Page 1 of 13

Archives of Business Research – Vol. 10, No. 8

Publication Date: August 25, 2022

DOI:10.14738/abr.108.12816. Marui, J., & Jing, S. (2022). A Brand Concept Strategy Model of T-Shirts for Young People Using Selected Conjoint Analysis. Archives

of Business Research, 10(8). 49-61.

Services for Science and Education – United Kingdom

A Brand Concept Strategy Model of T-Shirts for Young People

Using Selected Conjoint Analysis

Junpei Marui

Department of Social Engineering, Nagoya Institute of Technology, Japan

Sun Jing

Department of Social Engineering, Nagoya Institute of Technology, Japan

ABSTRACT

Today's T-shirt market of young people is very competitive. As the brand of T-shirt,

UNIQLO、GU、Nike、Adidas、AZUL are known. This paper aims to propose a

method of designing an optimal brand concept for T-shirts. In the modern highly

competitive business world, marketing strategies are required to meet the

increasingly complex and diverse needs of customers. Therefore, it is very

important for companies to have a brand strategy in order to increase their

competitiveness in the market. On the other hand, a general decline in consumer

confidence among young people has begun to appear since the early 2000s. The

Nikkei Regional Industry Research Institute's "Youth Attitude Survey" also shows a

decrease in the ownership rate of famous foreign brand clothing in 2007 compared

to that of 2000. Kuga's analysis of the Ministry of Internal Affairs and

Communications' "National Survey of Consumption" also indicates that young

people's interest in fashion has declined. In 2017~2019 years, GDP of Japan was

increasing. However, GDP decreases greatly in 2020 because of COVID-19. In the

market of Japan, Young person's spending intention in recent years continues

decreasing. In this paper, we consider the case of launching a new T-shirt brand for

university students of Japan. Also, we study the marketing strategy of T-shirts for

young people who are considered to have decreased their interest in fashion, taking

into account the complexity and diversification of their needs, using a choice-type

conjoint analysis.

Keywords: New product concept; Brand strategy; Fashion

INTRODUCTION

In recent years, the fashion retail industry in Japan has been expanding its business due to the

globalization of the economy, with distributors from all over the world entering the Japanese

market and opening new stores. Numerous companies have opened stores, ranging from world- famous haute couture brands such as Chanel, Yves Saint Laurent, and Christian Dior to fast

fashion brands such as ZARA and H&M, which are popular among young people. On the other

hand, UNIQLO, GU, UNITED ARROWS, BEAMS, etc. are developing their stores while creating

their own originality. Thus, the battle in the fashion retail industry is intensifying day by day.

The fashion retail industry is also undergoing major changes in marketing strategies, such as

SNS-based advertising and online shopping. Therefore, it is important for a company's brand

strategy to increase its competitiveness in the market. Especially in the global market,

Page 2 of 13

50

Archives of Business Research (ABR) Vol. 10, Issue 8, August-2022

Services for Science and Education – United Kingdom

marketing strategies are required to meet the increasingly complex and diverse needs of

customers.

Under such circumstances, a general decline in consumer confidence among young people

began to appear in the early 2000s. The Nikkei Research Institute for Regional Industry's

"Survey of Youth Attitudes" also shows a decrease in the ownership rate of famous foreign

brand clothing in 2007 compared to 2000.

Kuga's analysis focuses on the Ministry of Internal Affairs and Communications' "National

Survey of Current Consumption" until 2009, and reports a decrease in expenditures on

"clothing and footwear. However, he attributes this to the appearance of inexpensive products

rather than a decline in young people's interest in fashion, as there were changes in the

purchasing environment in the late 2000s with the arrival in Japan of foreign fast fashion

brands such as H&M of Sweden in 2008 and Forever21 of the United States in 2009. In the

second half of the 2000s, there was a change in the purchasing environment.

