The effects of packaging attributes on Vietnamese consumers purchase intention

Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (35.06 MB, 141 trang )

(1)<div class='page_container' data-page=1>

VIETNAM NATIONAL UNIVERSITY, HANOI 
VIETNAM JAPAN UNIVERSITY

... & ...

HO PHUONG HONG

THE EFFECTS OF PACKAGING

ATTRIBUTES ON VIETNAMESE

CONSUMERS PURCHASE INTENTION

MASTER’S THESIS BUSINESS

ADMINISTRATION

</div>
(2)<div class='page_container' data-page=2>

VIETNAM NATIONAL UNIVERSITY, HANOI 
VIETNAM JAPAN UNIVERSITY

... & ...

HO PHUONG HONG

THE EFFECTS OF PACKAGING

ATTRIBUTES ON VIETNAMESE

CONSUMERS PURCHASE INTENTION

MAJOR: BUSINESS ADMINISTRATION 
Code: 60340102

Research Supervisors 
Assoc. Prof. Kodo Yokozawa

Assoc. Prof. Pham Thi Lien

</div>
(3)<div class='page_container' data-page=3>

ACKNOWLEDGEMENT

First of all, I would like to I want to send my sincere thanks to my Vietnamese and
Japanese professors, Associate Professor Pham Thi Lien and Associate Professor
Kodo Yokozawa who constantly giving me advices and orientation during my thesis
process.

Secondly, I would like to warmly thank to Hanh sensei, Matsui Sensei, Morita Sensei
and Hino Sensei for giving me many meaningful recommendations helping me to
improve my thesis writing in our Research Proposal Reports (on 25th December,

2018).

Thirdly, I would like to thank to participants who took part in my online survey and
gave me many valuable recommends to help me complete my questionnaire
successfully.

Additionally, I would like send a big thank to Huong san who always support and
remind us all procedure we need to do to complete your thesis smoothly.

</div>
(4)<div class='page_container' data-page=4>

ABSTRACT

Purpose – This study aims to measure Vietnamese instant oats packaging and its

influences on Vietnamese consumer’s buying intention.

Design/ methodology/ approach – A quantitative research was conducted using an

online survey to collect primary data for hypothesis testing. The questionnaire was
transferred successfully to 147 respondents.

Findings – The findings indicated the positive relationship between packaging

elements (i.e. graphic, structural and verbal attributes) and consumer purchase
intention based on the literature reviews and data analysis discussed precisely in this
research. Additionally, the moderation role of involvement level to the interaction of
visual elements and purchase intention was also proved in this paper.

Research limitation/ implication – the main limitation of this study is excluding

other important factors impact purchase intention such as price, promotion, etc. The
findings provide knowledge related to consumer behavior for relevant companies to
increase them to design an effective communication tool – package.

Practical implication - The findings of this study can be used by managers and

marketers to create an effective packaging to ensure their products stand out among
competitors.

Originality/ value – This study is one of the few quantitative researches measure the

impacts of graphic, structural and verbal elements on purchase intention
simultaneously. Furthermore, it emphasizes the communication role graphic,
structural and verbal attributes which has been ignore in previous studies.

Additionally, this study also indicated that involvement level positively strengthens
the interaction between structural package design and purchase intention which did
not quantitatively proved in previous studies.

Keywords Consumer’s purchase decision, packaging design, graphic elements,

structural element, verbal elements, involvement level

</div>
(5)<div class='page_container' data-page=5>

Table of Contents

CHAPTER 1: INTRODUCTION ... 1

1.1. Research Motivation ... 1

1.2. Research Objectives ... 3

1.3. Research Scope ... 3

1.4. Research Structure ... 3

CHAPTER 2: LITERATURE REVIEW ... 5

2.1. Consumer Behavior... 5

2.1.1. Customer behavior definition ... 5

2.1.2. Purchase decision making processes ... 6

2.2. Packaging ... 10

2.2.1. Package and packaging design ... 10

2.2.2. The role of packaging ... 11

2.2.3. Packaging Elements... 13

2.2.4. Product involvement ... 13

2.3. Research Gap and Research Questions ... 14

2.4. Theoretical framework and research hypotheses... 16

2.4.1. Variable definition ... 16

2.4.2. Measurement of Variables ... 18

2.4.3. Integration of literature review and hypothesis ... 19

CHAPTER 3: METHODOLYGY AND RESEARCH DESIGN ... 25

3.1. Research approach ... 25

3.2. Research design ... 26

3.3. Data selection ... 28

3.3.1. Secondary Data ... 28

3.3.2. Primary Data ... 28

3.3.2.1. Sampling design... 28

3.3.2.2. Data Collection and participants characteristics ... 29

3.3.2.3. Questionnaire and Experimental Design ... 30

3.4. Data Analysis ... 33

3.4.1. Reliability Analysis ... 33

3.4.2. Continuous improvement cycle ... 33

3.4.3. Survey data analysis ... 34

3.4.3.1. Exploratory factor analysis (EFA) ... 34

3.4.3.2. Confirmatory factor analysis (CFA) ... 36

3.4.3.3. Structural equation modelling (SEM) ... 37

</div>
(6)<div class='page_container' data-page=6>

CHAPTER 4: DATA ANALYSIS RESULTS ... 40

4.1. Measurement Scale Test ... 40

4.1.1. Cronbach’s Alpha ... 40

4.1.2. Exploratory Factor Analysis (EFA) ... 41

4.1.3. Confirmatory Factor Analysis (CFA)... 44

4.2. Research Model Test ... 47

4.2.1. Hypothesis testing without moderation of involvement level ... 47

4.2.2. Involvement Level effect testing by SEM ... 49

CHAPTER 5: CONCLUSION ... 52

5.1. Discussion of findings ... 52

5.2. Managerial implication ... 54

5.3. Practical implication ... 55

5.4. Limitation and further research ... 55

REFFERENCES ... 57

APPENDIXES 1: ONLINE SURVEY ... 63

APPENDIXES 2: Descriptive analysis ... 109

APPENDIXES 3: Cronbach’s Alpha ... 110

APPENDIXES 4: EFA ... 113

APPENDIXES 5: CFA... 115

APPENDIXES 6: SEM ... 122

</div>
(7)<div class='page_container' data-page=7>

LIST OF TABLES

Table 2. 1: Summarized literature review ... 14

Table 2. 2: Measurement of variables ... 19

Z
Table 3. 1: Frequency of demographic information of respondents ... 30

Table 3. 2: Study variable areas and corresponding section of the questionnaire .. 32

Table 3. 3: Suggested procedure for improve measurement construct validity ... 34

Table 3. 4: EFA requirement assumptions ... 35

Table 3. 5: Model diagnostics in CFA ... 37

Table 3. 6: Model fit indices ... 37

Z
Table 4. 1: Cronbach’s Alpha Results ... 40

Table 4. 2: Removed variables ... 41

Table 4. 3: Exploratory Factor Analysis Results ... 41

Table 4. 4: Rotated components results ... 42

Table 4. 5: The new latent variables ... 46

Table 4. 6: Model fit indices ... 44

Table 4. 7: Confirmatory Factor Analysis results ... 44

Table 4. 8: Composite reliability and AVE results... 45

Table 4. 9: The model fit test of structural model ... 47

Table 4. 10: Research model without moderator tested by SEM ... 47

Table 4. 11: Hypotheses testing without moderator results ... 48

Table 4. 12: Coding of variable computing ... 49

Table 4. 13: Moderator effect model fit ... 50

</div>
(8)<div class='page_container' data-page=8>

LIST OF FIGURES

Figure 2. 1: The factors impact customer purchase intention ... 9 
Figure 2. 2: The proposed research framework ... 16 
Z

Figure 3. 1: The research process flow chart ... 27 
Figure 3. 2: Observed variables example... 39 
Z

</div>
(9)<div class='page_container' data-page=9>

LIST OF ABBREVIATIONS

VS: Visual Elements

VB: Verbal Elements

SEM: Structural Equation Modeling

PI: Purchase intention

</div>
(10)<div class='page_container' data-page=10></div>
(11)<div class='page_container' data-page=11>

CHAPTER 1: INTRODUCTION

1.1. Research Motivation

</div>
(12)<div class='page_container' data-page=12></div>
(13)<div class='page_container' data-page=13>

consider “The effects of packaging attributes on Vietnamese consumers Instant Oats
purchase intention” as personal thesis topic.

1.2. Research Objectives

There exist very limited comprehensive practical studies have precisely
analyzed packaging as a customer’s purchase communication tool in Vietnam from
customer perspectives. Hence, this paper aims to quantitatively analyze the impact of
packaging design on consumer purchase intention then provide Vietnamese managers
a better understanding about the importance of packaging in differentiating among
competitors on the shelves. Particularly, the study has the following sub-objectives:
• To measure precisely the influences of packaging elements (i.e. visual and verbal

elements) on buying intention under the moderation of involvement level.
• To identify which attributes should be concentrated while designing packaging.

1.3. Research Scope

Due to the purpose of this research is focusing on the package influences only
on purchase intention, to control the price influences and promotion impacts, author
chose Vietnamese instant oat brands which have the same price range is 100.000

VND for a 400 grams package without any promotion at the moment. There have
three Vietnamese instant oats brands including Xuan An, Vinamit and Union were
used to packaging assessment. In addition, to determine the suitable targeted group,
author chose participants who have already consumed Vietnamese instant oats before
collecting their feedback about a certain product.

1.4. Research Structure

</div>
(14)<div class='page_container' data-page=14>

Chapter 1: Introduction

This chapter provides an introduction about study motivation, research
objectives, research scope and research structure.

Chapter 2: Literature Review

This chapter provides the summary and reviews existing research related to
literature of packaging design and customer behavior and their relationship under the
influence of mediator “Involvement level”. Based on literature reviews, author
propose research conceptual framework, research question and research hypothesis.

Chapter 3: Research Methodology

In this chapter, author discussed about research design, data collection process
and data analysis methods were used to measure the conceptual framework proposed
in chapter 2.

Chapter 4: Analysis results

This chapter indicated research results was analyzed from data collection.

Chapter 5: Conclusion

</div>
(15)<div class='page_container' data-page=15>

CHAPTER 2: LITERATURE REVIEW

2.1. Consumer Behavior

2.1.1. Customer behavior definition

According to AMA (American Marketing Association), customer behavior is
defined as “The dynamic interaction of affect and cognition, behavior, and
environmental events by which human beings conduct the exchange aspects of their
lives". In the other words. In other words, customer behavior is consideration,
emotion and reaction of customer in the consumer process (Satish K. Batra and S. H.
H. Kazmi, 2004). The behavior of customer is affected by many factors such as the
opinions from family or friends, social media, advertising, product information,
prices, packaging, product appearance can all affect the feelings, thoughts and
behavior of customers.

Research on consumer behavior is an important task that has a great influence
in the decision-making process of marketing strategies (Philip Kotler, 2001, p.
197-198). Previously, marketers could understand consumers through their exposure,
transaction and daily sales experiences. However, due to the growth of market size,
marketing managers no longer have direct contact with their customers and the
information from sales department is becoming subjective. As a result, many
managers started using consumer behavioral research to have an appropriate and
accurate information in order to attract more customers. According to Peter Drucker,
who is considered the "father" of modern business management said: "The ultimate
goal of all business activities is to create customers. And only two business tools can
do this is marketing and creativity ”(Vneconomy).

</div>
(16)<div class='page_container' data-page=16>

2.1.2. Purchase decision making processes

Purchase decision making process is defined as the stage where the consumer
actually purchase the product (Amstrong, 2012). The first purchase decision making
processes, were introduced by Engel, Blackwell & Kollat, includes 5 stages:
“Problem recognition, Information search, Evaluation of alternatives, Purchase
intention and Post-purchase behavior”. These five stages are a good framework for
assessing customer buying behavior. However, customers do not always go through
these five stages, they may skip or reverse one. For example, if a customer feels they
need to buy chocolate to eat, they can go to a store immediately to buy a chocolate
bar without searching any information or considering alternatives in advance.
Meanwhile, in case of car’s purchase decision making, customer will thoroughly
research product information as well as comparing to others similar car brands before
making final buying decision. In conclusion, the purchase decision making processes
is different among the amount of effort a consumer puts into a product while
purchasing it.

Problem recognition

</div>
(17)<div class='page_container' data-page=17>

At this stage, marketers need to identify what situations that often make
consumers quickly recognize their problem. They should study consumers behavior
to find out what kinds of feelings have generated problems or needs, explain what
makes them, and how they impact consumers to choose to buy a certain product.

Information search

After recognizing needs stage, a consumer starts looking for information. If
the desire of buyers is strong enough and the products are available at that time, they
tend to buy them immediately. In contrast, if the desired products are not within the
reach, the

If consumers' impulse is strong, and desired products are within reach,
consumers will most likely buy immediately. Otherwise, consumers will keep their
needs in their subconscious. Consumers may refuse to search for information, or find
out basically products information, or actively seek information related to their needs.
In case they want to search for information, there are usually the following sources
of information.

• Family, friends, neighbors and acquaintances.

• Commercial information collected through advertising, salesman, merchants,
packaging or product displays.

• Public information obtained from mass media and organizations.

• Personal experience obtained through the interaction in daily life, survey or
product usage.

</div>
(18)<div class='page_container' data-page=18>

computer software products through commercial information sources, but discuss to
programming experts about software products before final buying decision.

