Factor Structure of the MCMI Basic Personality Scales and Common-Item Artifact

The eight basic personality scales of the Millon Clinical Multiaxial inventory (MCMI) were derived from Millon's theory of personality, but the adequacy of the MCMI for measuring Millon's personality constructs has never been assessed. One major problem with using factor analysis to illuminate the structure of the MCMI personality scales is that artifactual structure may result from item overlap among the scales. To analyze this, item-overlap coefficients were factored and compared to the factor structures of five subject samples. For the eight basic personality scales, three factors emerged for the overlap matrix and each of the five sample matrices: Aloof-Social, Aggressive-Submissive, and Lability-Restraint. It was concluded that these three factors are inconsistent with Millon's theory and that they will be found artifactually across a wide variety of populations due to overlapping items.     

Factor structure of the MCMI basic personality scales and common-item artifact

The eight basic personality scales of the Millon Clinical Multiaxial Inventory (MCMI) were derived from Millon's theory of personality, but the adequacy of the MCMI for measuring Millon's personality constructs has never been assessed. One major problem with using factor analysis to illuminate the structure of the MCMI personality scales is that artifactual structure may result from item overlap among the scales. To analyze this, item-overlap coefficients were factored and compared to the factor structures of five subject samples. For the eight basic personality scales, three factors emerged for the overlap matrix and each of the five sample matrices: Aloof-Social, Aggressive-Submissive, and Lability-Restraint. It was concluded that these three factors are inconsistent with Millon's theory and that they will be found artifactually across a wide variety of populations due to overlapping items.     

Development and psychometric properties of a new measure for memory phenomenology: The Autobiographical Memory Characteristics Questionnaire

The purpose of this study is to develop a valid and reliable measure for the phenomenology of autobiographical memories. The psychometric properties of the Autobiographical Memory Characteristics Questionnaire (AMCQ) were tested in three studies: the factor structure of the AMCQ was examined for childhood memories in Study 1 (N = 305); for autobiographical memories related to romantic relationships in Study 2 (N = 197); and for self-defining memories in Study 3 (N = 262). The explanatory factor analyses performed for each memory type demonstrated the consistency of the AMCQ factor structure across all memory types; while a confirmatory factor analysis on the data garnered from all three studies supported the constructs for the autobiographical memory characteristics defined by the researchers. The AMCQ consists of 63 items and 14 factors, and the internal consistency values of all 14 scales were ranged between .66 and .97. The relationships between the AMCQ scales related to gender and individual emotions, as well as the intercorrelations among the scales, were consistent with both theoretical expectations and previous findings. The results of all the three studies indicated that this new instrument is a reliable and robust measure for memory phenomenology.     

Are All Hostility Scales Alike? Factor Structure and Covariation Among Measures of Hostility1

Few studies have examined the correlational and structural relationships between measures of hostility that may predict coronary heart disease. Using data from a population-based study (N= 2,682), we examined the factor structure among the five subscales from Siegel's (1986) Multi-dimensional Anger Inventory, the three subscales from Spielberger et al.'s (1985) anger expression scales, a Cook-Medley (Cook & Madley, 1954) hostility subscale that measures cynicism, and items measuring hostility from four scales that assess Type A behavior. Eight separate factors were identified: hostile anger expression, perceived control over the expression of one's anger, frequency of anger, ease of anger provocation, brooding, hostile outlook, cynicism, and sullenness. These results suggest that the structure of hostility measures is more complex than previous factor analyses have suggested. Our findings point to the dangers in assuming that a factor structure can be generalized beyond the data on which it is based to provide theoretical statements about psychological structure. The new scales appear to be valid because they correlate with demographic variables and indicators of social support in the expected directions.     

Reliability and concurrent validation of the IPIP Big-Five factor markers in China: Consistencies in factor structure between Internet-obtained heterosexual and homosexual samples

Previous studies have suggested the cross-cultural generalizability of a 5-factor structure for personality traits. In this article, we analyzed the utility of 2 versions (100-item and 50-item) of the IPIP Big-Five factor markers in both heterosexual (N = 633) and homosexual (N = 437) samples in China. Factor analysis within versions showed that both versions of these IPIP measures showed clear 5-factor orthogonal structures that were nearly identical to the American structure in both subject samples. The reliabilities of the five factors were quite high except for the 50-item measure of Agreeableness. The part-whole correlations between the 100-item and 50-item factors were high, as were the factor congruence coefficients between the heterosexual and the homosexual samples. Both versions of the IPIP Big-Five factor markers were strongly correlated with the scales from the Big-Five Inventory (BFI: John, Donahue, & Kentle, 1991), thus providing some concurrent validation in a Chinese context.     