After Kuga's analysis, the 2014 results of the National Survey of Family Income and Expenditure

were released. Looking at the changes in the average monthly amount of "clothing and

footwear" in 1999, 2004, 2009, and 2014, the largest amount for single person households

under 30 years old was 9,285 yen in 2009 for men, and the largest amount for women was

22566 yen in 2004. However, in 2014, the amount for men was 5124 yen and that for women

was 9114 yen, showing a marked decrease compared to the previous three surveys. However,

the consumer price index for "clothing and footwear" was 102.2 in 2014, when the base year

was 2000 (100.0), showing no decline. In other words, Kuga's interpretation that there was an

overall decrease in purchases is not valid, although there may have been a phenomenon in

which prices fell due to the availability of inexpensive products in some areas. It is more

reasonable to assume that young people's interest in fashion has declined [1].

Based on the above background, we will study the marketing strategies of fashion brands to

young people, whose interest in fashion is considered to have declined.

In this paper, we study the marketing strategies of fashion brands using a choice-type conjoint

analysis. Selective conjoint analysis has been used in various studies. Conjoint analysis was

developed in the field of quantitative psychology in the 1960s and has been widely used in the

field of marketing since then. The reason for its popularity in the field of marketing is that it can

estimate consumers' evaluations of the level of each attribute because it combines multiple

attributes and levels of a product and can directly present a concrete image to consumers as a

hypothetical situation in advance. In other words, this method is considered to be suitable for

linking the needs for each level obtained from consumers to technology when companies are

developing products.

In fact, selective conjoint analysis has been used in a wide range of research fields, such as

estimating the brand power of fruit and vegetable production areas [2], examining the

sequential effects of environmental risk information [3], evaluating multidimensional

management of agricultural technology [4], and evaluating the attributes of rice produced in

Hokkaido [5].

Page 3 of 13

51

Marui, J., & Jing, S. (2022). A Brand Concept Strategy Model of T-Shirts for Young People Using Selected Conjoint Analysis. Archives of Business

Research, 10(8). 49-61.

URL: http://dx.doi.org/10.14738/abr.108.12816

The most referenced previous study in this study is "A Study of Marketing Strategies for Late

Entrants in the Tea Market Using Selective Conjoint Analysis" by Ouyangni [6].

The purpose of this paper is to consider the marketing strategy of T-shirts for young people

who are considered to have decreased interest in fashion, taking into account the complexity

and diversification of their needs, and using a choice-type conjoint analysis.

Therefore, in this paper, we consider a marketing strategy for a hypothetical case where a new

T-shirt brand for students is to be launched.

The paper is organized as follows: Chapter 1 describes the social background, previous studies,

and objectives of this study. Chapter 2 presents the mathematical model of choice-type conjoint

analysis, the calculation method of marginal willingness to pay, and the orthogonal table. In

Chapter 3, we describe the method of designing an optimal brand concept for T-shirts. In

Chapter 4, a choice-type conjoint analysis is conducted based on the results of the questionnaire

for choice-type conjoint analysis, and the utility values for each attribute level, the total utility

value and choice probability for each product concept, and the marginal willingness to pay are

calculated. In Chapter 5, marketing strategies are presented based on the results of the choice- type conjoint analysis conducted in Chapter 4. Finally, Chapter 6 provides a summary of this

study.

MATHEMATIC MODEL

The mathematical model of choice-type conjoint analysis is shown below.

The total utility gained from the purchase of a given good is

� = �!�! + �"�" + �#�# +...+�$�$ + � (1)

�$(� = 1,2,3,...) are the attributes of the goods, � is the coefficient parameter of the attribute,

and � is the stochastically varying error term.

Then, the total utility �!and �" of commodity 1 and commodity 2 with different attribute levels

are

�! = �!�!! + �"�"! + �#�#! +...+�$�$! + �! (2)

�" = �!�!" + �"�"" + �#�#" +...+�$�$" + �" (3)

If � is a variable that takes 0 when product 1 is chosen and 1 when product 2 is chosen, then

the relationship between the magnitude of utility and product choice is

� = /

0, if �! ≥ �"

1, if �! < �"

(4)

� = /

0, if ∆� ≥ 0

1, if ∆� < 0 (5)

In addition, ∆� is expressed as

∆� = �!(�!! − �!") + ⋯ + �$(�$! − �$") + (�! − �") (6)

