The present study has three major aims: to design a new framework and analytical tool for Referential Cohesion Analysis (RCA); to justify the validity and reliability of the instrument for RCA through its implementation on a corpus of academic texts; to explore problem areas of novice writers and identify expert strategies in the use of reference to provide an empirical foundation for teaching this aspect of academic writing. These aims were accomplished through a series of five empirical studies, which focused both on piloting and refining the analytical tool used and on exploring cohesive reference in academic writing. This study uses a mixed approach and is built up from smaller-scale studies, which contributed to achieving the research aims in different ways. Although the study is largely exploratory in nature, at some stages of the research process, it was necessary to rely on hypotheses. Other stages were data-driven; that is, the analytical problems or emerging data provided the input for subsequent stages of research. Therefore, at this point, the research questions for the sequence of five smaller scale studies are listed here, but they will be supplemented with further questions and hypotheses in the relevant chapters discussing each.

 

 

Table 10 summarizes the research questions for the five small-scale studies that contributed to achieving the first two aims of the research; that is, designing and validating the Referential Cohesion Analysis instrument, and the third, pedagogical aim of this research (stated at the beginning of this section). The gradual expansion of the corpus is shown under the heading “data sources” in Table 10.