Millon Clinical Multiaxial Inventory (MCMI) scores with a collegiate sample: Long-term stability and self-other agreement

This study examined (a) the stability of the personality and symptom scales of the Millon Clinical Multiaxial Inventory (MCMI) for a sample of university undergraduates and (b) the correspondence between MCMI scores from self-report versus scores obtained from knowledgeable others who responded by describing the subject rather than themselves. The MCMI was administered to subjects during their freshman year and then again during their senior year. As in clinical populations, stability estimates in this collegiate sample were greater for the basic personality scales than for the symptom scales. Comparison of the results with those from previous studies further showed that the stability coefficients for the collegiate sample were somewhat lower overall than those from treatment follow-up studies with clinical populations. The results also showed that reasonable agreement between self- and others' ratings was obtained on most MCMI scales. The greatest agreement was found for scales that reflect more observable behaviors and relate to an introversion-extroversion dimension, whereas the least agreement was found for scales related to a psychoticism dimension. Lower levels of agreement were also found for scales having a lower mean proportion of items endorsed.This study was sponsored in part by PHS Grant R01 MH31750-01-6, by PHS Grant 5R01 AA06754-01-03, and by funds from the University of Connecticut Research Foundation and Computer Center.     

Student-teacher concordance on dimensions of student social competence: A multitrait-multimethod analysis

Teacher and self-ratings of social competence were obtained for 669 (341 males, 328 females) grades 3–6 students using the Social Behavior Assessment—Revised (SBA-R) and a self-report scale consisting of the identical items reworded in first person. Factor analysis and multitrait-multimethod matrix analysis were used to validate construct interpretations. The previous four-factor structure of the SBA-R was not replicated. However, the three-factor structure that did emerge was better defined. Although student-teacher concordance was moderate for the Self-Control/Social Convention and Academic Responsibility factors, it was minimal for the Social Participation factor. Evidence of discriminant validity was weak. As with other studies using rating scales, method bias was in evidence. Sex differences were found in terms of variance explained by each factor and convergent validity. Discussion focuses on problems associated with construct validation in general and validation of school-related social competence in particular.     

An examination of the factor structure of the Sociocultural Attitudes Towards Appearance Questionnaire-3 in Malaysia

The present study examined the factor structure of a Malay translation of the Sociocultural Attitudes Towards Appearance Questionnaire-3 among a community sample of 554 Malaysian women. Results of an exploratory factor analysis revealed the existence of four factors, two of which (Information and Internalization-Athlete) mirrored those found among Western samples. An additional factor was an amalgamation of two factors reported in the West, namely Pressure and Internalization-General. A fourth factor consisted of six items, four of which cross-loaded onto previous factors, and was consequently dropped from analyses. Cronbach's α coefficients for the three retained factors were all above .82, and the three factors were significantly correlated with each other and with participants’ body mass index. The results of this study stress the need for locally developed scales in the study of body image and a shift away from reliance on scales developed in the West.     

The structure and reliability of the teacher temperament questionnaire

The Teacher Temperament Questionnaire is a recently developed instrument designed for personality assessment with preschool and early school children. The present study employed a sample of 116 in order to study the reliability and structure of the TTQ. A principal-components analysis yielded four well-defined components, which were labeled Compliance, Extraversion, Interpersonal Affect, and Environmental Sensitivity. These results did not support the theoretical scales. The component patterns were well defined and stable across the two administrations of the inventory. Internal consistency and test-retest reliability coefficients were determined for each of the four new scales. A second administration involving only items from the first three components provided a partial replication.An initial version of this paper was presented at the Northeastern Educational Research Association Meeting, Ellenville, New York, October 1979.     

A Moderated Nonlinear Factor Model for the Development of Commensurate Measures in Integrative Data Analysis

Integrative data analysis (IDA) is a methodological framework that allows for the fitting of models to data that have been pooled across 2 or more independent sources. IDA offers many potential advantages including increased statistical power, greater subject heterogeneity, higher observed frequencies of low base-rate behaviors, and longer developmental periods of study. However, a core challenge is the estimation of valid and reliable psychometric scores that are based on potentially different items with different response options drawn from different studies. In Bauer and Hussong (2009) we proposed a method for obtaining scores within an IDA called moderated nonlinear factor analysis (MNLFA). Here we move significantly beyond this work in the development of a general framework for estimating MNLFA models and obtaining scale scores across a variety of settings. We propose a 5-step procedure and demonstrate this approach using data drawn from n = 1,972 individuals ranging in age from 11 to 34 years pooled across 3 independent studies to examine the factor structure of 17 binary items assessing depressive symptomatology. We offer substantive conclusions about the factor structure of depression, use this structure to compute individual-specific scale scores, and make recommendations for the use of these methods in practice.     

