In this study, it was aimed to determine the problem solving skills of six years old (60-72 months) children and determine the difference between the children who received experiment based education and those who did not. Pretest, posttest, and control group experimental design was employed in the study. This study was carried out with 42 children in preschools in the 2012-2013 academic year and 22 of them (11 girls, 11 boys) were in the experimental group and 20 of them (11 girls, 9 boys) were in the control group. The children's mean age was 64.38 months and they were members of middle income families. Following the pretest applications, children in the experimental group received "Experiment Based Education Program" for two days a week during a ten week period. The experiment based science education program was consisted of experiments aiming to improve children's use of scientific process skills, independent thinking, decision making, and problem solving process. In the study, the "General Information Form" to have general information about the children and the "Problem Solving Scale in Science Education (PSSSE)" to identify the children's problem solving skills were utilized. The reliability and validity studies of the PSSSE were achieved by the researchers. The data were analyzed with the Mann-Whitney U Test, Independent Samples T Test, and Wilcoxon Signed Rank tests. According to the results, there was a statistically significant difference at the 0.05 significance level between the mean scores of PSSSE for children in the experimental group who received the experiment based education program and for those in the control group. The children in the experimental group had higher scores compared to the controls. This finding revealed that the Experiment Based Science Education Program was effective on the improvement of problem solving skills in preschool science education. (author abstract)