Validity of Stimulus and Response Components in a Video Test of Social Competence

Various stimulus components (video, orally-presented questions) and response components (multiple-choice, written, orally-given replies) of situational judgement tests of occupational social competency were investigated as to their impact upon the validity for a behavior-oriented role playing criterion while keeping test content constant. The stimulus component video alone had no impact upon validity. The response components contributed to validity. Validity increased with improved fidelity of response components. Concerning stimulus-response-combinations, the validity of two video tests (r = 0.17 and r = 0.36) was not higher than the validity of similar oral questioning (r = 0.13 and r = 0.37) but was significantly lower than a situational interview (r = 0.59). Response fidelity proved to be a bottleneck regarding validity of video tests. As a result, it is recommended that, in order to maximize validity of video and multimedia efforts, the developers of video- and multimedia tests focus special attention on response fidelity.     

A multivariate investigation of the reinforcer structure of occupations

Earlier research based on 81 Occupational Reinforcer Patterns (ORPs) suggested a nine-category classification of occupations. The present research, based on 148 ORPs, investigated the ability of that classificatory system to assimilate new information. Cluster analysis of the 148 ORPs yielded an eight-category classification of occupations in which five of the original occupational clusters appeared virtually unchanged; the other four original clusters were combined into two occupational clusters. One new cluster was identified. Principal factors analysis was employed to study the factor structure of the occupational reinforcers and to highlight the essential nature of the classificatory system. Five reinforcer factors were identified.     

Quantifying private events: A functional measurement analysis of equisection

Functional measurement theory was applied to bisection, trisection, and quadrisection of grayness. Theoretically, these judgments should obey an averaging model. But the overt responses are not a valid measure of subjective magnitude (since they are made on the physical stimulus continuum), and so they cannot be used directly to test the model. However, scaling and model testing can both be accomplished simultaneously using functional measurement theory. If the subject is indeed averaging, then there exists a monotone transformation that makes the data additive; and this transformation can be computed with FUNPOT, Weiss’s (1973a) computer program which finds polynomial transformations that reduce selected effects. Further, determination of this transformation also reveals the psychophysical function, because it gives the relation between subjective magnitude and overt response. For bisection, the averaging model was successful; it was possible to find a monotone transformation that made the data additive. This psychophysical function differed somewhat in form from the Munsell scale. It gained cross-task validity from its agreement with a grayness scale obtained from rating data (Weiss, 1972). For trisection and quadrisection, the averaging model was not accepted; it was not possible consistently to find transformations which induced additivity.     

Supervisor estimation of abilities required in jobs

The Minnesota Job Requirements Questionnaire (MJRQ) was developed and administered to supervisors of 11 selected jobs who rated the ability requirements of the job they supervised. The MJRQ represented each of nine GATB aptitudes by five items in a Likert-type format. MJRQ ratings were compared with expert ratings in the Dictionary of Occupational Titles (DOT) and Occupational Ability Patterns (OAPs) derived from administration of the General Aptitude Test Battery (GATB). Reliability of MJRQ ratings was shown by high Hoyt reliability coefficients. Comparison of mean ratings and variabilities of ratings by supervisors in a job showed construct validity for the MJRQ approach, as did intercorrelations of MJRQ ratings. OAPs derived from the MJRQ compared favorably with those derived from the GATB and DOT, showing promise for the parsimonious MJRQ approach to the development of OAPs.     

Reflections on a Community Psychology Setting and the Future of the Field

The 50th anniversary of the Swampscott Conference offers an opportunity to reflect on a community psychology setting, The Consultation Center at Yale, that was formed in response to the 1963 Community Mental Health Act and the 1965 Swampscott Conference. The Center has flourished as a community psychology setting for practice, research, and training for 39 of the 50 years since Swampscott. Its creation and existence over this period offers an opportunity for reflection on the types of settings needed to sustain the field into the future.     

Pathologisches Glücksspiel und Delinquenz

Pathological gambling is characterized by excessive gambling behavior which influences the life and well-being of a person in a dysfunctional way. It often leads to severe consequences in social and family life, career and material matters. This can also include illegal acts. In the Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV) excessive gambling was classified under impulse-control disorders but was renamed as gambling disorder (GD) and reclassified under the category “substance-related and addictive disorders” in DSM-5, due to parallels to substance use disorders in the clinical characteristics as well as genetic and neurobiological factors. In Germany, approximately 0.5?% of the adult population suffer from GD. Especially gambling machines are associated with an increased risk of gambling-related problems. Common comorbidities are substance-related and affective disorders. The most important risk factors are young age, male gender, a history of migration, gambling-related problems in the family, unemployment, a low educational level and low household income. Delinquent behavior was not considered to be a sufficiently discriminative factor for GD, although there is a correlation between delinquency and gambling behavior. In addition, those GD subjects who show delinquent behavior also show a specific risk profile. Thus, mental health professionals should be aware of possible psychosocial consequences in GD, such as illegal acts, and this should be addressed in the effective treatment strategies, such as cognitive behavioral therapy. Moreover, it is necessary to improve the access routes for affected people into healthcare system in order to ensure early detection and treatment of the disease.     

Optimal designs for the generalized partial credit model

Analysing ordinal data is becoming increasingly important in psychology, especially in the context of item response theory. The generalized partial credit model (GPCM) is probably the most widely used ordinal model and has found application in many large-scale educational assessment studies such as PISA. In the present paper, optimal test designs are investigated for estimating persons’ abilities with the GPCM for calibrated tests when item parameters are known from previous studies. We find that local optimality may be achieved by assigning non-zero probability only to the first and last categories independently of a person's ability. That is, when using such a design, the GPCM reduces to the dichotomous two-parameter logistic (2PL) model. Since locally optimal designs require the true ability to be known, we consider alternative Bayesian design criteria using weight distributions over the ability parameter space. For symmetric weight distributions, we derive necessary conditions for the optimal one-point design of two response categories to be Bayes optimal. Furthermore, we discuss examples of common symmetric weight distributions and investigate under what circumstances the necessary conditions are also sufficient. Since the 2PL model is a special case of the GPCM, all of these results hold for the 2PL model as well.