User Experience Over Time - Modeling Users' Experiences with Interactive Systems - page 64

as arousal-heightening attributes and their impact on the appraisal of art (Machotka,
1980). Furthermore, time seems to have an impact on the importance we attach to
different qualities of the experience with interactive products. For example, despite
the crucial importance of usability in a product's initial acceptance, aspects of prod-
uct ownership (and not use) are essential for a user to resonate with a product and
value it in the long term.

4.3.4 Limitations of the Study

A number of limitations of the study have to be noted before proceeding to general
conclusions. First, an inherent limitation of the study was the evaluation of only one
product and only by a small number (10) of participants. As a result, these findings
cannot be generalized to a wider population and range of products; they provide,
however, some initial insights that can be falsified or corroborated in subsequent
studies with different products and participants. However, it highlights an inherent
limitation of longitudinal studies in producing generalized knowledge due to their
labor intensive nature which evidently restricts the sample of products, participants,
or time studied. Alternative techniques such as the one we will propose in the next
chapter, and others proposed by von Wilamowitz Moellendorff et al. (2006) and
Fenko et al. (2009) are essential for overcoming the labor-intensive nature of longi-
tudinal studies.
Second, the two measurements (1 st and 4 th week of use) assessed users' percep-
tions at each time (current state) rather than directly assessing how their perceptions
changed over time. One could be concerned about the sensitivity as well as the re-
liability of such absolute measures where judgments do not take place in contrast
to some specified reference point. As discussed in chapter 2, paired comparison
scales that define explicitly both reference points, have been shown to be less sensi-
tive to contextual effects than single-stimulus scales. Thus, one would assume that
by explictly asking participants to assess how their opinion on a given quality has
changed over the course of time, may increase the technique's sensitivity in assess-
ing this change.
Third, one major limitation of the study was the use of a-priori defined measure-
ment model in a pre-post design (with only two measures being taken). Although in
such a study design, measurement has the least influence on users' attitudes, it pro-
vides a rather limited insight into the actual patterns of change over time. First, as
only two measures were taken, one could not know whether the identified changes
were an effect of time, or of random contextual variation. Second, as we employed
an a-priori defined measurement model (Hassenzahl, 2004), we might have evi-
dently missed constructs that dominate prolonged use and are not apparent in this
model. Secondly, due to the quantitative nature of the study, we gained no insight
as to the reasons for these changes in users' experiences. In study 2, we attempt a
qualitative understanding into the differences between initial and prolonged use.