1
Means for measuring people’s preferences for
visual wood with aid of Internet
Olof Broman
Luleå University of Technology
Division of Wood Technology, Sweden
Aims
• To develop a methodology for preference studies
• To show how the ranking and interview results could be analysed
3
Limitations
• Only wood floors of parquet type
• Only one type of room (however there were two different looks)
• The interview concept should be self
instructive and possible to run with aid of Internet
• Attractive and take no more than 10 minutes
The interview concept
Front Page
Instructions
Ranking procedure
Preference questions
Personal questions
5
Frontpage
Olof Broman LTU, division of Wood Technology Sweden
Instructions
7
Ranking procedure
Comparison in pair
Which floor suits best to the room?
Alternate picture by clicking the button above
When the better one is shown then press the button below
Change image
This floor is better
Three interview rounds
Test 1.
Test 2. Two weeks later
Test 3. Additional two weeks later with altered showroom
9
7 different looks of parquet floorings
Results
50 persons made all three tests
• Tables
• Principal Component Analysis
11
Alder Birch Oak Cherry Maple Oakplank Walnut Test 1 Average rank 5,0 4,4 4,1 5,8 4,0 3,2 1,6 Score 7 12 14 7 43 10 14 0
Score 6 26 7 19 24 17 7 0 (%) Score 5 38 21 10 17 5 5 5 Score 4 10 29 19 7 21 12 2 Score 3 10 17 29 5 29 7 5 Score 2 2 10 17 5 14 31 21 Score 1 2 2 0 0 5 24 67
Test 2 Average rank 5,2 4,8 3,9 5,5 4,4 3,3 2,1 Score 7 23 13 8 30 18 10 0 Score 6 23 20 8 30 15 5 0 (%) Score 5 33 18 20 10 13 5 3 Score 4 5 35 13 18 13 15 3
Score 3 13 8 33 13 33 3 0 Score 2 5 8 20 0 3 45 20 Score 1 0 0 0 0 8 18 75
Test 3 Average rank 5,3 4,0 3,8 6,2 3,8 3,3 1,6 Score 7 12 6 2 53 12 14 2
Score 6 35 10 14 29 6 6 0 (%) Score 5 31 22 16 6 14 10 2 Score 4 12 27 18 6 25 6 6 Score 3 10 22 33 4 12 14 6 Score 2 0 8 16 2 27 35 12 Score 1 0 6 2 0 4 16 73
PCA loadings
-0,30 -0,20 -0,10 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70
p[2]
Alder
Birch Oak
Cherry Oakplank
Walnut
13
-3 -2 -1 0 1 2 3
3 4 5 6 7 8 9 10
t[2]
t[1]
PCA scoreplot
-0,50 -0,40 -0,30 -0,20 -0,10 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70
0,00 0,10 0,20 0,30 0,40 0,50
p[2]
Alder
Birch Oak
Cherry
Maple Oakplank
Walnut
-2 -1 0 1 2 3
t[2]
15
-3 -2 -1 0 1 2 3
3 4 5 6 7 8 9 10
t[2]
t[1]
Maple Maple
Maple
Birch
Birch Birch
Alder Alder
Oak
oakplank
oakplank oakplank Walnut
Walnut Walnut
Maple Maple Maple
Maple Birch
Alder
Alder
Alder Alder
Alder
Alder
Cherry Cherry
Cherry
Cherry Cherry
Cherry Cherry
Cherry Cherry
Cherry
Cherry Cherry Cherry
Cherry
Cherry Cherry
Cherry
Cherry Cherry Cherry
Cherry Cherry
Cherry
Cherry
Cherry Cherry
Cherry Cherry
Cherry
Cherry Cherry
Cherry
Cherry
Cherry
Cherry
Cherry
Cherry Cherry
Cherry
Cherry Cherry
Cherry Cherry
Cherry Cherry
Cherry
Cherry Cherry Cherry
Cherry
Cherry Cherry
Cherry
Cherry Cherry Cherry
Cherry Cherry
Cherry
Cherry
Cherry Cherry
Cherry Cherry
Cherry
Cherry Cherry
Cherry
Cherry
Cherry
Cherry
Cherry Oak oakplank
oakplank oakplank
oakplank oakplank
oakplank
oakplank
oakplank
Each persons favourite
-2 -1 0 1 2 3
t[2]
Maple
Maple Maple
Maple
Maple
Birch
Birch
Birch Birch
Birch
Alder Alder
Oak
oakplank
oakplank
oakplank
oakplank oakplank
oakplank oakplank oakplank
oakplank
oakplank
oakplank
Walnut Walnut
Walnut
Walnut
Walnut Walnut Walnut
Walnut Walnut
Walnut Walnut
Walnut
Walnut Walnut
Walnut Walnut
Walnut Walnut
Walnut
Walnut Walnut
Walnut
Walnut Walnut Walnut
Walnut
Walnut Walnut
Walnut
Walnut Walnut
Walnut
Walnut
Walnut
Walnut Walnut
Walnut
Walnut
Walnut Walnut Walnut Walnut
Walnut Walnut
Walnut Walnut
Walnut
Each persons ”worst floor”
17
-3 -2 -1 0 1 2 3
3 4 5 6 7 8 9 10
t[2]
t[1]
2 0
0
1
0
21 0
0 0
1 1 -2
1
1
2 2
1 1
2
2 1
1 1
1 0
1 1 1
2
2 2 2
2
2 0
1 0
1
2 1 2
1 1 2
0
2
2
0 0
1
2
2 2
0
0
2
2 2
1 1
1 1
2
2
-2 1
0
0 0
0
I want it ( 2 ) ( 1 ) ( 0 ) (-1 ) (-2 ) I do not want it
Distribution of answers on the question:
“Please express Your appreciation for this look of floor”
Answer alternatives
Most preferred floor (%)
Least preferred floor (%)
2 43 1
1 31 1
0 20 1
-1 3 15
-2 4 83
19
Change in the ranking of the Comparison
between test- round nr:
Best floor (%)
Worst floor (%)
T1 <-> T2 42 (6.5) 6.5 (0) T1 <-> T3 54 (43) 26 (8.6) T2 <-> T3 52 (30) 24 (8.7)
* Distribution of change in preference
Maple Birch Alder Cherry Oak Oakplank Walnut
( Example of a radical change)
Conclusions
• Digital images and the Internet can be used for preference studies when aesthetic features is in focus.
• Ranking procedure worked well.
• PCA suitable for thorough understanding.
• The look of the showroom or environment important and affect people’s choice
21
Future work
• Large interview test with a more realistic room that lie in line with the product examined.
• Combine this preference mapping with the answers of detailed questions (or sensory
aspects) to find generic parameters that explain the reasons in differences in preference