Marketers need to understand the importance of the information source which
is usually referred by their target customers to strategically format information
sources. As a result, in order to create a marketing content which is effectively
communicates to target markets, marketers should interview consumers to collect
their first opinion about products, brand image and what sources they received
information and the how consumers react to the differences between each source.

Evaluation of alternatives

Before making a purchase decision, buyers process the collected information
and then evaluate others similar brand. The evaluation process is usually done based
on the following principles and sequences.

Firstly, consumers consider a product with a set of attributes. In particular,
each attribute is assigned to a useful function that can bring satisfaction to consumers
when they use it.

</div>
(19)<div class='page_container' data-page=19>

Purchase intention

After the evaluation, the intention to purchase will be formed by the brand that
received the highest rating and becoming to the purchase decision. In addition,
according to Aizen (1991), “intention represent moticational components of a
behavior, that is, the degree of conscious effort that a person will exert to perform a
behavior”. There are two factors that can be intervened before consumers make the
purchase intention as follows (Philip Kotler & Kevin Keller, 2011, Marketing
management, p.170):

(Sources: Phillip Kotler & Kevin Keller, Marketing management, p.170)

Figure 2. 1: The factors impact customer purchase intention

</div>
(20)<div class='page_container' data-page=20>

packaging draws attention primarily through color and shape (Khan, Lee and
Lockshin, 2017); The appearance of packaging may also affect the evaluation of core
products (Kumar and Kapoor, 2017); and product packaging also affects consumers'
perceived quality and their purchase intentions (Rundh, 2016).

Post-purchase behavior

After purchasing, consumers can feel satisfied or dissatisfied with some

products aspects and then take actions after buying or reacting to certain products or
the using process. If the characteristics and functions of the product meet their
expectations, consumers will be likely to buy products again and introduce it to others.
On the contrary, dissatisfied customer tend to return or say bad thing about the
products with others and switch to products of competitors.

According to an overview of consumer behaviors, with the affecting of others
factors, such as price, promotion, distributions, cultural context, etc, packaging place
a main role in communicating to customer in the store (Gonzalez et al, 2007).
Obviously, among five stage of purchase decision making process, packaging has
significant influence on buyer purchase intention which is the fourth stage of the
process.

2.2. Packaging

2.2.1. Package and packaging design

Products package is known as a portion of the product itself as well as brand
recognition. Packaging has an important role in presenting products features and
providing product information to customers. From buyers’ perspectives, both
package and products are the same on the shelves. During the purchase decision
making process, customer uses package as a supportive tool in evaluating products
quality and its functions to make a right choice.

</div>
(21)<div class='page_container' data-page=21>

have a significant role in product perception. Meanwhile, verbal package design
elements importantly detailed product information. At the point of purchase, the
fundamental role of package and packaging design is to make buyers’ attention and
to be highlighted among compertitors in the store.

Effective packaging and package design are the outcome of the collaboration

between designers, marketers and customers. Thus, packaging design is one of the
marketing strategy tools for products itselfs. According to Prone (1993), customer’s
impulse buying is made by packaging which boosts products function, connects to
company’s image and helps company brand stands out among competitors (Gaber &
Jones, 2000). Hence, in the buying decision making process, packaging has an
significant role in communicating with customers by providing goods-related
information.

Besides, packaging is also defined as product positioning which creates the
company brand in the customers mind and emphasized the added value that
differentiate products from alternatives. Maggard (2976) claimed that product
positioning induces marketing mix where the portions including pricing policy, place,
products and promotion are involved”. Considering positioning elements and
competitors capability helps marketers to conduct an appropriate marketing strategy
(Ampuero & Vila, 2006). The differences of the relevant element’s category depend
on positioning strategy’s aim i.e. globalization or localization. As the result,
packaging performs in different functions to reach the different goals, however,
stimulating customer purchasing a particular products is the main role of positioning.
Hence, while marketers use positioning to place products in the market, packaging
and package design are used as assistant of company to attract customers’s attention.

2.2.2. The role of packaging

</div>
(22)<div class='page_container' data-page=22>

function but also psychological function i.e. packaging communicate to customer.
Accrording to Bill Stewart (2004), there has three main functions of package is
mentioned as below.

To contain: Packaging acts in covering role which ensure the visual features,

original functions and the quality of product during the lead time. A package, which

has a good containing function, enhance the trust of customer about products.

To protect: Packaging acts a protection part to keep actual products against

external effects including temperature, moisture, light, etc. In this term, designers
choose package material based on the characteristics of the goods, the transit process
and the environmental risks that it will faced with. Accordingly, if package performs
this function well, the shelf life and the freshness of goods will be extended.

To Identify: Packaging plays an important role in reminding customer about

the available of goods and providing products information to buyers. Customers can
easily access product information e.g. ingredient, country of origin, production and
expiry date, etc. In addition, this function can stimulate customers actual purchasing
and assist product standing out among alternatives.

Marketing tool

Product design is described as an decisive tool to create marketing strategy for
consumer goods (Rundh, 2009). To enhance a competitive marketing strategy, along
with SWOT analyzing, the product design should involve the references from
customers (Creusen et al, 2010). Packaging is an effective communication method
helping marketers to show their product information and message to consumers
(Silayoi & Speece, 2007).

</div>
(23)<div class='page_container' data-page=23>

packaging assist product in differentiating consumer good from alternative products
(Holmes et al, 2012).

Additionally, as an element of product design, packaging design has a
significant communitive role and a strong impact on consumer purchase intention

(Salem, 2017), thus, packaging is an important instrument in marketing strategy.

2.2.3. Packaging Elements

Based on previous related literature (Sonsino, 1990; Hine, 1995; Vila &
Ampuero, 2007; Underwood, 2003), visual package design consists of two major
attributes, such as graphic and structural elements. Graphic component includes color,
typographic, shape, image while the structural component consists of size and
material used to cover actual products (Hine, 1995). Additionally, Silayoi and Speece
(2004, 2007) added verbal component involving factors related to information or
words such as brand name, product information, language used on package (Salem,
2017). In some aspects, three of them have influences customers purchase intention.

2.2.4. Product involvement

</div>
(24)<div class='page_container' data-page=24>

2.3. Research Gap and Research Questions

Table 2. 1: Summarized literature review

Author Review of article Research Gap

Underwood (2001) The theoretical framework was conducted to understand the communicative effects of package image on brand attention. According to the virtual reality simulation results,
package image positively influences consumer's brand attention in private label brand.

These qualitative
papers mainly aim to
explore the initial
understanding of the
relationship between
packaging and

purchase intention
with the considering
involvement level as
its moderator.
However, they did not
have any precise
measurement the
interaction among
them. Additionally,
most of these papers
just focused on the
impact of visual
element only, did not
mentioned much
Underwood (2003) The exploratory qualitative study confirmed the role of packaging in enhancing the

relationship between consumer and brand in low involvement products.

Silayoi (2004)

This is a qualitative approach research adopting a focus group method to understand
consumer response to product packaging and packaging design influences on the
consumer purchase decision. The findings are that customers decision mostly affected by
visual elements when they considered low involvement products under high time
pressure.

Ampuero (2007)

This paper study relationship between packaging graphic elements and positioning
strategies from related previous study and found that the existing literatures mainly

concentrated on the impacts of visual elements including color, shape, typography and
image which is determined by corresponding positioning strategy.

Butkeviciene et al.

(2008) An empirical research found that non-verbal components enhances consumer impulsive purchasing while verbal components did not impact on repeated purchasing.
Mutsikiwa (2014) This paper aims to evaluate the influences of aesthetics package design on buyer's purchase decision in daily products consuming. The analysis focused on package color,

</div>
(25)<div class='page_container' data-page=25>

Ford et al (2015)

Collected information from in-depth interviews and observation of 11 older participants
(in range 58 - 85 years old)'s behavior with fast-moving consumer products packaging.
The results indicated that customers aging has positive relationship with their perceived
risk in packaging interaction.

about the role of
verbal elements.

Sarker (2015)

A qualitative study examines the influences of package design and naming strategies on
perceived quality and purchase intentions. The results show that while packaging
positively effects on perceived quality and purchase decision, the naming strategy did not
have any significant impacts.

These quantitative
had objectively
analyze the impacts
of package attributes
on purchase intention

but there has no
positive direct
relationships among
them.

Muhammad (2014) This study used quantitative analysis to test the influence of verbal elements (i.e. nutritional information, product information, country-of-origin) on consumer buying
behavior. The findings revealed that product information has negative impact.

Imiru (2017) The paper used correlation and regression to analysis the relationship between packaging attributes on consumer purchase decision. As the result, there has no relationship between
package color and material.

To fill these research gap, this study aims to investigate the positive relationship between packaging elements and purchase
intention by adding involvement level as moderator of their interactions. According to Quester and Smart (1998), involvement level
strongly influences consumer buying decision making processes. Additionally, in case of low involvement level, consumer normally
affected by visual packaging design while buyers pay more intention to the product itself in case of high involvement (Grossman and
Wiseblit, 1999). In general, this paper purpose to precisely measure the relationship between packaging design and purchase intention
under moderated by involvement level with following research questions.

1. Is there a positive relationship between packaging design and consumer purchase intention?

</div>
(26)<div class='page_container' data-page=26>

2.4. Theoretical framework and research hypotheses.

Figure 2. 2: The proposed research framework 
2.4.1. Variable definition

2.4.1.1. Packaging Elements

</div>
(27)<div class='page_container' data-page=27>

There have many different definitions about the packaging elements. In the
related research of Smith and Taylor (2004), they divided packaging design into six
parts including “color, size, form, materials, graphic and flavor”. Besides, Kotler (2003)

claims that product package involves similar six elements, such as “color, size, form,
material, text and brand”. These elements should be considered in package design
process (Vila and Ampuero, 2007). Differently, Underwood (2003) separate packaging
into two main attributes i.e. structural attributes (“form, size of the containers and
materials”) and graphic attributes (“color, shape, typography and image”). Obviously,
similar to Underwood (2003), Smith & Taylor (2004) do not mention about verbal
elements of packaging which has decisive influences on actual purchase of
high-educated consumer (Mutsikiwa et al, 2013). To fill this gap, Rettie and Brew (2000)
mentioned about positioning function of packaging, thus, they separated packaging to
main groups: visual (e.g. color, image, shape, font, size) and verbal (e.g. brand name,
slogans) elements. Additionally, according to Silayoi and Speece (2004), packaging has
two major elements: visual and informational elements. Visual elements, includes
graphics, size and shape of packaging, influence the affective aspect of purchase
decision making of consumer. Meanwhile, informational elements, which impact
cognitive aspects, consists of product information (e.g. the name of the firm, address,
country of origin, production and expiry date).

Though both Rettie & Brew (2000) and Silayoi & Speece (2004) considered the
importance of informational elements, they did not mention about the environmental
effects of package materials which significant influence on customer’s food purchase
intention (Rundh, 2005) due to the growth of environmental concerned consumers. Not
only that, they also did not indicate the role of printed language which may affect on
the willingness to purchase of buyers (Salem, 2017).

</div>
(28)<div class='page_container' data-page=28>

color, shape, font, picture), structural element (e.g. size, material) and verbal elements
(e.g. brand name, product information, language).

Graphic elements: Graphics elements, are factors can be seen, comprised of

color combination, shape, background image and the font style used on packaging (Hine,

1995).

Structural elements: structural elements, are factors simultaneously designed

to display and protect the products effectively, consist of product size design and
material should be used.

Verbal elements: Verbal elements, are factors related to information or words,

include brand name, the information of product and the language used on packaging. In
the decision-making process, the verbal attributes influence the cognitive.

2.4.1.2. Product involvement

Product involvement, is the level of interest or effort that consumers put in
purchasing a certain product, divided into two type including low and high involvement.
High involvement products, are perceived as having high value with high cost
and provide long term benefits and buyers tend to carefully evaluation before
purchasing it. Meanwhile, low involvement products have cheaper cost, thus, buyers do
not need much time to deeply research or consider before chasing its.

2.4.1.3. Purchase Intention

According to Sproles & Kendall (1986), purchase intention is a “mental
orientation characterizing a customer’s approach to making choice”. Purchase intention
associates with cognitive and affective process in decision making process.
Additionally, consumer purchase intention is likely influenced by three main packaging
design components including graphic, structural and verbal elements

2.4.2. Measurement of Variables

</div>
(29)<div class='page_container' data-page=29>

Table 2. 2: Measurement of variables

Variable No. of questions Measurement sources 
Graphic

elements

Color 3 Olawepo (2015)
Shape 2 Olawepo (2015)
Font 3 Olawepo (2015)
Picture 2 Olawepo (2015)

Structural 
elements

Size 1 Salem (2017)
Material 3 Salem (2017)

Verbal 
elements

Brand name 3 Salem (2017)
Product information 4 Salem (2017)
Language 1 Salem (2017)

Involvement level 3 Mittal (1989)

Purchase intention 4 Weisstein (2017) Pei (2014) and

2.4.3. Integration of literature review and hypothesis 
2.4.3.1. Graphic package elements and purchase intention 
Color

Color has significant influence on consumer emotion (Salem, 2017), thus, color
selection process is very importance to design an attractive package (Cheskin, 1957).
Package color assists product differentiate from other brand and enhance consumer’s
long-lasting memories about products. Color is an effective design tool without cost,
product attributes and function adjustments (Garber et al, 2000).

