ArticlePDF Available

A comparative study of the emotional assessment of automotive exterior colors in Asia

Authors:
  • Nippon Paint Co., Ltd.

Abstract and Figures

This article compares color assessments of automotive exteriors at four locations in China, Thailand and Japan. This study employed painted panels of twelve colors, thirty sensory words and fifteen pairs of emotion variables. It also used reference figures of nine cars that included four styling categories, namely sedans, compacts, sport utility vehicles and sports cars. The results of like–dislike tests revealed that basic automotive exterior colors, such as black, white and red, were highly preferred across locations. However, geographical differences are apparent in the preference for some non-basic colors. To elucidate these geographical differences, a Chi-square test and principal component analysis were performed on the data on emotion variables. The results revealed both similarities and differences in color preferences among the four locations. Color preferences were similar in Hong Kong and Shanghai when simply evaluated by color samples. However, similar preferences for automotive exterior colors were observed in Hong Kong, Kyoto and Bangkok, but not in Shanghai. These results suggest that there is fluctuation in the emotion evoked by a color when the color is considered in the context of automotive exteriors. Therefore, this study indicates that automotive sales can be secured in the global marketplace by creating a single palette of exterior colors based on similar color preferences across countries. However, these results also indicate that preparing market-specific automotive-color line-ups can be an effective marketing strategy.
Content may be subject to copyright.
Progress
in
Organic
Coatings
72 (2011) 528–
540
Contents
lists
available
at
ScienceDirect
Progress
in
Organic
Coatings
j
ourna
l
ho
me
p
ag
e:
www.elsevier.com/locate/porgcoat
A
comparative
study
of
the
emotional
assessment
of
automotive
exterior
colors
in
Asia
Izumi
Satakea,f,,
John
H.
Xinb,
Tu
Tianmingc,
Aran
Hansuebsaid,
Kenji
Andoa,
Tetsuya
Satoe,
Kanji
Kajiwarae,
Seiji
Ohsawaf
aResearch
&
Development,
Nippon
Paint
Co.,
Ltd.,
4-1-15
Minamishinagawa,
Shinagawa-ku,
Tokyo
140-8675,
Japan
bThe
Hong
Kong
Polytechnic
University,
Hong
Kong,
China
cDong
Hua
University,
Shanghai,
China
dChulalongkorn
University,
Bangkok,
Thailand
eKyoto
Institute
of
Technology,
Kyoto,
Japan
fOtsuma
Women’s
University,
Tokyo,
Japan
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
15
February
2011
Received
in
revised
form
23
May
2011
Accepted
21
June
2011
Keywords:
Automotive
color
Color
emotion
Color
preference
Color
perception
Asian
automotive
color
Cross-culture
a
b
s
t
r
a
c
t
This
article
compares
color
assessments
of
automotive
exteriors
at
four
locations
in
China,
Thailand
and
Japan.
This
study
employed
painted
panels
of
twelve
colors,
thirty
sensory
words
and
fifteen
pairs
of
emotion
variables.
It
also
used
reference
figures
of
nine
cars
that
included
four
styling
categories,
namely
sedans,
compacts,
sport
utility
vehicles
and
sports
cars.
The
results
of
like–dislike
tests
revealed
that
basic
automotive
exterior
colors,
such
as
black,
white
and
red,
were
highly
preferred
across
locations.
However,
geographical
differences
are
apparent
in
the
preference
for
some
non-basic
colors.
To
elucidate
these
geographical
differences,
a
Chi-square
test
and
principal
component
analysis
were
performed
on
the
data
on
emotion
variables.
The
results
revealed
both
similarities
and
differences
in
color
preferences
among
the
four
locations.
Color
preferences
were
similar
in
Hong
Kong
and
Shanghai
when
simply
evaluated
by
color
samples.
However,
similar
preferences
for
automotive
exterior
colors
were
observed
in
Hong
Kong,
Kyoto
and
Bangkok,
but
not
in
Shanghai.
These
results
suggest
that
there
is
fluctuation
in
the
emotion
evoked
by
a
color
when
the
color
is
considered
in
the
context
of
automotive
exteriors.
Therefore,
this
study
indicates
that
automotive
sales
can
be
secured
in
the
global
marketplace
by
creating
a
single
palette
of
exterior
colors
based
on
similar
color
preferences
across
countries.
However,
these
results
also
indicate
that
preparing
market-specific
automotive-color
line-ups
can
be
an
effective
marketing
strategy.
© 2011 Elsevier B.V. All rights reserved.
1.
Introduction
Recently,
automotive
manufacturers
have
begun
to
pay
more
attention
to
Asian
markets,
including
China
and
Thailand,
in
response
to
the
increasing
use
of
automobiles
in
that
region.
Auto-
motive
sales
for
2001
were
5.9
million
in
Japan,
2.3
million
in
China
and
three
hundred
thousand
in
Thailand
[1].
To
expand
automo-
bile
sales
at
the
global
level,
automotive
products
must
be
adjusted
to
reflect
the
tastes
and
customs
of
each
nation’s
market
because
each
country
has
its
own
preferences
for
color,
style
and
function.
Because
color
is
the
most
important
factor
in
catching
the
con-
sumer’s
eye,
this
paper
primarily
addresses
color
assessments
of
automotive
exteriors
in
different
countries.
Corresponding
author
at:
Research
&
Development,
Nippon
Paint
Co.,
Ltd.,
4-1-
15
Minamishinagawa,
Shinagawa-ku,
Tokyo
140-8675,
Japan.
Tel.:
+81
3
3740
1197;
fax:
+81
3
3740
1306.
E-mail
address:
SATAKE NP5746@npc.nipponpaint.co.jp
(I.
Satake).
