Food Research International 85 (2016) 266–272

Contents lists available at ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Consumer acceptance of a quick response (QR) code for the food

traceability system: Application of an extended technology acceptance
model (TAM)
Yeong Gug Kim, Eunju Woo ⁎
Department of Tourism Administration, Kangwon National University, 24341 Chuncheon-si, South Korea

a r t i c l e i n f o a b s t r a c t

Article history: The objectives of this study are to apply the TAM using the addition of perceived information to individuals' be-
Received 15 February 2016 havioral intention to use the QR code for the food traceability system; and to determine the moderating effects of
Received in revised form 18 April 2016 food involvement on the relationship between perceived information and perceived usefulness. Results from a
Accepted 1 May 2016
survey of 420 respondents are analyzed using structural equation modeling. The study findings reveal that the
Available online 3 May 2016
extended TAM has a satisfactory fit to the data and that the underlying dimensions have a significant effect on
Keywords:
consumers' intention to use the QR code for the food traceability system. In addition, food involvement plays a
QR code significant moderating function in the relationship between perceived information and perceived usefulness.
Food traceability system The implications of this study for future research are discussed.
Perceived information © 2016 Elsevier Ltd. All rights reserved.
Technology acceptance model

1. Introduction
With the development of information technology, the food trace-
ability system which can reduce individuals' concerns about food safety
by providing unambiguous information about the safety and quality of
the whole process, from producers to consumers—has been broadly dis-
seminated in the food industries (e.g., Badia-Melis, Mishra, &
Ruiz-García, 2015; Cozzolino, 2014; Lin, Mark, Zetian, Trebar, &
Xiaoshuan, 2014; Melo, Andrew, & Faleiro, 2015).
The policy makers in the public health sectors of many countries
have accepted the food traceability system. For instance, the EU Rapid
Alert System for Food and Feed (RASFF) employed the tracking software
system Grapenet, an internet-based electronic service for certification of
grapes for export from India to the EU (Badia-Melis et al., 2015). In the
US, food traceability information has been mandatory since 2002, and
facilities, transport organizations, storage facilities, and other food han-
dlers have been responsible for recording information, including prod-
uct descriptions and providers' and recipients' addresses and phone
numbers if the goods cross state lines (US FMSA, 2011). South Korea
also introduced a beef traceability system, which traces the distribution
channel for beef (Badia-Melis et al., 2015).
The Quick Response (QR) code, one of the traceability systems, has
been introduced to the food industries as a two-dimensional barcode
(e.g., Shin, Jung, & Chang, 2012; Tarjan et al., 2011; Tarjan, Senk,
⁎ Corresponding author.
E-mail addresses: yeongkim@kangwon.ac.kr (Y.G. Kim), ejwoo@kangwon.ac.kr
(E. Woo).
Srdjan, Stankovski, & Ostojic, 2015). The QR code can hold considerably
more information than the one-dimensional code, as it can embed text,
video, advertisements, personal information, etc.
The QR code can be integrated into users' smartphone applications;
that is, the smartphone can scan and decode information and messages
about products that the QR code provides. The use of the QR code is in-
creasing globally (Shin et al., 2012; Tarjan et al., 2011), but even with its
introduction for food traceability in the food industry (Fig. 1), there has
been limited research on consumer acceptance of its usefulness for pro-
viding food information or the use of the QR code for the food traceabil-
ity system in the context of food research.
A significant number of studies have indicated that the technology
acceptance model (TAM) is a suitable psychometric tool with which to
assess consumers' acceptance of technology, determined by the individ-
ual's perception of the new technology's usefulness (e.g., Venkatesh &
Davis, 1996). Therefore, the first purpose of this study is to apply the
TAM to individuals' acceptance of the QR code for the food traceability
system.
While, Chen and Huang (2013) examined the moderating effect of
involvement between uncertainty, formed by the lack of information
on foods, and consumer behavior. They found that the higher one's de-
gree of involvement and the more the food traceability system mitigates
their uncertainty, the greater their intention to buy a food. Other previ-
ous studies have also stressed that the function of consumer involve-
ment in foods is a topic worth investigation (Karlsen, Sorensen, Foras,
& Olsen, 2011; Verbeke & Vackier, 2004). Therefore, the other purpose
of the present study is to determine the moderating effects of food in-
volvement on the relationship between constructs of the TAM.