According to Steward (2004), each packaging color of a particular product has a
transferred message to consumers. Especially, for food products, package color has
strong impact on customers perception about the food taste (Kauppinen, 2010;
Koch&Koch, 2003; Gaber et al, 2000).

</div>
(30)<div class='page_container' data-page=30>

Therefore, to make the right choice of color, marketers should fully understand the
meanings of each colors in different cultural context (Salem, 2017).

Shape

Interestingly, many customers confirm that they have potential to purchase a
certain product without reading the label or product information (Salem, 2017).
Previous marketing literature proved that the shape of package associating with
message affect consumer feeling and perceived quality (Abdelsamie et al, 2013;
Ruumpol, 2014). For instance, while male is impressed by linear angular shapes, female
prefer curving line and round shape (Shimp, 1990). When considering two products
with the same weight, buyers tend to choose the product which has taller shape because
in buyer’s mindset, the higher has larger volume (Silayoi et al, 2007). Additionally, an
innovative package helps products enhance the attractiveness and stand out among
similar brands (Salem, 2017). Unique packaging is a competitive tool used for

differentiating and consequently increase products sales volume (Sherwood, 1999).

Background picture

According to Salem (2017), pictures and graphics affect consumer sensory. the
pictures, are printed on product packaging, describes the information related to the
goods, such as products usage instructions and its functions where consumer can
generally imagine what the product is (Pensasitorn, 2015). Hence, pictures on package
places an important role in communicating to customer through transferring products
information and imagined stimuli about products (Salem, 2017). In the other words,
packaging image is an effective instrument to convey the functions of product and assist
goods to be different from alternatives (Meyers and Lubliner, 1998).

Font Style

</div>
(31)<div class='page_container' data-page=31>

have an innovative font style, many companies hire experts to design a creative and
attractive font style used for their logos, slogans and product package (Imiru, 2017).
Therefore, font style, is a powerful tool to draw consumers attention, has positive impact
on consumers purchase decision (Imiru, 2017).

In general, the graphic elements have significant impacts because they have a
capability to influence the emotion and feeling of targeted customers (Silayoi and
Speece, 2004). Thus, graphic elements positively impact on buyer’s purchase intention
perceived quality (Saker, 2015). Based on discussion above, the first hypothesis was
established as below:

H1: Graphic packaging elements positively influence consumer’s instance oats 
purchase intention

2.4.3.2. Structural package elements and purchase intention

Size

As package shape, buyers consider package sizes to make volume perception.
Hence, the size design should meet consumer’s demand (Makanjuola and Enujiugha,
2015). According to Benedetti et al (2014), marketers should understand target’s
customers behavior before making product sizes decision. Additionally, package sizes
strongly influence on consumer buying choice when buyers cannot clearly evaluate
product quality, thus, they are potential to buying smaller one for trial usage (Ksenia,
2013). Meanwhile, some buyers prefer the large size of products for saving cost.
Therefore, to meet different type of consumer demands, it would be better if goods are
sold in various packaging size (Rundh, 2005).

Material

</div>
(32)<div class='page_container' data-page=32>

products (Silayoi & Speece, 2004). In addition, many today consumers more concern
about environmental issues (Rundh, 2005). Accordingly, buyers potentially choose
products which have environment friendly, recycle and ease-reuse packaging (Rundh,
2005).

Based on previous literature review, to understand more the influences of
structural elements will be thoroughly analyzed in the following section; thus, the
second hypothesis was conducted as following:

H2: Structural packaging elements positively influence consumer’s instance oats 
purchase intention

2.4.3.3. Verbal package elements and purchase intention 
Product information

Packaging places a major role in conveying the information related to products

which enhance the cognitive process of consumers (Salem, 2017). In some country,
there exist the regulations about packaging printed content, such as the name of the firm,
address, country of origin, production and expiry date (Ahmed et al, 2014). Due to the
increase of healthy eating trend, customer concern more about food ingredients and
nutrition (Coulson, 2000), thus, they more carefully consider product information
available on package. The package has appropriate products information assist enhance
consumers’ reliability and boost purchase decision making process (Silayoi and Speece,
2004). In the store, package maybe the only instruments helping product communicate
with buyers (Gonzalez et al, 2007). Accordingly, packaging is considered as the
fundamental elements in consumer purchase intention.

Brand elements

</div>
(33)<div class='page_container' data-page=33>

elements of product information (Silayoi and Speece, 2004). Brand identification helps
buyers to reduce searching time when they purchase food products (Bassin, 1988). With
other information-related-to products, brand elements visibly evoke consumers interests
and boost their purchase decision process (Mutsikiwa et al, 2013).

Language

De Run and Fah (2006) indicated that using mother language on package assists
buyer easily understand products functionalities and usage instructions, thus, buyers are
likelihood to purchase products has printed information in mother language. Hence, to
globalize effectively, marketers should eliminate the language barrier which is one of
the major of cultural barriers (Hall & Hall, 1987). Additionally, under cultural context,
the ways to describe a sentences or words are different among local languages (Doole
& Lowe, 1999). Local people will have a positive sense that the certain foreign firm is
seriously doing business in targeted country if their language is used on package
(Hollensen, 1998). Adopting a suitable language for information contents or packages
is a very necessary mission to increase consumer purchase intention (Salem, 2017).

Whilst visual elements (e.g. color, shape, size, font, picture, material) evoke
buyer’s emotion and feeling, verbal elements (e.g. products information, brand
elements, language) significant impact on the cognitive process of customer (Silayoi &
Speece, 2004). As a result, verbal elements also have impacts on buyer’s purchase
intention (Salem, 2017). This led to the establishment of the third hypothesis:

H3: Verbal packaging elements positively influence consumer’s instance oats 
purchase intention

2.4.3.4. The effect of involvement level

</div>
(34)<div class='page_container' data-page=34>

process shows its limitation. Recognizing that limitation, based on Kotler’s literature
review, Patty & Cacioppe (1981) proposed two accessing contexts for high and low
involvement products. With high involvement products, buyers are likelihood to
carefully search the information related to products, which follows the Kotler’s
purchase decision making process (Patty and Cacioppe, 1981). For low involvement
level, buyers easily are drawn by packaging design (Solomon, 2002; Silayoi & Speece,
2004).

On one hand, according to Silayoi and Speece (2004), packaging assist low
involvement products evoking customer’s emotional action. The expected outcome of
purchase decision and the element evaluation of low involvement products are less
important, thus, the role of packaging graphic and structural elements become more
critical (Grossman & Wisenblit, 1999). Thus, it is possible the level of involvement has
influenced the interaction between graphic or structural package elements and purchase
intention, based on it, the fourth and fifth hypothesis were established as below:

H4: Involvement level is the moderator of the relationship between graphic 
package elements and purchase intention

H5: Involvement level is the moderator of the relationship between structural 
package elements and purchase intention

On the other hand, buyers do not care much about visual aspects when they are
considering to purchase high involvement products (Kupiec & revell, 2001).
Accordingly, they pay more attention with package informational elements (Silayoi &
Speece, 2004). Obviously, level of involvement has impacted the relationship between
verbal package elements and purchase intention, based on it, authors hypothesize the
following:

</div>
(35)<div class='page_container' data-page=35>

CHAPTER 3: METHODOLYGY AND RESEARCH DESIGN

Research methodology is the detailed procedures used in identifying, selecting,
processing and analyzing data for answering research questions. Research methodology
is the fundamental chapter of research writing. This chapter describe specifically the
study processes to address research issues and examine research hypotheses. For
instance, in this chapter, author discussed the methods related to research methodology
determination as well as collecting data and analyzing instruments to conduct this
research.

3.1. Research approach

</div>
(36)<div class='page_container' data-page=36>

3.2. Research design

According to Souna (2007), research design is defined as “the framework or
guide used for the planning, implementation, and analysis of a study”. The different of
research questions or hypotheses lead to the different of research design, thus,
understanding and distinguishing exactly the types of research design is an important
mission while conducting a study. Based on the researcher’s variables controlling level,

quantitative research design is divided into 4 main types: descriptive, correlational,
causal-comparative and experimental design.

Causal-comparative design is adopted to create the cause-effect relationship
among variables. In this research design, independent variables are considered as causal
factors while dependent variables are affected factors. Causal-comparative design is a
popular design used in social science to analyze human behavior through assessing the
cause-effect relationship among groups.

</div>
(37)<div class='page_container' data-page=37>

Figure 3. 1: The research process flow chart 
L
ite
ra
tur
e
R
ev
ie
w

Research problem definition
- Question

- Objectives
- Hypotheses

Theory exploration
- Theory framework
- Model building

Sampling (survey)

Questionnaires developing

Pilots study

Refinement questionnaire

Data collection

Selection of basic research methods: Questionnaires/ survey

Editing/ coding data

Quantitative analysis
- EFA
- CFA
- SEM

Interpretation of results and findings

</div>
(38)<div class='page_container' data-page=38>

3.3. Data selection

Research data has been collected from both secondary and primary sources.
According to Saunders (2009), secondary data is the information collected from existing
sources including company reports, academic journals, scientific articles and media
information. Otherwise, primary data is gathered by researchers themselves to answers
their research questions (Hox & Boeije, 2005). Saunders (2009) claimed that the
reliability of data collection is very important to have a valuable result.

3.3.1. Secondary Data

Secondary data has been gathered mainly from academic journals and scientific
articles. The secondary is used as references for gathering primary data and research
questions. Author purpose to find the suitable secondary data related to packaging
design and its impacts on consumer purchase intentions. In addition, to solve research
problem, author used key words such as packaging elements, purchase decision and
consumer involvement level to search expected secondary data. The used database is
belonging to Yokohama National University, Vietnam National University and Google
Scholar.

3.3.2. Primary Data

According to Churchill and Lacobucci (2010), researchers should consult the
secondary research first then conduct the primary data to achieve a general knowledge
about research topic. The primary data was collected according to specific research
purpose and research question (Mark Saunder, 2009). To gather primary data, this study
used online survey built by google forms.

3.3.2.1. Sampling design

</div>
(39)<div class='page_container' data-page=39>

each of them consists of more than 3 items with high item communalities (> 0.6); 150
in case of model with at most 7 constructs have modest communalities (0.5); 300 in
case of models involves at most 7 constructs with low item communalities (0.45) and
over 500 for models consists a large number of constructs with low item communalities
and under 3 measurements items. Generally, 100 is the practical and acceptable size for
SEM (Hair et al., 2010).

Furthermore, for SEM in using AMOS, Pallant (2005) claimed that the sample
size should be “at least five times the number of question items. Accordingly, the

proposal model involves 2 independent variables, 1 moderate variable, 1 dependent
variable containing 30 question items. Specifically, 10 for “Graphic elements” variable,
4 for “Structural elements”, 8 for “Verbal elements”, 4 for “Involvement level” and 4
for “Purchase intention”. Hence, there has 30 items in total (30 x 5 = 150) which mean
that 150 is the minimum acceptable sample size. Therefore, this study considered 150
for minimum sample size estimation.

3.3.2.2. Data Collection and participants characteristics

Due to the conveniences and objectivity, survey became a popular tool to gather
information from respondents. Especially, along with the spreading of internet, online
survey tends to be the more cost-effective than traditional method such as paper surveys
or face-to-face interviews. Online survey assists researchers reduce the geographical
dependence and connect to more hard-to-reach respondents in less developing time and
money.

Therefore, data of this study was collected by online surveys from March 29th to

April 20th, 2019. The criteria for participating in this study was that the respondents

</div>
(40)<div class='page_container' data-page=40>

Table 3. 1: Frequency of demographic information of respondents 
Items Number of

respondents Percentages

Total 147 100

1. Gender

- Male 45 30.6

- Female 102 69.4

2. Age

- Under 18 3 2
- 18 – 30 101 68.7
- 30 – 50 40 27.2

- Over 50 3 2

3. Income

- Under 3 million/ month 18 12
- 3 – 7 million/ month 37 24.7
- Over 7 million/ month 92 61.3

4. Living City

- Hanoi 88 58.7

- Hue 20 13.3

- Danang 5 3.3

- Ho Chi Minh 23 15.3
- Others 11 7.3

3.3.2.3. Questionnaire and Experimental Design

</div>
(41)<div class='page_container' data-page=41>

the reliability of questionnaire, research used Cronbach’s alpha test. As a result, the

Cronbach’s alpha coefficients each variable was greater than 0.7 meaning that the
questionnaire is significant and high reliable coefficients. Therefore, the questionnaire
was adjusted to avoid the probabilities of misinterpretation, distraction and partiality
through ensure the design of questionnaire assists to collect the expected responses from
participants. The final questionnaire is shown at appendixes section.

