The aim of this article is to analyze the psychometric properties of an abbreviated version of the Teaching Problems Inventory. For this purpose, the scale is subjected to a previous exploratory factor analysis to statistically examine the underlying constructs. Previously, a confirmatory factor analysis was performed and a new grouping into dimensions was generated. To analyze the properties of the items, descriptive statistics were performed: mean, standard deviation and inter-item correlations. To study the psychometric properties of the scale, Cronbach's alpha, Exploratory Factor Analysis (EFA) -to identify the latent structure of the variables-, and Confirmatory Factor Analysis (CFA) were used as a procedure to check the results of the EFA. As recommended, participants were randomly divided into two groups to perform the AFE with one group and the AFC with the other to avoid distortions Participant: the proportion of items, sample size, normality, linearity, and correlation between variables were checked as assumptions for the factor analysis. The sample consisted of 992 undergraduate teacher training and master's degree students from different Spanish regions: Valencia, Andalusia, Catalonia, Madrid, Castilla la Mancha, and Castilla y León. The results of the different reliability and validity analyses showed that the reduced scale has higher reliability values and is still sensitive to capture gender and age scoring difference. As a conclusion, the consistency of the instrument to evaluate the teaching difficulties of students of different degrees and postgraduate degrees in education is highlighted. As for the limitations, the elimination of 2 items relevant at a theoretical level stands out: Motivating students in school tasks and Discipline problems with students/groups of students.
Read full abstract