This study examined the frequency of educational games applied on problem-solving skills in children of the age group 11-12 years. The experimental and control groups consisted of 30 students each. Educational games were played for 3 days a week, for 10 weeks in the experimental group, and pretest and posttest were applied to both groups as Problem Solving Inventory for Children. Data was analyzed using the Mann Whitney Test, U Test and Wilcoxon signed ordinal numbers test. While no significant difference between the groups was found in preliminary tests, a significant difference was found as p = 0.008 <0.05 in confidence sub-dimension between the posttests, and as p= 0.016 < 0.05 in self-control sub-dimension. In the avoidance sub-dimension, no statistically significant difference was found as p = 0.132> 0.05. As a result, the conclusion was reached that the 10-week educational game practices created a significant difference in confidence and self-control that were the sub-dimensions of problem-solving skills.