EDUCATIONAL RESEARCH FROM INTERNET LATENT DATA

Abstract

This study aims to analyse the use of latent information in the Internet, as basis for the research in Educational and Social Sciences. Such work is distinguished from those that collect the data “through Internet”, e.g. with questionnaires or forms. It also differs from those using data stored in the Internet. We work with studies that analyse documents in the Internet, extracting data from the analysis of these documents. These documents were once distributed through the Web not knowing this future use as research objects. We can say that it is a research model based on trails left by Internet users. For this work we have revised the literature with research work based on this latent corpus, addressing four questions: What can be investigated with this data? In which context this data was found? How to select samples? And what ethics aspects should be considered? We have found very different kind of contents. In fact we have distinguished between studies on content and studies on interaction. Also we have found differences between websites oriented to news and oriented to knowledge. About samples, we have considered three kinds of population: closed-finite, open-infinite and The Web. We have also studies sample correction techniques. Conclusions of this work detail possibilities and opportunities as well as the limits of this research method.

Keywords: Research; latent Corpus; Internet; Education.

DOI: http://dx.doi.org/10.14244/19827199713