Jerry W. O'Dell

The Validity and Use of the Family-of-Origin Scale.

This study explores the Family-of-Origin Scale (FOS) with regard to its validity and potential use. Subjects included 100 adults currently in psychotherapy, 100 who were not, 32 who previously had sought treatment, and 41 college students. Between-group comparisons, correlations with both the 16PF and a scale of authoritarianism, factor analyses of both the patient and nonpatient FOS responses, and a review of patient feedback revealed several significant results. The FOS differentiates between patients and nonpatients, but it measures only one general factor and delivers little information. Therefore, it has little value as a research instrument, but it may be a useful adjunct to psychotherapy in individual cases.

Eliza as a "Therapeutic" Tool.

Allowed 70 undergraduate students to interact for 44 minutes with ELIZA--a computer simulation of psychotherapy. Their responses during the first and second halves of the session were compared, and several differences were found. Removing the effect of the simple volume of responses (using the analysis of covariance) did not change the significance of many of the differences.

Is the 16PF Factorially Valid

Summary The factor structure of the items of the 1962 16PF Form A was examined on a sample of 499 air traffic controllers. It was found that the test does indeed measure the primary factors it purports to measure although several of the factors found did not correspond well with the ideal pattern to be expected with the 16PF on this special, highly selected sample of American males. The results were not as discouraging to the 16PF user as would be indicated by other recent studies which shared certain methodological shortcomings.

Rogers, Lazarus, and Shostrom in Content Analysis.

Subjected Shostroms films of Rogers, Lazarus, and Shostrom working with the patient, Cathy to a content analysis using the system developed by Zimmer and Cowles. Differences in verbal activity on the part of both the therapist and patient were noted, which seem consistent with the principles of the various therapies.

Performance ratings and personality factors in radar controllers

Abstract : The purpose of the study was to determine whether primary or second-order personality questionnaire factors were related to job performance ratings on the Employee Appraisal Record in a sample of 264 radar controllers. A Pearson correlation matrix was computed based on 19 variables which included EAR part II and part IV, a motivational distortion score and the primary factors measured by the Sixteen Personality Factor Questionnaire. A principal axis factor analysis was completed with varimax rotation which yielded eight second-order factors which were identified as follows: I. Anxiety-versus-Dynamic Integration; II. Subduedness-versus-Independence; III. Criterion; IV. Cortertia-versus-Pathemia; V. Exvia-versus-Invia; VI. Intelligence; VII. Obsessive-Compulsive; VIII. Rebelliousness. While all of these second-order factors have been identified previously, the criterion variables were not related to any of the personality measures. (Author)

SPECTRAL ANALYSIS OF INTERVIEWS CONDUCTED BY ROGERS AND ELLIS

Recordings of famous therapeutic interviews were digitized by computer, then subjected to time-series analysis. Spectrograms produced by the same therapist seeing different patients were similar, while spectrograms of different therapists were rather divergent. Further analyses were carried out on the digitized data. The techniques used offer promise as an objective way of quantifying therapeutic interview data.

Reassessment of factor solution of Halstead's Matrix.

Halsteads correlational matrix of 13 neuropsychological tests was reanalyzed with the objective of obtaining a more neuropsychologically meaningful solution. The emphasis was on solving for different numbers of factors (three and five). The three-factor solution supported and perhaps refined two of Halsteads original factors (P and A). The third factor loadings resulted in a considerably different concept than Halsteads D. The three factors (psychological vigilance, abstraction, and perception and memory for incidental stimuli) may be an essential factor pattern in recovery from head traumas. It was concluded that the investigator should not a priori be committed to a structure involving a particular number of factors. Experimentation with factor number or method would allow evaluation of the best psychological fit to the data.

Evaluation of a content-analysis system for therapeutic interviews

The Zimmer and Cowles system for content analysis was tried out on psychotherapeutic sessions and on interviews with disturbed patients. It was found that the system indeed did seem to pick up improvement in psychotherapy, and to a lesser degree, differences in speech behavior between patients with radically differing pathologies. It also was concluded that the computer program for the content analysis system is not easily portable, but a SNOBOL equivalent is made available.

The Epsat as a Predictor

Summary Two independent studies were done, both indicating that the Engineering and Physical Science Aptitude Test (EPSAT) and the Graduate Record Examination (GRE) offer promise in prediction of performance in graduate statistics courses.

Letters from Jenny revisited: The computer content analysis redone

The text of Allports Letters from Jenny was reanalyzed by computer. The analysis was accomplished by splitting the letters at Rosss death and through the use of significance tests. Surprising differences were found between Paiges (1964) results and those of the current study: We found substantially fewer words in the text than he did, and our count of number of letters written was higher. It was found that there were significantly more references to the topics of Positives, Art, Paranoid matters, Death, and Illness after Rosss death. References to Ross, Children, and Money decreased. Certain statistical and methodological problems in analysis of data of this type were commented upon.