Multimodal data analysis - hints

Free WiFi and internet at Auckland public library, so updating with my notes on meeting with Dr. Sigrid Norris at AUT on multimodal analysis.


A good chat to share research interests and to learn from Sigrid’s experiences. Also affirmation of the things I had planned to put in place would work.

Firstly, point of view & perspective important, use of overhead cameras ineffectual for capturing the nuances of interpersonal & individual gestural clues. If using multiple cameras, each camera should focus on one unit of analysis (an individual, or a group or the total group). Best for researcher to do the filming or to be taking field notes – with timer- which can then be used to track back on data stream later during analysis.

Data analysis best done as individual annotated still shots rather than a stream of images. Transcribe only the portions which may illuminate research question!! Use of atlasti recommended.

Sigrid also provided invitation to share pertinent data once we have collected some so that a joint data analysis session may be convened to work through the data. Also recommended touching base and reading the work of other researchers including Kay O-Halloran in Singapore and Gunther Kress's team in London.