CAADRIA 2001

1
A Methodology for Measuring Preferences of Design
Alternatives Using Internet
Jan Dijkstra, Jos van Leeuwen and Harry
Timmermans
2
overview

1. Research Context
2. Introduction
3. Conjoint Measurement
4. Experimental Design
5. Session
6. Discussion

3
1
Research Context
4

• Research project is part of the VR-DIS platform.
• Given the acronym ICARUS
• A System for Interactive Conjoint-Based Analysis
  in Virtual Reality of User
• Satisfaction and (Complex) Decision Making

5
2
Introduction
6
Conjoint Analysis is used to describe a technique
for estimating the value people place on the
attributes or features which define products,
services and design alternatives.

• In a conjoint study, a researcher
• selects attributes (characteristics, features)
• classifies attributes into levels
• combines attribute levels into profiles

Is it possible to develop a VR based system of
conjoint analysis?
7
3
Conjoint Measurement
8

• Verbal

• Pictorial
• Multi media

• Virtual reality

9
4
Experimental Design
10
Intention

• A sever hosts the experiment and supports a
  series of page accesses by the same client.
• General information about the subjects is
  logged.
• Session state information is collected during
  the course of the experiment.

11
Designed Profiles
Attribute Values
Wall between workplace and open space Transparency (50) Transparency and no transparency alternated by vertical strips Closed
Wall between workplaces Transparency and no transparency alternated by vertical strips Closed
Dividing wall between workplaces in the open space No Yes
12
Profile Preference Measurement
13
Database Structure
14
Session
5
15
6
Discussion
16

• Data is being collected.
• This methodology offers the potential of an a
  priori evaluation of design performance and an
  opportunity for generalizing the findings.

17
Evaluate our future Faculty Building http//www.ds
.arch.tue.nl
18
(No Transcript)
19
mediumlarge
Number of shops small Travel time 15
min Parking search time 4 min
3045
1220
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)