SciELO - Scientific Electronic Library Online

 
vol.9 special issue 1Contributions of phenomenology to organizational studies author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Cadernos EBAPE.BR

On-line version ISSN 1679-3951

Abstract

MATTOS, Pedro Lincoln C. L. de. "The results of this (qualitative) research cannot be generalized": dotting the i's on this caveat. Cad. EBAPE.BR [online]. 2011, vol.9, n.spe1, pp. 450-468. ISSN 1679-3951.  http://dx.doi.org/10.1590/S1679-39512011000600002.

In articles, dissertations and theses in administration that involving case studies and analyses based on the "qualitative research" tradition, authors warn readers that their conclusions are restricted and cannot be generalized. However, on the contrary, they would be able to claim this authority by using some statistical analysis methods. The term "generalization" is understood as the conferring of qualities of cases or singular units of analysis to others or combinations of them, when these latter were not infact the object of observation and analysis. The aim of this text is to critically clarify the caveat-title in which a wide methodological issue remains implicit. After posing the question (Introduction), some assumptions of it are examined, particularly the paradigmatic force of positivism. A quite different perspective of research by Max Weber is visited in order to experience a healthy epistemic shift in relation to the previous methodological practice. The article then returns to the issue of generalization and critically examines some more elaborate variations of it. Finally, putting the dots on the is, the article concludes by suggesting the purely valuable character of the caveat-title and takes a stance on bout both inductive generalization and qualitative research, two contextual, central points of this essay.

Keywords : Generalization; Qualitative research in administration; Post-positivist analysis.

        · abstract in Portuguese     · text in Portuguese     · pdf in Portuguese