This study aims to determine the impact of the ECIRR(Elicit, Confront, Identify, Resolve, Reinforce) learning model on students' mathematical reasoning abilities in terms of student motivation. The research method used was a quasy-experimental method with a post-test only control design research design. The population of this study was all students in five classes XII Private School. The Samples were taken at class XII AP-2 and XII MM-1 as the experimental class, and class XII AP-1 and XII MM-2 as the control class. The data analysis technique used is hypothesis testing using ANAVA 2 paths. Based on the research results obtained that (a) There is an influence of the ECIRR (Elicit, Confront, Identify, Resolve, Reinforce) learning model on mathematical reasoning abilities. (b) There is an influence of student learning motivation on mathematical reasoning abilities. (c) There is no interaction between the treatment of learning models and categories of students' learning motivation towards mathematical reasoning abilities. So as a whole it can be concluded that the ECIRR (Elicit, Confront, Identify, Resolve, Reinforce) learning model influences the ability of mathematical reasoning and can increase students' learning motivation.

Diagnostic tests are generally two or three-tier and based on classical test theory. In this research, the Four-Tier Diagnostic Test (FTDT) was developed based on modern test theory to determine understanding of physics levels: scientific conception (SC), lack of knowledge (LK), misconception (MSC), false negatives (FN), and false positives (FP). The goals of the FTDT are to (a) find FTDT constructs, (b) test the quality of the FTDT, and (c) describe students' conceptual understanding of physics. The development process was conducted in the planning, testing, and measurement phases. The FTDT consists of four-layer multiple-choice with 100 items tested on 700 high school students in Yogyakarta. According to the partial credit models (PCM), the student's responses are in the form of eight categories of polytomous data. The results of the study show that (a) FTDT is built on the aspects of translation, interpretation, extrapolation, and explanation, with each aspect consisting of 25 items with five anchor items; (b) FTDT is valid with an Aiken's V value in the range of 0.85-0.94, and the items fit PCM with Infit Mean Square (INFIT MNSQ) of 0.77-1.30, item difficulty index of 0.12-0.38, and the reliability coefficient of Cronbach's alpha FTDT is 0.9; (c) the percentage of conceptual understanding of physics from large to small is LK type 2 (LK2), FP, LK type 1 (LK1), FN, LK type 3 (LK3), SC, LK type 4 (LK4), and MSC. The percentage sequence of MSC based on the successive material is momentum, Newton's law, particle dynamics, harmonic motion, work, and energy. In addition, failure to understand the concept sequentially is due to Newton's law, particle dynamics, work and energy, momentum, and harmonic motion.