JOSEPH EDWARD CONBOY (**)
RESUMO
As medidas de magnitude do efeito representam uma resposta eficaz às diversas críticas às metodologias de investigação tradicionalmente utilizadas nas ciências sociais. Tais medidas evitam algumas das dificuldades mais evidentes dos testes de hipótese; são de interpretação simples, sem o recurso a valores de p, que podem ser enganadores; e fornecem uma base para a síntese de grande número de estudos. Este artigo descreve o cálculo e interpretação das formas mais comuns desta abordagem alternativa, ou complementar, aos testes de hipótese. Pressupõe-se que o leitor dispõe de um conhecimento básico de estatística aplicada sem ter, necessariamente, uma forte formação em estatística matemática.
Palavras-chave: Magnitude do efeito, Métodos de Investigação, Análise de Dados, Ciências Sociais.
]]> ABSTRACT
Measures of effect size represent an alternative to traditional methods of data analysis in the social sciences. These metrics avoid some of the most evident problems associated with tests of hypotheses, are more simply interpreted than sometimes misleading p values, and provide the basis for the synthesis of large numbers of research studies. This text assumes that the reader has a basic understanding of inferential statistics and describes the calculation and interpretation of common forms of effect size that can be used either as alternatives, or complements, to classical hypothesis testing.
Key words: Effect size, Research methods, Data analysis, Social Sciences.
Texto completo disponível apenas em PDF.
Full text only available in PDF format.
REFERÊNCIAS
]]>APA (1994). Publication manual of the American Psychological Association (4ª edição). Washington, DC: Autor. [ Links ]
APA (2001). Publication manual of the American Psychological Association (5ª edição). Washington, DC: Autor. [ Links ]
Barnette, J., & McLean, J. (2002a, Abril). Shedding light on the eta-square and omega-square relationships with the standardized effect size. Paper presented at the annual meeting of the American Educational Research Association. New Orleans, LA. [ Links ]
Barnette, J., & McLean, J. (2002b, Abril). The need to abolish the arbitrary effect size standards. Paper presented at the annual meeting of the American Educational Research Association. New Orleans: LA. [ Links ]
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum. [ Links ]
Conboy, J. (2002). A magnitude de efeito como forma de análise em ciências sociais. Lusiada Psicologia, 1 (1), 187-210. [ Links ]
Conboy, J. (2000). Estimando a dimensão do N de uma amostra. INUAF Studia, 1 (2), 227-274. [ Links ]
Cooper, H. (1997). Some finer points in meta-analysis. In M. Hunt (Ed.), How science takes stock: The story of meta-analysis (pp. 169-181). New York: Russell Sage Foundation. [ Links ]
Cooper, H., & Hedges, L. (1994). The handbook of research synthesis. New York: Russell Sage Foundation. [ Links ]
Cortina, J., & Nouri, H. (2000). Effect size for ANOVA designs. Thousand Oaks, CA: Sage. [ Links ]
Cumming, G., & Finch, S. (2001). A primer on the understanding, use, and calculation of confidence intervals that are based on central and non-central distributions. Educational and Psychological Measurement, 61, 532-575. [ Links ]
Fidler, F. (2002). The 5th edition of the APA Publication Manual: Why its statistics recommendations are so controversial. Educational and Psychological Measurement, 62 (5), 749-770. [ Links ]
Glass, G. (2000). Meta-analysis at 25. Retirado do World Wide Web em 1 de Fevereiro de 2002: http://glass.ed.asu.edu/gene/papers/meta25.html [ Links ]
Glass, G. (1976). Primary, secondary, and meta-analysis of research. Educational Researcher, 10, 3-8. [ Links ]
Glass, G., McGaw, B., & Smith, M. (1981). Meta-analysis in social research. Beverly Hills, CA: Sage. [ Links ]