</div>
(42)<div class='page_container' data-page=42>

Table 3. 2: Study variable areas and corresponding section of the questionnaire

No. Variables Items Sources

1

Graphic 
Packaging

Design

The color combination on the packaging draws my attention

Olawepo (2015)

2 The color combination can easily be remembered

3 The color combination makes product stands out among another brand

4 The shape of packaging is unique compared to another brand

5 The shape of packaging is comfortable to use

6 The Font used on the product is legible and can be understood

7 The Font used in writing Ingredient composition is legible and could be interpreted

8 The Font used on the product attracts attention from distance

9 The picture quality of the product packaging draws my attention

10 The picture of the product packaging reflects the fact that it is healthy

11 Structural

Packaging 
Design

The size of packaging meets my demand

Salem (2017)

12 Packaging material is made from recycle materials

13 Packaging material has high quality

14 Packaging material is environmentally friendly

15

Verbal 
Packaging

Design

Brand name on packaging draws my attention

Salem (2017)

16 Brand name on packaging is unique compared to another brand

17 Brand name on packaging is easy to remember

18 Product information on packaging is described clearly

19 Product information on packaging effects trust for the product

20 Storage information on packaging is easy to follow

21 I react more favorably to product packaging imprinted in Vietnamese

22 Product information on packaging (such as: the name of the firm, address, country of origin, production and expiry date) is clearly printed 
23

Involvement 
Level

In selecting from the many types and brands of Instant Oats available in the market, I will care a great deal as to which one I
buy.

Mittal (1989) 
24 I think that the various types and brands of Instant Oats available in the market are all very different.

25 To me, making a right choice of instant oats is very important

26 In making my selection of Instant Oats, I concern about the outcome of my choice

27

Purchase 
Intention

I would be willing to buy Instant Oats of this brand

Weisstein (2017) 
& Pei (2014) 
28 If I were going to buy Instant Oats, the probability of this brand is high

29 The probability that I would consider buying the instant oats of this brand is high

</div>
(43)<div class='page_container' data-page=43>

3.4. Data Analysis

3.4.1. Reliability Analysis

The Cronbach’s alpha value is a popular tool to purify research measurements.
Nunnally and Bemstein (1994) suggested that the Cronbach’s Alpha value of each
variable should be greater than .70 threshold. To gain the possible highest reliability
coefficient, the variables are purified by deleting items which have the lowest
item-to-total correlation or items which have the value “Cronbach’s alpha if item deleted”
is greater than total Cronbach’s Alpha value. In this study, there has 4 variables with
30 items were measured in this section. When the Cronbach’s Alpha coefficient
reaches to expected value, the analysis goes to the continuous improvement cycle
stage.

3.4.2. Continuous improvement cycle

</div>
(44)<div class='page_container' data-page=44>

Table 3. 3: Suggested procedure for improve measurement construct validity 
Test Procedure

Internal

consistency Cronbach’s Alpha
Construct validity

(EFA approach)

Unidimensional: Factor loadings

Convergent Validity: Eigen value, Variance Extracted-VE,
Reliability

Construct validity
(CFA approach)

Convergent Validity: t-values, squared correlations
Fits and unidimensional assessment: Fits and indices
Discriminant Validity: constrained model pairs; Variance
Extracted versus squared correlation between factors
Composite Reliability; Variance Extracted

3.4.3. Survey data analysis

The research data was analyzed through statistical techniques including
descriptive statistics, quantitative data analysis (e.g. EFA, CFA, SEM, etc). This

study adopted CFA to validate construct measurements while structural equation 
modeling (SEM) for assess research hypotheses.

3.4.3.1. Exploratory factor analysis (EFA)

Exploratory factor analysis (EFA) aim to explore the underlying structure of a
certain set of variables. In the other words, EFA clarified the pattern of correlations
among variables by discovering underlying factors. According to Gorsuch (1983),
EFA is adopted to following below reasons:

• To narrow down the large quantity of items to smaller one for modelling purposes
where larger group of items may interrupt modelling process of all measurements
individually.

• To select a subgroup of factors from larger group by identifying highest
correlations with the principal component factors.

• To identify uncorrelated items to avoid multicollinearity while adopting multiple
regression.

</div>
(45)<div class='page_container' data-page=45>

EFA includes three fundamental stages: “(1) assessment of suitability of data
for factor analysis, (2) factor extractions and (3) factor rotation”. Hence, preliminary
analysis should involve some assumptions (indicated in table 3.3) to tested data
suitability before conducting EFA. To check the linear relationship between variables,
Hair et al (2010) recommend the usage of Plotted-Point (P-P plots) corresponding
with the ideal line for linearity to exist. In addition, multicollinearity occurs when
among independent variable has high intercorrelation level and leads to unreliable
probability values (P-value) and larger confidence intervals of independent variables.
The value of variance inflating factor (VIF) is used to identify multicollinearity
occurrences (VFI > 10).

Table 3. 4: EFA requirement assumptions

Condition Requirement References

Outliers No outliers accepted (Hair et al., 2010)
Linearity No multicollinearity (Hair et al., 2010)
Normality Should be Normally distributed (Hair et al., 2010)

Sample size Minimum: 5 cases to each study items (Pallant, 2005; Tabachnick&Fidell, 2007) 
Bartlett’s test

of sphericity Be significant (p < 0.5) (Tabachnick&Fidell, 2007)

Kaiser-Meyer-Olkin (KMO)

Index ≥ 0.5

(Hair et al., 2010; Malhotra,
2007)

The Kaiser-Myer-Olkin (KMO) is used to measure the adequacy of sampling.
The KMO, is considered as the best method for determining the acceptability of data
for subsequent factor analysis. Tabachnick & Fidell (2007) indicated that if the KMO
values is too small or accounts in the range 0 to 1.0, the factor analysis should not be
operated. The KMO need to be 0.6 or higher to run factor analysis.

</div>
(46)<div class='page_container' data-page=46>

Factor extraction is the important analysis way of EFA. Researchers use factor
extraction to identify what factor can summarize the interrelationship among

variables. Hair et al (2010) indicated that tactor extraction consists of 7 methods
which are “principal components analysis (PCA), unweighted least squares,
generalized least squares, maximum likelihood, principal axis factoring, alpha
factoring, and image factoring”. This study used principle component analysis which
is the most popular used in EFA. Tabachnick & Fidell (2007) claimed that PCA is
the good choice for researchers who are interested in empirical summary rather than
theoretical solution.

Meanwhile, Hair et al (2010) indicated that the items which have factor
loading greater than 0.50 are valuable for further analysis. To select which factors
remained in scale, Hair et al (2010) also suggests that researchers should keep factors
have eigenvalue greater than 1.0 to use in further examination. When the quantity of
retained factors were addressed, researcher aim to determine the pattern of loading
for interpretation by rotation. There exist 2 direction for rotation including orthogonal
(varimax) and oblique rotation. This study adopted varimax rotation which purpose
to “simplify factors by maximizing the variance of the loadings within factors, across
variables” (Tabachnick & Fidell, 2007).

3.4.3.2. Confirmatory factor analysis (CFA)

</div>
(47)<div class='page_container' data-page=47>

Table 3. 5: Model diagnostics in CFA

Model Diagnostic Requirement References

Modification Index (MI) ≥ 4

(Hair et al., 2010)
Standardized Residual < 2.5 no problem

> 4 possible problem

Path Estimate (Construct

to Indicator) ≥ 5 ; ideally ≥ 0.7; and be significant
Square Multiple

correlations (SMC) or
reliability

≥ 0.3

Table 3. 6: Model fit indices

Fit Indexes Acceptable level

Chi-square Value with non-significant p-value 
Normed Chi-square (CMIN/df) ≤ 3

Goodness-of-fit Index (GFI) 0 < GFI ≤ 1
Adjusted Goodness-of-fit Index (AGFI) 0 < AGFI ≤ 1
Tucker-Lewis Index (TLI) 0 < TLI ≤ 1
Comparative Fit Index (CFI) 0 < CFI ≤ 1
Root Mean Square of Error of Estimation

(RMSEA) 0 < RMSEA ≤ 1

3.4.3.3. Structural equation modelling (SEM)

Author used structural equation modelling (SEM) to describe specifically the
research model and analyze the relationship between independent and dependent
variables as well as interaction effect of moderator addressed in this study.

</div>
(48)<div class='page_container' data-page=48>

• Identify the coeficient values in the linear model framework
• Test the model acceptability for representing the studied processes

• According to model acceptability testing, conclude that the reliability of variables
relationship

Particularly, SEM assist researches conduct the hypothesis models of marker
behavior and assess research model statistically. Furthermore, SEM predicts the
unknown coefficients in a group of linear structural equation system which normally
involves observed variables and related latent variables (unobserved variables). In
addition, during analysis process, SEM assumes there exists a causal-relation
between a certain number of observed variables and latent variables and observed
variables play as indicators role.

Latent Variables

Latent variable defined as unobserved or unmeasured variable, which has
theoretical constructs and can be directly measurable, normally be referred as “factors”
or “common factors” which are assumed than can be observed when they have the
significant influence on the outcome resulted by observed variables. Additionally,
latent variables can directly affect other latent

Observed Variables

</div>
(49)<div class='page_container' data-page=49>

Figure 3. 2: Observed variables example

In the figure 3.2, the linking between latent and observed variables was
connected by the single-headed arrow indicating the influences of latent variables on
the outcome from observed variables under a regression relationship.

Structural equation modelling does not only include latent and observed
variables but also has the “residual” and “error” elements, attached to each variable,
are the key elements of the general model. In conclusion, SEM is “a complex
interplay between a large number of observed and latent variables with residual and
error terms” (Maclean, 1998).

3.4.3.4. Hypothesis testing by SEM

According to Kaplan (2001), there has two stage in SEM analysis including
structural part and measurement part.

Structural stage: connecting the constructs of each variables. This stage shows

the dependent constructs as linear functions of the independent constructs (Kaplan,
2001).

Measurement stage: connecting the constructs of observed measurements.

This stage is similar to CFA model and uses the same cut-off points value and
P-value (> 0.05) to assess the significant relationship among variables (Kaplan, 2001).

</div>
(50)<div class='page_container' data-page=50>

CHAPTER 4: DATA ANALYSIS RESULTS

4.1. Measurement Scale Test 
4.1.1. Cronbach’s Alpha

Cronbach’s Alpha coefficients are coefficients in statistical tests that are used
to check the correlation between observed variables, in order to analyze the scale
reliability assessment. To purify measurement, this method allows analyst to remove

unsuitable variables and unreliable variables in the research measurement. The
variables which have corrected items-total correlation values greater than 0.3 and
Cronbach's Alpha coefficient greater than 0.6 are acceptable and appropriately used
in the next stage (Nunnally and Burnstein, 1994). Many researchers agree that when
Cronbach’s Alpha is between 0.7 and 1, the scale of measurement is good and the
correlation will be higher.

Table 4. 1: Cronbach’s Alpha Results 
No. Variables Number

of Items

Cronbach’s 
Alpha 
Coefficient

1 Graphic Element 10 0.718
2 Structural Element 4 0.924
3 Verbal Element 6 0.837
4 Involvement Level 4 0.795
5 Purchase Intention 4 0.964

(Data analysis by SPSS 20)

</div>
(51)<div class='page_container' data-page=51>

Table 4. 2: Removed variables 
Variables Deleted items

Verbal
Elements

VB2: “Brand name on packaging is unique compared to another
brand”

VB3: “Brand name on packaging is easy to remember”

Overall, after eliminating 2 item, the satisfied 5 variables with 28 items were
used for exploratory factor analysis step.

4.1.2. Exploratory Factor Analysis (EFA)

In this research, factor analysis will help us consider the possibility of reducing
28 observed variables down to a smaller number to reflect particularly the impact of
packaging elements on the consumer buying intention. The original model has 5
latent variables with 28 observed variables which affected the intention to buy
Vietnamese brand of instant oats, is used for EFA stage.

As mentioned in the previous chapter, remained items were analyzed by
exploratory factor analysis (EFA) with Principal components method for extraction
and Varimax method for rotation. The results (table 4.3) indicated that 4 factors were
extracted from measurement scales with extraction sum of squared loadings being
about 63.24% (greater than 50%). The KMO index was significant at 0.853 and the
Bartlett’s Test of Sphericity had chi- square= 2540.660, df= 378 and sig= .000.

Table 4. 3: Exploratory Factor Analysis Results

Condition Value Requirement

KMO index 0,853 0,5 < 0,853 <1
Sig. (Bartlett’s Test) 0,000 0,000 < 0,05
Total Variance Explained 63,240 63,240 > 50%

Eigenvalue 1,146 1,146 > 1

(Data analysis by SPSS 20)

Accordingly,

• KMO = 0.801 proved that factor analysis is appropriate;

• Sig. (Bartlett’s Test) = 0,000 (Sig <0.05) indicates that observed variables are
correlated in the overall;

• Eigenvalue = 1.168> 1 represented the variation explained by each variable and
the collected variables has the best summary information;

</div>
(52)<div class='page_container' data-page=52>

Table 4. 4: Rotated components results

Component

1 2 3 4 5

The color combination on the packaging draws my

attention .713

The color combination can easily be remembered .803

The color combination makes product stands out

among another brand .828

The shape of packaging is unique compared to another

brand .809

The shape of packaging is comfortable to use .682

The Font used on the product is legible and can be

understood .767

The Font used in writing Ingredient composition is

legible and could be interpreted .772

The Font used on the product attracts attention from

distance .644

The picture quality of the product packaging draws my

attention .764

The picture of the product packaging reflects the fact

that it is healthy .786

Brand name on packaging draws my attention .675

Product information on packaging is described clearly .733

Product information on packaging effects trust for the

product .738

Storage information on packaging is easy to follow .748

I react more favorably to product packaging imprinted

in Vietnamese .747

Product information on packaging (such as: the name of
the firm, address, country of origin, production and

expiry date) is clearly .733

I would be willing to buy Instant Oats of this brand .854

If I were going to buy Instant Oats, the probability of

this brand is high .846

The probability that I would consider buying the instant

oats of this brand is high .820

The probability that I would purchase the instant oats of

this brand is high .819

In selecting from the many types and brands of Instant

Oats available in the market, I will care a great deal as
to which one I buy.

