This study was carried out to examine the relationship between secondary school students' problem-solving skills and scientific attitudes in terms of gender, class level and education level of the parents. In the study, correlational research model, which is among the general survey models, was employed. The sample of the study consisted of 560 students selected from the secondary schools in Afyonkarahisar Province, Turkey by using convenience sampling method. In the study, Problem Solving Inventory for Children (PSIC) and Scientific Attitude Scale (SAC) were applied to collect the data. In the analysis of the data, Pearson correlation analysis, one-way variance analysis and two-way variance analysis were performed. As a result of the analyses, a negative and low-level relationship was found between secondary school students' problem-solving skills and scientific attitudes. While there was a negative and low-level relationship in terms of female students, it was found that this relationship was not significant for male students. When analyzed in terms of class level and maternal education level, it was concluded that this relationship was not significant. It was concluded that problem solving skills and gender did not have a significant and common effect on students' scientific attitudes. Similarly, it was concluded that problem solving skills and class level did not have a significant and common effect on students' scientific attitudes. There was no significant difference between middle school students' scientific attitudes in terms of problem-solving skill levels. In addition, it was determined that students' problem-solving skills were not a significant predictor of their scientific attitude.