Glass, G., & Hopkins, K. (1996). Statistical methods in education and psychology, Third Edition. Needham Heights, MA: Allyn & Bacon. [ Links ]
Haddock, C., Rindskopf, D., & Shadish, W. (1998). Using odds ratios as effect sizes for meta-analysis of dichotomous data: A primer on methods and issues. Psychological Methods, 3 (3), 339-353. [ Links ]
Hays, W. L. (1981). Statistics for the social sciences (3rd ed.). New York: Holt, Rinehart and Winston. [ Links ]
Hays, W. L. (1972). Statistics for the social sciences (2nd ed.). New York: Holt, Rinehart and Winston. [ Links ]
Hedges, L. (1981). Distribution theory for Glass's estimator of effect size and related estimators. Journal of Educational Statistics, 6, 107-128. [ Links ]
Hedges, L., & Olkin, I. (1985). Statistical methods for meta-analysis. New York: Academic Press. [ Links ]
Henkel, R. (1976). Tests of significance. Beverly Hills, CA: Sage. [ Links ]
Huberty, C. (2002). A history of effect size indices. Educational and Psychological Measurement, 62 (2), 227-240. [ Links ]
Huberty, C. (1994). A note on interpreting an R-squared value. Journal of Educational and Behavioral Statistics, 19, 351-356. [ Links ]
Hunt, M. (1997). How science takes stock: The story of meta-analysis. New York: Russell Sage Foundation. [ Links ]
Jacobson, N., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12-19. [ Links ]
Kendall, P. (Ed.) (1999). Clinical significance [Special section]. Journal of Consulting and Clinical Psychology, 67, 283-339. [ Links ]
Kirk, R. E. (1996). Practical significance: A concept whose time has come. Educational and Psychological Measurement, 56, 746-759. [ Links ]
Lipsey, M., & Wilson, D. (2001). Practical meta-analysis. Applied Social Research Methods Series (Vol. 49). Thousands Oaks, CA: Sage. [ Links ]
Roberts, J. K., & Henson, R. (2002). Correction for bias in estimating effect sizes. Educational and Psychological Measurement, 62 (2), 241-253. [ Links ]
Rosenthal, R., & Rubin, D. (1982). A simple general purpose display of magnitude of experimental effect. Journal of Educational Psychology, 74, 166-169. [ Links ]
Rudas, T. (1998). Odds ratios in the analysis of contingency tables. Thousand Oaks, CA: Sage. [ Links ]
Siegel, S. (1975). Estatística não-paramétrica. São Paulo: McGraw-Hill. [ Links ]
Smith, M., & Glass, G. (1977). Meta-analysis of psychotherapy outcome studies. American Psychologist, 32 (9), 752-760. [ Links ]
SPSS. (2001). Statistical package for the social sciences (Versão 11.0). [Computer Software]. Chicago: IL: SPSS. [ Links ]
Steiger, J., & Fouladi, R. (1997). Non-centrality interval estimation and the evaluation of statistical models. In L. L. Harlow, S. A. Mulaik, & J. Steiger (Eds.), What if there were no significance tests? (pp. 221257). Mahwah, NJ: Erlbaum. [ Links ]
Tallmadge, G. (1977). The joint dissemination review panel ideabook. Washington, DC: National Institute of Education and the US Office of Education. [ Links ]
Thompson, B. (2002a). «Statistical», «practical», and «clinical»: How many kinds of significance do counselors need to consider? Journal of Counseling and Development, 80 (1), 64-71. [ Links ]
Thompson, B. (2002b). What future quantitative social science research could look like: Confidence intervals for effect sizes. Educational Researcher, 31 (3), 25-32. [ Links ]
Wilkinson, L., & APA Task Force on Statistical Inference (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54, 594-604. [ Links ]
(*) Agradecimentos: Gostava de agradecer à Professora Jesuína M. B. Fonseca, da Universidade do Algarve, e A. K. Conboy da Universidade de Bowdoin que comentaram uma versão anterior do trabalho. Gostava igualmente de mencionar o meu reconhecimento aos responsáveis das Bibliotecas Guttman da Harvard Graduate School of Education e Milbank da Columbia University que possibilitaram o acesso a algumas das referências incluídas no artigo.
(**) Instituto Superior Dom Afonso III, Loulé, Portugal.
]]>