Naslov
Evaluation of factor analysis methods: Student feedback case study

Autori
Novak, Matija ; Šlibar, Barbara ; Kermek, Dragutin ; Begičević, Nina

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
INTED2020 Proceedings / - Valencia : International Academy of Technology, Education and Development (IATED), 2020, 8208-8216

ISBN
978-84-09-17939-8

Skup
14th International Technology, Education and Development Conference (INTED 2020)

Mjesto i datum
Valencia, Španjolska, 02.03.2020. - 04.03.2020

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
course evaluation, efa, principal axis factoring, principal component analysis, promax rotation
(course evaluation, efa, principal axis factoring, principal component analysis, promax rotation.)

Sažetak
The questionnaire for acquiring student feedback about course satisfaction which have been used in a graduate course for more than ten years is analysed within this study. The questionnaire contains different type of questions, but only multiple choice questions are analysed. Questions were analysed using Exploratory Factor Analysis (EFA) in order to find the core factors which are sufficient for assessment of overall student satisfaction with the course. The process of choosing multivariate analysis technique was automated using the R system. The main motivation to conduct this research is reduction of questions in the questionnaire to those that are the most important. Before we can start analysing one must decide which EFA method to use. Therefore, the research is directed to the comparison of different EFA methods in combination of different types of rotations for finding the most suitable EFA method for analysing student questionnaires. According to the results of the analysis, all methods (i.e. PCA, PAF, ML) perform better in combination with Promax rotation. Although, the Common Factor Analysis methods (i.e. PAF, ML) group the variables better into the factors, the Component Model method (i.e. PCA) explains more of the cumulative variance.

Izvorni jezik
Engleski

Znanstvena područja
Kliničke medicinske znanosti, Pedagogija

POVEZANOST RADA