Surprising finding: similarity in cognitive maps and cause maps

When you build cognitive maps and cause maps you may notice more or less duplicate statements. Depending on how similar these statements are you will either merge them together because they are equal, link them together because their similarity is due to a shared theme or leave them be since their similarity is just superficial.




Whatever the case how do you find similar statements in the first place? On the fly while mapping with your interviewees or while you weave the cognitive maps together? But that is quite random. A more systematic way is to think of comparisons. You start with statement 1 and compare it to every other statement except to itself. You continue with statement 2 and compare it to all other statements except to itself. Rinse and repeat for 3, 4, etc. Easy, right?

However, take for instance a standard job interview. This may produce a map of about 50 statements. You need to compare every one of the 50 statements. You need to compare these to every other statement except to themselves. So you have 50 statements times 49 statements which gives 2450 comparisons. Ok, right, since comparing, e.g., statement 2 to 23 is the same as comparing statement 23 to 2 you can halve the comparisons and arrive at 1225. That is still quite amazing actually. A relatively small map of 50 statements already requires 1225 comparisons in order to make sure you have considered every combination that might yield a merger or link.

But do not worry. If you thought that was bad news, there is something way more terrible coming in the next post.

(C) CC BY (https://creativecommons.org/licenses/by/4.0/), Jo. Richter, http://causal-cognitive-mapping.blogspot.com/2017/01/surprising-finding-similarity-in.html