Psychometric Analysis of an Instrument Evaluating Online Efl Teaching Presence

Abstract

This research intended to examine the psychometric characteristics of the Indonesian version of the instrument used in evaluating the performance of pre-service teachers when practicing online teaching which was adapted from Howard. This quantitative study used convenience sampling technique by conducting a survey to collect respondent data. A total of 13 questionnaire items were developed using a 5-point Likert scale ranging from Strongly Disagree (1) to Strongly Agree (5). Data were collected from 88 students who were carrying out online teaching practices or who had finished carrying out these activities. Several statistical analyzes were performed using Rasch analysis to assess unidimensionality, reliability, item statistic, person and item measure, and wright maps. The findings of this study revealed that the items in Howard's Indonesian version of the questionnaire are less valid and less reliable with the Indonesian sample, and the items do not retain the psychometric characteristics of the original scale because it failed to meet the minimum criteria of 0.80 logits. Implications and recommendations were discussed and offered with reference to the findings.