To analyze the video, we used a qualitative approach based on open coding and axial coding, as described by Creswell et al.. The coding of the video was performed in three passes, using an iterative process and three coders. The first pass involved open coding, allowing us to identify several categories of interest. For the second pass, we decided on common codes to compare activities. We used terms from Jansen et al. InfoVis pipeline to relate our coding to previous thinking about InfoVis authoring.
In this page you can access :
We used chronoviz to annotate the video. The annotated data is available for download in JSON and XML chronoviz format.
We transformed the chronoviz annotation files into Json files for visualizing the data with d3.js. The script to perform this transformation is available here.
To better understand the annotation data, we created an interactive visualization of the annotations. Below you can see annotations for each participant. Roll over a category label to highlight the segments from that category. To open this visualization into an independant window click here .
To better understand the differences in actions between conditions, we created a small multiples version of the previous visualization. To open this visualization in a separate window, click here.