Use of Computers in the Grid Interview

This is a highly edited summary of an aspect of a discussion about the use of computers and grid software for the application and analysis of grid held on the Repgrid User Group, starting at message 10.


Repertory Grid Interview Application

The Grid interview has many applications in both business and psychology. Many applications provide good results without computers or statistical analysis.

And, provided you go about it the right way (which means having regard to the basic principles of personal construct theory), you can elicit a grid very easily indeed on the back of an old envelope. Moreover, you can do a variety of simple analyses of both element and construct relationships on the reverse of that envelope. Choose a slightly larger envelope, apply an hour's-worth of patience, and you can do a cluster analysis by hand to a 10 x 10 grid if you want to and the information you obtain in a cluster analysis is sufficient for many, many applications.

You do not have to have a computer.

Nevertheless a computer provides a means of conducting the interview while, at the same time, using the results of analysis as feedback. Feedback can obviously be done some time after analysis has been completed but computer analysis integrated with the interview process provides the ability to do cluster analysis of the data while the interviewee is present and to use that analysis as input to further construct development.

That is not always necessary or appropriate but it does provide an additional powerful tool that allows interactive interviewing, free from interviewer bias, while developing very large grids. Analysing a grid manually while having an ongoing interviewing conversation with a client is difficult - if not impossible.

Computer Based Interview Advantages

The advantages in computer based interviewing are:

  1. A less professional interviewer can be supported by software-based routines (which of course has the disadvantage of restriction for more advanced interviewers).
  2. A computer-based interview can do the analysis much quicker, because the data is already entered into the computer.
  3. Most of clients want feedback immediately, during an interview and don't want to wait for an hour.
  4. Immediate feedback can confim that the client's construct system is being captured accurately.
  5. Cluster analysis done during the interview can be used as input for further construct development as part of an iterative process.


But it is important that computerised routines don't act as a substitute for professional experience among "less professional interviewers". A reasonable knowledge of the underpinnings of the technique in personal construct psychology is very important. And the bells and whistles of computerised techniques can simply distract and get in the way - computers are a bonus but very rarely essential.

On the other hand the prime thing that computer analysis gives us is immediacy of feedback. A computer can give a subject feedback faster than any manual technique. This is important because we can confirm the construct system with the subject as it is being developed.

Computer analysis can improve the performance of a less experienced interviewer using a grid, simply because it may alert them to some of the questions they need to ask and wouldn't have thought of otherwise. This is not substitution for understanding PCT or the theory behind grids, which also improves performance for everyone on using grids. If a computerised grid elicitation and analysis program were simply bells and whistles it wouldn't be a very good program. The best grid analysis programs embody the best human insights.

Note 1: The Easy Guide to Repertory Grids by Devi Jankowicz is a user-friendly introduction to the powerful mental mapping tool of repertory grid technique. Refer to related.htm.

Note 2: There was much of value in this discussion that I believe warrants treatment in this way in order to help others understand the role of computers in Repertory Grid interviews. There was a lot more content. This is not intended to be a well-constructed information module but is, instead, my summary of that part of the discussion on that issue. I have not acknowledged the participants, as, what they actually said has been substantially modified and is available at the Rep Grid group. I thank the participants for their input.


Search for results from the Enquire Within site.





Contact Us