In this study, the effect of fluent reading (speed, reading accuracy percentage, prosodic reading), comprehension (literal comprehension, inferential comprehension) and problem solving strategies on classifying students with high and low problem solving success was researched. The sampling of the research is composed of 279 students at elementary school 4 th grade. In the research, in order to figure out reading accuracy percentage and reading rate, 5 scales were used: a reading text, prosodic reading scale, literal comprehension scale, inferential comprehension scale and problem solving scale. In order to see the effect of fluent reading and comprehension skills on classifying students with high and low problem solving success, logistic analysis was conducted while discriminant analysis was conducted to determine the effect of problem solving skills. At the end of the study, it was seen that fluent reading skills had no effect on classifying students according to their problem solving success. It was concluded that both comprehension skills are 77% effective in classifying problem solving success, but inferential comprehension is more effective than literal comprehension. It was found that problem solving strategies were effective on classifying students with high and low problem solving success problem at 88% level; that the most important factors while classifying were estimation and control, systematic listing, looking for a pattern and drawing figures and diagrams respectively; and that backward-studying strategies were inadequate in classifying successful and unsuccessful students. At the end of the study, mathematical sentence writing strategy appeared to be the most important strategy in classifying students with high and low problem solving success, but it had a negative correlation. In other words, a rise in the usage rate of this strategy increased the likelihood for individuals to be in the group with low problem solving success.