Researchers
have
conducted
cross-cultural
studies
of
color
pref-
erences
among
Asian
countries
in
the
past
by
using
colored
paper
chips
as
stimuli.
With
respect
to
color
tastes
in
various
coun-
tries,
Saito
compared
color
preferences
in
three
Asian
cities
(Tokyo,
Taipei
and
Tianjin)
and
found
that
white,
vivid
blue,
vivid
green,
light
violet
and
light
blue
were
commonly
preferred
[2].
Saito
also
confirmed,
in
later
work,
that
white
is
a
favorite
color
in
Jakarta
and
Seoul
[3,4].
Chijiiwa
surveyed
the
color
preferences
of
sev-
eral
Asian
countries,
including
Japan,
China,
Korea
and
Taiwan,
by
using
47
colored
paper
samples.
That
study
concluded
that
vivid
purplish
blue,
vivid
red
and
white
were
commonly
pre-
ferred
in
those
countries
[5].
However,
there
are
some
colors
that
induce
different
emotions
in
different
countries.
For
example,
according
to
the
Chijiiwa
survey,
purple
and
black
have
different
traditional
meanings
in
Europe
and
some
parts
of
Asia,
whereas
Saito
remarked
that
purple
elicits
a
different
emotion
in
Japan
and
Taiwan.
Researchers
have
also
conducted
many
cultural
and
geograph-
ical
studies
of
color
emotion
among
Asian
countries.
Xin
et
al.
compared
emotional
responses
to
color
in
Hong
Kong,
Japan
and
0300-9440/$
see
front
matter ©
2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.porgcoat.2011.06.013
I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540 529
Thailand
using
218
color
samples
and
twelve
pairs
of
color
emo-
tions.
That
study
concluded
that
the
influences
of
lightness
and
chroma
were
much
more
important
than
that
of
hue
on
color
emo-
tions
[6,7].
In
the
study
conducted
by
Gao
and
Xin,
a
total
of
218
color
samples
were
assessed
by
seventy
subjects
in
terms
of
twelve
emotion-linked
variables.
The
results
indicated
that
color
emotion
can
be
specified
in
terms
of
three
indices
that
are
dependent
on
lightness
and
chroma.
These
results
also
illustrated
that
the
influ-
ence
of
hue
on
emotional
response
is
less
significant,
even
for
the
variable
“warm–cool,”
than
previous
studies
had
indicated
[8].
In
the
study
by
Gao
et
al.,
subjects
from
seven
different
regions
were
asked
to
evaluate
214
color
samples
on
twelve
emotion
variables.
Its
authors
observed
that
chroma
and
lightness
were
the
most
important
factors
triggering
color
emotions
and
that
the
influences
of
hue
and
cultural
background
were
very
limited
[9].
However,
the
survey
by
Chijiiwa
did
indicate
a
regional
difference
in
color
emo-
tion
when
evaluated
in
terms
of
“somber–gaudy”,
“light–heavy”
and
“cool–warm”.
While
color
is
a
definitive
factor
in
consumers’
decisions,
little
is
known
regarding
how
people
assess
automotive
exterior
colors
in
different
countries.
Although
some
companies
handling
durable
goods,
such
as
automobiles,
have
surveyed
consumers’
color
prefer-
ences
regarding
specific
products,
the
results
of
those
surveys
have
never
been
disclosed
because
of
the
companies’
need
to
secure
a
competitive
advantage.
Pigment
companies
and
color-trend
fore-
casting
companies
report
and
discuss
color
marketing
from
the
perspective
of
consumers’
color
preferences
when
they
provide
details
on
the
number
of
cars
sold
with
specific
exterior
colors
in
various
markets.
Color
trends
were
surveyed
by
counting
the
exterior
colors
of
passenger
cars
on
the
heavily
trafficked
streets
in
Asian
cities.
Such
surveys
confirm
that
50%
of
car
exteriors
are
monochromatic,
whereas
medium
and
dark
colors
are
preferred
among
achromatic
colors
[10].
Japan
Color
Research
Institute
has
surveyed
the
favorite
colors
for
a
sedan
and
a
mini-vehicle.
Subjects
were
asked
to
choose
a
favorite
color
for
a
sedan
and
mini-vehicle
from
among
75
colored
paper
chips.
The
results
showed
the
dependence
of
favorite
colors
on
car
category,
with
black
being
the
favorite
color
for
a
sedan
and
vivid
red
for
a
mini-vehicle
[11].
Saito
compared
color
preferences
among
50
colored
paper
sheets
both
with
and
without
revealing
the
shape
of
a
passenger
car
on
the
template.
No
particular
preference
was
observed
among
neutral
colors,
such
as
black
and
white,
but
neutral
colors
were
slightly
preferred
when
the
shape
was
not
shown.
Light,
vivid
and
deep
colors
were
preferred
when
the
shape
was
not
shown,
but
darker
colors
were
favored
when
the
shape
was
shown
[12].
From
these
findings,
it
is
possible
to
conclude
that
color
preferences
for
automotive
exteriors
are
influenced
by
the
shape
and
category
of
an
automobile.
An
automobile
design
is
specific
to
its
intended
use,
as
represented
by
styling
categories
that
include
sports
cars
and
sedans
among
others.
Consumers
identify
the
function
and
purpose
of
an
automobile
properly
by
its
styling
category,
not
by
its
square
or
round
shape.
For
example,
a
zippy
compact-type
automobile
is
used
for
day-to-day
purposes,
whereas
a
sport
utility
vehicle-type
automobile
is
used
to
take
a
family
on
a
vacation.
However,
even
if
people
from
different
countries
have
a
similar
image
regarding
the
function
of
a
certain
styling
category,
consumers
may
form
a
particular
image
of
an
automobile
and
its
possession
by
reference
to
their
lifestyles,
cultural
traditions
and
the
types
of
automobiles
to
be
found
in
their
country.
