The Text Matrix as a Tool to Increase the Cohesion of Extensive Texts José Osvaldo De Sordi Faculdades Metropolitanas Unidas (FMU), rua Iwakumi, 236, Jundiaí, SP 13211-424, Brazil. E-mail: osdesordi@gmail.com Manuel Meireles Faculdade Campo Limpo Paulista (FACCAMP), rua Guatemala, no. 167, Campo Limpo Paulista, SP 13231-230, Brazil. E-mail: profmeireles@uol.com.br Osvaldo Luiz de Oliveira Faculdade Campo Limpo Paulista (FACCAMP), rua Guatemala, no. 167, Campo Limpo Paulista, SP 13231-230, Brazil. E-mail: osvaldo@faccamp.br The significant amount of time needed to prepare concise and cohesive texts is among the reasons why extensive texts proliferate in organizations. Extensive documents are more likely to have low cohesion among their various sections, which may lead the reader to perceive the information as being of low quality. This research addresses this issue by presenting a tool (Text Matrix) composed of procedures and algorithms coded in software, with the aim of analyzing cohesion between the text sections in extensive documents. A design science research approach was applied to develop, test, and prove the usefulness of the Text Matrix. A total of 127 academic advisors and advisees were trained to use the Text Matrix and apply it in the development of the extensive text of dissertations. An analysis both of those users’ perceptions and of variations in network density among 41 extensive texts that they improved through the analyses performed by the artifact proposed in the present study demonstrated the ease of operation and effectiveness of the Text Matrix to identify parts of exten- sive texts associated with cohesion failures that could be connected by cohesion devices or even excluded from the extensive text. The Challenge of Extensive and Cohesive Texts If I had more time, I would have written a shorter letter. This aphorism concerning information quality highlights the dynamics between the time for the elaboration of a composition, the scope of its content, and text cohesion. The original idea is attributed to the French mathematician and philosopher Blaise Pascal, according to the text found in one of his letters of 1656: “The present letter is a very long one, simply because I had no leisure to make it shorter” (Pascal, 2013, p. 240). Contrary to common perceptions, communi- cating through a long text is faster and easier than transmit- ting the same content using a concise and cohesive text. The Portuguese writer, journalist, diplomat, and medical doctor Julio Dantas expressed his perception of writing with the following: “What is difficult is not to write too much; it is to say everything writing a little” (Victoria, 1966, p. 92). Eva (2010, p. 432) describes one of the difficulties in developing concise and cohesive texts: “Using fewer words requires that authors think long and hard about the core messages they wish to express. Every word must be chosen deliberately, every thought linked and articulated in a way that conveys exactly what is meant and nothing more.” Conciseness is required to make a text more appealing to readers not only by making it easier to read (Eva, 2010), but also by preventing difficulties that reduce the quality of extensive texts, for example, the high likelihood of dis- connected or even contradictory ideas. In information science, “concise representation” is present in most tax- onomies associated with the dimensions of information quality (Huang, Lee, & Wang, 1999). It is noteworthy to highlight the existence of various dimensions that aid in the analysis of quality information and the large diversity of models proposed for analyzing those dimensions. These models differ from one another in terms of denomination and the number of the dimensions of information quality (Arazy & Kopak, 2011; Stvilia, Gasser, Twidale, & Smith, 2007). In the organizational environment, the lack of Received January 12, 2014; revised August 11, 2014; accepted September 10, 2014 © 2015 ASIS&T • Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/asi.23434 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY, ••(••):••–••, 2015 V C 2015 ASIS&T  Published online 17 March 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/asi.23434 JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY, 67(4):900–914, 2016