Interpretation of a Full-Information Item-Level Factor Analysis of the MMPI-2: Normative Sampling and Nonpathognomonic Descriptors

An exploratory item-level full-information factor analysis was performed on the normative sample for the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). This method of factor analysis, developed by Schilling and Bock (Bock & Schilling, 1997) and based on item response theory, works directly with the response patterns and avoids the artifacts associated with phi coefficients and tetrachoric coefficients. Promax rotation of the factor solution organizes the clinical scale items into 10 factors that we labeled Distrust, Self-Doubt, Fitness, Serenity, Rebelliousness, Instrumentality, Irritability, Artistry, Sociability, and Self-Reliance. A comparison was made to the results of Johnson, Butcher, Null, and Johnson (1984), who performed a principal-component analysis on an item set of 550 items from the previous version of the MMPI (Hathaway & McKinley, 1943). Along with version changes and sampling differences, the essential differences between Johnson et al.'s results and ours may be attributed to differences between the Schilling and Bock method, which uses all information in the item responses, and the principal-component analysis, which uses the partial information contained in pairwise correlation coefficients. This study included 518 of the complete 567 items of the MMPI-2, versus Johnson et al.'s retention of 309 of the initially included 550 items of the previous MMPI. The full-information analysis retained all 518 initially included items and more evenly distributed the items over the 10 resulting factors, all sharply defined by their highest loading items and easy to interpret. Sampling effects and factor label considerations are discussed, along with recommendations for research that would validate the clinical utility of the implied scales for describing normal personality profiles. The full-information procedure provides for Bayes estimation of scores on these scales.     

FACTORIAL INVARIANCE OF PERSONALITY RATINGS

K uusinen , J. Factorial invariance of personality ratings. Scand. J. Psychol ., 1969, 10 , 33–44—Three studies employing the same set of 33 personality rating scales are described. (I) Five factor structures of the scales, derived from ratings of five groups of stimuli, were compared. (2) One subject rated different personality concepts three times at one week's intervals. (3) 12 individual factor structures were compared to the factor structure computed from peer ratings of a group of 39 subjects. Results: (a) the factor structure of the scales was largely independent of stimuli, (b) the structure for an individual was stable, (c) the group structure represented the individual's structure, ( d ) differences between group and the individual structure and between individuals were negligible.     

A note on the discovery of ag factor by means of Thurstone's centroid method of analysis

A fictitious factor matrix including 16 tests and 3 factors, one of which was ag factor, was prescribed. From it two typical factor problems, including errors of sampling, were derived. Students in training, without awareness of the factor patterns, arrived at essentially correct solutions by the use of Thurstone's centroid method with rotation of axes. Errors in the calculated factor matrix were very close in size to the sampling errors in the correlation coefficients. It is concluded that ag factor need not escape detection by Thurstone's procedures if the criteria of complete simple structure are not demanded.     

Vulnerability factors in depression: The facets of sociotropy and autonomy

Sociotropy-autonomy (Beck, 1983) describes a set of personality dimensions that relate to an individual's vulnerability to depression. Two recently developed scales, the Sociotropy-Autonomy Scale (Clark & Beck, 1991) and the Personal Style Inventory (Robinset al., 1994), have been developed in order to assess these personality dimensions. Typically, these measures are used in isolation and little published information is available concerning their interrelationship. The present study examined the relationship between the two scales and specified the various factors that emerge when the items of the two scales are integrated. Six hundred fifty-two participants responded to the Personal Style Inventory (Robinset al, 1994) and the Sociotropy-Autonomy Scale (Clark & Beck, 1991). A factor analysis on all of the items of the two scales revealed a five-factor structure (two sociotropy, two autonomy, and one achievement factor). The relationships among depression, the five derived factors, and the original scales developed by Clark and Beck (1991) and Robinset al. (1994) were critically examined. The results are discussed in the context of vulnerability to depression.     

