U4102167175

Kakuro Amasaka et al Int. Journal of Engineering Research and Applications
ISSN : 2248-9622, Vol. 4, Issue 1( Version 2), January 2014, pp.167-175
４１０１２１５２３

RESEARCH ARTICLE

www.ijera.com

OPEN ACCESS

Attention-Grabbing Train Car Advertisements
Motoi Ogura1, Takayuki Hachiya2, Kenta Masubuchi3, Kakuro Amasaka4
Graduate School of Science and Engineering Aoyama Gakuin University 5-10-1, Fuchinobe, Chuo-ku,
Sagamihara-shi, Kanagawa, 252-5258 JAPAN

Abstract
Younger people have changed the way they access information in recent years, increasing the number of
opportunities to take advantage of transit advertising. Train car advertising in particular boasts a high contact rate
and extended length of contact. This study seeks to understand the correlations between passenger information
and riding conditions when it comes to train car advertising (hanging posters, above-window posters, and sticker
ads) in order to discover the ideal way to advertise inside passenger trains. Specifically, the study first tries to find
the correlations among three factors: (1) whether or not passengers pay attention to train car advertisements, (2)
basic passenger information (age and gender), and (3) riding conditions (riding time, average number of trips,
etc.). Once data on riding conditions is collected, it is then grouped using a cluster analysis and Quantification
Theory Type III in order to establish the relationship of the data to passenger attention towards train
advertisements once again. Then, in order to conduct an in-depth analysis, a Categorical Automatic Interaction
Detector (CAID) analysis is done on each group, using passengers pay attention to train advertisements as the
objective variable and basic passenger information as the explanatory variable. The results of the CAID analysis
are then used to study the format of existing and future train car advertisements.
Keywords: Train Car Advertisements, AIDA model, CAID analysis

I.

Introduction

Younger adults, particularly men in their 20s,
have been changing the way they access information
and participate in city life. Because they are so
frequently on the move for their jobs, there are more
opportunities for them to come in contact with forms
of outdoor media. Longer times spent traveling and on
the internet have been shown to decentralize media
contact. Essentially, transit ads are a more effective
form of advertising these days since they have a high
contract rate among people with hectic modern
schedules.
This study deals with the in-car train
advertisements that have been garnering attention in
recent years with research that focuses on attention:
the first stage in the AIDA model (A : Attention, I :
Interest, D : Desire, A : Action) of consumer behavior.
The goal is to first quantify the way passengers pay
attention to train car advertisements, and then propose
the ideal form that in-car train advertising should take
based on a visual representation of passenger
information.
Specifically, the study first tries to find the
correlations among three factors: (1) whether or not
passengers pay attention to train car advertisements,
(2) basic passenger information (age and gender), and
(3) riding conditions (riding time, average number of
trips, etc.). Once data on riding conditions is collected,
it is then grouped using a cluster analysis and
Quantification Theory Type III in order to establish
www.ijera.com

the relationship of the data to passenger attention
towards train advertisements once again.
Then, in order to conduct an in-depth
analysis, a Categorical Automatic Interaction Detector
(CAID) analysis is done on each group, setting
whether passengers pay attention to train
advertisements as the objective variable and basic
passenger information as the explanatory variable.
The results of the CAID analysis are then used to
study the format of existing and future train car
advertisements.

II.

Background

2.1 The advertising industry: Current status and
issues
Over the last ten years, Japan has consistently
spent around 6.5 trillion yen on advertising. The
companies have no choice but to spend money on
advertising, and are always looking for a format
whose performance will justify their significant
investment (Dentsu, 2011). The kinds of advertising
that companies use to target consumers can be divided
into three major categories: (1) mass media
advertising, including TV commercials and
newspaper ads; (2) direct mail (DM) advertising; and
(3) newer forms of advertising media, such as
advertising on public transportation and the internet.
In terms of money spent on each category,
mass media advertising accounts for as much as
two-thirds of company advertising budgets (Dentsu,
2011). It is thought that companies blindly spend
167 | P a g e

