A 35·item questionnaire concerning writing habits, experiences of writing and productivity was sent to 228 full·time, U.K. domiciled, social science research students. One hundred and one complete responses were received. Cluster analysis was used to identify three distinct groups of students in terms of the strategies they used when writing: "Planners", who planned extensively and then made few revisions. "Revisers", who developed content and structure through extensive revision, and "Mixed Strategy" writers. who both planned before starting to write and revised extensively as part of their writing processes. The Planners reponed higher productivity than both the Revisers and Mixed Strategy Writers. Planners and Revisers did not differ significantly in how difficult they found writing to be; Planners found writing less difficult than did the Mixed Strategy Writers. We conclude that working from a plan can be an effective writing strategy for some, but that planning is neither a necessary nor a sufficient condition for writing success.Academic writing is difficult. It requires a complex combination of generating ideas, selecting the ideas that are appropriate to the writing task, translating these into text and polishing the text to produce a presentable document. In doing this the writer has to attend not only to his or her own thoughts, but also to the content and style conventions of the community for whom the piece is being written.For all but the simplest writing task, it is probably not possible to manage this number of constraints simultaneously (Bereiter 1980). Writing most documents will only be possible if the task is first divided up into more manageable sub-tasks. These sub-tasks may then be performed in series (rather than concurrently) to produce a finished piece of text. For the purposes of this paper we will call the way in which a particular writer partitions and sequences the writing process his or her "writing strategy". This paper examines the writing strategies of graduate research students with a view to exploring the relationship between writing strategy and success at thesis writing. Few new research students will have had previous experience of writing a document as long or as complex as a research thesis. Also, more so than with most undergraduate writing, a thesis should be written in a style that conforms to that expected by the academic audience at which it is aimed. It is likely, therefore, the process of writing a thesis will present a major challenge to most research students, and research suggests that an appreciable number of students find thesis writing very difficult (Rudd 1985;Torrance, Thomas and Robinson 1992). Despite this, however, writing instruction for graduate research students is often afforded a low priority within doctoral degree courses. StUdying the writing strategies of research students is interesting, therefore, for two reasons. It offers insight into the writing