Misconceptions are a common problem that still often occurs in students both at elementary school, high school, and even in college students. Especially science learning with many abstract concepts has a large chance of misconception. Therefore, it is necessary to detect student misconceptions early on. There are several ways to detect misconceptions, one of which is a multilevel diagnostic test that has developed into two tiers, three tiers, and four tiers. This diagnostic test can detect student misconceptions so that later teachers can set the right learning strategy to correct misconceptions and slowly improve student learning outcomes. The type of research used is descriptive research with a quantitative approach. The location of the research is SMP NWDI Pancor with a sample of 27 students. The data analysis technique uses the Certainty of Response Index (CRI) analysis. The results of the analysis show that the number of students who experience misconceptions in science lessons is more than 70% of the total students. Misconceptions occur most in the material of Work and Energy.