The identification of students' mistakes on mathematical communication ability in three-dimensional shapes of geometry: Cube and cuboid

Abstract

Geometry is one of the mathematical sciences that study point, line, 2D shapes, 3D shapes, and their properties and parts. It shows that how important geometry to be learnt by students. But in reality, students consider geometry to be the most difficult material to understand in mathematics subject because geometric objects are abstract. This research method is a descriptive study, with the aim of the study is to find out the number of percentage and the level of students' mistakes using Newman's Error Analysis on mathematical communication ability in three-dimensional shapes of geometry: cube and cuboid. The subjects used in this study were students of Pre University students (class entry 2018) in STKIP Surya as many as 9 students coming from Lani Jaya Regency, Papua Province. The instrument used was type description items, as many as 5 numbers about mathematical communication ability. In giving a score in student work, researcher focus on the results of student answers that are wrong (error), 1 score for a wrong answer and 0 for a right. The analysis of the errors uses five errors according to Newman's Error Analysis, namely (1) reading error; (2) comprehension error; (3) transformation error; (4) process skill errors; and (5) encoding error. Here are the results that have been obtained based on five proficiency level errors, namely reading errors with a percentage of 37.78%, comprehension error with a percentage of 86.67%, transformation error with a percentage of 82.22%, process skill error with a percentage of 71.11 %, and encod- ing error with a percentage of 80%. Based on the results obtained in this study, the researchers conclude that the large percentage of errors made by Pre University students in STKIP Surya Tangerang at the stage of comprehension error, transformation error, and encoding are categorized very high, then at the stage of process skills error can be categorized high, while at the reading error stage can be categorized low.