Team science, Decision Making, and Diagrams - Ohh My!
Friday, December 26th 2014
this post is a work in progress
Group Concept Mapping
Earlier this year, Sept 4th according to my brain book, Pips was explaining a method for collecting and organizing information from a group with goal of improved decision making.
The process is called Group Concept Mapping. While Pips explained the methodology I was seeing the data gathering step as web application and the flow users and researchers would have with it. I captured this in my brain book (#21). The following is the how I see the users (green) contributing concepts to facilitators. The collecting, rating, and bucketing process is the easy part of this (relatively). The post-processing, analysis, and reporting (the mapping) is where things get more complicated.
Many time since Pips explained this method I’ve found myself wishing I had the tool in hand for projects at work.
I saw this process as a set of steps like this:
- Collect on a Topic - web page presents the users topic and ask them for what they recommend for solutions to address the topic. Encourage them to provide three or more answers.
Once the user group has finished submissions the resulting entries are available for the admins (facilitators or researchers) to clean up the data:
- Edit, De-dupe, Split - Users submissions need to be clarified, de-duplicated (if they provided the same concept twice but phrased differently), and split apart (make sure an entry is not a multi-part entry - if so break it apart into it’s own parts)
At this point we’ve got a new dataset that is the scrubbed set of response entries from users. The next phase is to get the scrubbed dataset rated and bucketed. This could be done by the same user group that submitted it or perhaps a newly selected group.
- Rate & Bucket - This is a two part step where you provide all the scrubbed entries back to users and have them do the following to each individually:
- Rate - 5 star system like Amazon or Netflix.
- Bucket - Let the user define a set of buckets (as many or few as they need) to group the ideas together. Let the user name the buckets at they please to assist in their grouping of similiar entries into a category of their choosing.
Depending on how automated the analysis and reporting process below is I see this step as being useful both the a group of users providing a single rating&bucketing session and for an individual doing their own
At this point the data collection from users is complete.
The edited, de-duplicated, split, rated, and bucketed dataset is available for the researcher to do Analysis on.
The social research methods has an old, but good, overview of the concept mapping process in this Introduction to Concept Mapping for Planning and Evaluation by William M.K. Trochim. I love the fact they talk about the compute technology needed to do this - check out these quotes:
- “On a mainframe computer, brainstormed statements can be entered into any standard editor.”
- “The concept mapping analyses can also be accomplished on microcomputers, although there are some trade-offs involved.”
- “The SYSTAT program, available for both MS/DOS and Macintosh environments is one of the few microcomputer programs at this time which includes both multidimensional scaling and cluster analysis”
I think revisiting this in a modern, mobile friendly, web application would provide, at least myself, a handy tool.
Looking around for open source tools around this I find things that are similiar but aren’t quiet what I’d like to see available. There are a variety of commercial and closed source platforms available on a subscription basis.
Brain book #21 sketch entry
