TrUMPo members Philippe Hambye and Barbara De Cock co-authored with fellow UCLouvain colleague Andrea Pizarro Pedraza a chapter on Annotation and mark up for representation analysis for newly published Analysing Representation: A Corpus and Discourse Textbook , edited by Charlotte Taylor and Frazer Heritage. As opposed to academic handbooks, that are often less readily accessible to students and rarely include practical exercises, the editors of this textbook wanted to offer a book that combines a solid academic foundation with hands-on exercises and that is readily usable for (self-)study by students or by anyone less familiar with corpus and discourse analysis. The exercises were tested by students prior to publication.
In the chapter on Annotation and mark up, De Cock, Hambye and Pizarro Pedraza show step by step how to develop a solid annotation, starting with the conception of the categories used for annotating in view of a specific research question related to the annotation of representation. In line with the aims of this very hands-on textbook, the chapter furthermore gives practical suggestions to set up an annotation in Excel that is methodologically sound, reliable and easily searchable. It exemplifies these steps through case studies, one of which being the annotation grid developed by the TrUMPo team. However, the chapter aims to help readers develop their own annotation methodology, showing how the experience gained through the TrUMPo project can help others to analyse the data relevant for their research question and offering hands-on exercises to prepare the readers to do so. To give a gist of other topics that can be analysed, a second case study discussed in this chapter concerns the representation of ‘to abort’ and the chapter further also mentions the representation of the 2021 riots at the US Capitol.
From a broader perspective, this chapter further discusses frequent challenges and pitfalls related to inter-annotator agreement (viz. the degree to which different annotators would reach the same results) and/or transferability of annotation categories to other corpora.
Thus, this chapter – and indeed the textbook as a whole – is a must-have for readers who seek to apply a reliable method to analyse representation in discourse.
コメント