massive amounts of money on mass media advertising
in order to get consumers’ attention—which they
believe is the only function of their promotional
efforts. What is needed is for companies to
quantitatively grasp the amount of publicity generated
by different forms of advertising media (Amasaka,
2005; Amasaka, 2007; Amasaka, 2010).
2.2 The study of advertising performance
In the past, many researchers have studied
consumer response patterns as a way of quantitatively
grasping advertising performance (Amasaka, 2009).
Numerous consumer response models, which predict
consumer behavior from the time they learn about a
new product through an ad until the time they
purchase it, have been offered. The most well-known
of these is probably the AIDA (Attention, Interest,
Desire, Action) model, which outlines a four-stage
consumer response pattern.
This model has been further developed
through applied research in recent years, with
alternative models such as the AIDMA (Attention,
Interest, Desire, Motive, Action) and AISAS
(Attention, Interest, Search, Action, Share) also being
proposed. Despite these variations, the one thing that
all of these models have in common is that the initial
step in consumer purchase behavior is always the
attention phase. Because of this, attention can be
considered the most critical element in the process
whereby consumers move towards actually making a
purchase (Ferrell and Hartline, 2005; Mehta, 2000;
Shimizu, 2004).
2.3 Transit advertisements: Current status
The number transit advertising opportunities

www.ijera.com

has increased in recent years. One of the major
characteristics of this form of ads is an extended
contact period. As Figure 1 shows, railway usage
peaks in the morning and evening. Between these two
periods, in the hours following 12 noon, train usage
and shopping activities follow essentially the same
line.
What this means is that the form of ads
consumers are most likely to come in contact with
while they are shopping is transit advertising, a
situation that creates major opportunities for these
advertisements to lead to purchases. Three key
characteristics of transit advertising are listed below
(The Kanto Transit Advertising Council, 2009).
1. For male and female teenagers (ages 10–19), group
1 men (age 30–34), and group 2 men (age 35–49),
transit advertising is a powerful form of media on par
with traditional mass-media advertising in terms of
having a good contact rate, long-term contact, and a
frequent contact rate.
2. After television, transit advertising scores higher
than all other forms of media in terms of natural,
habitual viewing, enjoyable viewing, generating
surprise or interest in viewers, and other positive
characteristics. For this reason, transit advertisements
have a powerful ability to draw consumers in, making
it difficult for viewers to overlook them.
3. Because they make appeals to consumers while they
are out of the house, transit advertisements are known
for having a powerful link to outside purchasing
behavior. The third most common source of new
product information after TV and shop displays,
transit advertising is a form of media that positions
itself
within
everyday
situations
and
locations—including buying situations.

Figure 1 The media contact situation according to the time on weekdays
www.ijera.com

168 | P a g e

The above characteristics give transit
advertising enormous potential when it comes to
influencing purchase behavior. This study takes the
insights gained from previous advertising research and
uses it to lay the foundation for future inquiries on this
important topic.
2.4 Prior research on transit advertisements
Even within the category of transit
advertising, in-car train advertisements are a
powerfully influential form of media that is placed
right in front of viewers for a certain period of
time—whether or not it captures their interest. Still,
the authors are not aware of that many studies that
have been done on train advertising relative to the
amount of research done on TV or newspaper
advertising.
Two examples of prior research on transit
advertising follow. The first study, A Look at
Subconscious Awareness in Train Car Advertising by
Gonda (2004), focused on the subconscious and saw
train advertising as a locally based form of media that
was both impactful and could be customized to suit
individual train lines or regions. Gonda felt that in-car
train advertising was a form of information that
improved the quality of life in local areas, and studied
the role of transit advertising as a conduit of local
information for passengers.
A second study by Takenaka and Harada
(2003) entitled A Study on Watching Professional
Baseball and Information Contact via Transit
Advertisements, focused on transit advertisements
that promoted professional baseball games, grouping
subjects into segments based on how frequently they
rode trains and looking at differences in the level of
information contact among the groups. The study then
examined how those differences affected game

www.ijera.com

watching.
The studies above did not look statistically at
the correlations between passenger information and
riding conditions. Noting this gap, the authors of this
study decided to examine those relationships and how
they relate to train car advertising with the aim of
defining the ideal format for this type of media.

III.

Proposing transit advertisements for
boosting ad awareness of passengers

3.1 Clarifying the relationships between basic
passenger information and riding conditions
As mentioned in the previous chapters, the
purpose of this study was to examine the correlations
between passenger information and riding conditions
in train car advertising in order to discover the ideal
way to advertise inside passenger trains. For the
purpose of this study, in-train advertisements were
divided into three different types based on external
appearance.
The
types
(No1.hanging
posters,
No2.above-window posters, and No6.stickers placed
to the side of doors) are shown in Figure 2. The
number of trains featuring digital advertisements is
increasing, but the authors chose to limit their
selection to these three based on an examination by a
professional market survey company, which indicated
that these ads are used on all major railways in the
Kanto region, marking them as representative forms of
media on every train line.
3.1.1 Field survey
In order to suggest ideal positioning for train
advertisements, the authors teamed up with the market
survey company mentioned above and a related
committee to produce a survey to collect information
from actual passengers. The interview-style survey

