Great reliance is now placed by many educational researchers on audio- and video-recordings, and generally speaking these need to be transcribed before the data can be analysed. Transcription is not simply a matter of ‘writing down what was said’, a range of decisions is involved about what to include and how to represent it. This chapter reviews these decisions, and the options available. However, while transcripts are constructions, this does not mean that the resulting data are simply ‘made up’. Indeed, great care is required to try to ensure their accuracy. The range of forms that transcription can take is emphasised, and it is suggested that the mode of transcription adopted should depend upon the nature of the data collected, and the purpose and character of the analysis proposed.