In
fact,
as
shown
by
previous
studies,
each
country
is
characterized
by
its
own
most
popular
color
for
an
automotive
exterior.
The
most
popular
color
in
each
country
might
be
largely
determined
by
practical
considerations,
including
the
influence
of
the
strength
of
sunlight
on
the
appearance
of
different
colors.
There
are
also
other
emotional
factors
for
color
determina-
tion.
In
the
context
of
such
factors,
color
images
will
be
different
and
the
styling
and
function
of
an
automobile
may
impact
people
differently
in
different
countries.
To
elucidate
geographical
differences
in
preferences
for
and
the
emotion-linked
variables
of
exterior
automotive
colors,
this
study
observed
responses
to
categories
of
cars
in
Hong
Kong,
Shanghai,
Thailand
and
Japan.
2.
Methodology
2.1.
Selection
of
colors
and
emotion
variables
2.1.1.
Subjects
One
hundred
and
one
subjects
were
selected
from
among
the
students
of
Hong
Kong
Polytechnic
University
in
Hong
Kong
and
the
Kyoto
Institute
of
Technology
in
Japan,
as
well
as
from
among
the
research
staff
of
Nippon
Paint
Co.
Ltd.,
Japan.
At
Hong
Kong
Polytechnic
University,
the
subjects
were
20.7
±
0.86
years
old
(mean
±
SD,
n
=
40),
including
24
males
(20.7
±
0.92
years
old)
and
16
females
(20.6
±
0.81
years
old).
At
the
Kyoto
Institute
of
Tech-
nology,
the
subjects
were
21.4
±
1.25
years
old
(n
=
40),
including
20
males
(21.7
±
1.53
years
old)
and
20
females
(21.0
±
0.79
years
old).
At
Nippon
Paint
Co.
Ltd.,
the
subjects
were
38.3
±
10.4
years
old
(n
=
21),
including
11
males
(43.2
±
11.4
years
old)
and
10
females
(32.3
±
4.44
years
old).
2.1.2.
Survey
conditions
Colored
sheets
of
paper
were
placed
on
a
desk
covered
with
gray
color
cloth
facing
a
north
window
with
natural
sunlight
(which
shows
less
diurnal
variation
of
color
and
quantity
of
light
in
fine
weather).
The
colored
sheets
consisted
of
pieces
(12
cm
×
17.4
cm)
of
the
104b
Color
Chart,
purchased
from
Japan
Color
Enterprise
Co.,
Ltd.,
that
were
applied
to
the
viewing
objects.
In
addition,
the
figures
of
fifteen
automobiles
in
four
different
styling
categories,
including
sedan,
sports
car,
sport
utility
vehicle
and
compact,
were
shown
to
the
subjects
to
encourage
them
to
engage
in
a
survey
of
automotive
exterior
colors.
2.1.3.
Questionnaire
The
subjects
were
asked
to
select
their
favorite
automobiles
by
viewing
a
set
of
randomly
arranged
black-and-white
photographs
of
fifteen
commercial
automobiles.
The
subjects
were
then
asked
to
categorize
the
color
pieces
described
above
as
“suitable”,
“unsuit-
able”
or
“neither”
for
a
selected
automobile,
and
then
to
choose
adjectives
expressing
the
suitability
or
unsuitability
of
the
speci-
fied
automotive
shape
using
the
local
language
(Cantonese
in
Hong
Kong
and
Japanese
in
Japan).
Twelve
colors
were
chosen
for
coating
automobile
exterior
body
panels,
with
four
colors
picked
for
each
of
the
following
categories:
the
most
suitable
color
common
to
all
countries,
the
most
unsuitable
color
common
to
all
countries
and
the
color
specified
as
the
most
unsuitable
by
a
paint
company
but
as
the
most
suitable
by
university
students.
The
colors
were
also
specified
vice
versa.
2.1.4.
Selected
colors
and
emotion
variables
for
the
main
experiment
The
selected
colors
vary
either
noticeably
or
subtly
according
to
the
objects
and
the
subjects
of
the
three
categories.
For
the
main
experiment,
the
selected
colors
were
coated
on
metal
plates
with
automotive
paint
and
were
evaluated
in
terms
of
the
CIELAB
val-
ues
by
using
a
multi-angle
spectrophotometer
(X-Rite,
MA68II),
as
shown
in
Table
1.
The
adjective
words
were
composed
into
the
pairs
of
antonyms,
as
shown
in
Fig.
1.
Fifteen
pairs
of
emotion
vari-
ables
were
composed
from
the
adjectives
listed
by
the
subjects
in
the
questionnaire
for
purposes
of
evaluating
degrees
of
preference
on
a
seven-point
scale.
The
six
similar
side
shapes
were
removed
from
the
original
chart
and
nine
monochromatic
side
views
were
530 I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540
Table
1
Measured
CIELAB
values
of
twelve
painted
panels-colors
evaluated
for
emotional
assessment
using
multi-angle
spectro-photo-meter
(X-rite,
MA68II).
Color
name
L*
a*
b*
c*
Light
green 80.99
35.38
19.74
40.52
Vivid
red 43.59
59.12
32.59
67.51
Black
12.46
0.38
2.2
2.24
Bright
reddish
orange 60.68
39.86
34.74
52.88
Yellowish
brown
49.77
25.53
42.21
49.34
White
93.55
0.53
1.23
1.34
Dull
greenish
blue
40.96
15.63
15.09
21.72
Deep
blue
green 32.29
35.1
4.33
35.37
Vivid
purple
33.11
26.99
24.59
36.51
Vivid
blue 39.6
3.88
43.96
44.13
Light
yellow
84.41
1.57
46.85
46.87
Vivid
violet
37.1
18.58
42.87
46.72
L*
represent
the
lightness
of
the
color.
a*
represent
its
position
between
red/magenta
and
green.
b*
represent
its
position
between
yellow
and
blue.
c*
represent
the
chroma.
chosen
for
the
main
experiment
from
the
fifteen
automobiles
used
in
questionnaire.