.844
I think that the various types and brands of Instant Oats

available in the market are all very different. .812

To me, making a right choice of instant oats is very

important .753

In making my selection of Instant Oats, I concern about

the outcome of my choice .702

The size of packaging meets my demand .753

Packaging material is made from recycle materials .738

Packaging material has high quality .699

Packaging material is environmentally friendly .618

</div>
(53)<div class='page_container' data-page=53>

According to the table “Rotated components results”, after exploratory factors
analysis, all of 28 items were sorted in to 5 groups with factor loading greater than
0.5.

Accordingly, the 5 latent variables were renamed as below:

• LATENT VARIABLES 1, named as “GRAPHIC ELEMENT”, has eigenvalue
equal to 7.992 > 1. There have 10 observed variables were used to measure this
latent variable.

• LATENT VARIABLES 2, named as “VERBAL ELEMENT”, has eigenvalue
equal to 3.480 > 1. There have 6 observed variables were used to measure this
latent variable.

• LATENT VARIABLES 3, named as “PURCHASE INTENTION”, has eigenvalue
equal to 3.054 > 1. There have 4 observed variables were used to measure this
latent variable.

• LATENT VARIABLES 4, named as “INVOLVEMENT LEVEL”, has eigenvalue
equal to 2.432 > 1. There have 10 observed variables were used to measure this
latent variable.

</div>
(54)<div class='page_container' data-page=54>

4.1.3. Confirmatory Factor Analysis (CFA)

Along with purifying observed variable discussed in chapter 3, confirmatory
factor analysis is also used to measure the relevance of the model to primary data.
Chi-square (CMIN); Normed Chi-square (CMIN /df); CFI - Comparative Fit Index;
TLI - Tucker & Lewis index); RMSEA index - Root Mean Square Error
Approximation are the values used for test the fitness of research model.

Additionally, according to Nguyen Khanh Duy (2009): The model is
considered suitable for primary data if the Chi-square test has P-value > 0.05; If the
model receives a probability value of Chi-square greater than 0.08 or GFI and CFI
index close to 1 and RMSEA index below 0.08 (Browne and Cudek, 1992). In the
research which has CMIN/df < 3 (with sample n <200), the model is considered to be
a good fit (Kettinger and Lee, 1995). The above rules are summarized in the table as

below:

Table 4. 5: Model fit indices

(Source: Nguyen Khanh Duy, 2009)

In CFA analysis, based on standardized weights of the variables, there exit no
variables were removed because the standardized weight of the observed variables
had the allowable weights (> = 0.5) and statistically significant p is equal to 0.000.
The specific results are shown in the following table:

Table 4. 6: Confirmatory Factor Analysis results

CMIN/df RMSEA GFI TLI CFI

1.413 0,053 0,826 0,934 0,941

(Data analysis by AMOS 22)

</div>
(55)<div class='page_container' data-page=55>

Convergent validity

The standardized regression weights of 28 measurements were used to check
this criterion. As a result, the loadings are greater than 0.05 – recommended values
by Anderson & Gerbing (1884) with the highest and the lowest values corresponding
to 0.955 and 0.528. Additionally, the composite reliability (CR) and average variance
extracted (AVE) were calculated. All 5 latent variables had CR values greater than
0.6 (Bagozzi & Yi, 1988) and AVE value above 0.5 (Fornell & Larcker, 1981).
Particularly, the AVE values of verbal element and structural element were less than
0.5 at 0.462 and 0.401 respectively, however, according to SheuFen et al (2012) that
values were acceptable. In general, all those values of 5 variables achieved the

convergent validity requirements. The specific results were indicated in following
table.

Table 4. 7: Composite reliability and AVE results 
Latent Variables Composite reliability AVE

Graphic Elements 0.925 0.553
Structural Elements 0.726 0.401
Verbal Elements 0.837 0.462
Involvement Level 0.826 0.545
Purchase Intention 0.965 0.873

Discriminant validity

</div>
(56)<div class='page_container' data-page=56>

Table 4. 8: The new latent variables

No. Observed variables Latent variables

1 The color combination on the packaging draws my attention

GRAPHIC 
ELEMENT

2 The color combination can easily be remembered

3 The color combination makes product stands out among another brand

4 The shape of packaging is unique compared to another brand

5 The shape of packaging is comfortable to use

6 The Font used on the product is legible and can be understood

7 The Font used in writing Ingredient composition is legible and could be interpreted

8 The Font used on the product attracts attention from distance

9 The picture quality of the product packaging draws my attention

10 The picture of the product packaging reflects the fact that it is healthy

1 The size of packaging meets my demand

STRUCTURAL 
ELEMENT

2 Packaging material is made from recycle materials

3 Packaging material has high quality

4 Packaging material is environmentally friendly

1 Brand name on packaging draws my attention

VERBAL 
ELEMENT

2 Product information on packaging is described clearly

3 Product information on packaging effects trust for the product

4 Storage information on packaging is easy to follow

5 I react more favorably to product packaging imprinted in Vietnamese

6 Product information on packaging (such as: the name of the firm,

address, country of origin, production and expiry date) is clearly

1

In selecting from the many types and brands of Instant Oats
available in the market, I will care a great deal as to which one I
buy.

INVOLVELMENT 
LEVEL

2 I think that the various types and brands of Instant Oats available in the market are all very different.
3 To me, making a right choice of instant oats is very important

4 In making my selection of Instant Oats, I concern about the outcome of my choice
1 I would be willing to buy Instant Oats of this brand

PURCHASE 
INTENTION

2 If I were going to buy Instant Oats, the probability of this brand

is high

3 The probability that I would consider buying the instant oats of

this brand is high

4 The probability that I would purchase the instant oats of this brand is high

</div>
(57)<div class='page_container' data-page=57>

4.2. Research Model Test

This study analyzed two times including (1) the model without moderator and
(2) the model with moderator effects.

4.2.1. Hypothesis testing without moderation of involvement level

The structural model contained all of three fundamental interaction had degree
of freedom = 245. The results consist of Chi-Square = 388.004, Chi-square/df = 1.584,
p-value = 0.000, GFI = 0.831, TLI = 0.926, CFI = 0.934, RMSEA = 0.063 (Table
4.9). The Chi-square/df (CMIN/df) was less than 2 and GFI, TLI, CFI indices
achieved satisfied values. Hence, the structural model has satisfactory fit to collected
data.

Table 4. 9: The model fit test of structural model

CMIN/df RMSEA GFI TLI CFI

1.584 0,063 0,831 0,926 0,934

(Data analysis by AMOS 22)

The influences of packaging element on consumer purchase intention tested

by SEM had analyzed results as following table.

Table 4. 10: Research model without moderator tested by SEM

Interaction Estimate S.E. C.R. P-value

INTENTION ß GRAPHIC 0.634 0.115 5.509 0.000
INTENTION ß STRUCTURE 1.332 0.221 6.025 0.000
INTENTION ß VERBAL 0.470 0.227 2.072 0.038

(Data analysis by AMOS 22)

</div>
(58)<div class='page_container' data-page=58>

verbal elements at , were 1.332 and 0.470, respectively. The hypothesis testing
results was summarized in the table as below.

Table 4. 11: Hypotheses testing without moderator results

H1 Graphic packaging element positively influence consumer’s instance oats purchase intention Supported 
H2 Structural packaging element positively influence consumer’s instance oats purchase intention Supported 
H3 Verbal packaging element positively influence

consumer’s instance oats purchase intention Supported 
In general, visual package elements have stronger impact on customer
purchase intention than verbal elements.

Figure 4. 1: Research Hypothesis structural equation modeling

</div>
(59)<div class='page_container' data-page=59>

4.2.2. Involvement Level effect testing by SEM

To measure the moderation impacts of involvement level, three interactions

were calculated by multiplying independent variables and moderators (Sauer&Dick,
1993). They were presented as ZMGRxZMIL, ZMTRxZMIL and ZMVBxZMIL.

Table 4. 12: Coding of variable computing

Coding Equation

ZMGR The mean value of variable “Graphic Element”

ZMST The mean value of variable “Structural Element”

ZMVB The mean value of variable “Verbal Element”

ZMIL The mean value of variable “Involvement Level”

ZMPI The mean value of variable “Purchase Intention”

ZMVBxZIL The multiple value between the mean value of independent variable

“Verbal element” and moderator “Involvement Level”

ZMSTxZIL The multiple value between the mean value of independent variable

“Structural element” and moderator “Involvement Level”

ZMGRxZIL The multiple value between the mean value of independent variable

“Graphic element” and moderator “Involvement Level”
The moderator effect on three interaction between packaging elements and
purchase intention were tested by SEM in AMOS as the following figure

Figure 4. 2: Moderator Effect

</div>
(60)<div class='page_container' data-page=60>

In this model, ZMGRxZMIL, ZMTRxZMIL and ZMVBxZMIL were
considered as the multiple results of moderator “Involvement level” and graphic,
structural, verbal elements, respectively. To investigate the moderator effects on each
package elements on purchase intention, researcher tested the interaction of above
multiple variables and purchase intention by P-value consideration.

The results of SEM analysis, indicated that research model achieved the
requirement of model fit, specifically shown in the table 4.12.

Table 4. 13: Moderator effect model fit

CMIN/df RMSEA GFI TLI CFI

1.824 0,075 0,960 0,796 0,896

(Data analysis by AMOS 22)

Thereafter, the moderator effects of involvement level on the interaction
between packaging elements and purchase intention was tested through SEM in
AMOS had findings as below.

Interaction Estimate S.E. C.R. P-value

INTENTION ß ZMVBxZMIL 0.043 0.062 0.683 0.495

INTENTION ß ZMSTxZMIL 0.114 0.057 1.998 0.045

INTENTION ß ZMGRxZMIL -0.042 0.055 -0.756 0.450

(Data analysis by AMOS 22)

</div>
(61)<div class='page_container' data-page=61>

Generally, the results of hypothesis testing are shown in following table:

Table 4. 14: Hypothesis testing results

H1 Graphic packaging element positively influences consumer’s instance oats purchase intention Supported 
H2 Structural packaging element packaging positively

influences consumer’s instance oats purchase intention Supported

</div>
(62)<div class='page_container' data-page=62>

CHAPTER 5: CONCLUSION

5.1. Discussion of findings

Based on the analysis results from gathered data, this chapter indicated the
findings discussion and some recommendation for future related study. Researcher
has considered two main elements of packaging design (i.e. visual and verbal
attributes) to advance the argument in this study. The visual elements consist of can
be seen attributes such as color, shape, size, picture, material, whilst verbal elements
associated with words including brand name, product information and language.

The analysis results proved that graphic attributes have a positive impact on
consumer’s buying intention for following reasons: the graphic package design has
attractive and easily memorable color. These findings are similar to previous
researches (e.g. Salem, 2017; Pensasitorn, 2015; Krimi et al., 2013; etc) which
claimed that package color and image evoke consumer attention and easy to
remember. Thus, understanding buyer’s response to package color assist marketers

and designers enhance their product’s competitiveness on the shelves. Besides,
package shape can make product more attractive and additionally enhance the
convenience in usage, thus, convenience package shape can boost buyers purchase a
certain product. Besides, for the font style used on package, it can make product more
reliable and highlighted among alternative, thus, the font needs to be legible and have
the appropriate size to attract buyers from distance. The findings also proved that the
picture on the package has ability to evoke buyer’s feeling and play a main role in
transferring product’s usage instruction as well as its function.

</div>
(63)<div class='page_container' data-page=63>

package. Today, consumers concern more about environmental issues, thus, they
prefer environment friendly packaging along with requiring a package material is
good enough to protect products itself from external attacks.

The findings also indicated that verbal elements have positive influences on
consumer purchase intention due to following reasons: the brand name draw buyer’s
attention; the products information is clearly described; the printed information
storage is easily to follow and consumers prefer package is printed in local language,
thus, selecting a suitable language for product label plays an important role in
transferring effectively message to consumers. These findings align to existing
studies (e.g. Salem, 2017; Adam & Ali, 2014; Mutsikiwa, 2013; etc) indicated that
informational element is a decisive factor while making a buying decision. Buyers
often make purchase intention based on printed package information. Reading
information on package makes buyers evaluate product quality even though visual
elements draw their attention at the beginning.

</div>
(64)<div class='page_container' data-page=64>

Differently, involvement level is found that have no influence on the
interaction between graphic element and purchase intention and the relationship of
verbal element and purchase intention. This result, is in contrast with existing study
(Silayoi and Speece, 2004, 2007; Imiru, 2017), which caused by the different in
products brand collection methods. Specifically, related previous study quantitatively

did research in total food industry with various kinds of product, differently, this
study focused on a certain product type – instant oats with a certain number of
selected brands. Besides, participants were asked to evaluate only one brand they
selected leading to the inconsistency of data which distorted the research accuracy
which caused the contrary results to previous research. Additionally, the different of
respondent demographic characteristics might be another reason. For instance, in
Vietnamese’s mindset, package is the products image which reflect quality and the
thoroughness manufacturers put into products, thus, there is no difference to the
degree of graphic package design requirement for low or high involvement products.
In the other words, Vietnamese consumers always pay attention to graphic package
design no matter how much involvement level is; thus, their purchase intention is
easily affected by them (graphic design). Similarly, in Vietnam, most of buyers,
purchase instant oats, have high health consciousness and safe consciousness,
obviously, reading product information becomes their positive habits during
shopping. Their purchase intention is likelihood to be influenced by verbal element
in case of consuming both low and high involvement products.

