Método de análise e validação nas investigações em Educação em Ciências e Matemática na REAMEC : método Delphi como critérios de triagem

Abstract

There is a wide array of options of scientific methods for researchers on Education in Science and Mathematics that are possible to be adopted in research on analysis of records and validation of results, and among these options there is the Delphi Method, which since World War II, significantly contributes to the studies that adopt it in order to obtain estimates raised from interactive assessments carried out by a specialized set of anonymous people. However, it is observed in the method‘s procedures the absence of a systemic and methodological structure in the analysis and validation and the accuracy of the reliability level. Therefore, this study aims to develop some screening criteria with systemic and methodological structure to get results from the confluence of the valuations performed by experts with a certain degree of reliability. For the validation of the screening criteria, a survey of qualitative and quantitative approach in the Education in Science and Mathematics‘ framework was held, in which questionnaires were submitted to a group of experts selected randomly so they would evaluate five of the indicators obtained from doctoral work in the Education in Science and Mathematics‘ area. The structural elements that have been taken into consideration to build the indicators were the problem, the general objective, the methodology and results. To evaluate the contribution of doctoral works, and indicator (also) called contribution was prepared. The group that issued the valuations, through e-mail, was structured by 16 experts. Open and closed-questions questionnaires were used for register of the valuations during the three interactions needed to obtain an answers‘ confluence. The valuations carried out by experts take into consideration the consistency between the elements that make up each indicator and their contribution to Science in the Education in Science and Mathematics‘ area. The statistical content used to elaborate the screening criteria made it possible to monitor the confluence of valuations with a determined degree of well-known reliability. The implementations of a screening criteria enabled the development of methodological guidance activities with the experts and provided the individual and collective reflection during the interactions. The screening procedure made It possible to register and to minimize the variation range of 5% and to reliably affirm that the 5 indicators and their contributions are extremely relevant to the Education in Science and Mathematics‘ area.