Analyzing Personal Attribute Judgments - Modeling Users' Experiences with Interactive Systems - page 40

Chapter 3
Analyzing Personal Attribute Judgments

Abstract. This chapter presents a second Multi-Dimensional Scaling procedure that
aims at identifying diverse views even within single individuals. The technique is
applied on an existing dataset (Heidecker and Hassenzahl, 2007). It is illustrated
that the - traditional - averaging analysis provides insight to only 1/6th of the total
number of attributes in the example dataset. The proposed approach accounts for
more than double the information obtained from the average model, and provides
richer and semantically more diverse views on the set of stimuli.

3.1 Introduction

In the previous chapter a simple and effective technique was presented for acquiring
diverse views on Repertory Grid data. This technique, however, relies on a number
of assumptions. First, it assumes that all attributes of an individual may be analyzed
by a single 2D view, and second, that individuals are able to consistently summarize
all attribute judgments in a single judgement of overall dissimilarity. As such, it dis-
cards diversity existing within a single individual and does not explicitly optimize
the goodness of fit of the diverse models for the analyzed attributes. It was shown
that while the technique succeeded in identifying the differences in individual's per-
ceptions in terms of overall dissimilarity, it failed in accounting for more attributes
than the traditional “averaging” analysis.
This chapter suggests a quantitative, exploratory Multi-Dimensional Scaling pro-
cedure to account for the diverse views that one or more individuals may have on a
set of products. It will be demonstrated that even single individuals can handle more
than one view on a set of stimuli. It will be shown that by averaging interesting
views are overlooked due to majorization bias. The insights strongly advocate the
view that the analysis of quality judgments of interactive products should not stop on
a group level, but must be extended to the relations between the attribute judgments
within an individual. The Repertory Grid combined with the suggested technique
to analyze the resulting quantitative data is an important step towards the adequate
account of homogeneity and especially diversity in individual quality judgments.