These
automobile
styles
included
three
com-
pact
cars,
two
sedans,
three
SUVs
and
one
sports
car,
as
shown
in
Fig.
2.
2.2.
Emotional
assessments
of
color
in
the
main
experiment
2.2.1.
Subjects
The
subjects
in
this
phase
of
the
study
were
composed
of
the
fol-
lowing:
students
from
Hong
Kong
Polytechnic
University
in
Hong
Kong,
who
were
21.8
±
1.30
years
old
(mean
±
SD,
n
=
50),
includ-
ing
29
males
(21.5
±
1.09
years
old)
and
21
females
(22.3
±
1.45
years
old);
students
from
Dong
hua
University
in
Shanghai,
who
were
23.6
±
1.58
years
old
(mean
±
SD,
n
=
54),
including
29
males
(23.8
±
2.02
years
old)
and
25
females
(23.2
±
0.78
years
old);
students
from
Chulalongkorn
University
in
Bangkok,
who
were
20.9
±
2.57
years
old
(n
=
66),
including
43
males
(20.8
±
2.85
years
old)
and
23
females
(21.0
±
1.97
years
old),
and
students
from
Kyoto
Institute
Technology
in
Kyoto,
who
were
21.5
±
1.63
years
old
(n
=
75),
including
54
males
(21.5
±
1.72
years
old)
and
21
females
(21.5
±
1.44
years
old).
Fig.
1.
Fifteen
pairs
of
emotion
variables
used
for
color
emotion
assessment
of
automotive
exteriors.
2.2.2.
Conditions
during
the
survey
of
color
assessments
Replicated
automobile
side-door
panels
in
12
colors
were
prepared
by
coating
them
with
commercial
automotive
paints
(specified
in
Table
1)
to
facilitate
the
visualization
of
automotive
exteriors.
Generally,
automobiles
are
viewed
in
the
open
air,
so
that
sunlight
irradiated
onto
the
exterior
color
is
reflected
on
the
curved
side
door,
producing
a
complex
pattern
composed
of
high-
light
and
shade.
The
panels
presented
in
the
experiment
simulate
the
shape
of
a
side-door
(30.5
cm
×
38
cm,
with
a
fold
line
in
the
center
section).
The
subjects
were
asked
to
specify
an
automo-
Fig.
2.
Monochromatic
automotive
shapes
used
for
color
assessment
of
automotive
exteriors.
Four
categories
were
used.
Compact
types
are
Nos.
1,
4
and
7.
Sedan
types
are
Nos.
6
and
9.
Sport
utility
vehicle
types
are
Nos.
2,
5
and
8.
The
sports
car
type
is
No.
3.
I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540 531
tive
monochromatic
side
shape
in
Fig.
2
and
then
to
answer
the
questionnaire.
The
painted
panels
were
hung
on
gray-colored
boards
at
a
height
of
1.5
m
from
the
ground.
This
was
done
to
reduce
the
influence
of
the
landscape
and
also
to
allow
the
subjects
to
see
comfortably
the
light
reflection
at
eye
level.
The
painted
panels
were
placed
outside
for
inspection
in
random
order
under
soft
and
stable
sunlight
from
31st
October
to
27th
December
2002.
2.2.3.
Questionnaire
The
observers
were
first
asked
to
judge
each
color
as
an
auto-
mobile’s
exterior
color,
using
the
labels
“like”
or
“dislike”.
They
were
then
requested
to
provide
a
reason
for
choosing
that
label
and
to
select
sensory
words
induced
by
each
painted
color
from
a
list
of
30
adjectives.
This
list
consisted
of
the
following:
bright,
dull,
fast,
pretty,
beautiful,
light,
cool,
heavy,
fresh,
calm,
sporty,
cheerful,
sharp,
dark,
gaudy,
shiny,
dirty,
trendy,
plain,
old,
vivid,
fashionable,
colorful,
up-to-date,
young,
interesting,
luxury,
slow,
feminine
and
out-of-date.
Next,
the
observers
were
asked
to
rate
fifteen
pairs
of
adjectives
on
a
seven-point
scale,
as
shown
in
Fig.
1.
Finally,
the
observers
were
asked
to
identify
attractive
painted
col-
ors
for
each
automotive
shape
by
referring
to
the
automobile
in
nine
monochromatic
side
shapes
(shown
in
Fig.
2).
The
experiments
were
conducted
at
four
locations
(in
Kyoto,
Hong
Kong,
Shanghai
and
Bangkok)
using
the
appropriate
local
language.
2.3.
Statistical
analysis
The
statistical
analysis
was
performed
using
an
SPSS
program
obtained
from
SPSS
Japan
Inc.
This
analysis
included
frequency
distributions,
Chi-square
tests,
residual
analysis
and
principal
com-
ponent
analysis.
The
results
are
corrected
with
respect
to
sampling
number,
as
the
proportion
of
automobile
categories
was
different.
3.
Results
and
discussion
3.1.
Color
preferences
for
automotive
exteriors
The
color
preferences
at
each
location
are
summarized
in
Table
2,
where
the
ranking
indicates
the
frequency
of
each
color
being
labeled
“like”.
The
most
preferred
color
in
all
locations
was
black.
White
and
vivid
red
were
also
popular,
whereas
bright
red-
dish
orange
and
yellowish
brown
were
the
least
popular
colors.
