Personal Attribute Judgments - Modeling Users' Experiences with Interactive Systems - page 32

while Groups 2, 3 and 4 perceive relatively strong differences between TouchToPrint
and Badge.

2.6.3.2 Interpreting Product Dissimilarities

As mentioned all perceptual spaces show two main clusters of products. One may
notice a strong preference for products in the first cluster (i.e. only two participants
prefer products from the second cluster). This can be further explained by analyzing
the directions and the type of attributes. It is evident that most attributes point to
product cluster 1 while only three attributes point to cluster 2. A close look at the
content of these three attributes provides some more insight. Two out of the three
attributes are essentially the contrast poles to a negative attribute of the comparison
products. This is also evident when looking into the element differentiators. Partic-
ipants indicated that “TouchToPrint is not reliable” and that “it is privacy threaten-
ing”, thus other products will score better on reliability and privacy. It seemed that
participants could more easily identify qualities in products of cluster 1 that were
less familiar to them than in the traditional products from cluster 2.
Within product cluster 2 one can notice differences among groups. Groups 1
and 4 perceive differences between Scroll List and Direct Print while Groups 2, 3
and 5 perceive these two products as relatively similar. Group 1 seems to perceive
differences between Scroll List and Direct Print in two attributes: control and per-
sonal. Apparently, for Group 1, Scroll List provides more control in the interaction
as documents are only printed when the user is identifying himself to the printer.
Moreover, it is more personal than Direct Print as every user has a personalized ac-
cess to the printer. More interesting though is the observation that Group 3 does not
regard scroll list as personal despite having this personalization feature. Apparently,
users from Group 3 distinguish between the personalization features of Scroll list
and those of TouchToPrint, Badge and Touch & Select. One explanation could be
that the perception of personalization for Group 3 is influenced by the perception
of security (i.e. with scroll list one can identify himself as another person). Further,
group 4 seems to perceive differences between Scroll List and Direct Print in two
attributes: environmental friendly and fast. One can notice that group 4 differs from
other groups in the perception of the attribute fast. For group 4 the two extremes
are Scroll List and Direct Print where the second is fast as it involves no interaction
with the product and the first involves tedious interaction. It is also evident from
preferences that a product quality for this group is a minimum interaction with the
product. The other groups however perceive that direct print requires additional ef-
fort for identifying the documents among the pile of all printed ones, thus Direct
Print is not fast. It is evident that the attribute fast is not statistically consistent be-
tween groups but it is consistent within groups. This demonstrates the value of the
two-step classification process as it enables us to gain insight in the different per-
ceptions of an indistinguishable product quality.
Further, within cluster 1 one can notice differences in the relation between Badge
and TouchToPrint as compared to Touch and Select. Groups 1 and 5 perceive
TouchToPrint and Badge as very similar to Touch & Select while Groups 2, 3 and