 |
 |
Hi All, Sorry it’s taken me a while to get back to this. You asked: > have you ever tried mathematical models > (such as cluster analysis, multiple correspondence analysis, some forms of > machine learning etc.) to see whether there are some hidden patterns in how > the performance deteriorates? Or how these new rules are invented? No, we have not. But not because of lack of understanding of such things. We are in the main dealing with very small numbers and the reasons for the deterioration are fairly obvious, particularly if we have a good understanding of the context in which the information is used because of our earlier design work. If we do repeat testing of designs with people, we typically use small numbers of participants: between about 6 and 15. We used to do larger sample sizes when we used to do error analysis of paper forms: a sample of 200 was quite common. (Sless, David, and P Fisher. “Improving Information Management in the Insurance Industry.” Information Design Journal 6, no. 2 (1990): 103–29.) But the reasons for many of the changes we see over time are fairly obvious when we detect them. As an example, in the area of design of utility bills, banking statements, and financial documents such as insurance and superannuation (retirement savings), patterns of interaction and expectations of customers have changed—at least in Australia. The standard used in most of these documents is to provide customers with a single contact number for any inquiry. When you ring the number as a customer, you have to listen to a long list and choose a number from the offered menu. We see a growing number of customers complaining about this and asking for the menu numbers to be spelt out on the bill itself so that they don’t have to rely on their memory of the options as they are recited by the ‘voice’ when the phone is answered. Also, they can then go directly to the option they want, or complain if the option they want is not available. This type of small dynamic changes in complex systems, and the rules people invent to deal with them, are an ongoing feature of all ‘language’ communities. The new conventions arise out of the dynamic interaction of participants. In hindsight they often seem obvious, but that is the nature of hindsight, not the situation itself. Moreover, before the introduction of recorded voice messages, such a convention or rule would be meaningless. Hope this helps. David -- blog: http://communication.org.au/blo <http://communication.org.au/blo>g/ web: http://communication.org.au <http://communication.org.au/> Professor David Sless BA MSc FRSA CEO • Communication Research Institute • • helping people communicate with people • Mobile: +61 (0)412 356 795 Phone: +61 (03) 9005 5903 Skype: davidsless 60 Park Street • Fitzroy North • Melbourne • Australia • 3068 ----------------------------------------------------------------- PhD-Design mailing list <[log in to unmask]> Discussion of PhD studies and related research in Design Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design -----------------------------------------------------------------