Prepared By
Greg Andersen - E-Mail:
andersen@cpsc.ucalgary.ca
March 25, 1996
For my first WebGrid experiment, I chose "employment" as my domain and "careers" as the context of my grid. Several common careers make up the elements of my grid and then constructs were built with the triad elicitation tool. This tool takes three careers and prompts the user for information on what makes one career stand out from the other 2. All the elements are then rated on this newly created construct.
The constructs were fairly easy to build using the triad feature once you got going on it. I did not find the program all that intuitive however, and played around for a little while before I figured out what was going on. To experiment with my grid, click on the Link to Grid.
Part 2: Elicitation of Constructs on CPSC547 Topics
[TOC]
The constructs were much more difficult to create on this grid. This is due to the fact that the elements were not as easily distinguishable as my own grid elements were. I had to cancel out from the triad facility a number of times when I couldn't think of any original ways in which one element stood out from the others.
The elements used in this grid were the topics presented in class by the students. As above, the constructs were created using the triad function and include:
The comparison between my constructs and those of Dr. Gains can be seen in the following graph.
There appears to be absolutely no relationships here between our constructs and yet I would have to say that most of them are in a correspondence relationship according to the 4-way classification of constructs given by Dr Gains and Dr Shaw. While our terminology is different, the distinctions match up quite well. For an interactive look, click on the Link to Grid.
Part 3: Elicitation of Constructs on Presentation Topic
[TOC]
For my presentation WebGrid experiment, I chose "Wide Area Networks" as my domain and "Applications of Broadband Networks" as the context of my grid. The elements of my grid include Video on Demand, Distance Learning, Video Conferencing, Home Shopping and Banking, Medical Imaging, CAD/CAM, and Animation. The constructs were:
The above grid shows how each of the elements rated (ie. on a scale from 1 to 5) on each of the constructs. For example, using the Development tool -- Application tool construct, Video on Demand was rated at 5 while Animation was rated at 1. This means that Video on Demand is an Application tool while Animation is a Development tool
The FOCUS graph shows the model of the relations between elements and constructs in my data set. From the graph, we can see that the constructs educational purpose -- professional purpose and cheap -- expensive are related and that elements CAD/CAM and Animation are also related.
The graph below shows the PrinCom principal components analysis. From this graph, we can see that there are two clusters of related constructs typified by Development tool -- Application tool and expensive -- cheap respectively, plus two isolated constructs, platform dependant -- platform independant and educational purpose -- professional purpose. It would appear that one of these isoloated constructs is providing the "z" dimension (ie "coming out of the screen") and therefore would not really be isolated. I suspect it is the platform dependant -- platform independant construct.
[ TOC | Exploration | Elicitation of Constructs on CPSC547 Topics | Elicitation of Constructs on Presentation Topic ]