For decades, studies have shown the existence of a stereotypical image of scientist between children among the world using the Draw-a-Scientist-Test (DAST) method.However, the methodological limitations associated with the DAST question the results obtained and highlight the need to develop new protocols to evaluate students' conceptions about scientists.In this paper the methodological deficiencies of the DAST instrument are reviewed and some changed made to the original protocol are described.The pilot test carried out with 149 students of cultural diversity enrolled in 2 nd to 6 th grades of Elementary School shows that this modification allows a more reliable identification and analysis of students' conceptions, reducing some of the methodological problems identified in the use of DAST.