5.2. Managerial implication

</div>
(65)<div class='page_container' data-page=65>

implication resulted from this study to communicate with their target customers and
help their products stand out in highly competitive market. Last but not least, in term
of marketing implication, this study provides the understanding of instant oats
consumer’s response to graphic, structural and verbal packaging design. Thus, it may
help the relevant companies to increase their knowledge about consumer behavior to
design an effective communication tool – package.

5.3. Practical implication

The findings of this study can be used by managers and marketers to design
an effective packaging to ensure their products stand out among competitions. Today

consumers are becoming more careful when purchasing food products, thus, graphic,
structural and verbal elements should be designed accurately and informatively
provide ingredient contents. Specifically, based on findings, this study provides some
practical implication to marketers and designers in designing an effective package:
• The package design (color, shape, font, picture, brand name) should be different

and memorable. For example, we easily recognize the red can of Coke or the blue
can of Pepsi on the shelf without reading its brand name.

• The package design should truly reflect the product information with clear font
style and easy-to-follow storage information.

• Product should be sold in varied package size to meet different quantity demand
of buyers.

• Package material should be appropriate with product shelf life and transportation
condition. Due to the increased number of environmental consciousness
consumers, designers should consider the environmentally friendly condition of
package material.

5.4. Limitation and further research 
Limitations

</div>
(66)<div class='page_container' data-page=66>

part of the target group. Besides, convenient random sampling method will also
reduce the accuracy of research results.

Limitations on research subjects: This paper only analyzes the impact of
packaging design of three Vietnamese brands of instant oats, so it does not reflect
results objectively.

Limitation on research brand selection: due to the participants were asked to
choose then evaluate different brands, thus, there exist the inconsistency of collected
data which might affect research accuracy.

Limitations on the scope of the study: The study is only conducted based on
cultural, economic and social factors in some city in Vietnam with very small quantity,
thus, the objectivity of the topic is also limited due to each area has different buying
intentions.

In addition to the above limitations, this study excluded other important factors
affecting consumers' buying intention (such as price, promotion, taste, experiences
etc.,) which may affect more or less the accuracy of the research topic.

Recommend for future research

Future studies need to have a more general view of the market and understand
the related research demand to provide more detailed and accurate factors that are
likelihood to influence intention buying, avoiding the omission of the influencing
factor that reduces the accuracy of the study.

</div>
(67)<div class='page_container' data-page=67>

REFFERENCES

1. Estiri, M., Hasangholipour, T., Yazdani, H., Nejad H.J., and Rayej, H., 2010.
Food Products Consumer Behaviors:

2. Rettie, R. and Brewer, C. (2000) “The verbal and visual components of package
design”, Journalof Product & Brand Management, Vol. 9 No. 1, pp. 56-70.
3. Silayoi, P. and Speece, M. (2004), “Packaging and purchase decisions: a focus

group study on the impact of involvement level and time pressure”, British Food

Journal, Vol. 106 No. 8, pp. 607-28.

4. Silayoi, P. And Speece, M. (2005) The importance of packaging attributes: a
conjoint analysis approach; European Journal of Marketing, Vol. 41 No. 11/12,
2007, pp. 1495-1517, Emerald Group Publishing Limited

5. Silayoi, P., & Speece, M. (2007). The importance of packaging attributes: a
conjoint analysis approach. European Journal of Marketing, 41(11/12),
1495-1517

6. Underwood, R.L. (2003), “The communicative power of product packaging:
creating brand identity via lived and mediated experience”, Journal of Marketing
Theory and Practice, Vol. 11 No. 1, pp. 62-77.

7. Underwood, R. L., Klein, N. M., & Burke, R. R. (2001). Packaging
communication: attentional effects of product imagery. Journal of Product &
Brand Management, 10(7), 403-422.

8. S.T. Wang Edward, (2013) "The influence of visual packaging design on
perceived food product quality, value, and brand preference", International
Journal of Retail & Distribution Management, Vol. 41 Issue: 10, pp.805-816
9. Nicholas Ford, Paul Trott & Christopher Simms (2016) Exploring the impact of

packaging interactions on quality of life among older consumers, Journal of
Marketing Management, 32:3-4, 275-312

10. Lockamy, A. (1995), “A conceptual framework for assessing strategic packaging
decisions”, International Journal of Logistics Management, Vol. 6 No. 1, 7th
edition.

11. McDonagh, D., Bruseburg, A. and Halsam, C. (2002), “Visual product
evaluation: exploring users’ emotional relationships with products”, Applied
Ergonomics, Vol. 33 No. 3, pp. 231-240.

</div>
(68)<div class='page_container' data-page=68>

13. Deliya, M. M., & Bhaveshkumar, P. (2012). Role of Packaging on Consumer
Buying Behavior—Patan District. Global Journal of Management and Business
Research, 12(10).

14. Vila, N., & Ampuero, O. (2007). The role of packaging in positioning an orange
juice. Journal of Food Products Marketing, 13(3), 21-48.

15. Zafar U. Ahmed, J. P. (2004). Does country of origin matter for low-involvement
products? International Marketing Review, 21 (1), 102-120.

16. Wang, E. S. (2013). The influence of visual packaging design on perceived food
quality, value and brand preference. International Journal of Retail &
Distribution Management, 41 (10), 805- 816.

17. Rundh, B. (2005). The multi-faceted dimension of packaging marketing logistic
or marketing tool? British Food Journal, 107 (9), 670-684.

18. Rundh, B. (2009), “Packaging design: creating competitive advantage with
product packaging”, British Food Journal, Vol. 111 No. 9, pp. 988-1002.

19. Rundh, B. (2016), “The role of packaging within marketing and value creation”,
British Food Journal, Vol. 118 No. 10, pp. 2491-2511.

20. Zimbabwe. IOSR- Journal of Business and Management \, 8 (5), 64-71.

21. Ayu, K., & Harimukti, W. (2012). Analyzing the factors that affecting

consumer’s purchase intention. Journal of Consumer Research, 16(4), 461-471. 
22. Emily, M. (2010). Importance of packaging branding in marketing of color

meaning and preferences. Journal of International Marketing, 8(4), 90-107. 
23. Mohammed Z. Salem, (2018) "Effects of perfume packaging on Basque female

consumers purchase decision in Spain", Management Decision.

24. Heide, M., Olsen, S.O.: Influence of packaging attributes on consumer evaluation
of fresh cod. Food Qual. Prefer. 60, 9–18 (2017)

25. Chang, T.-Z., & Wildt, A. R. (1994). Price, product information, and purchase
intention: An empirical study. Journal of the Academy of Marketing Science, 22,
16–27.

26. Clement, J., Kristensen, T., & Grønhaug, K. (2013). Understanding consumers’
in-store visual perception: The influence of package design features on visual
attention. Journal of Retailing and Consumer Services, 20, 234–239.

</div>
(69)<div class='page_container' data-page=69>

28. Gomez, M., Martín-Consuegra, D., & Molina, A. (2015). The importance of
packaging in purchase and usage behaviour. International Journal of Consumer
Studies, 39, 203–211.

29. Murphy, M., Cowan, C., Henchion, M. and O’Reilly, S. (2000), “Irish consumer
preferences for honey: a conjoint approach”, British Food Journal, Vol. 102 No.
8, pp. 585-98.

30. Mitchell, V.W. and Papavassiliou, V. (1999), “Marketing causes and
implications of consumer confusion”, Journal of Product & Brand Management,
Vol. 8 No. 4, pp. 319-39.

31. Hair, F.J., Anderson, E.R., Tatham, L.R. and Black, C.W. (1998), Multivariate
Data Analysis, 5th ed., Prentice-Hall, Englewood Cliffs, NJ.

32. Lowe, A.C.T. and Corkindale, D.R. (1998), “Differences in ‘culture value’ and
their effects on responses to marketing stimuli: a cross-cultural study between
Australians and Chinese from the People’s Republic of China”, European Journal
of Marketing, Vol. 32 Nos 9/10, pp. 843-67.

33. Grossman, R.P. and Wisenblit, J.Z. (1999), "What we know about consumers’
color choices"-Journal of Marketing Practice: Applied Marketing Science, Vol.
5 No. 3, pp. 78-88.

34. Philip Kotler, Veronica Wong, John Saunders, Gary Armstrong (2011)"
Principles of Marketing", 11th European edition, Pearson Education Limited.
35. Natalia Vila & Olga Ampuero (2007) The Role of Packaging in Positioning an

Orange Juice, Journal of Food Products Marketing, 13:3, 21-48.

36. Hui, M. K., Zhao, X., Fan, X., & Au, K. (2004). When does the service process
matter? A test of two competing theories. Journal of Consumer Research, 31(2),
465–475.

37. Mohammed Ali Alkhchrom & Solomon, (2011), the perceived impact on the
perceived quality and consumer

38. loyalty difference Brand, Syria, Issue 4, Vol. 27, Damascus University Journal
of Economic and Legal Sciences”.

</div>
(70)<div class='page_container' data-page=70>

40. Olsson, A. and Györei, M., 2002, "Packaging Throughout the Value Chain in the

Customer Perspective Marketing Mix", Packaging Technology and Science - An
International Journal, vol. 15, pp. 231-239

41. Pei, Z., Paswan, A. and Yan, R. (2019). E-tailer׳s return policy, consumer׳s

perception of return policy fairness and purchase intention.

42. Weisstein, F., Song, L., Andersen, P. and Zhu, Y. (2017). Examining impacts of
negative reviews and purchase goals on consumer purchase decision. Journal of

Retailing and Consumer Services, 39, pp.201-207.

43. Prashant Kumar Bhimrao M Ghodeswar, (2015),"Factors affecting consumers’
green product purchase decisions", Marketing Intelligence & Planning, Vol. 33
Iss 3 pp. 330 - 347

44. Mittal, Banwari. (1989). "Measuring Purchase-Decision Involvement."
Psychology & Marketing, 6, 147-162.

45. Randall E.Schumacker, Richard G.Lomax (2008), “A beginner’s guide to 
structural equation modeling”, 2nd edition, Lawrence Erlbaum Associates, Inc.,
Publishers.

46. Muthen, B. O., & Asparouhov, T. (2003). Modeling interactions between latent
and observed continuous variables using maximum-likelihood estimation in
mplus. Mplus Web Notes #6.

47. Jaccard, J., & Wan, C. K. (1995). Measurement error in the analysis of interaction
effects between continuous predictors using multiple regression: Multiple indi-
cator and structural equation approaches. Psychological Bulletin, 117, 348-357.

48. Joreskog, K. G., & S ̈orbom, D. (1996). LISREL 8: Structural equation modeling.

Chicago: Scientific Software International.

49. Klein, A., & Moosbrugger, H. (2000). Maximum likelihood estimation of latent
interaction effects with the lms method. Psychometrika, 65, 457-474.

50. Klein, A., & Muthen, B. O. (2002). Quasi maximum likelihood estimation of
struc- tural equation models with multiple interaction and quadratic effects.
Unpublished manuscript, Graduate School of Education, University of California,
Los Angeles.

</div>
(71)<div class='page_container' data-page=71>

52. Chou, C.-P. & Bentler, P.M. (1995). Estimates and tests in structural equation
modeling. In: Hoyle, R.H. (ed.). Structural equation modeling: concepts, issues

and applications. Newbury Park, CA: Sage.

53. Cohen, J. (1982). Statistical power analysis for the behavioral sciences. San 
Diego, CA: Academic Press.

54. Hoyle, R.H. (1995) (ed.). Structural equation modeling: concepts, issues and

applications. Newbury Park, CA: Sage.

55. Wright, S. (1921). Correlation and causation. Journal of Agricultural Research, 
20, 557- 585.

56. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: indeed, a silver bullet.
Journal of Marketing Theory and Practice, 19(2), 139–151

57. Rigdon, E. E. (1998). Structural equation modeling. In G. A. Marcoulides (Ed.),
Modern method.

58. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equations with
unobservable variables and measurement errors. Journal of Marketing Research,
18(1), 39-50.

59. Hopper, D., Coughlan, J., & Mullen, M. R. (2008). Structural Equation
Modeling: Guidelines for determining model fit. Electronic Journal of Business
Research Method, 6(1), 53-60.

60. Mutsikiwa et al., (2013) “The Impact of Aesthetics Package Design Elements on
Consumer Purchase Decisions: A Case of Locally Produced Dairy Products in
Southern Zimbabwe”, IOSR Journal of Business and Management, Volume 8,
Issue 5 (Mar. - Apr. 2013), PP 64-71.

61. Vitalija Butkeviien, Jurgita Stravinskien telion (2008), “Impact of Consumer
Package Communication on Consumer Decision Making Process”, European
Journal of Business and Management ISSN 2222-1905 (Paper) ISSN 2222-2839
(Online) Vol.5, No.8, 2013

62. Giang Thu Quan (2018), “Vietnamese consumers tend to choose healthy drinks”,
Article of Vietnam Finance Magazine, published in 07/02/2018,

</div>
(72)<div class='page_container' data-page=72>

64. Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a
determinant of argument-based persuasion. Journal of Personality and Social

Psychology, 41(5), 847-855.

