The aim of the study is to investigate the effects of gender, type of the school attended and mother/father's educational background on middle school students' problem-solving, science process skills and learning styles. The population of the present study consists of all the 4th-grade students in the secondary schools in the city of Mugla in the 2012-2013 school year while the sample consists of 569 middle school fourth grade students. As the data collection tools, "The Problem-Solving Inventory", "The Science Process Skills Test", "The Kolb Learning Styles Inventory" and a student information form were used. The data were analyzed by using a variety of statistical techniques such as descriptive statistics, independent sample t-test, One way ANOVA and Chi-Square test. As a result of the study were found that middle school students mostly use the "Avoidance" sub-dimension. Also, there isn't any difference was found between problem-solving skills and gender, the school attended, mother and father's education level. Also, it was found that the students frequently used "Basic Science Process Skills" and that there is not any statistically significant difference between science process skills and gender and maternal education level yet there is a significant difference between science process skills and the school attended and father's education level and "Integrated Process Skills" were found to be correlated with father's education level. On the other hand, learning styles were found to be varying significantly depending on the school attended but not depending on gender and maternal education level and father's education level and it was also found that the highest number of students has the "Diverging" learning style while the lowest number of students has the "Accommodating" learning style. Thus, it can be said that planning instructional and educational activities in such a way as to give feedbacks to individual students can increase efficiency of learners.