Background In order to define a criterion to assign funds, the evaluation of the quality of the Italian uni-versity education is nowadays compulsory after the official publication of the law 537. Materials and Methods A questionnaire including 32 items was submitted to students during the semester 1997-1998 and a total of 1.118 questionnaires have was collected. By applying multivariate analysis using HOMALS (Homogeneity Analysis) tecnique, the questionnaire was then modified. Two scoring systems, a 'crude' one (Index A) and an 'optimal' one (Index B) were calculated and compared through simple and weighted regression analysis. Re-sults A Cronbach's alpha coefficient as high as 0.90 was found, A brief version of 20 items was selected. The comparison between Index A and B showed that the former was not a good predictor of the latter. Conclusions Both indexes have advantages and disadvantages, however the Index B is to be preferred because based on 'optimal values' starting from the item frequencies.
Premessa La legge 537/1993, Art 5, ha reso obbligatoria la valutazione dell'attività di ricerca e didattica ai fini della ripartizioni dei fondi destinati all'Università. Materiali e Metodi Un questionario originario com-posto da 32 items è stato sottoposto al giudizio degli studenti. Tra Novembre 1997 e Marzo 1998 sono stati compilati 1.118 questionari di tutti gli anni del Corso di Laurea. È stata calcolata l'omogeneità dello strumento e poi modificata la struttura originaria. Si è applicata una metodologia multivariata HOMALS (Homogeneity Analysis). Due distinti scores, denominati indice B 'ottimale' ed indice A 'grezzo', sono stati proposti e confron-tati mediante modelli di regressione semplice e ponderata. Risultati Lo strumento ha mostrato una buona omogeneità (? di Cronbach > 0.90). La versione ridotta, più maneggevole, contiene 20 items. Il confronto fra i due scores ha mostrato che l'indice A non risulta un buon predittore dell'indice B. Conclusioni L'indice B è da preferirsi perché costruito a posteriori in base a valori ottimali riferiti alle frequenze di risposta degli items.
Validazione e Scoring di un Questionario per la Valutazione della Qualità Didattica
Biino G;
1999
Abstract
Background In order to define a criterion to assign funds, the evaluation of the quality of the Italian uni-versity education is nowadays compulsory after the official publication of the law 537. Materials and Methods A questionnaire including 32 items was submitted to students during the semester 1997-1998 and a total of 1.118 questionnaires have was collected. By applying multivariate analysis using HOMALS (Homogeneity Analysis) tecnique, the questionnaire was then modified. Two scoring systems, a 'crude' one (Index A) and an 'optimal' one (Index B) were calculated and compared through simple and weighted regression analysis. Re-sults A Cronbach's alpha coefficient as high as 0.90 was found, A brief version of 20 items was selected. The comparison between Index A and B showed that the former was not a good predictor of the latter. Conclusions Both indexes have advantages and disadvantages, however the Index B is to be preferred because based on 'optimal values' starting from the item frequencies.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
