Implementing Analysis of Ordinal Regression Model on Student’s Feedback Response

Abstract

Instruction is a multidimensional procedure including a quantity of features, e.g., tutor qualities, that occasionally are hard to assess. In certain points, education efficiency, that is a part of instructing, is affected by a combination of teacher features for example, capacity and clarity to encourage the students to make them study of his subjects, capacity to establish the lesson also with trainings and lectures. These aspects are not only attributable to motivate students, but also age, gender, prior experiences, As more and more the effectiveness of teaching is becoming even more significant in school evaluation system, it is, indeed, essential to discover how to assess it and find the significant related factors for giving the best rank of feedback to their tutors. This paper focuses the assessment of teaching effectiveness forum from students’ perspective “Feedback”, examine the questionnaires provided to the students of Cihan University at the end of their courses and discover the most effected teacher’s characteristics. The outcome variable (Student’s Feedback) was recorded on well-ordered, five-point scale of Likert provided by the students, to a set of independent variables to teacher level. The major techniques elaborated in the model fitting for ordinal regression were stating which independent variables are most likely to be kept in the model and selecting the link function such as, (logit link, complementary log-log link, negative log-log link and probit link) which verified the model suitability. Further to that, several statistical diagnoses have conducted like the model fitting, classification accuracy and the validity assumptions of the model, which is parallel lines, were fundamentally calculated in choosing the best fitted model. The dataset implemented in the analysis entails of almost (21566) respondents to the formed questionnaire in relation to courses of Cihan University for the 2018 – 2019 academic year.

An application of ordinal regression to extract social dysfunction levels through behavioral problems

“…Norusis (2012) [21] suggests the choice link function should be based on the distribution of the response variable. In this study, we have used a negative log-log link function [22] .…”

Compliance with COVID-19 preventive measures among chronic disease patients in Wolaita and Dawuro zones, Southern Ethiopia: A proportional odds model

“…Since our outcome variable is measured at ordinal level, we have chosen an ordinal regression model to identify the independent predictors of compliance level. However, evidences have suggested and used additional assumptions that need to be fulfilled before running the model, so as to have a valid result [39][40][41][42][43]. Accordingly, independent variables should only be treated as either categorical or continuous variable which was done in our study.…”

The Use of Factor Analysis and Cluster Analysis Methods to Identify the Most Crucial Key Factors Influencing the Psychological Stability of University Students

“…Even though there were much discussion and research on how to improve the university education system and determine the attitudes of teachers and students toward the factors that significantly impact this issue, concentrating more attention on students' psychological issues, which may be caused by the atmosphere of the university or any other condition that may affect achievement in the long run, is crucial if we would like to improve the education system's success and teaching methods in universities (Mahmood et al, 2018;Kadir and Omer, 2021;Mahmood et al, 2022).…”