This is qualitative research with a descriptive-explorative approach that aims to determine students' mathematical representation skills in solving Geometry problems. The population is all students in the mathematics education department who study Basic Geometry. The main subject was chosen by stratified technique and purposive sampling. The instruments are placement tests, diagnostic tests, and guided interviews. Data collection uses a triangulation method that aims to examine legitimate data. The results showed that 1) subjects with high skills always present visual representations and mathematical expressions, 2) subjects with intermediate skills present visual representations and mathematical expressions, and verbal but less representative representations, 3) subjects with low skills don’t present visual representations and mathematical expressions and verbal representations of problem-solving steps; and 4) there are several factors that influence the ability of mathematical representation to solve geometric problems, among others, subjects are less able to present problems in geometric patterns, because the subject does not understand the problem.