The quantitative Knowledge Space Theory is applied here for the first time to specific science concepts in order to evaluate its feasibility as a general assessment method for science learning. In particular, the knowledge states of students in the 4th through 12th grades are suggested for three science concepts before and after classroom teaching/learning experiences. The theory belongs to the vast category of latent trait models based upon the Rasch model, but differs in all its most important characteristics by having the power to extract the cognitive organization of the material. Student responses to the pre-tests and the post-tests were analyzed for the pressure concept, the density concept and the conservation of matter, and tentative critical learning pathways for each concept are suggested which describe the most probable sequences of learning steps. In all three cases the quantitative results show a change in the students’ logic structure after the classroom experience. Knowledge Space Theory shows considerable promise for providing a useful quantitative analysis technique for assessing and analyzing the knowledge of students when used with other assessment methods such as concept mapping and clinical interviews. Ultimately, exemplary strategies and materials may be identified by this procedure for helping teachers effectively move students along the most probable critical learning pathways as suggested by the students themselves.