Introduction: One of the difficulties associated with administering the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is that it requires dichotomous responses (T/F; 0/1). The authors present this pilot study as a means of confronting this problem by proposing a new version of the test: MMPI-2-FUZZY, which allows subjects being tested to respond in a nuanced manner while following the principles of fuzzy logic. Materials and Methods: Fuzzy logic is based on assigning a degree of truth to a particular variable ranging between 0 and 1 (In our case items on the test). Computerized administration of the MMPI-2 was converted in such a way so as to allow the subject to express a level of agreement (or disagreement) for each test item on the MMPI-2 by moving a cursor along a continuous line that contains values from 100% true to 100% false. This new version of the test was preliminarily administered on a sample of 26 healthy male subjects. As an external validity criterion, the diagnostic accuracy of the MMPI-2-Fuzzy was then compared with that of the classic MMPI-2 (bivalent) by using the diagnoses of 20 male subjects with psychotic disorders upon discharge from the Psychiatric Clinic at the University of Bari (Italy), each of whom was administered the computer-based MMPI-2-Fuzzy. Results: Notwithstanding the preliminary nature of the study and the limitations associated with the small sample size, it is nonetheless possible to affirm a higher validity of the MMPI-2-Fuzzy and the two modes of administration and scoring by utilizing ROC curve (fuzzy 0 0.82; bivalent = 0.80), and the study of sensitivity, specificity and global diagnostic accuracy (fuzzy = 0.71; bivalent = 0.60) when discriminating between those who are healthy and those who are ill. Conclusions: Such promising results may indicate the need for further study regarding alternate versions of administering the MMPI-2 in both clinical and legal-medical settings.
Il test MMPI-2 FUZZY. Incremento dell'accuratezza diagnostica del MMPI-2 per mezzo di un innovativo metodo di somministrazione. Dati preliminari di una ricerca
GRATTAGLIANO, IGNAZIO
;CATANESI, Roberto
2013-01-01
Abstract
Introduction: One of the difficulties associated with administering the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is that it requires dichotomous responses (T/F; 0/1). The authors present this pilot study as a means of confronting this problem by proposing a new version of the test: MMPI-2-FUZZY, which allows subjects being tested to respond in a nuanced manner while following the principles of fuzzy logic. Materials and Methods: Fuzzy logic is based on assigning a degree of truth to a particular variable ranging between 0 and 1 (In our case items on the test). Computerized administration of the MMPI-2 was converted in such a way so as to allow the subject to express a level of agreement (or disagreement) for each test item on the MMPI-2 by moving a cursor along a continuous line that contains values from 100% true to 100% false. This new version of the test was preliminarily administered on a sample of 26 healthy male subjects. As an external validity criterion, the diagnostic accuracy of the MMPI-2-Fuzzy was then compared with that of the classic MMPI-2 (bivalent) by using the diagnoses of 20 male subjects with psychotic disorders upon discharge from the Psychiatric Clinic at the University of Bari (Italy), each of whom was administered the computer-based MMPI-2-Fuzzy. Results: Notwithstanding the preliminary nature of the study and the limitations associated with the small sample size, it is nonetheless possible to affirm a higher validity of the MMPI-2-Fuzzy and the two modes of administration and scoring by utilizing ROC curve (fuzzy 0 0.82; bivalent = 0.80), and the study of sensitivity, specificity and global diagnostic accuracy (fuzzy = 0.71; bivalent = 0.60) when discriminating between those who are healthy and those who are ill. Conclusions: Such promising results may indicate the need for further study regarding alternate versions of administering the MMPI-2 in both clinical and legal-medical settings.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