Figure 2 The type of in-train advertisements
www.ijera.com

169 | P a g e

was conducted multiple times on two urban train lines
(Line A and Line B) between late December 2005 and
early January 2006.
The number of valid response was 50 people
(26 were men and 24 were women ranged from 15 to
54 years). The survey asked passengers a series of
multiple-choice questions designed collect the basic
information needed to construct personal profiles of
the test subjects. Respondents were asked to select
their age, gender, time spent riding trains, and the
average number of weekly trips they made. Subjects
were also shown a diagram of a train interior and
asked to point out where they were in the train, what
position they took when riding, and whether they were
aware of the train advertisements (hanging posters,
above-window posters, and stickers to the sides of
doors).
The authors looked at the format of
questionnaires conducted by the marketing survey
company for reference (The Kanto Transit
Advertising Council, 2006). The data they collected
was analyzed and closely examined in order to
pinpoint trends that affected passenger’s attention
towards hanging posters, above-window posters, and
stickers. For each type of advertisements, subjects
were asked to respond in one of three ways: ―I saw it‖,
―I think I saw it‖, and ―I didn’t see it‖. The results
were used to assign an attention ratio to each form of
advertising by taking the total number of ―I saw it‖
and ―I think I saw it‖ responses as a percentage of the
total. The authors defined this value as the ―attention
rate‖ for that form of media.
3.2 Visualizing causal relationships between basic
passenger information and riding conditions using
multivariate analysis
The survey data used to examine the current
state of train car advertising in the previous section
was then subjected to a statistical analysis in order to
visually represent the correlations between passenger
information and riding conditions. The procedure and
goals of this analysis follow.
1. Look at overall trends in passenger information
using a cross tabulation method that focuses on
whether passenger attention turned to hanging posters,
above-window posters, or sticker advertisements.
2. Perform a cluster analysis on riding conditions and
group the results using Quantification Theory Type
III. Then, look at the relationship between (1) the
riding conditions grouped in the cross tabulation and
(2) whether passenger attention turned to hanging
posters,
above-window posters,
or
sticker
advertisements.
3. Run a CAID analysis on the grouped riding
condition data and basic passenger information to
determine how it relates to attention rates established
for the three types of advertising.
www.ijera.com

www.ijera.com

3.2.1 Visualizing causal relationships between
basic passenger information and ad awareness of
passengers (Analysis step 1)
In step 1 of the analysis, the authors
performed a cross tabulation on the survey data
collected in a previous section and researched the
correlations between whether passengers notice each
form of train car advertising (hanging, above-window,
and sticker) and passenger information (age and
gender) in order see how passenger information
relates to attention rates. As part of this process, the
age information from the surveys was grouped into
young (15–25), intermediate (26–40), and older (41
and up) age brackets.
Table 1 shows the results of the cross
tabulation for passenger information (age and gender)
and attention rates for each form of train car
advertising. The analysis revealed that younger
passengers looked at the hanging posters, while the
attention rate for older passengers was low. The
results were reversed for advertisements placed above
train windows, with older passengers paying close
attention to them. Sticker advertisements tended to be
noticed by women, and younger passengers in general
also tended to look at sticker advertisements.

Table 1 The cross tabulation for passenger
information and attention rates for each form of train
car ad
riding conditions and ad awareness of passengers
(Analysis step 2)
In this step, the authors used the survey data
gathered to represent current in-train advertising
conditions, subjected it to a cluster analysis, and then
used Quantification Theory Type III to group riding
conditions. They then looked for correlations between
the resulting data and whether passengers paid
attention to different types of train advertising.
The binary data on riding time (0–15 minutes,
16–45 minutes, 46 minutes or more), average number
of weekly trips (Less than two, two or three, four or
more), and riding position (standing or sitting) that
passengers provided during the survey was subjected

170 | P a g e


www.ijera.com

Figure 3 Cluster analysis of the ride condition

Figure 4 Quantification Theory Type III of the ride condition
to a Euclidean distance cluster analysis. The results of
this analysis are shown in Figure 3. It was found that
the eight types indicated in this dendrogram could be
separated into two groups.
www.ijera.com

Similar results were obtained when the
Quantification Theory Type III was used on the survey
data (Figure 4). Since the first group consisted of
standing passengers who ride the train for 0–15
171 | P a g e

minutes, they were labeled ―short-distance
passengers‖. The second group rode the train for a
longer period of time and tended to sit, so this group
was called the ―long-distance passengers‖.

