NVivo7 Test Drive Report #2

November 20, 2007 – Creating my ‘project’

I followed the step-by-step instructions in the Getting Started guide for setting up a new project. As I worked through the steps, I noted some things I liked and disliked about NVivo7:

Likes:
- You can use regular Word files (.doc) or even text files (.txt) instead of having to convert everything to Rich Text files (.rtf), although .rtf files are acceptable. This is not really a very big deal I suppose but always felt cumbersome to me when working with other QDA software (e.g. MaxQDA and Transana).
- The ability to import multiple documents at once rather than having to select and import them one at a time.

Dislikes:
- You can’t use punctuation in naming nodes, e.g. I wanted to name a node “everything ought to be together” including the quotation marks because I was quoting a participant.
- (this one started out as a like) There doesn’t appear to be a limit to the length of description that one can give to nodes like there is in Transana…whoops! I was wrong
- No spell check
- While coding I can only look at a list of EITHER my free nodes or my tree nodes but not both at once, that’s awkward
- The little window that opens up every 15 minutes to remind you to save your work in annoying, wonder if there’s a way to change that setting to auto-save without asking me…there is! I found it by using Help but it was exactly where I would have expected it to be had I been thinking more clearly about the similarities between NVivo7 and Outlook.


I used the online help screens to find out how to add attributes to cases after reading that this was possible in the Quick Start Guide. I decided to identify each of my participants as a case since we had collected data describing characteristics like age, gender, perceived computer skills, internet use, library web site use, academic status, major, etc. I created attributes for each piece of data we collected.

Next, I added some tree nodes and free nodes based on codes I had already created and been using with this data in another piece of QDA software. One of the things I’m hoping to find out is whether one can create a tree node at a higher level of a hierarchy and then assign existing, hierarchically lower nodes to the new hierarchically higher level node.

I read through one entire transcript and coded it. During the process I created some new codes. These transcripts accompany a video recording of the screen of the workstation on which the participant is completing tasks. This is the first time I’ve analyzed a transcript without simultaneously watching the video recording of the sequence of pages visited. It’s a different experience focusing completely on the text. I can’t tell yet whether I’ll get more or less or just different results doing it this way.