The
present
study
confirms
that
black,
red
and
white
are
favorite
colors
for
automotive
exteriors
at
all
locations.
These
constitute
the
basic
colors
for
automotive
exteriors.
Although
Saito’s
survey
using
colored
paper
chips
indicated
that
the
most
commonly
preferred
color
was
white,
the
present
study
indicates
that
black
is
the
most
preferred.
One
possible
reason
for
this
result
may
be
that
this
sur-
vey
was
strongly
associated
with
automotive
exteriors.
Black
has
historically
been
one
of
the
most
popular
colors
for
automotive
exteriors.
Because
basic
colors
are
frequently
observed,
people
are
accus-
tomed
to
linking
them
with
automotive
exteriors.
Reactions
to
non-basic
colors,
such
as
vivid
purple,
light
green,
dull
greenish
Table
2
Ranking
indicates
the
frequency
of
each
color
labeled
as
“like”:
percent
responding.
Color
Kyoto
(%)
Hong
Kong
(%)
Shanghai
(%)
Bangkok
(%)
Basic
color
Black 83.8
84
83
76.4
White
80
84
77.4
58.2
Vivid
Red
70.7
68
81.1
69.1
Non-basic
color
Vivid
blue
82.7
76
56.6
43.6
Dull
greenish
blue 64 52 34 41.8
Deep
blue
green 57.3
50
34
36.4
Vivid
violet
50.7
52
54.7
40
Light
yellow
48
50
43.4
36.4
Light
green
37.3
56
43.4
40
Vivid
purple
26.7
62
37.7
29.1
Yellowish
brown 25.3
10 13.2
18.2
Bright
reddish
orange
21.3
10
22.6
23.6
blue
and
deep
blue
green,
exhibit
geographical
differences
in
pref-
erence.
Although
there
may
be
various
reasons
for
this
difference
in
color
preference,
two
main
factors
should
be
considered:
sex
dif-
ference
and
the
characteristics
of
color
emotion
for
each
country.
Some
colors
for
automotive
exteriors
might
be
intentionally
chosen
for
these
reasons.
To
elucidate
these
sex
and
geographical
differences,
several
statistical
analyses
were
performed.
Initially,
a
Chi-square
test
was
applied
to
male
and
female
responses
to
evaluate
regional
differences
in
the
tendency
to
like
or
dislike
each
color.
The
Chi-
square
values
indicate
regional
differences
regarding
bright
reddish
orange,
white,
dull
greenish
blue,
vivid
purple
and
vivid
blue
in
the
case
of
males,
and
for
vivid
purple
and
vivid
blue
in
the
case
of
females,
as
shown
in
Table
3.
Those
differences
in
color
preference
were
further
investigated
through
residual
analysis
as
summarized
in
Table
4.
This
test
analyzes
the
difference
between
the
value
mea-
sured
in
the
experiment
and
the
theoretical,
or
true,
value.
White,
a
common
color
for
automotive
exteriors,
was
liked
by
a
majority
at
all
locations,
especially
in
Kyoto,
because
it
is
regarded
as
“pure”
and
“clean”
according
to
the
responses.
However,
it
is
also
disliked
by
many
Bangkok
males
because
it
is
easily
soiled
and
gives
an
automobile
the
appearance
of
a
commercial
vehicle.
These
reac-
tions
account
for
the
regional
difference
in
color
preference
judged
by
the
2test.
White
is
a
favorite
color
in
general
and
is
frequently
observed
as
an
automotive
exterior;
accordingly,
evaluations
of
its
suitability
as
an
automotive
exterior
color
are
divided.
In
Thai,
“dis-
like”
and
“unsuitable”
are
often
mixed
in
responses;
therefore,
the
regional
difference
in
the
preferences
for
white
is
due
to
more
prac-
tical
reasons.
Bright
reddish
orange
is
mostly
disliked
in
Hong
Kong
and
Kyoto,
but
less
so
in
Shanghai,
and
is
even
liked
in
Bangkok.
Vivid
blue
is
a
favorite
color
in
Kyoto
and
Hong
Kong
but
is
less
favored
in
Shanghai
and
Bangkok.
These
two
colors
exhibit
a
sim-
ilar
pattern
of
like/dislike
with
respect
to
location.
Cooler
colors
are
preferred
in
Kyoto
and
Hong
Kong,
whereas
warmer
colors
are
preferred
in
Shanghai
and
Bangkok.
Dull
greenish
blue
is
a
favorite
color
in
Kyoto
and
regarded
neutrally
in
Hong
Kong,
but
is
disliked
in
Shanghai
and
Bangkok.
The
regional
differences
for
Table
3
2test
results
for
likes
and
dislikes
at
four
locations.
The
p-value
under
0.05
is
indicated
in
bold
face.
Light
green
Vivid
red
Black
Bright
reddish
orange
Yellowish
brown
White
Male
(p-value)
0.364
0.464
0.446
0.023
0.118
0.008
Female
(p-value) 0.607
0.585
0.453
0.923
0.565
0.16
Dull
greenish
blue
Deep
blue
green
Vivid
purple
Vivid
blue
Light
yellow
Vivid
violet
Male
(p-value)
0.038
0.073
0.030
0.010
0.059
0.467
Female
(p-value) 0.171
0.119
0.013
0.015
0.934
0.261
532 I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540
Table
4
Results
of
residual
analysis
comparing
frequency
of
likes
and
dislikes
of
males
at
four
locations.