Interpretation of a full-information item-level factor analysis of the MMPI-2: normative sampling and nonpathognomonic descriptors

An exploratory item-level full-information factor analysis was performed on the normative sample for the MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). This method of factor analysis, developed by Schilling and Bock (Bock & Schilling, 1997) and based on item response theory, works directly with the response patterns and avoids the artifacts associated with phi coefficients and tetrachoric coefficients. Promax rotation of the factor solution organizes the clinical scale items into 10 factors that we labeled Distrust, Self-Doubt, Fitness, Serenity, Rebelliousness, Instrumentality, Irritability, Artistry, Sociability, and Self-Reliance. A comparison was made to the results of Johnson, Butcher, Null, and Johnson (1984), who performed a principal-component analysis on an item set of 550 items from the previous version of the MMPI (Hathaway & McKinley, 1943). Along with version changes and sampling differences, the essential differences between Johnson et al.'s results and ours may be attributed to differences between the Schilling and Bock method, which uses all information in the item responses, and the principal-component analysis, which uses the partial information contained in pairwise correlation coefficients. This study included 518 of the complete 567 items of the MMPI-2, versus Johnson et al.'s retention of 309 of the initially included 550 items of the previous MMPI. The full-information analysis retained all 518 initially included items and more evenly distributed the items over the 10 resulting factors, all sharply defined by their highest loading items and easy to interpret. Sampling effects and factor label considerations are discussed, along with recommendations for research that would validate the clinical utility of the implied scales for describing normal personality profiles. The full-information procedure provides for Bayes estimation of scores on these scales.     

Adolescent offenders and stages of change

Two hundred twenty-five adolescent offenders completed the Stages of Change Scales (SOCS; E. A. McConnaughy, J. O. Prochaska, & W. F. Velicer, 1983), which were designed to measure 4 stages of psychological and behavioral change. The goals of this research were to obtain norms among adolescent offenders for the SOCS, to examine the factor structure of the SOCS, and to investigate the convergent and discriminant validity of the SOCS by examining associations among these scales and the Paulhus Deception Scales (PDS; D. L. Paulhus, 1994, 1999) and Multidimensional Anger Inventory (MAI; J. M. Siegel, 1986). Means and alpha coefficients for the SOCS were similar to those reported in a previous normative clinical sample of adult patients. However, relatively large intercorrelations among the 4 scales of the SOCS, and results from factor analyses of SOCS items, suggested that fewer than 4 factors represented the SOCS items among our sample of adolescent offenders. Patterns of correlations with the PDS and with the MAI provided convergent and discriminant validity for the SOCS. Implications of the findings for research and clinical practice are discussed.     

A critical appraisal of the PRF ANDRO scale

A factor analysis of the PRF ANDRO reveals a structure reflecting the PRF scales from which the items came. It is argued that there is little value in combining a conglomeration of items from different Personality Research Form scales as has been done with the MASCUL and FEMIN scales of the PRF ANDRO. Rather, relevant intact PRF scales should be considered as separate predictors, for example in regression equations. Alternatively, if single scales are required to represent masculinity and femininity, it is suggested that PRF dominance and nurturance, the scales from which MASCUL and FEMIN draw a plurality of their items respectively, are more suitable than the two PRF ANDRO scales. These latter scales are less internally consistent and show less convergence with other sex-role scales than PRF dominance and nurturance.     

The Anxiety Depression Distress Inventory–27 (ADDI–27): New Evidence of Factor Structure,Item-Level Measurement Invariance,and Validity

Three studies examining the factor structure and psychometric properties of the Anxiety Depression Distress Inventory–27 (ADDI–27) extended the initial instrument development studies for this recently introduced inventory. The ADDI–27 is an empirically derived short form of the Mood and Anxiety Questionaire–90 (MASQ–90) comprising three scales: Positive Affect, Somatic Anxiety, and General Distress. The main objectives of Study 1 (N = 700) were to examine the factor structure of the ADDI–27 and its measurement invariance across gender at the item level. The objective of Study 2 (N = 538) was to examine evidence for the convergent and discriminant validity of scores on the ADDI–27. The objective of Study 3 (N = 240) was to assess further evidence for the nomological network and convergent and discriminant validity of the ADDI–27 scores. Results of exploratory structural equation modeling yielded strong support for a 3-factor model, with approximate fit indexes meeting or exceeding the conventional cutoffs. With p ≤ .001 as the criterion for detecting noninvariance, results of measurement invariance analysis suggested that all of the ADDI–27 items were invariant across gender. Results of multivariate validity analyses across 2 studies provided support for the convergent and discriminant validity of scores on the ADDI–27 scales.     

A critical appraisal of the PRF ANDRO scale

A factor analysis of the PRF ANDRO reveals a structure reflecting the PRF scales from which the items came. It is argued that there is little value in combining a conglomeration of items from different Personality. Research Form scales as has been done with the MASCUL and FEMIN scales of the PRF ANDRO. Rather, relevant intact PRF scales should be considered as separate predictors, for example in regression equations. Al ternati vely, if single scales are required to represent masculinity and femininity, it is suggested that PRF dominance and nurturance, the scales from which MASCUL and FEMIN draw a plurality of their items respectively, are more suitable than the two PRF ANDRO scales. These latter scales are less internally consistent and show less convergence with other sex-role scales than PRF dominance and nurturance.