Learning geometry is one that students fear. They have difficulty understanding the concepts of geometry. Students experience problems with Lobachevsky’s axiom alignment, as well as Riemann’s axiom alignment. Therefore, we need learning that makes it easy for students to imagine these axioms more concrete, although geometry cannot always be carried in a concrete world. The purpose of this study is to describe students’ cognitive processes in understanding the properties of triangles in Riemann Geometry through local content. This is a part of development research. We conducted a needs analysis of the ability of students to understand new concepts in geometry. The subject of this research was chosen from students of mathematics education at one of the tertiary institutions in Bengkulu City, Indonesia. Subjects were interviewed in-depth through task-based interviews. His job is to solve the problem of the properties of triangles in Riemann Geometry. Interview data were analyzed through genetic decomposition analysis. The results of this study were subjects able to analyze the interrelationship between concepts, axioms, and basic properties of geometry through orange peel media. He can attribute and organize. The attributes are obtained from three segments that meet at the ends. Subjects can organize the elements of the triangle so that the nature of a large number of angles in the triangle is more than 180. The conclusions of this study are the subjects can achieve the cognitive process of analysis, and can arrange the correct population.

Algebra is mandatory learning in junior high school. That is a continuation of arithmetic with the bridge being variable. Students often have difficulty in understanding it, especially quadratic equations. The purpose of this study is how students’ cognitive processes in understanding the principles of quadratic equations. It was an exploratory study conducted on research subjects. The subjects were junior high school students selected based on their ability to understand concepts and principles about quadratic equations. That’s done through tests about understanding quadratic equations. There were 16 selected research subjects. They were interviewed in-depth by researchers using interview guides. Interview data were analyzed qualitatively based on genetic decomposition. The results of this study are that students can use square, rectangular and unit square cards to determine the principle of factorization of quadratic equations. Students can generalize until completing a perfect quadratic equation. The conclusion is that the cognitive process of students who learn to use learning media based on square and rectangular able to reach a high level (trans-level).

This study aims to determine whether there is an effect of the learning model Blended Learning with the Realistic Mathematics Education (RME) approach on students' conceptual understanding. This type of research is an experimental study with a quasi-experimental design which was carried out at The Junior High School Student Number 11 Bengkulu City in the even semester of the 2020/2021 school year with the research population being all students of class VIII totaling 248 students and the research sample totaling 60 students consisting of 2 classes namely class VIII.E as the experimental class and VIII.F as the control class. Based on the Mancova analysis test, it shows that there is a significant effect of thelearning model Blended Learning with the approach Realistic Mathematics Education on understanding concepts. The understanding of students' mathematical concepts in the class taught with thelearning model Blended Learning with the approach Realistic Mathematics Education is higher than the class taught with the conventional model. The average understanding of concepts in the experimental class is 75.67 while in the control class it is 67.83. ABSTRAKPenelitian ini bertujuan untuk mengetahui apakah terdapat pengaruh model pembelajaran Blended Learning dengan pendekatan Realistic Mathematics Education (RME) terhadap pemahaman konsep siswa. Jenis penelitian ini adalah penelitian eksperimen dengan desain quasi eksperimental yang dilaksanakan di SMPN 11 Kota Bengkulu pada semester genap tahun pelajaran 2020/2021 dengan populasi penelitian adalah seluruh siswa kelas VIII yang berjumlah 248 siswa dan sampel penelitian berjumlah 60 siswa yang terdiri dari 2 kelas yaitu kelas VIII.E sebagai kelas eksperimen dan kelas VIII.F sebagai kelas kontrol. Berdasarkan uji analisis Mancova, menunjukan bahwa terdapat pengaruh signifikan model pembelajaran Blended Learning dengan pendekatan Realistic Mathematics Education terhadap pemahaman konsep. Pemahaman konsep matemaika siswa pada kelas yang diajar dengan model pembelajaran Blended Learning dengan pendekatan Realistic Mathematics Education lebih tinggi dibandingkan kelas yang diajar dengan model konvensional. Rata – rata pemahaman konsep pada kelas eksperimen sebesar 75,67 sedangkan pada kelas kontrol sebesar 67,83.