∆� = ∑ �%��% $

%&! + ∆� (7)

In the present model, since the error terms are independent and identically distributed, the

density function and distribution function of �$are respectively

Page 4 of 13

52

Archives of Business Research (ABR) Vol. 10, Issue 8, August-2022

Services for Science and Education – United Kingdom

�(�$) = �'(!�')"#! (8)

�(�$) = �')"#! (9)

And ∆� = �$ − �* is the logistic distribution

�(�$) = )#

!+ )# (10)

then the selection probability is

�(�|�) = -./01+2�$4

∑ -./ !

%&' (1+2�%) , ∀� ∈ � (11)

When the multinomial logit model is applied to the brand selection model, the maximum

likelihood method is usually used to estimate the parameters.

In the maximum likelihood method, the probability that product � is selected from among �

products is first formulated as equation (11), and the results of � selections are assumed to

have been observed. Let �8

9 = 1 if product � is selected on the � th selection opportunity, and

�8

9 = 0 otherwise. Furthermore, let �8

9 be the marketing variable vector of product � at the � th

selection opportunity, and let �t (�|�) be the selection probability in equation (11)

accordingly. The likelihood � under which �8

9 and the corresponding marketing variable vector

�8

9 are obtained is

� = ∏ ∏ �9(�|�)

:$

( $

9&! ;

8&! (12)

With �8

9 and �8

9 obtained, � in equation (12) is considered to be a function of the parameter

vector �. The value that maximizes the likelihood function � of � is the maximum likelihood

estimator of �.

The estimation method of the marginal willingness to pay is shown below.

Here, we consider the following main effect model for the utility function.

�8=�8(�8, �8) + �8 (13)

�8 is the attribute vector of commodity � and �8 is the price of commodity �.

�(�, �) = ∑* �*�* + �<� (14)

� is a parameter estimated by the multinomial logit.

Here, the total differentiation of equation (14) yields

∑ =>

=�)

* ��* + =>

=< �� = �� (15)

Assume that the utility level is fixed at the initial level (�� = 0) and attributes other than

commodity attribute �! are also fixed at the initial level (��* = 0 , � ≠ 1). In this case, the

marginal willingness to pay when commodity attribute �! increases by one unit is

�����' = @<

@�'

= −

*+

*�)

*+

*-

= − 2'

2-

(16)

Page 5 of 13

53

Marui, J., & Jing, S. (2022). A Brand Concept Strategy Model of T-Shirts for Young People Using Selected Conjoint Analysis. Archives of Business

Research, 10(8). 49-61.

URL: http://dx.doi.org/10.14738/abr.108.12816

The result can be obtained by the above equation.

RESEARCH DESIGN

In order to conduct a conjoint analysis, it is necessary to select the attributes and levels to be

measured and to set the combinations to be evaluated. There are many ways to select

attributes, but in this study, we conducted a questionnaire survey of students and set attributes

and levels by referring to the results of the survey. The results of the survey are shown in Table

1 below.

Table 1: Results of the responses to the Attribute Questionnaire

The percentages of the responses selected as very important and important are shown here.

The results of this survey were used to determine the attributes and levels as shown in Table 2

below.

Very Important Important Sum

Price 42.47% 46.58% 89.04%

Easy to wear 32.88% 39.73% 72.60%

The quality and functionality of the clothes 24.66% 38.36% 63.01%

Be yourself 12.33% 41.10% 53.42%

Can be worn for a long time 20.55% 31.51% 52.05%

TPO 12.33% 39.73% 52.05%

Eyes from the opposite sex 9.59% 39.73% 49.32%

Don't stand out too much from the people around you 2.74% 43.84% 46.58%

Eyes from the same sex 6.85% 36.99% 43.84%

Don't Suffer with People 15.07% 21.92% 36.99%

word of mouth 5.48% 21.92% 27.40%

Being a trend 4.11% 23.29% 27.40%

Brand 2.74% 21.92% 24.66%

Brand Visibility 2.74% 20.55% 23.29%