<div dir="ltr"> Colegas, A CAPES, em algumas áreas, por exemplo Engenharias I, calcula a média ponderada de dados categóricos ordinais, quando associa um número às avaliações dos periódicos classificados como A1, A2, etc. Não está cometendo um erro elementar? O seguinte foi transcrito da página 5 do documento “Requisitos para a Apresentação de Propostas de Cursos Novos” <a href="http://www.capes.gov.br/images/documentos/Criterios_apcn_2semestre/Crit%C3%A9rios_de_APCN_2017_-_Engenharias_I.pdf" rel="noreferrer" target="_blank" style="color:rgb(17,85,204);font-size:12.8px">http://www.capes.gov.br/images/documentos/Criterios_apcn_2semestre/Crit%C3%A9rios_de_APCN_2017_-_Engenharias_I.pdf</a> Calcula-se a produtividade pela relação: (1,0 × A1 + 0,85 × A2 + 0,70 × B1 + 0,50 × B2) / (4 * número de docentes do programa); onde A1, A2, B1 e B2 corresponde ao número de publicações em cada um dos estratos Qualis Periódicos e 4 corresponde ao número de anos inteiros considerados para a quantificação. Luis Braga, Vinicius Arcaro STATISTICS: BASIC PRINCIPLES AND APPLICATIONS William J. AdamsIrwin KabusMitchell P. PreissFebruary 3, 2009, Xlibris Corporation Example 8 A rating Situation We return to Example 3 which had nine students in an introductory psychology course who were asked to express their satisfaction with their instructor's teaching quality in terms of the rating 20 (Outstanding), 10 (Good), 5 (Satisfactory), 2 (Poor), -7 (Terrible). The results obtained are reproduced in table 3. Arranging the 9 ratings in increasing order yields: -7,2,2,2,5,5,10,10,20 The median is the fifth value, 5; half of the students judge the instructor's teaching as satisfactory or below satisfactory; half judge his teaching satisfactory or better than satisfactory. There is no question about the interpretation of the median of ordinal data being meaningful. This cannot be said about the mean of ordinal data. It is worthy of note that while the median is independent of the rating system used, this is not the case with the mean. PUBLICAÇÃO ONLINE DA AMERICAN STATISTICAL ASSOCIATION <a href="https://ww2.amstat.org/publications/jse/v3n2/datasets.short.html" rel="noreferrer" target="_blank" style="color:rgb(17,85,204);font-size:12.8px">https://ww2.amstat.org/publications/jse/v3n2/datasets.short.html</a> 2. Measuring Center 10. As the students discover that the mean of ordinal data is not meaningful, some will probably suggest comparing the medians.<div class="gmail-yj6qo" style="font-size:12.8px;text-decoration-style:initial;text-decoration-color:initial"></div> </div>