The problem of text in evaluative research: A case study

Abstract

Based on a rhetoric of the evidence, technocratic discourses construct narrative plots "starring" by unquestioned measures and indicators before which counter arguments do not fit. Thus, certain decisions are presented as the only possible ones, as the one best way of government action. Before the text of a report presented in terms of scientific and political truth, we propose an analysis of its elaboration processes, its forms of appropiation and its uses; a critique of its conditions of possibility whose object is the text of the report itself. This article provides a sample of this critical work. Through intensive study of the evaluation of a public intervention, we analyze the production process of an evaluation report, its discursive result and its institutional reception. From that analysis, we derive some implications about research as writing process and about the coupling and decoupling games between research and government discourses and practices.