Students’ Confidence Instrument Analysis in Poetry Learning through Rasch Model

Abstract

Background: Students' low confidence becomes the reasons for conducting this research whereas it is one of the most important skills needed to develop ability and potency. Objective: This research aims to obtain information about Students' confidence levels in Bahasa Indonesia's subject, especially in poetry learning. Materials and Methods: This research employed a cross-sectional study with the quantitative approach. Participants of this research were 45 fifth grade students in Cibeureum Elementary School and 50th grade students in Kotabaru Elementary School 2019/2020 academic year. The self-report questionnaires were used as a research instrument to measure Students' confidence levels with the cluster sampling technique. Data were analyzed using the Partial Credit Model (PCM). These parameters are identified with the category coefficients and the scoring function of the Rasch model for polychotomous responses in which the latent trait is assumed uni-dimensional. Results: The trial results showed that as many as 13 items proved to be compatible with PCM instruments, the reliability of the instrument was 0.74, the item difficulty index started from -0.36 to 0.36 which meant that all items in the category were sufficient. Conclusion: This development instrument is valid and really measures the effect of student confidence on poetry learning, so it can be used to measure student confidence in poetry learning.