Educating the graduates with cognitive potentials and emotional tendencies for research and innovation has been the ideal for all the education systems throughout the world. It has been also considered a necessity and a national will for our country (Iran). There is no doubt that this important goal should become a reality by the most fundamental element of education i.e.The Teacher. The aim of the present research is investigating the effectiveness of teaching by the problem solving method. It is also to evaluate establishment and raising of research moral among the school students based on the research indexes. A pilot project took place consisting of two test groups and one control group. We selected a total of 68 students from two schools in Tehran (Iran). The students were from three classes in fourth grade. The education year was of 2005-2006. The three independent variables were randomly applied on the mentioned population in the science and geography courses. These variables were: a) teaching by problem solving method along with the evaluation based on research indexes, b) teaching by problem solving method without the evaluation based on research indexes, and c) teaching based on traditional method. Research moral, as the dependent variable, was evaluated on the three indexes of A) attitude toward research, B) the pupils’ performance in the examinations containing the questions about researching issues, and C) research oriented class activities. The obtained results showed that teaching by problem solving with/without the evaluation based on research indexes is more efficient than teaching by the traditional method. This was with the probability of 0.99. We found, there was no significant difference between the teaching with the evaluation based on research indexes and teaching without the evaluation based on research indexes. This similarity was in terms of their efficiency in raising research moral among the research population.