The handling of ordinal variables presents many difficulties in both the measurements phase and the statistical data analysis. Many efforts have been made to overcome them. An alternative approach to traditional methods used to process ordinal data has been developed over the last two decades. It is based on a fuzzy inference system and is presented, here, applied to the student evaluations of teaching data collected via Internet in Modena, during the academic year 2009/10, by a questionnaire containing items with a four-point Likert scale. The scores emerging from the proposed fuzzy inference system proved to be approximately comparable to scores obtained through the practical, but questionable, procedure based on the average of the item value labels. The fuzzification using a number of membership functions smaller than the number of modalities of input variables yielded outputs that were closer to the average of the item value labels. The Center-of-Area defuzzification method showed good performances and lower dispersion around the mean of the value labels.
The ordinal controversy and the fuzzy inference system through an application and simulation to teaching activity evaluation / Lalla, Michele; Pirotti, Tommaso. - STAMPA. - 620:(2016), pp. 189-211. [10.1007/978-3-319-26393-9_12]
The ordinal controversy and the fuzzy inference system through an application and simulation to teaching activity evaluation
LALLA, Michele;PIROTTI, Tommaso
2016
Abstract
The handling of ordinal variables presents many difficulties in both the measurements phase and the statistical data analysis. Many efforts have been made to overcome them. An alternative approach to traditional methods used to process ordinal data has been developed over the last two decades. It is based on a fuzzy inference system and is presented, here, applied to the student evaluations of teaching data collected via Internet in Modena, during the academic year 2009/10, by a questionnaire containing items with a four-point Likert scale. The scores emerging from the proposed fuzzy inference system proved to be approximately comparable to scores obtained through the practical, but questionable, procedure based on the average of the item value labels. The fuzzification using a number of membership functions smaller than the number of modalities of input variables yielded outputs that were closer to the average of the item value labels. The Center-of-Area defuzzification method showed good performances and lower dispersion around the mean of the value labels.File | Dimensione | Formato | |
---|---|---|---|
Lalla-Pirotti Springer.pdf
Accesso riservato
Tipologia:
Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione
2.05 MB
Formato
Adobe PDF
|
2.05 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
lalla2016.pdf
Accesso riservato
Tipologia:
Versione pubblicata dall'editore
Dimensione
764.14 kB
Formato
Adobe PDF
|
764.14 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris