2 as the recognized version for use in clinical practice and in professional personality testing. Challenges remain, however, as the decision to use the MMPI-2 was based in large part on user preference and observation among certain professionals that the second instrument was superior. Questions over the comparability of profiles generated by the two versions of the test were never answered to the satisfaction of many leading authorities who worked directly with the MMPI and the MMPI-2 (Edwards, Morrison, & Weissman, 1993). Their concerns clarify the goals of the current research effort, as they highlight why the original version of the MMPI was considered a successful instrument. This literature review will familiarize the reader with issues involved in the evolution of the MMPI and provide a working foundation for the research efforts contained in the following chapters. Creation of the MMPI The MMPI was created to augment mental health care by providing a targeted, focused process that could be dedicated to each person's unique case needs. The initial concept was guided by the view that accurate and effective psychiatric care could only be 10

provided after the patient's condition was appropriately and accurately assessed (Butcher, 1990; Butcher & Williams, 2000). In 1942 and 1943, the original MMPI was released as a multi-level diagnostic tool that required a minimum sixth grade literacy level for successful completion, thus allowing more widespread application. The instrument was a questionnaire through which the respondent could provide responses that accurately reflected his or her personal beliefs, attitudes, and experiences. Scale development used an empirical keying methodology, an approach that identifies items that distinguish between normal and criterion groups empirically without regard to the obvious content of the test item (Nichols, 2001). Empirical keying was derived from mental health research in general and was not established on the basis of any single subset of the mental health population. Consequently, empirical keying could be used in the assessment of clinical cases in which patients did not have abnormal pathology. This broadened the applications of the MMPI for use in multiple dimensions of mental health and personality assessment, as empirical keying reduced ambiguity, improved clarification of responses, and was not limited to subjects who were pathological. The MMPI was also unique in that its developers recognized that the majority of other diagnostic instruments contained leading questions (i.e., items that cued a specific response from the patient) (Butcher & Williams, 2000; Nichols, 2001). The MMPI addressed this problem through the use of questions designed to provide a broader, more generalized environment in which multiple concepts were given equal positioning (Graham, 1999; Nichols, 2001). It was believed that this positioning would improve the overall representation of mental health issues and would facilitate improved accuracy in self-reporting responses (Megargee, 2006; Pope et al., 2006). 11

Eight core categories were initially used for the coding processes (Megargee, 2006). However, as the MMPI was more broadly used, an additional two categories were added to the core categories: masculinity-femininity and introversion (Nichols, 2006; Pope et al., 2006). The mental health outcomes determined by the coding process in the original MMPI were derived from scores in these initial eight categories (Greene, 1999; Megargee, 2006; Nichols, 2001; Pope et al., 2006). Mental health professionals could isolate the domains in which mental health issues were evident (e.g., high scores) or appeared likely to manifest (e.g., scores indicating the presence of corresponding symptoms). The MMPI enabled co-morbid scoring, where the presence of two or more mental health conditions could be determined based upon high scores in multiple categories. The interpretation of co-morbid scores also facilitated accurate diagnosis of mental illness, where the mental status of the patient was assessed by contrasting the scores to known types of diseases or disorders. While MMPI interpretation of co-morbid profiles fomented accurate diagnostic impressions, the scoring process in the original MMPI was achieved through determining normal scale scores from persons considered to be free from mental health issues or chronic mental illnesses. The mean in the eight categories was determined on the basis of scores obtained by a normative, nonpathological sample, which created a representative coding key (Greene, 1999). Deviation from this key suggested deviation from the norm in one or more of the eight categories, and it was inferred that the higher the deviation from the normal mean, the more severe the mental illness (Nichols, 2001). Scoring was also cumulative: Not only could assessment of the individual categories be used to indicate the potential type of mental illness of the patient, but also the scores themselves 12

could be compiled to indicate the overall mental health of the patient. Code types in the MMPI evolved as the next step in the application of the MMPI (Greene, 1999; Nichols, 2001). Code types are generated by identifying the most highly elevated scales and are connected to specific personality profiles (Greene, 2010; Nichols, 2001). Challenges to the Validity of the MMPI While the MMPI was one of the foremost psychometric tests in the domain of mental health, problems in the use of the instrument emerged after nearly 4 decades of use (Butcher, 1990, 2004; Butcher & Williams, 2000). Foremost among these were issues corresponding to its utility in co-morbid diagnoses, where the terminology and values found in the MMPI could yield confusing results with patients who had multiple clinical pathologies (Megargee, 2006; Pope et al, 2006). It was suggested that efforts to clarify the scoring process would reduce ambiguity when the MMPI was used in clinical practice. Another serious issue with the MMPI was the concept of deviation from a standard of normalcy. As previously stated, the mean score was derived from scores generated by test-takers who were considered to be normal or who represented behaviors and attitudes that were situated within a range common to the majority of mentally healthy persons. Yet, the diagnostic criteria assigned to one patient who tested within normal limits on one or more categories could not be transferred to other categories. A normal patient had acceptable scores in all eight categories, but a patient who had an abnormal score in one category was deemed to be mentally ill. Ongoing research, however, suggested that most persons had scores that deviated from the normal range in at least one category, suggesting that they suffered from mental illness. Such findings 13

Full document contains 115 pages
Abstract: The Minnesota Multiphasic Personality Inventory (MMPI) is recognized as one of the most widely researched and utilized personality instruments in clinical and professional settings (Greene, 2010; Hathaway & McKinley, 1989; Megargee, 2006). Concerns over dated norms and terminology, poor ethnic representation, and other issues led to the creation of a second version, the MMPI-2, in which the perceived flaws of the MMPI were eliminated and the accuracy of the instrument and its administration procedures were improved. However, after the MMPI-2 became public, controversy emerged among users over whether the MMPI-2 was superior to the original MMPI or whether it integrated new errors into a different testing format. While the MMPI-2 is officially recognized by the publisher as the superior instrument of the two, debate emerged concerning whether or not decades of research used to interpret code types could still be accessed using the code types that emerged from the new norms in the revised MMPI-2. The present research sought to make a systematic quantitative analysis of T -scores between the two versions of the MMPI to compare elevations and code patterns. This analysis clarifies how the MMPI clinical scales and corresponding code types change when they are converted to MMPI-2 norms and, conversely, how MMPI-2 code types change when they are converted to MMPI norms. The frequency and significance of these changes are explored with respect to the impact on the use of the MMPI or the MMPI-2 in clinical or professional practice.