65. Smith, P. and Taylor, J. (2004), Marketing Communications: An Integrated
Approach, 4th ed., Kogan, London.

66. Solomon, M. (2012), Consumer Behavior, 10th ed., Pearson Education, Upper
Saddle River, NJ.

67. Pilditch, J. (1957), The Silent Salesman: How to Develop Packaging that Sells,
B.T. Batsford Limited, London.

</div>
(73)<div class='page_container' data-page=73>

APPENDIXES

APPENDIXES 1: ONLINE SURVEY

Link:

</div>
(74)<div class='page_container' data-page=74></div>
(75)<div class='page_container' data-page=75></div>
(76)<div class='page_container' data-page=76></div>
(77)<div class='page_container' data-page=77></div>
(78)<div class='page_container' data-page=78></div>
(79)<div class='page_container' data-page=79></div>
(80)<div class='page_container' data-page=80></div>
(81)<div class='page_container' data-page=81></div>
(82)<div class='page_container' data-page=82></div>
(83)<div class='page_container' data-page=83></div>
(84)<div class='page_container' data-page=84></div>
(85)<div class='page_container' data-page=85></div>
(86)<div class='page_container' data-page=86></div>
(87)<div class='page_container' data-page=87></div>
(88)<div class='page_container' data-page=88></div>
(89)<div class='page_container' data-page=89></div>
(90)<div class='page_container' data-page=90></div>
(91)<div class='page_container' data-page=91></div>
(92)<div class='page_container' data-page=92></div>
(93)<div class='page_container' data-page=93></div>
(94)<div class='page_container' data-page=94></div>
(95)<div class='page_container' data-page=95></div>
(96)<div class='page_container' data-page=96></div>
(97)<div class='page_container' data-page=97></div>
(98)<div class='page_container' data-page=98></div>
(99)<div class='page_container' data-page=99></div>
(100)<div class='page_container' data-page=100></div>
(101)<div class='page_container' data-page=101></div>
(102)<div class='page_container' data-page=102></div>
(103)<div class='page_container' data-page=103></div>
(104)<div class='page_container' data-page=104></div>
(105)<div class='page_container' data-page=105></div>
(106)<div class='page_container' data-page=106></div>
(107)<div class='page_container' data-page=107></div>
(108)<div class='page_container' data-page=108></div>
(109)<div class='page_container' data-page=109></div>
(110)<div class='page_container' data-page=110></div>
(111)<div class='page_container' data-page=111></div>
(112)<div class='page_container' data-page=112></div>
(113)<div class='page_container' data-page=113></div>
(114)<div class='page_container' data-page=114></div>
(115)<div class='page_container' data-page=115></div>
(116)<div class='page_container' data-page=116></div>
(117)<div class='page_container' data-page=117></div>
(118)<div class='page_container' data-page=118></div>
(119)<div class='page_container' data-page=119>

DATA ANALYSIS

APPENDIXES 2: Descriptive analysis

Gender

Frequency Percent Valid Percent Cumulative
Percent
Valid

Male 45 30.0 30.6 30.6
Female 102 68.0 69.4 100.0
Total 147 98.0 100.0

Missing System 3 2.0
Total 150 100.0

Age

Frequency Percent Valid Percent Cumulative
Percent

Valid

Under 18 3 2.0 2.0 2.0
18-30 101 67.3 68.7 70.7
30-50 40 26.7 27.2 98.0
Over 50 3 2.0 2.0 100.0
Total 147 98.0 100.0

Missing System 3 2.0
Total 150 100.0

Income

Frequency Percent Valid
Percent

Cumulative
Percent

Valid

Under 3 millions/

month 18 12.0 12.2 12.2

3-7 million/ month 37 24.7 25.2 37.4
Over 7 trieu/

month 92 61.3 62.6 100.0

Total 147 98.0 100.0

Missing System 3 2.0

</div>
(120)<div class='page_container' data-page=120>

City

Frequency Percent Valid Percent Cumulative
Percent

Valid

Hanoi 88 58.7 59.9 59.9

Hue 20 13.3 13.6 73.5

Danang 5 3.3 3.4 76.9

Ho Chi Minh 23 15.3 15.6 92.5

Khac 11 7.3 7.5 100.0

Total 147 98.0 100.0
Missing System 3 2.0

Total 150 100.0

APPENDIXES 3: Cronbach’s Alpha 
GRAPHIC ELEMENTS

Reliability Statistics

Cronbach's
Alpha

N of Items
.924 10

Item-Total Statistics

Scale Mean if
Item Deleted

Scale Variance
if Item Deleted

Corrected
Item-Total
Correlation

Cronbach's

Alpha if Item

</div>
(121)<div class='page_container' data-page=121>

STRUCTURAL ELEMENT 
Reliability Statistics

Cronbach's
Alpha

N of Items
.718 4

Item-Total Statistics

Scale Mean if
Item Deleted

Scale Variance
if Item Deleted

Corrected
Item-Total
Correlation

Cronbach's
Alpha if Item

Deleted
ST1 10.85 3.717 .440 .698
ST2 10.75 3.573 .497 .662
ST3 10.61 4.062 .516 .657

ST4 10.61 3.403 .588 .603

VERBAL ELEMENTS 
TIME 1

Reliability Statistics

Cronbach's
Alpha

N of Items
.664 8

Item-Total Statistics

Scale Mean if
Item Deleted
Scale
Variance if
Item Deleted
Corrected
Item-Total
Correlation
Cronbach's
Alpha if Item

</div>
(122)<div class='page_container' data-page=122>

TIME 2

Reliability Statistics

Cronbach's

Alpha N of Items
.837 6

Item-Total Statistics

Scale Mean if
Item Deleted

Scale Variance
if Item Deleted

Corrected
Item-Total
Correlation

Cronbach's
Alpha if Item

Deleted
VB1 19.86 6.740 .560 .821
VB4 19.89 6.495 .623 .808
VB5 19.98 6.595 .607 .811
VB6 19.84 6.425 .634 .806
VB7 19.84 6.503 .614 .810
VB8 19.95 6.216 .632 .807

INVOLVEMENT LEVEL 
Reliability Statistics

Cronbach's
Alpha

N of Items
.795 4

Item-Total Statistics

Scale Mean if
Item Deleted

Scale Variance
if Item Deleted

Corrected
Item-Total
Correlation

Cronbach's
Alpha if Item

Deleted
IL1 11.96 3.341 .507 .789
IL2 11.95 2.868 .657 .718
IL3 11.94 2.784 .703 .694
IL4 12.11 2.961 .564 .766

PURCHASE INTENTION 
Reliability Statistics

Cronbach's
Alpha

</div>
(123)<div class='page_container' data-page=123>

Item-Total Statistics

Scale Mean if
Item Deleted

Scale Variance
if Item Deleted

Corrected
Item-Total
Correlation

Cronbach's
Alpha if Item

Deleted
PI1 10.26 17.714 .918 .952
PI2 10.04 18.505 .930 .947
PI3 10.01 19.390 .912 .953
PI4 9.94 19.647 .892 .959

APPENDIXES 4: EFA

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .853

Bartlett's Test of Sphericity

Approx. Chi-Square 2540.660

df 378

Sig. .000

Total Variance Explained

Component Initial Eigenvalues Extraction Sums of

Squared Loadings

Rotation Sums of Squared
Loadings
Tota
l
% of
Varianc
e
Cumulativ
e %

Total % of
Varianc

Cumulat

ive %

Total % of
Varianc

Cumulati
ve %

1 7.99

2 28.545 28.545

7.99

2 28.545 28.545
6.22

0 22.216 22.216

2 3.48

0 12.428 40.973

3.48

0 12.428 40.973
3.52

8 12.601 34.817

3 3.05

4 10.907 51.879

3.05

4 10.907 51.879
3.36

9 12.031 46.848

4 2.43

2 8.687 60.566

2.43

2 8.687 60.566

2.59

1 9.255 56.103

5 1.14

6 4.094 64.660

1.14

6 4.094 64.660

2.39

6 8.557 64.660

6 .910 3.251 67.912

7 .878 3.136 71.048

8 .820 2.930 73.978

9 .710 2.535 76.513

</div>
(124)<div class='page_container' data-page=124>

11 .665 2.376 81.295

12 .565 2.017 83.312

13 .506 1.807 85.120

14 .496 1.770 86.889

15 .466 1.663 88.553

16 .428 1.529 90.082

17 .382 1.364 91.445

18 .347 1.241 92.686

19 .336 1.202 93.888

20 .315 1.126 95.014

21 .298 1.066 96.080

22 .252 .899 96.979

23 .217 .777 97.756

24 .169 .604 98.360

25 .147 .527 98.886

26 .140 .501 99.388

27 .111 .397 99.785

28 .060 .215 100.000

Extraction Method: Principal Component Analysis.
Component Matrixa

Component

1 2 3 4 5
GR3 .775

PI1 .768

PI3 .756
PI4 .733
PI2 .733
GR4 .726
GR1 .717
GR7 .716
GR2 .689
GR5 .678
GR6 .677
GR9 .675
GR10 .657
GR8 .595

ST4 .599
ST3 .516
ST2

</div>
(125)<div class='page_container' data-page=125>

VB7 .655
VB5 .641
VB8 .620
VB4 .612
VB1 .569

IL3 .758

IL2 .700

IL4 .692

IL1 .638

ST1

Extraction Method: Principal Component Analysis.
a. 5 components extracted.

APPENDIXES 5: CFA

Notes for Model (Default model)

Computation of degrees of freedom (Default model)

Number of distinct sample moments: 406
Number of distinct parameters to be estimated: 67
Degrees of freedom (406 - 67): 339

Result (Default model)

Minimum was achieved
Chi-square = 478.959
Degrees of freedom = 339
Probability level = .000

Model Fit Summary 
CMIN

Model NPAR CMIN DF P CMIN/DF
Default model 67 478.959 339 .000 1.413
Saturated model 406 .000 0

Independence model 28 2730.820 378 .000 7.224

</div>
(126)<div class='page_container' data-page=126>

Model RMR GFI AGFI PGFI
Default model .057 .826 .791 .689
Saturated model .000 1.000

Independence model .375 .285 .232 .265

Baseline Comparisons

Model Delta1 NFI rho1 RFI Delta2 IFI rho2 TLI CFI
Default model .825 .804 .941 .934 .941
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000

Parsimony-Adjusted Measures

Model PRATIO PNFI PCFI
Default model .897 .740 .843
Saturated model .000 .000 .000
Independence model 1.000 .000 .000

NCP

Model NCP LO 90 HI 90
Default model 139.959 86.069 201.877
Saturated model .000 .000 .000
Independence model 2352.820 2190.616 2522.433

FMIN

Model FMIN F0 LO 90 HI 90
Default model 3.281 .959 .590 1.383
Saturated model .000 .000 .000 .000
Independence model 18.704 16.115 15.004 17.277

RMSEA

Model RMSEA LO 90 HI 90 PCLOSE
Default model .053 .042 .064 .311
Independence model .206 .199 .214 .000

</div>
(127)<div class='page_container' data-page=127>

Model AIC BCC BIC CAIC
Default model 612.959 646.172 813.318 880.318
Saturated model 812.000 1013.265 2026.116 2432.116
Independence model 2786.820 2800.700 2870.552 2898.552

ECVI

Model ECVI LO 90 HI 90 MECVI
Default model 4.198 3.829 4.622 4.426
Saturated model 5.562 5.562 5.562 6.940
Independence model 19.088 17.977 20.250 19.183

HOELTER

Model HOELTER
.05

HOELTER

.01
Default model 117 123
Independence model 23 24

Estimates (Group number 1 - Default model)

Scalar Estimates (Group number 1 - Default model) 
Maximum Likelihood Estimates

Regression Weights: (Group number 1 - Default model)

Estimate S.E. C.R. P Label
GR3 <--- GRAPHIC 1.000

GR4 <--- GRAPHIC .870 .073 11.975 ***
GR2 <--- GRAPHIC .852 .078 10.887 ***
GR10 <--- GRAPHIC .739 .068 10.826 ***
GR7 <--- GRAPHIC .864 .076 11.335 ***
GR6 <--- GRAPHIC .861 .083 10.336 ***
GR9 <--- GRAPHIC .815 .083 9.808 ***
GR1 <--- GRAPHIC .701 .070 9.996 ***
GR5 <--- GRAPHIC .712 .076 9.342 ***
GR8 <--- GRAPHIC .711 .095 7.484 ***
VB6 <--- VERBAL 1.000

</div>
(128)<div class='page_container' data-page=128>

Estimate S.E. C.R. P Label
VB5 <--- VERBAL .936 .130 7.203 ***
VB1 <--- VERBAL .849 .128 6.611 ***
PI2 <--- INTENTION 1.000

PI1 <--- INTENTION 1.056 .043 24.423 ***
PI4 <--- INTENTION .893 .042 21.046 ***
PI3 <--- INTENTION .925 .039 23.856 ***
IL3 <--- INVOLVEMENT 1.000

IL2 <--- INVOLVEMENT .744 .082 9.062 ***
IL1 <--- INVOLVEMENT .626 .081 7.755 ***
IL4 <--- INVOLVEMENT .747 .092 8.083 ***
ST4 <--- SRUCTURE 1.000