Table 2 The cross tabulation for passenger
information and attention rates for each form of train
car ad after grouping
The passengers in each group were then
summarized, and this data was subjected to a cross
tabulation with awareness rate for hanging posters,
above-window posters, and sticker advertisements to
find the correlations (Table 2). The highest awareness
rates for the short-distance passenger group were on
hanging advertisements, followed by above-window
and sticker ads.
These short-term riders had higher awareness
rates overall than those in the long-distance group,
who showed similar results in that they were most
aware of hanging posters, followed by above-window
posters and stickers. They paid less attention to ads
overall than their short-term passenger counterparts.

www.ijera.com

basic passenger information, riding conditions and
ad awareness of passengers (Analysis step 3)
The
authors
then
examined
the
characteristics of the short-term and long-term
passenger groups defined in the previous section,
conducting a CAID analysis (Murayama, 1982) in
order to understand in-depth correlations involving
passenger information. The CAID analysis is to search
which variable is strongly correlated with criterion
variable on the basis of simultaneous distribution of
frequency data, and divide a sample population by the
category of the related high criterion variable.
Specifically, a CAID analysis was conducted
for each form of advertising (hanging, above-window,
and stickers next to doors) as it related to the two
groups discussed in the last section. The data collected
in the survey was used for the analysis, with
awareness rates for the different forms of advertising
used as criterion variables. Passenger information (age
and gender) was used for the explanatory variables. In
this way, the relationship between on-board riding
conditions and passenger information was analyzed.
Figure 5 shows some of the results from the
analysis. These are the results of the awareness rate for
hanging posters among short-term riders is used

Figure 5 Quantification Theory Type III of the ride condition

www.ijera.com

172 | P a g e


www.ijera.com

Table 3 The results of the CAID analysis on the short-term passenger group

as a criterion variable using the CAID analysis. The
results indicate that the highest awareness rate in this
group is among men in the youngest age category
(15–25). Table 3 shows the results of the CAID
analysis on the short-term passenger group.
A CAID analysis like the one in Table 3 was
similarly conducted for each type of media among the
long-term passenger group, with passenger
information designated as the explanatory variable.
Barring a few exceptions, the results indicate an
overall trend of awareness rates being higher among
the younger generations. Comparing the two groups,
the authors found that the short-term passengers (who
tended to stand when riding) had higher awareness
rates in general.
Passengers who sat, on the other hand, had
more opportunities to engage in different activities
during their ride, such as reading or doing work, which
probably contributed to their paying less attention to
advertisements than the passengers who were standing.
The analysis revealed that women passengers 26 and
older in particular did not look at in-car
advertisements. It also indicated that older passengers

www.ijera.com

frequently paid attention to advertisements located
above windows.

IV.

Proposing train car advertisements

Based on the results from the previous
section, the authors attempted to provide specifics on
an ideal format for train car advertisements. Figure 6
uses a picture to indicate the results of the CAID
analysis in a way that is easy to understand.
Awareness rates for hanging posters were high among
younger passengers and standing passengers, and the
difference between men and women was slight.
Both male and female members of the older
age group frequently paid attention to above-window
posters. Men 41 and over who sat while riding the
train, as well as women who stood while riding,
tended to view these advertisements the most.
Awareness rates for sticker advertisements were low
overall; however, younger riders frequently looked at
them, as did standing passengers. To summarize, train
car advertisements are most likely to catch the eye of
standing passengers, and the younger generation tends
to look at them the most.

173 | P a g e


www.ijera.com

Figure 6 Visualizing causal relationships between basic passenger information, riding conditions and ad
awareness of passengers
Based on these conclusions, the authors’ first
recommendation is for existing train car advertising.
Because hanging posters, stickers, and other in-car
advertisements are likely to attract younger riders, this
space should be used to advertise weekly manga
magazines, fashion magazines, sales, or other
products likely to appeal to this generation. Another
important consideration is using popular celebrities to
catch the eye of these passengers.
Above-window advertising space, on the
other hand, may be better used to appeal to those of the
older generation. These passengers are more likely to
be married and have children, so it may be beneficial
to feature family-friendly topics. Specifically, posters
advertising events for families or travel may be ideal
in this location. Also, because it was found that
standing passengers tend to look at advertisements
frequently, riders may pay attention not only to ads
that help them pass the time while standing, but also
those that stimulate their interest or desire. Instead of
showing just a picture, a magazine ad, for example,
could feature headlines or other clever designs aimed
at stimulating purchase behavior. It is important that
other advertisements do not simply catch the eye, but
encourage viewers to linger.
Secondly, the authors suggest that trains
adopt new forms of advertising media. The analysis
results indicated that passengers who stand tend to
have high awareness rates when it comes to in-car
advertising, but advertisements on the floor may be
www.ijera.com