Bright
reddish
orange
Dislike
Like
Total
White
Dislike
Like
Total
Kyoto
Frequency
43
11
54
Kyoto
Frequency
8
46
54
Adjusted
residual 0 0
Adjusted
residual
2
2
Hong
Kong Frequency
28
1
29
Hong
Kong
Frequency
4
25
29
Adjusted
residual
2.6
2.6
Adjusted
residual
1.4
1.4
Shanghai
Frequency
21
7
28
Shanghai
Frequency
9
19
28
Adjusted
residual
0.7
0.7
Adjusted
residual
1.2
1.2
Bangkok
Frequency
13
8
21
Bangkok
Frequency
10
11
21
Adjusted
residual 2.2
2.2
Adjusted
residual 2.8
2.8
Total
Frequency
106
27
132
Total
Frequency
31
101
132
Dull
greenish
blue
Dislike
Like
Total
Vivid
purple
Dislike
Like
Total
Kyoto
Frequency
20
34
54
Kyoto
Frequency
41
13
54
Adjusted
residual
2.6
2.6
Adjusted
residual
1.7
1.7
Hong
Kong
Frequency
15
14
29
Hong
Kong
Frequency
14
15
29
Adjusted
residual 0.1
0.1
Adjusted
residual
2.5
2.5
Shanghai
Frequency
19
9
28
Shanghai
Frequency
17
11
28
Adjusted
residual
2
2
Adjusted
residual
0.9
0.9
Bangkok Frequency 13 8 21 Bangkok Frequency
17
4
21
Adjusted
residual
1.1
1.1
Adjusted
residual
1.4
1.4
Total
Frequency
67
65
132
Total
Frequency
89
43
132
males
are
exhibited
especially
in
the
preference
of
white,
bright
reddish
orange
and
dull
greenish
blue.
Color
preference
tendencies
are
similar,
but
less
marked,
for
females.
An
interesting
color
is
vivid
purple,
which
is
liked
in
Hong
Kong
but
disliked
everywhere
else.
With
regard
to
vivid
purple
in
particular,
Saito’s
survey
con-
firmed
that
it
is
a
favorite
color
in
Taiwan
because
it
is
regarded
there
as
an
elegant
and
mystical
color
[2].
However,
it
gives
a
neg-
ative
impression
as
being
“cheap”
or
“indecent”
in
Japan.
However,
many
Japanese
people
accept
vivid
purple
as
a
fashionable
color.
The
present
survey
shows
that
vivid
purple
is
a
favorite
color
for
automotive
exteriors
in
Hong
Kong
because
it
is
a
mystical,
noble
and
feminine
color
but
is
disliked
in
Kyoto
because
it
is
perceived
there
as
too
loud
and
crass.
Geographical
differences
in
color
preference
may
reflect
the
social
image
of
automobiles
in
each
location.
The
population
per
passenger
car
in
the
year
1998
was
2.5
in
Japan,
20.2
in
Hong
Kong,
29.8
in
Thailand
and
444.5
in
mainland
China
[13].
In
Japan
and
Hong
Kong,
the
automobile
is
considered
a
life
necessity
that
har-
monizes
with
peoples’
lifestyles.
Consequently,
the
youth
in
these
areas
are
familiar
with
the
newest
models
and
have
no
hesitation
in
accepting
new
colors,
such
as
vivid
blue
or
dull
greenish
blue.
These
youth
are
capable
of
imagining
any
automobile
and
choosing
a
color
for
its
exterior.
In
contrast,
the
automobile
is
still
a
symbol
of
a
social
status
in
Shanghai,
and
the
youth
there
cannot
afford
to
purchase
one.
We
discuss
later
the
image
of
the
automobile
in
each
location
through
color
emotion
analysis.
At
the
time
of
this
survey
(around
2000),
high-class
recreational
vehicles
were
very
popular
in
Japan,
and
color
designs
for
private
compact
cars
were
becoming
fashion
driven.
Lifestyle
changes
in
Japan
during
the
1990s
ignited
color
designs
of
various
scenes
and
promoted
the
development
of
paint
materials
and
coating
technol-
ogy.
Life
has
become
a
full
of
color
and
people
enjoy
the
choice
of
colors.
The
market
also
requires
the
availability
of
colors
associ-
ated
with
high-class
goods
as
well
as
high-quality
coating.
These
trends
are
reflected
in
the
color
preferences
for
automotive
exteri-
ors
in
Japan,
where
people
accept
such
new
colors
as
vivid
blue
and
dull
greenish
blue.
The
automobile
market
in
Thailand
was
expanding
as
a
result
of
the
rapid
economic
growth
that
had
been
ongoing
since
the
1980s.
The
introduction
of
a
preferential
tax
system
in
rural
areas
for
the
transportation
of
agricultural
prod-
ucts
stimulated
the
demand
for
pick-up
tracks.
Thus,
sales
volume
for
pick-up
trucks
was
56.8%
in
2001,
whereas
that
for
sedans
was
35.2%
[14].
The
exterior
color
of
private
sedans
in
Thailand
is
mostly
a
monotone
of
dark
red,
blue
or
green,
chosen
in
order
to
distinguish
such
vehicles
from
the
high-chromatic
body
col-
ors
that
characterize
commercial
vehicles,
including
taxis
[15].
This
tendency
confirms
that
Thais
are
relatively
conservative
with
respect
to
their
choice
of
color
for
automotive
exteriors.
The
sit-
uation
was
similar
in
China,
where
the
black
sedan
has
been
a
symbol
of
social
power.
Sedans
used
as
taxis
or
for
business
pur-
poses
were
commonly
the
only
cars
to
be
seen
in
China,
as
the
general
public
had
no
opportunity
to
own
a
private
automobile.
Although
residents
of
Hong
Kong
could
potentially
own
private
cars,
the
government
introduced
a
law
to
restrict
the
total
number
of
automobiles
as
a
solution
to
traffic
jams
and
air
pollution.
