Le Roy A. Stone

The MEDEX medical occupation concept in subjective medical occupation-profession multidimensional space.

Members of a MEDEX training staff (N = 13) estimated the similarity of 16 medical occupations-professions (including the MEDEX concept). Evaluative dimensions utilized by these judges to form their judgments of scale similarity were extracted using a recently proposed multidimensional scaling method-model. Five judgmental factor-dimensions were identified (tentative labels: I. Amount of training—degree of responsibility—prestige; II. Therapist; III. Nurse; IV. Technician; and V. Hospital nursing staff authority) which accounted for up to 86% of the variance of the averaged similarity estimates. These factor-dimensions as well as the methodology employed were discussed. Future research was suggested.

Multidimensional Scaling of Persons in Groups: Progress Report

The use of multidimensional scaling applied to persons, in groups, as stimuli was historically traced. 18 investigations which employed the Stone-Coles paradigm for multidimensional scaling of judgments pertaining to such kinds of stimuli were described. The apparent initial success of this type of methodology in the sphere of clinical-social psychological research appears to justify further use.

Instructors' Multidimensional Perceptions of Their Students.

Two instructors of a single psychology course made similarity estimations of students who were enrolled in this course. This judgmental data was multidimensionally scaled using the Stone-Coles paradigm. Three judgmental dimensions were extracted: academic achievement, sex membership, and liberal-conservative attitudes and behaviors. A suggestion was made that multidimensional scaling may represent an approach to grading if grading is to be based solely on observations-evaluations of teachers.

CONGRUENT MULTIDIMENSIONAL SCALING RESULTS OBTAINED USING THE HALO-MODEL AND THE STONE-COLES METHOD-MODEL

Comparability of results from the Micko and Stone-Coles approaches to judgmental data is discussed.

TRAINING STAFF'S MULTIDIMENSIONAL PERCEPTIONS OF A CLASS OF MEDEX (PHYSICIAN'S EXTENSION) TRAINEES: A METHOD OF GRADING

A recently proposed multidimensional similarity judgment analysis methodology (the Stone-Coles method) was used to analyze the dimensionality of a class of Medex (physicians extension) trainees as perceived by the staff of the involved Medex training project. The extracted dimensions associated with the “early-in-program” and “about-to-be-graduated” perceptions accounted for 76 and 79%, respectively, of the judgmental variances; all dimensions appeared to be readily interpretable. The major dimension for the last obtained judgments clearly seemed to be some kind of “quality of trainee” characterization. This particular evaluation judgment dimension was accepted by the involved training staff, the judges, as a satisfactory “grading” measure for the trainees. A conclusion reached is that multidimensional scaling can be used, with advantage, in the academic grading of students when the basis for such grading is limited to evaluative opinion of others (e.g., teachers).

Another Law of Clinical Judgment: A Psychological Mechanism Based on the Logic of Psycho-Dynamics:

Following a “psycho-dynamics” paradigm, two psychiatric direct-estimation scales (each pertaining to classification names for psychotic disorders as the stimuli and each based on judgments from two independent groups of psychiatrists, Ns = 29 and 33) were shown to be systematically interrelated. A power function was preferred to show the relation. The form of the relationship between judged susceptibility to external stress and judged prognostic favorability was discussed. Suggestions for further research were made.

Corrected (for curtailment) correlations between the Marlowe-Crowne social desirability scale and the MMPI.

It has been observed that correlations between the Marlowe-Crowne Social Desirability Scale (M-C SDS) and the standard MMPI scales were higher for a psychiatric patient sample than for college students. The suggestion had been made that the correlational differences with these two types of samples were due to the more curtailed distributions of MMPI scores with college students. These previously reported correlations, based on college students, were corrected for curtailment regarding the MMPI variables. The corrections utilized variability estimates based on psychiatric patient samples. The corrected (for curtailment) values were noticeably larger and resembled in magnitude those reported for psychiatric patients.

Reliability of a Utility for Risk Measure

Since the publication of Von Neurnann and Morgensterns book (1944), risk-taking has received intensified attention as a factor in psychological behavior. Recently, Ziller (1957) has developed a technique for estimating utility for risk which supposedly measures individual decision-making response under conditions of uncertainty and risk. Zillers Test Risk score is obtained by the following formula developed by him and Roby ( 1957 ) : R = 2\W/ ( 2 W + 0 ) , where R = Risk, W = Number of wrong responses, and 0 = Number of items omitted. The measure of utility for risk may then be described as the ratio of the number of items on which Ss hazard guesses to the number of items not known by Ss. However, the reliability of this type of measure has not been established. A 15-item quiz,l designed on the Ziller paradigm for an introductory psychology class at the University of North Dakota was administered to 279 students. Their scores from the odd-numbered items of the quiz were used in the Ziller formula in order to obtain Test Risk scores. The same procedure was followed, using information from the even-numbered items. A split-half reliability coefficient was computed, based on these two sets of Test Risk scores. The obtained split-half coefficient was .77, when corrected by the Brown-Spearman formula was 3 7 . The split-half technique generally is believed to result in spurious estimates of the reliability of a measure. However, the results suggest that chis parcicular measure possessed acceptable split-half reliability.

Another look at reanalysis of the multidimensional scaling of Rorschach percepts.

Reanalysis of Ekehammars judgmental data from Rorschach cards produced two judgmental dimensions.

Perceived dimensions for combinations of drug substances.

8 experienced users of “street-drugs” estimated the over-all similarity of 11 drug-substance combinations based on their own experiences with such combinations. Their similarity estimations were analyzed using multidimensional scaling procedures (namely, the Stone-Coles method). Four judgmental factor-dimensions were extracted and identified (tentative labels: I. Psychedelic drugs in combination with other substances vs pot in combination with downers; II. stimulant-depressant; III. psychedelic drugs vs other drugs, IV. LSD in combination with other substances) which accounted for 92% of the variance of the averaged similarity estimates. These particular judgmental factor-dimensions were similar to those extracted from judgments pertaining to single drugs as stimuli.