The concurrent validity of three self-report measures of borderline personality

The recent accumulation of self-report measures of borderline personality disorder (BPD) affords the opportunity to evaluate both the construct validity of the concept and the quality of these measures. This study examines the relationship among three recently developed self-report instruments for assessing BPD from the Personality Assessment Inventory (PAI; Morey, 1991), the MMPI Personality Disorders Scales (MPD; Morey, Waugh, & Blashfield, 1985), and the Bell Object Relations Inventory (BORI: Bell, Billington, & Becker, 1986). Data on the three measures were provided by 119 undergraduate subjects from a southeastern university. A correlational analysis addresses the convergence of these measures of BPD, their divergence from measures of different but related traits, and their independence from variance due to method. Application of the Campbell-Fiske (1959) criteria indicates adequate convergence for all the BPD measures but a lack of discriminant validity for the BORI scales. The fit of the data to a structural model of construct validity is tested using confirmatory factor analysis, and these results are consistent with the hypothesis of a latent borderline trait factor independent of measurement method factors. In sum, the construct validity of the borderline personality concept using self-report methodologies receives support, and a strong association between borderline personality and paranoid phenomena is also suggested.     

Effects of vowel variety and sample lenght on identification of a speaker in a line-up

Subjects (average age 21 years, recruited by personal contact and through a school) were presented with a spoken sentence on tape and then heard six speakers of the same sex, including the original speaker, say the same sentence. They were required to indicate which was the original speaker. The task was repeated with seven different sentences and sets of speakers. One group of subjects heard short sentences containing an average of 2.14 different vowel sounds and 6.28 syllables, another group heard short sentences containing an average of 6.14 vowel sounds (7.28 syllables) and a third group heard longer sentences containing an average of 6.28 vowel sounds (11.00 syllables). Accuracy of speaker identification improved significantly when more vowel sounds were heard, but increased sentence length had no significant effect on performance. Performance was significantly better when the listener was the same sex as the speaker than when the listener was of the other sex.     

Interrelationships among students' study activities,self-concept of academic ability,and achievement as a function of characteristics of high-school biology courses

This investigation focused on the interrelationships among students' study activities, students' self-concept of academic ability ratings, students' academic achievement, and instructional practices in 12 high school biology courses. Using a framework derived from a previous investigation, course features were classified into those that appear to (a) place demands on, (b) support, or (c) compensate for student engagement in particular study activities. Students' study activities, self-concept of academic ability ratings, and achievement were measured with experimenter-developed instruments. Results are reported for (a) characteristics of instruments and course features, (b) relationships between central factors of the investigation, and (c) multi-level relationships between course features and student variables. Results at the student level indicated that self-concept of academic ability and, to a lesser extent, students' study activities were positively associated with student achievement. Students' self-concept of academic ability ratings were also linked to students' engagement in generative, proactive study activities. At the course level the supportive practices of providing challenging homework assignments and extensive feedback on student coursework were associated with student engagement in effortful, generative, proactive study activities. The provision of extensive feedback was also associated with high student achievement. Multi-level relationships were analysed using hierarchical linear modelling (HLM) analyses. These analyses revealed, for example, that in courses in which little or no feedback is given on homework assignments, the relationship between achievement and student engagement in diligent effort management activities was enhanced. Other HLM analyses were conducted to examine the mediating role of course features on the relationship between students' self-concept of academic ability and their study activities and achievement. For example, the presence of challenging course demands was associated with an enhancement of the relationship between self-concept of academic ability and achievement whereas the presence of instructor provisions (supports and compensations) designed to reduce course demands was associated with a reduction in this relationship.     

Relationship between trust in appraisal and appraisal effectiveness: A field study

The current study examined the influence of trust in the appraisal system on ratees' reactions to performance evaluations. One-hundred and sixty eight bank tellers completed an instrument that assessed their trust in the appraisal system, appraisal satisfaction, perceived utility of evaluations, behavioral intention to improve performance, and intention to leave the job. Trust and performance evaluations were significant predictors of ratee reactions to appraisals and intention to improve on performance weaknesses. In addition, performance ratings were a better predictor of intention to improve on performance weaknesses for ratees who had high trust in the appraisal system. Implications for future research and practice are discussed.An earlier version of this paper was presented at the Annual Meeting of the Society of Industrial and Organizational Psychology, Miami, Florida, April, 1990.     

Safety locus of control and employees' accidents

This study further established the criterion-related validity of the Safety Locus of Control Scale. The safety scale was designed to predict employment applicants' propensity for on-the-job accidents. In this study, 283 hospital employees with moreexternal safety locus of control orientations reported significantly more occupational accidents, as well as more severe and costly injuries, than workers with moreinternal safety attitudes (p<.05 in all cases). Implications of these results are discussed.Research study presented at the 1988 Annual Conference of the Association of Human Resources Management and Organizational Behavior, Long Beach, California, October 26–29.     

Validity of the DSM-III-R personality disorder Clusters

This study assessed the validity of the DSM-III-R personality disorder clusters (i.e., odd-eccentric, dramatic-emotional-erratic, and anxious-fearful) by examining the relationships between self-report measures that tap the core features shared by disorders from each cluster and Cluster scores established via a semistructured interview in a sample of 57 outpatients. Results indicated a high degree of correlation among the DSM-III-R personality disorder Cluster scores. In addition, a series of regression analyses revealed that self-report scores did not account for a significant amount of variance in their respective Cluster scores over and above that accounted for by other self-report measures and other Cluster scores. These results suggest that the current DSM-III-R cluster classification scheme may not be appropriate, and it is recommended that a more empirically justifiable classification of the personality disorders be adopted in DSM-IV.     

A comparison of Guided Imaginal Coping and imaginal exposure in the treatment of panic disorder

The present study evaluated the efficacy of a new intervention, Guided Imaginal Coping (GIC), compared to imaginal exposure (IEX), and a waiting-list control (WLC) condition among 24 clinic outpatients meeting DSM-III-R criteria for panic disorder (PD) or PD with agoraphobia. GIC was the more consistently effective intervention, with individuals demonstrating significant improvement in frequency and severity of panic attacks (Pas) as well as significant reductions in physiological and subjective arousal to provocative imagery. Individuals in the IEX intervention demonstrated less consistent improvement on the dependent measures, while individuals in the WLC condition essentially remained unchanged. Both GIC and IEX were effective according to a composite measure of clinically significant improvement. However, only GIC was superior to the WLC condition when clinical improvement was defined as zero PAs at follow-up. These findings, while preliminary, suggest that GIC can be added to the armamentarium of effective approaches to treating PD.