The
automobile
registration
tax
is
high
(100%),
and
gasoline
and
park-
ing
are
expensive.
However,
Hong
Kong’s
youth
are
well-informed
regarding
the
newest
car
models
by
means
of
Internet
and
satel-
lite
broadcasting
and
hope
to
possess
a
car
in
future.
As
discussed
below,
these
social
environments
may
influence
the
choice
of
auto-
mobile
exterior
colors.
3.2.
Emotion
variables
with
respect
to
automobile
exterior
colors
We
also
took
into
account
differences
in
color
emotion
for
each
location
to
elucidate
the
causes
of
likes
and
dislikes
for
automo-
bile
exterior
colors.
We
performed
a
Semantic
Differential
analysis
(hereinafter
referred
to
as
the
SD
method)
for
the
15
pairs
of
adjec-
tives
and
analyzed
them
according
to
standard
statistical
methods.
Fig.
3
shows
the
SD
profiles
of
the
average
evaluation
for
each
pair
of
emotion
variables.
3.2.1.
The
average
profiles
of
the
SD
evaluation
Fig.
3
shows
the
line-plot
profiles
of
average
SD
points.
There
is
little
recognizable
regional
difference
in
emotional
responses
to
such
basic
colors
as
black,
white
and
red.
However,
opposing
responses
were
observed
regarding
those
colors
in
Bangkok
and
Hong
Kong.
Black
was
considered
“dull”
and
white
deemed
“plain”
in
Bangkok,
whereas
black
was
“sharp”
and
white
was
“gaudy”
in
Hong
Kong.
Color
impressions
were
neutral
in
this
respect
in
other
locations.
Vivid
red
evoked
a
“settled”
impression
in
Bangkok,
although
it
was
deemed
“unsettled”
in
other
cities.
Non-basic
colors
exhibited
more
extensive
regional
variations
in
terms
of
color
emotions.
In
comparison
with
other
locations,
vivid
blue
elicited
a
relatively
neutral
impression
in
Bangkok.
Vivid
blue
was
considered
“light”
and
“sharp”
everywhere
but
in
Bangkok,
I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540 533
Black
f
ou
b
m
White
ou
b
m
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
ut of date-up to date
ad taste-good taste
cheap-luxury
melancholic-cheerful
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
ut of date-up to date
bad taste-good taste
cheap-luxury
melancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Vivid Red
fe
out
ba
me
Vivid blue
3
eminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
t of date-up to date
ad taste-good taste
cheap-luxury
elancholic-cheerful
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
out of date-up to date
bad taste-good taste
cheap-luxury
melancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Dull Greenish Blue
fe
out
ba
me
Deep Blue Green
o
b
m
3
eminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
t of date-up to date
ad taste-good taste
cheap-luxury
elancholic-cheerful
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
ut of date-up to date
bad taste-good taste
cheap-luxury
melancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Vivid Purple
fe
out
ba
me
Vivid Violet
fe
out
ba
me
3
eminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
of date-up to date
ad taste-good taste
cheap-luxury
elancholic-cheerful
3
eminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
of date-up to date
d taste-good taste
cheap-luxury
elancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Fig.
3.
SD
profiles
of
evaluation
averages
for
twelve
colors.
where
it
was
deemed
“heavy”
and
“dull”.
People
in
Shanghai,
Hong
Kong
and
Kyoto
exhibit
common
responses
to
vivid
blue,
although
people
in
Shanghai
are
less
sensitive
with
respect
to
the
“feminine–masculine”
and
“ugly–pretty”
dichotomies.
People
in
Hong
Kong
and
Shanghai
reacts
similarly
to
vivid
pur-
ple
and
vivid
violet
and
do
so
quite
differently
than
do
people
in
Kyoto
and
Bangkok.
The
color
emotion
associated
with
vivid
purple,
for
example,
is
described
in
the
terms
“pretty”,
“cheerful”,
“luxury”,
“good
taste”,
“up
to
date”,
“decent”,
“sharp”
and
“light”
in
Hong
Kong
and
Shanghai,
but
it
evokes
the
opposite
emotion
in
Kyoto
and
Bangkok.
Vivid
violet
generally
evokes
a
similar
color
emotion
across
these
countries,
but
the
emotional
reaction
in
Kyoto
and
534 I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540
Light Green
o
b
Light Yellow
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
ut of date-up to date
ad taste-good taste
cheap-luxury
melancholic-cheerful
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
out of date-up to date
bad taste-good taste
cheap-luxury
melancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Bright Reddish Orange
ou
b
m
Yellowish Brown
f
ou
b
m
3
feminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
t of date-up to date
ad taste-good taste
cheap-luxury
elancholic-cheerful
3
eminine-masculine
ugly-pretty
young-old
gaudy-plain
dark-bright
dull-sharp
fast-slow
light-heavy
indecent-decent
like-dislike
settled-unsettled
t of date-up to date
ad taste-good taste
cheap-luxury
elancholic-cheerful
-1
0
1
2
-1
0
1
2
-3
-2
-3
-2
Fig.
3.
(Continued
).
Bangkok
is
less
marked.
The
purplish
color
in
Shanghai
may
have
a
special
meaning
that
unconsciously
affects
its
valuation.
Dull
greenish
blue
and
deep
blue
green
elicit
similar
emotional
tendencies
in
each
location,
but
the
residents
of
Hong
Kong
dislike
both
colors.
Deep
blue
green
is
described
as
“plain”
and
“dark”
in
Hong
Kong,
but
dull
greenish
blue
evokes
no
significantly
different
color
emotion
than
in
other
cities.
The
reasons
for
disliking
any
particular
color
are,
therefore,
not
clear.
Light
green
exhibits
a
similar
pattern
for
color
emotion
in
respect
to
location.
