KU-Mapper  v 1.01 by Roy Clariana (RClariana@psu.edu)

Pathfinder Alternative (or Abbreviated) Rate data collection software that can be analyzed by KNOT (order KNOT 4.3 available from http://interlinkinc.net/).

Welcome! Here is a free research software tool for collecting proximity data in 3 different ways including (a) traditional Pathfinder rate pair-wise comparisons of all terms (whoa!), (b) an abbreviated approach that uses a list-wise comparisons, and (c) a semantic-map card-sorting tool that allows the participants to 'drag' the terms around the screen into a 'semantic-map'. All three approaches save the data as *.PRX files that can be analyzed by KNOT, and also as vectors that can be analyzed by standard MDS software, such as SPSS.

Why these 3 approaches? Using the 'traditional' Pathfinder rating approach is tough on the participants, since even a small number of terms results in a large number of pair-wise comparisons. Hopefully, these 2 alternative approaches provide a 'better' and not just different way to collect proximity data. The abbreviated approach (option b) is based on how KNOT converts proximity data into a PFNet, it is an attempt to give KNOT just the data it needs and no additional spurious data (which is a tremendous time saver for the participants). The semantic-map tool (option c) simply uses geometric distances as proximities.

Limitations - You can only use 15 terms (no more and no less). I was in a hurry when I programmed this software, so rather than making it open-ended, I hard-coded it with this limit to save bunches of time. If I have more time in the future (fat chance), I will add  more flexibility to the software. Alternately, if you are an Authorware programmer, here is 1 MB "raw" programming file KUmapper.a4p (yes, I'm still stuck at version 4 of Authorware) so that you can modify it yourself. Send me your improvements and I will post those here, etc.

(1) Create a folder on your local drive or network. 
(2) Download KUmapper.exe (1.3 MB) and terms.txt (1 KB) into that folder.
(3) Open terms.txt with notepad or another text editor and add your own 15 terms and then save it.
(4) Double-click KU-Mapper.exe. (KU Mapper will read your 15 terms, ask the participants for their names or IDs, and will provide ALL THREE rating approaches randomly, one after the other. Then it will create a new folder called /prxdata that contains a bunch of files (see below).

Data files created by KU-Mapper in the prxdata folder include
(a) ID.txt - contains the participants' names associated with their unique 5 digit number. This is the ONLY file with the participants' names, the rest of the files only use the 5-digit number. Obviously this is to maintain anonymity, if you lose or delete the ID.txt file, then there is no way to associate the participant with their data.
(b) Three *.PRX files for each user (these files can be analyzed by KNOT). For example, 12524.PRX, a12524.PRX, and m12524.PRX where the 'a' prefix refers to the 'abbreviated' approach and the 'm' to the mapper approach. The PRX file without a prefix is the 'traditional' approach.
(c) Three related text files, pfnet.text, apfnet.txt, and mpfnet.txt. These three files contain the 105 proximity data elements for every participant for each rating approach for use by MDS software, for cluster analysis, etc. Unfortunately, Authorware MUST write these text files horizontally, while SPSS needs the data in a vertical format.  I usually open these text files with Excel, copy all the data, then Paste Special, "transposed". This changes the data orientation from horizontal to vertical, and this file can then be opened directly by SPSS (using Excel is a real time saver).

Please send feedback, pro or con. Maybe we can start a KU-mapper ring, publish together, improve the software, or something... Go ye forth and collect data!

click here for a KU-Mapper manuscript (doc file)

Today's date is Easter Sunday, April 20, 2003; updated April 10, 2005.