Interactional metadiscourse markers in research articles

The aim of this study is to investigate the use of interactional metadiscourse markers (IMM) in research articles written by American academic writers and Turkish academic writers and to find out whether there are differences in the use of interactional metadiscourse markers between American and Turkish academic writers when they write research articles. With this purpose, on total 150 research articles on the field of teaching a foreign language were analyzed based on the taxonomy Hyland and Tse (2004) suggested. 50 English research articles written by American academic writers (AAWs), 50 English research articles written by Turkish Academic writers (TAWs) and 50 Turkish research articles written by TAWs from different journals were collected for the aim of the study. This study followed a qualitative study design and used a qualitative data analysis program NVivo10. English research articles written by AAWs and TAWs were collected from internationally published and refereed journals and the Turkish research articles were collected from journals published nationally. The biographies of AAWs were checked from their both personal and institutional websites in order to make sure that all the AAWs had their educational background completed in the USA (MA and PhD) and work at a university in the USA. The data were analyzed according to the taxonomy suggested by Hyland and Tse (2004) and some additional markers were considered after the pilot study such as the passive voice used as directives. All data were analyzed twice since extra markers came up during the analysis for constructing interaction with the readers on the writer's side. After the first analysis, extra markers were added to the taxonomy and second analysis was carried out. The frequencies for each marker under each category were found out and then statistical test (Binomial test) was run to find out whether significant difference among the use of interactional metadiscourse markers of AAWs and TAWs existed. The results of the binomial test showed that significant differences in the use of interactional metadiscourse markers existed among the data sets. AAWs used significantly more IMMs in English research articles compared to IMMs in English and Turkish research articles written by TAWs. TAWs used significantly more IMMs in their English research articles compared to IMMs in their Turkish research articles. The subcategories were analyzed and significant differences were found in terms of boosters, attitude markers, and self-mention. Key words: Academic writing, research articles, Interactional metadiscourse markers