ST1 <--- SRUCTURE .766 .144 5.319 ***
ST3 <--- SRUCTURE .724 .117 6.181 ***
ST2 <--- SRUCTURE .929 .149 6.249 ***

Standardized Regression Weights: (Group number 1 - Default model)

</div>
(129)<div class='page_container' data-page=129>

Estimate
IL1 <--- INVOLVEMENT .648
IL4 <--- INVOLVEMENT .673
ST4 <--- SRUCTURE .712
ST1 <--- SRUCTURE .528
ST3 <--- SRUCTURE .635
ST2 <--- SRUCTURE .644

Covariances: (Group number 1 - Default model)

Estimate S.E. C.R. P Label
GRAPHIC <--> VERBAL .120 .043 2.756 .006
GRAPHIC <--> INTENTION .647 .133 4.876 ***
GRAPHIC <--> INVOLVEMENT -.003 .078 -.040 .968

GRAPHIC <--> SRUCTURE .045 .056 .801 .423
VERBAL <--> INTENTION .220 .069 3.206 .001
VERBAL <--> INVOLVEMENT -.023 .043 -.544 .587
VERBAL <--> SRUCTURE .029 .031 .948 .343
INTENTION <--> INVOLVEMENT .230 .124 1.864 .062
INTENTION <--> SRUCTURE .532 .107 4.967 ***
INVOLVEMENT <--> SRUCTURE .092 .059 1.553 .121
e7 <--> e10 .278 .065 4.273 ***
Correlations: (Group number 1 - Default model)

Estimate
GRAPHIC <--> VERBAL .277
GRAPHIC <--> INTENTION .486
GRAPHIC <--> INVOLVEMENT -.004
GRAPHIC <--> SRUCTURE .081
VERBAL <--> INTENTION .320
VERBAL <--> INVOLVEMENT -.053
VERBAL <--> SRUCTURE .101
INTENTION <--> INVOLVEMENT .173
INTENTION <--> SRUCTURE .603
INVOLVEMENT <--> SRUCTURE .165
e7 <--> e10 .410

Variances: (Group number 1 - Default model)

</div>
(130)<div class='page_container' data-page=130>

Estimate S.E. C.R. P Label
VERBAL .224 .050 4.502 ***
INTENTION 2.113 .271 7.783 ***
INVOLVEMENT .840 .144 5.847 ***
SRUCTURE .369 .086 4.288 ***

e1 .342 .049 6.978 ***
e2 .323 .044 7.291 ***
e3 .432 .057 7.647 ***
e4 .331 .043 7.663 ***
e5 .388 .052 7.516 ***
e6 .520 .067 7.783 ***
e7 .544 .069 7.888 ***
e8 .381 .049 7.856 ***
e9 .477 .060 7.978 ***
e10 .840 .102 8.209 ***
e11 .231 .033 6.975 ***
e12 .248 .034 7.241 ***
e13 .263 .038 6.889 ***
e14 .227 .032 7.016 ***
e15 .228 .032 7.149 ***
e16 .260 .035 7.546 ***
e17 .203 .037 5.492 ***
e18 .308 .049 6.300 ***
e19 .362 .050 7.204 ***
e20 .258 .040 6.494 ***
e21 .298 .073 4.076 ***
e22 .351 .056 6.292 ***
e23 .456 .062 7.398 ***
e24 .567 .079 7.218 ***
e25 .359 .061 5.926 ***
e26 .558 .074 7.575 ***
e27 .287 .042 6.834 ***
e28 .448 .067 6.742 ***

Squared Multiple Correlations: (Group number 1 - Default model)

</div>
(131)<div class='page_container' data-page=131></div>
(132)<div class='page_container' data-page=132>

APPENDIXES 6: SEM

Computation of degrees of freedom (Default model)

Number of distinct sample moments: 300
Number of distinct parameters to be estimated: 55
Degrees of freedom (300 - 55): 245

Result (Default model)

Minimum was achieved
Chi-square = 388.004
Degrees of freedom = 245
Probability level = .000

Model Fit Summary 
CMIN

Model NPAR CMIN DF P CMIN/DF
Default model 55 388.004 245 .000 1.584
Saturated model 300 .000 0

Independence model 24 2457.623 276 .000 8.904

RMR, GFI

Model RMR GFI AGFI PGFI
Default model .060 .831 .793 .678
Saturated model .000 1.000

Independence model .434 .267 .203 .245

Baseline Comparisons

Model Delta1 NFI rho1 RFI Delta2 IFI rho2 TLI CFI
Default model .842 .822 .935 .926 .934
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000

Parsimony-Adjusted Measures

</div>
(133)<div class='page_container' data-page=133>

Model PRATIO PNFI PCFI
Saturated model .000 .000 .000
Independence model 1.000 .000 .000

NCP

Model NCP LO 90 HI 90
Default model 143.004 93.328 200.611
Saturated model .000 .000 .000
Independence model 2181.623 2026.825 2343.813

FMIN

Model FMIN F0 LO 90 HI 90
Default model 2.658 .979 .639 1.374
Saturated model .000 .000 .000 .000
Independence model 16.833 14.943 13.882 16.054

RMSEA

Model RMSEA LO 90 HI 90 PCLOSE
Default model .063 .051 .075 .037
Independence model .233 .224 .241 .000

AIC

Model AIC BCC BIC CAIC
Default model 498.004 520.732 662.478 717.478
Saturated model 600.000 723.967 1497.130 1797.130
Independence model 2505.623 2515.540 2577.393 2601.393

ECVI

Model ECVI LO 90 HI 90 MECVI
Default model 3.411 3.071 3.806 3.567
Saturated model 4.110 4.110 4.110 4.959
Independence model 17.162 16.102 18.273 17.230

</div>
(134)<div class='page_container' data-page=134>

Model HOELTER .05 HOELTER .01 
Default model 107 113
Independence model 19 20

Scalar Estimates (Group number 1 - Default model) 
Maximum Likelihood Estimates

Regression Weights: (Group number 1 - Default model)

Estimate S.E. C.R. P Label

INTENTION <--- GRAPHIC .634 .115 5.509 ***
INTENTION <--- VERBAL .470 .227 2.072 .038
INTENTION <--- STRUCTURE 1.332 .221 6.025 ***
GR3 <--- GRAPHIC 1.000

GR4 <--- GRAPHIC .870 .073 11.979 ***
GR2 <--- GRAPHIC .852 .078 10.885 ***
GR10 <--- GRAPHIC .739 .068 10.825 ***
GR7 <--- GRAPHIC .863 .076 11.318 ***
GR6 <--- GRAPHIC .862 .083 10.339 ***
GR9 <--- GRAPHIC .816 .083 9.813 ***
GR1 <--- GRAPHIC .701 .070 9.992 ***
GR5 <--- GRAPHIC .712 .076 9.326 ***
GR8 <--- GRAPHIC .712 .095 7.496 ***
VB6 <--- VERBAL 1.000

VB7 <--- VERBAL .944 .133 7.118 ***
VB8 <--- VERBAL 1.094 .146 7.509 ***
VB4 <--- VERBAL .973 .132 7.358 ***
VB5 <--- VERBAL .928 .129 7.189 ***
VB1 <--- VERBAL .846 .128 6.626 ***
PI2 <--- INTENTION 1.000

PI1 <--- INTENTION 1.057 .043 24.469 ***
PI4 <--- INTENTION .893 .042 21.037 ***
PI3 <--- INTENTION .925 .039 23.765 ***
ST4 <--- STRUCTURE 1.000

</div>
(135)<div class='page_container' data-page=135>

Standardized Regression Weights: (Group number 1 - Default model)

Estimate
INTENTION <--- GRAPHIC .399
INTENTION <--- VERBAL .153
INTENTION <--- STRUCTURE .555
GR3 <--- GRAPHIC .842
GR4 <--- GRAPHIC .814
GR2 <--- GRAPHIC .765
GR10 <--- GRAPHIC .762
GR7 <--- GRAPHIC .785
GR6 <--- GRAPHIC .738
GR9 <--- GRAPHIC .712
GR1 <--- GRAPHIC .721
GR5 <--- GRAPHIC .686
GR8 <--- GRAPHIC .580
VB6 <--- VERBAL .704
VB7 <--- VERBAL .669
VB8 <--- VERBAL .712
VB4 <--- VERBAL .695
VB5 <--- VERBAL .677
VB1 <--- VERBAL .618
PI2 <--- INTENTION .955
PI1 <--- INTENTION .941
PI4 <--- INTENTION .908
PI3 <--- INTENTION .935
ST4 <--- STRUCTURE .710
ST1 <--- STRUCTURE .530
ST3 <--- STRUCTURE .634
ST2 <--- STRUCTURE .645

Covariances: (Group number 1 - Default model)

Estimate S.E. C.R. P Label
GRAPHIC <--> VERBAL .120 .044 2.758 .006
GRAPHIC <--> STRUCTURE .045 .056 .802 .423
VERBAL <--> STRUCTURE .029 .031 .940 .347
e7 <--> e10 .277 .065 4.263 ***

</div>
(136)<div class='page_container' data-page=136>

Estimate
GRAPHIC <--> VERBAL .277
GRAPHIC <--> STRUCTURE .081
VERBAL <--> STRUCTURE .100
e7 <--> e10 .410

Variances: (Group number 1 - Default model)

</div>
(137)<div class='page_container' data-page=137>

Squared Multiple Correlations: (Group number 1 - Default model)

</div>
(138)<div class='page_container' data-page=138>

APPENDIXES 7: MODERATOR

Notes for Model (Default model)

Computation of degrees of freedom (Default model)

Number of distinct sample moments: 36
Number of distinct parameters to be estimated: 22
Degrees of freedom (36 - 22): 14

Result (Default model)

Minimum was achieved

Chi-square = 25.536
Degrees of freedom = 14
Probability level = .030

Model Fit Summary 
CMIN

Model NPAR CMIN DF P CMIN/DF
Default model 22 25.536 14 .030 1.824
Saturated model 36 .000 0

Independence model 8 138.533 28 .000 4.948

RMR, GFI

Model RMR GFI AGFI PGFI
Default model .069 .960 .897 .373
Saturated model .000 1.000

Independence model .151 .828 .778 .644

Baseline Comparisons

Model Delta1 NFI rho1 RFI Delta2 IFI rho2 TLI CFI
Default model .816 .631 .907 .791 .896
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000

</div>
(139)<div class='page_container' data-page=139>

Model PRATIO PNFI PCFI
Default model .500 .408 .448

Saturated model .000 .000 .000
Independence model 1.000 .000 .000

NCP

Model NCP LO 90 HI 90
Default model 11.536 1.114 29.755
Saturated model .000 .000 .000
Independence model 110.533 77.518 151.081

FMIN

Model FMIN F0 LO 90 HI 90
Default model .175 .079 .008 .204
Saturated model .000 .000 .000 .000
Independence model .949 .757 .531 1.035

RMSEA

Model RMSEA LO 90 HI 90 PCLOSE
Default model .075 .023 .121 .171
Independence model .164 .138 .192 .000

AIC

Model AIC BCC BIC CAIC
Default model 69.536 72.426 135.325 157.325
Saturated model 72.000 76.730 179.656 215.656
Independence model 154.533 155.584 178.457 186.457

ECVI

</div>
(140)<div class='page_container' data-page=140>

HOELTER

Model HOELTER
.05

HOELTER
.01
Default model 136 167
Independence model 44 51

Scalar Estimates (Group number 1 - Default model) 
Maximum Likelihood Estimates

Regression Weights: (Group number 1 - Default model)

Estimate S.E. C.R. P Label
ZMPI <--- ZMGR .409 .060 6.843 ***
ZMPI <--- ZMST .439 .058 7.505 ***
ZMPI <--- ZMVB .186 .060 3.106 .002
ZMPI <--- ZMIL .109 .058 1.871 .061
ZMPI <--- ZMSTxZIL .115 .057 2.007 .045

Standardized Regression Weights: (Group number 1 - Default model)

Estimate
ZMPI <--- ZMGR .419
ZMPI <--- ZMST .449
ZMPI <--- ZMVB .191

ZMPI <--- ZMIL .112
ZMPI <--- ZMSTxZIL .120

Covariances: (Group number 1 - Default model)

</div>
(141)<div class='page_container' data-page=141>

Correlations: (Group number 1 - Default model)

Estimate
ZMGR <--> ZMIL .014
ZMIL <--> ZMGRxZIL -.124
ZMGR <--> ZMGRxZIL -.042
ZMIL <--> ZMSTxZIL -.047
ZMST <--> ZMSTxZIL .038
ZMIL <--> ZMVBxZIL .130
ZMVB <--> ZMIL -.057
ZMVB <--> ZMVBxZIL -.015
ZMGR <--> ZMVB .216

Variances: (Group number 1 - Default model)

Estimate S.E. C.R. P Label
ZMGR .994 .116 8.545 ***
ZMST .993 .116 8.544 ***
ZMVB .994 .116 8.544 ***
ZMIL 1.003 .117 8.547 ***
ZMSTxZIL 1.033 .121 8.544 ***
ZMGRxZIL 1.124 .132 8.544 ***
ZMVBxZIL .887 .104 8.544 ***
e1 .495 .058 8.544 ***

Squared Multiple Correlations: (Group number 1 - Default model)

</div>