Hi!
I know that data from different sessions on the same computer can be exported in bulk as 3 csv(ego, edge, and attributeList) where each sheet has rows from multiple sessions. Since we have multiple computers which we use to conduct surveys through Network Canvas, we are wondering if there is any way we can export data from all computers at once, so we don’t have to merge the data later?
Thank you!
Jenny
Hi Jenny!
For the first part of your question, I believe you are referring to the “merge sessions” option, which will combine all interviews into a single “set” of export files (in the case of CSV) or a single file (in the case of GraphML). I would honestly not advise merging data in this way unless you have a specific need for it, especially in scenarios where you have multiple interview computers.
The reason for this is that it is usually better from a data management perspective to have a set of files per interview. For example, our R tutorials for data analysis assume that you have a directory of files, where each participant has an ego file, an alter list file (per node type), and an edge list file (per edge type). Working this way lets you structure your scripts so that they iterate over the files, and build a consolidated data frame, which you can then use for your analysis.
In some respects the “merge” option does that too, except that when you have multiple computers involved, you will end up with multiple consolidated session files - hence your question. It will be difficult to keep track of which interviews originated in which source files if you have issues, and we don’t have any documentation on how to work with this format.
My recommendation is for you to not use the “merge sessions” option at all, and to instead establish a workflow where your exported files are unzipped to a shared directory, and then merged using a scripting language such as Python, or in a statistics environment such as R. You can read more about this workflow in our documentation article on data workflows, and about the R side of things in our documentation article on analysis. I know you specifically said that you didn’t want to handle merging later, but being frank I think having a reproducible workflow that takes you from raw data to results is a good idea from a best-practices perspective.
To answer your second question, since Network Canvas cannot know about or control other computers, there is no way to trigger a simultaneous export that the software can merge for you.
Hope this information helps. Please feel free to ask any followup questions 