Doctoral students' research presentations were investigated for the way in which the students articulated their rationales for data selection for their research projects. Rather than look for genre-related characteristics, though still working within the area of academic discourse (spoken), the objective of this study was to identify a set of possible data selection rationales, for subsequent use by future doctoral students and their advisers. Presentations from two groups of doctoral students were analysed: part-time students on an Applied Linguistics 'thesis and coursework' PhD programme at a UK university, and student nurses on a doctoral programme at the University of Michigan (whose presentation transcripts were already publicly available). The result was a substantial set of data selection rationales, including availability of data, good quality (various types), professional concern, paradigm-relatedness, originality and generalisability. These were provisionally ranked in terms of 'research awareness'.