Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/13342
Title: Accounting for diversity in subjective judgments
Authors: Karapanos, Evangelos 
Martens, Jean Bernard O.S. 
Hassenzahl, Marc 
Major Field of Science: Natural Sciences
Field Category: Computer and Information Sciences
Keywords: Multi-dimensional scaling;Quantitative methods;Repertory grid;Subjective judgments;User experience
Issue Date: Apr-2009
Source: 7th International Conference on Human Factors in Computing Systems, 2009, 4-9 April, Boston, MA, United States
Conference: International Conference on Human Factors in Computing Systems 
Abstract: In this paper we argue against averaging as a common practice in the analysis of subjective attribute judgments, both across and within subjects. Previous work has raised awareness of the diversity between individuals' perceptions. In this paper it will furthermore become apparent that such diversity can also exist within a single individual, in the sense that different attribute judgments from a subject may reveal different, complementary, views. A Multi- Dimensional Scaling approach that accounts for the diverse views on a set of stimuli is proposed and its added value is illustrated using published data. We will illustrate that the averaging analysis provides insight to only l/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 diverse views on the set of stimuli.
DOI: 10.1145/1518701.1518801
Rights: Copyright ACM
Type: Conference Papers
Affiliation : Eindhoven University of Technology 
Folkwang University of the Arts 
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation

CORE Recommender
Show full item record

SCOPUSTM   
Citations

20
checked on Nov 6, 2023

Page view(s)

341
Last Week
0
Last month
2
checked on Nov 21, 2024

Google ScholarTM

Check

Altmetric


Items in KTISIS are protected by copyright, with all rights reserved, unless otherwise indicated.