easier for sitting passengers to see. Riders who sit
naturally allow their eyes to fall downward, making a
floor advertisement an eye-catching option. Those
who sit and read are also looking downward as well,
increasing the chances that they may see these
advertisements.

V.

Conclusion

Focusing on transit advertising which has a
good contact rate and long-term contact, the authors
decided to examine those relationships and how they
relate to train car advertising (hanging posters,
above-window posters, and sticker ads) with the aim
of defining the ideal format for this type of media.
Additional research must be done in order to
quantify how the attention captured through train car
advertising leads to other stages of purchasing
behavior. In addition, a strategic approach method that
actually drives consumers to purchase must be
considered. The effectiveness of other train-related
advertisements, such as in-car digital ads and exterior
train advertisements must be studied so that even
greater value can be obtained from transit advertising
media.

VI.

Acknowledgement

In this study, we appreciate the help received
from Nippon Research Center Co. Ltd. and the Kanto
Transit Advertising Council with cooperation.
174 | P a g e

References
Amasaka, K., (2005), ―Constructing a
Customer Science Application System
CS-CIANS - Development of a global
strategic vehicle Lexus utilizing New JIT -‖,
WSEAS Transactions on Business and
Economics, Issue 3, Vol.2, pp. 135-142.
[2]
Amasaka, K., (2007), ―The validity of
Advanced TMS - A strategic development
marketing system utilizing New JIT –‖, The
International Business & Economics
Research Journal, Vol.6, No.8, pp.35-42.
[3]
Amasaka, K., (2009), ―The effectiveness of
flyer advertising employing TMS: Key to
scientific automobile sales innovation at
Toyota‖, The Academic Journal of
China-USA Business Review, Vol.8, No.3,
pp.1-12.
[4]
Amasaka, K., (2010), ―Changes in marketing
process management employing TMS:
Establishment of Toyota Sales Marketing
System‖, The Academic Journal of
China-USA Business Review, Vol.10, No.7,
pp.539-550.
[5]
Dentsu, (2011), Advertisement expenses in
Japan (2011), http://www.dentsu.co.jp/
books/ad_cost/2011/index.html.
(in Japanese)
[6]
Dentsu, (2011), Advertisement expenses of
each media in Japan (2011), http:// www.
dentsu.co.jp/books/ad_cost/2011/media.html.
(in Japanese)
[7]
Ferrell, O.C., and Hartline, M., (2005),
Marketing Strategy, Thomson SouthWestern.
[8]
Gonda, M., (2004), (Master thesis) A study of
the latent function of train media: Focusing
on the index of the local society formed by
contact with train media, Tsukuba University
(Unpublished) (in Japanese).
[9] Murayama, T., et al. (1982), Analyzing CAID,
Marketing review: Japan research center, pp.
74-86 (in Japanese).
[10] Mehta, A., (2000), ―Advertising attitudes and
advertising effectiveness‖, Journal of
Advertising Research, Issue 3, Vol.40,
pp.67-72.
[11] Shimizu, K., (2004), Theory and Strategy of
Advertisement, Soseisha. (in Japanese)
[12] Takenaka, Y., and Harada, M., (2003), ―The
study of the degree of information contact by
the traffic advertisement and spectating
professional baseball games – Focusing on
Osaka Kintetsu Buffaloes-‖, Proceedings of
Osaka University of Health and Sport
Sciences, Vol.35, pp.15-23.

[13]

[1]

www.ijera.com

[14]

www.ijera.com

The Kanto Transit Advertising Council,
(2006), A Joint Survey of Transit
Advertising (March, 2006), http://www.
train-media.net/report/0502/adreport.pdf. (in
Japanese)
The Kanto Transit Advertising Council,
(2009), A Joint Survey of Transit
Advertising (March, 2009), http://www.
train-media.net/report/0612/0612.html. (in
Japanese)

175 | P a g e

U4102167175

More Related Content

What's hot

Viewers also liked

Similar to U4102167175

Recently uploaded

U4102167175