A Latent Class Analysis Approach to Challenges Experienced by Faculty Members in Online Assessment in Higher Education

Abstract

Online assessment is the use of computer technologies by faculty members to guide and check learning. Taking the advantage of technology, many universities have used online assessment applications to ensure sustainability in education due to the pandemic and natural disasters. The purpose of the current study is to explore challenges experienced by faculty members in online assessment, using latent class analysis. The descriptive design research was carried out with the participation of 105 faculty members. For the study, the number of latent classes was decided according to the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) and it was observed that the data structure was a good fit for a two-class model. According to the research results, the first class in online assessment applications was considered as the with-difficulty group (58.7 %) and the second as the without-difficulty group (41.3 %). When the conditional probabilities were examined, it was concluded that the observed variables that mostly contributed to the two-class model data structure were as follows, cheating, plagiarism and lack of education policies. It was found that the primary challenges in both groups (with or without difficulty) in online assessment applications were cheating, plagiarism and lack of education policies.