AbstractMany high school students experience serious learning difficulties in physics. This issue was examined and dealt with, within the context of an introductory physics course in Israeli high schools. Following a detailed task analysis, the cognitive entry requirements for this course were identified, and students tested for them. Secondly, specific difficulties students encountered during the study of the course, and prevailing misconceptions held by many of them, were identified. Based on all the above information, a remedial teaching method was developed. It consisted of supplying students with immediate and frequent feedback, to reinforce their understanding, correct misunderstandings, and fill in gaps in necessary background skills, while teachers could continuously monitor the progress of each individual student. The whole process took place within the natural classroom setting, without additional staff or time requirements. The method was implemented in a large number of schools, and its impact was assessed using an experimental versus control analysis of covariance design. Achievements in the experimental group were significantly higher. The statistical analysis used students' background as the covariate. Of the 47% of the variance, which the model accounted for, 24% were explained by the covariate (background), 16% were the effect of the treatment, and both were highly significant. An additional 7% were the teachers' effect (nested within treatments), but this effect did not reach statistical significance.