Research conducted at the University of Quality Jl Ngumban Surbakti No 18 Medan, North Sumatera Indonesia. This study aims to determine the influence of problem based learning models to improve learning quality. The type of research that researchers use is a draft of the research on quasi exsperimen with research draft using Non-Equivalent Control Group Design experiments. The population in the study is all students of the second generation of PGSD with a total of 7 classes with a student number of 278 people and as a sample of the class 2B21 (X-2) as the experimental class as many as 40 people and class 2B22 (X-1) as control class as many 35 people are acquired in a random way. Research instrument is a form test of the 35 description of the problem that has been validated by the validator with the level of breliability of Cronbach’s alphadan of 0.83 very high category. Hypothesis testing conducted with Independent sample T-Test significantly ∝ = 5% and test requirement analysis is data normality and variance homogenization.Based on the results of the test Independent Sample Test obtained t = 6.2706 and t at the degree of freedom (DF = 73) and a significant level of α 0.05 is 1.6660. It was therefore null Hypotesis rejected and received an alternate hypothesis. Thus the implementation of a problem-based learning model is influential and can improve the outcomes of low-grade mathematics teaching the primary school teacher Education course of Quality University
Keywords: Learning Outcomes, Problem-Based Learning, low-grade mathematics