People
in
Shanghai
react
favorably
to
light
yel-
low,
in
contrast
to
other
locations.
Light
yellow
is
regarded
as
“up
to
date”
and
in
“good
taste”
in
Shanghai.
Light
green
and
light
yel-
low
are
disliked
in
Kyoto
and
Bangkok,
respectively.
Light
green
gives
an
impression
of
slowness
only
in
Kyoto,
but
the
reason
why
light
yellow
is
disliked
in
Bangkok
is
unclear.
Yellow
most
likely
has
a
special
meaning
in
Thailand
and
is
not
considered
suitable
for
automotive
exteriors.
Bright
reddish
orange
was
disliked
in
all
loca-
tions
but
was
assigned
the
values
of
good
taste
and
cheerfulness
in
Shanghai.
Yellowish
brown
is
also
disliked
in
all
four
locations.
It
is
regarded
negatively
as
“slow”,
“old”,
“plain”,
“heavy”,
“unsettled”,
“out
of
date”,
“cheap”
and
“melancholic”.
As
a
whole,
the
color
emo-
tion
reaction
is
less
sensitive
in
Shanghai.
A
sense
of
speed
affects,
in
general,
the
criteria
for
liking
or
disliking
colors,
except
for
new
colors
(including
light
green,
light
yellow
and
vivid
purple
in
some
locations).
Here,
criteria
could
depend
on
local
traditions
and
color
emotion.
3.2.2.
Automotive
exterior
colors
characteristic
of
different
automobile
categories
As
seen
from
the
preceding
discussion,
color
taste
for
automo-
bile
exteriors
depends
on
various
factors,
including
environments
and
traditional
color
emotions.
The
automobile
signals
its
function
and
image
through
an
intentional
design.
The
color
taste
of
youth
also
depends,
accordingly,
on
the
styling
and
function
of
the
auto-
mobile,
but
reactions
could
be
different
from
that
intended
by
the
design.
In
other
words,
the
image
of
an
automobile
might
be
seen
through
a
color
emotion
that
deems
particular
exterior
colors
as
suitable
for
a
specific
style
of
automobile.
Fig.
4
shows
the
average
values
of
the
SD
evaluation
for
each
sensory
word
for
four
types
of
the
automobile
in
Kyoto,
Hong
Kong,
Shanghai
and
Bangkok.
Each
pair
of
sensory
words
was
translated
into
an
appropriate,
single-imaging
word.
In
the
case
of
the
sedan,
the
image
of
the
automobile
is
“official”
or
“business”.
In
Japan,
the
image
for
the
sedan
is
rather
negative,
being
deemed
as
“ugly”,
“old”,
“dark”,
“dull”,
“heavy”
and
“cheap”.
In
contrast,
its
image
is
quite
positive
in
Hong
Kong
and
particularly
so
in
Shanghai.
The
sedan
represents
social
status,
being
viewed
as
a
vehicle
for
a
high
government
official
in
Shanghai
or
for
a
member
of
the
upper
class
in
Hong
Kong.
In
Japan,
it
is
associated
with
taxis
or
Japanese
mafiosos.
Thai
youth
generally
react
to
the
sedan
as
being
“femi-
nine”
and
“settled”.
This
tendency
is
also
observed
regarding
sports
cars
and
SUVs.
That
is,
for
Thai
youth,
the
automobile
itself
should
be
feminine
and
settled.
The
compact
car
elicits
a
characteristic
pattern
of
color
emotion
in
each
country.
Shanghai
youth
seem
to
favor
the
compact
car
as
being
“young”,
“fast”,
“light”,
and
“feminine”.
The
negative
points
are
“unsettled”
and
relatively
“dull”.
In
contrast,
Thai
youth
are
not
favorably
disposed
to
compact
cars
and
seem
to
feel
that
they
are
“settled”.
The
reaction
of
the
Japanese
and
Hong
Kong
youth
is
sim-
ilar.
The
compact
car
is
“cheerful”,
“light”
and
“young”.
However,
the
compact
car
is
regarded
as
less
“fast”
in
Kyoto.
Because
compact
cars
were
not
popular
(except
in
Japan)
at
the
time
of
this
survey,
its
image
could
have
been
significantly
influenced
by
the
negative
views
of
sedans
in
Shanghai
and
Bangkok.
I.
Satake
et
al.
/
Progress
in
Organic
Coatings
72 (2011) 528–
540 535
Fig.
4.
Image
of
each
car
category
at
four
locations.
Thai
youth
react
with
more
consistency
to
SUVs
compared
to
young
people
in
other
locations.
The
SUV
is
“feminine”
and
“settled”
in
Bangkok,
but
“masculine”
and
“unsettled”
in
Kyoto.
The
image
of
the
SUV
in
Thailand
could
be
affected
by
its
differentiation
from
commercial
pick-up
trucks,
which
is
the
most
prevalent
type
of
vehicle
in
Bangkok
and
its
urban
areas.
3.2.3.
The
thirty
sensory
words
with
respect
to
the
automobile
exterior
colors
Fifteen
pairs
of
sensory
words
were
chosen
to
evaluate
auto-
motive
exterior
colors.
The
results
were
reduced
to
a
few
groups
of
sensory
words
by
using
principal
component
analysis.
This
analysis
was
performed
to
deduce
sensory
characteristics
in
each
coun-
try
and
to
determine
color
preferences.
Here,
principal
component
analysis
provides
the
measure
for
the
weight
of
the
significance
of
information
in
terms
of
eigenvalues
and
eigenvectors.
The
screeplot
of
the
principal
component
score
for
emotion
variables
by
the
eigenvalues
indicates
that
the
structure
was
mostly
composed
of
three
principal