http://dx.doi.org/10.1016/j.foodres.2016.05.002
0963-9969/© 2016 Elsevier Ltd. All rights reserved.
 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272
Fig. 1. An example of the QR code for the food traceability.
Source: 6 Degrees Counterfeit Prevention (2013) (www.6dcp.com).
2. Conceptual framework and hypotheses development
2.1. Technology acceptance model (TAM) and perceived information
Derived from the theory of human reasonability developed by Ajzen
and Fishbein (1980), the TAM explains the determinants of users'
acceptance of technology (Davis, 1989; Davis, Bagozzi, & Warshaw,
1989). The TAM shows that a user's attitude toward a particular technol-
ogy is determined by the individual's perceived usefulness (PU) and per-
ceived ease of use (PEOU) of the technology. The TAM is supported by the
relationships among belief, attitude, and behavior (Davis et al., 1989).
PU refers to the degree to which a person believes that using a par-
ticular technology will enhance his or her job performance. Davis et al.
(1989) found that there was a positive relationship between PU and
the intention to use a word processing system and indicated that PU ap-
peared to have a more significant relationship with usage behavior and
intention than with variables like attitude, satisfaction, and perception.
PEOU refers to the degree to which a person believes that using a partic-
ular technology will be effortless. A number of studies have found that
PEOU is significantly correlated with current usage and future usage
(Davis, 1989) and with a user's attitude toward acceptance of a system
(e.g., Venkatesh & Davis, 1996). Behavioral intention (BI) refers to the
strength of a user's intention to undertake a particular behavior, and
attitude refers to the user's overall feeling about performing the target
behavior. Hence, the TAM describes how an individual's beliefs, such
as PU and PEOU, affect his or her attitude toward using a system, and
BI is determined by those attitudes (Davis, 1989; Davis et al., 1989).
Given these findings, this study hypothesizes that:
H1. Perceived ease of use (PEOU) positively affects perceived useful-
ness (PU).
H2. Perceived usefulness (PU) positively affects attitude toward using
(ATT).
H3. Perceived ease of use (PEOU) positively affects attitude toward
using (ATT).
H4. Attitude toward using (ATT) positively affects behavioral intention
to use (BI).
While, research has noted the significance of perceived information
(PI) in the acceptance of innovative technology and its link to
267
consumers' intentions to use a new technology system (Amoako-
Gyampah, 2007). In the context of food goods and services, Cox and
Evans (2008) pointed out that the use of a new system is related to “cre-
dence qualities”; therefore, the purpose of a novel system is to provide
the food-related information—such as naturalness, brand, origin, pack-
aging, price, nutrition, ingredients, safety, sustainability, and environ-
mental effects that satisfies people's informational needs. Several
studies have also revealed that PI could have a significant effect on con-
sumers' food choices and how well they like particular food products
(Lee, Lusk, Mirosa, & Oey, 2016; Stolzenbach, Bredie, Christensen, &
Byrne, 2013).
PI has been linked in empirical studies to influencing individuals' PU
positively (e.g., Chen & Huang, 2013; Legris, Ingham, & Collerette, 2003;
Rese, Screnie, & Baier, 2014). More specifically, Chen and Huang (2013)
emphasized that the well-informed system must ensure that the infor-
mation asymmetry between people and providers of food products can
be eliminated, since people have begun to want information about what
they eat. In addition, Rese et al. (2014) found a positive influence of in-
formation on users' consumption choice and on their confidence in the
usefulness of the information provided by the innovative technology.
Thus, the present study adds PI to the original TAM as an important var-
iable to examine the PU of the QR code for the food traceability system.
Form this point of views, this study seeks to expand the TAM includ-
ing PI in order to determine the effect of the constructs on the intention
to use the QR code for the food traceability system, and given these find-
ings, hypothesis is further proposed.
H5. Perceived information (PI) positively affects perceived usefulness
(PU).
2.2. Food involvement
Involvement refers to the users' perceptions of the degree of goods'
and services' significance, interest, and relevance (e.g., Mollet &
Rowland, 2002; Pavlou & Gefen, 2004). According to Mollet and
Rowland (2002), people tend to believe that foods contribute either di-
rectly or indirectly to their health, so consumers are interested in infor-
mation on nutrition, calories, and ingredients. Involvement in food and
information is one of the key determinants in consumers' decisions re-
lated to purchasing foods (Karlsen et al., 2011). Verbeke and Vackier
268 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272
(2004) examined the relationship between consumer involvement and
fresh meat consumption. They found that a highly involved consumer
actively gathers related information, and the level of involvement
with foods varies from one individual to another. Thus the results indi-
cated that food involvement is a key element in the purchase decision.
Bell and Marshall (2003) investigated the effects of food
involvement on various relationships between the motives for
choosing foods and the consumer's BI to purchase foods. The authors
found that people's “food involvement” is a stable characteristic and
that people with a high level of food involvement discriminate
between food items in their sensory evaluations and hedonic ratings
better than those with a low level of food involvement do (Bell &
Marshall, 2003).
Bell and Marshall (2003) referred to food involvement as the extent
to which people enjoy talking about food, entertain thoughts about food
during the day, and engage in food-related activities. The authors devel-
oped a measure of food involvement using Goody's (1982) five phases
of the life cycle of food: food acquisition, preparation, cooking, eating,
and disposal. Acquisition relates to buying and shopping for food and
the associated decisions and responsibilities, while preparation is associ-
ated with the processes involved in preparing food, including washing,
cleaning, and cutting food. Cooking is the process of transforming the
product through the application of heat, either directly or indirectly, to
change the texture and palatability of the food in a way that makes it
more acceptable to the consumer. Eating involves ingesting the food,
sharing food, and serving food, and disposal is related to clearing up
the remnants of meals and snacks, washing up, and clearing the dishes.
These steps follow on sequentially as a series of behaviors.
Bell and Marshall (2003) contended that the level of involvement
can be influenced by the extent to which people take responsibility
and exercise control over these provisioning tasks, so they developed
the food-involvement scale (FIS), which consists of twelve items associ-
ated with the five activities of the food lifecycle. The authors found that
measures of food involvement were associated with discrimination and
hedonic ratings for a range of foods and suggested that food involve-
ment was a significant mediator to consider when undertaking food re-
search regarding food choice and preference. Therefore, the moderating
effect of involvement is investigated. Specifically, the study undertakes
a multiple-group analysis (i.e., a high-involvement group vs. a low-
involvement group) to measure the moderating influence of food in-
volvement on the relationships between PI and PU. From this perspec-
tive, hypothesis is proposed and the research model is designed as
shown in Fig. 2.
H6. Food involvement has a moderating effect on the relationship be-
tween PI and PU.
Fig. 2. Proposed model and hypotheses.
3. Materials and methods
3.1. Research instruments
The current study uses the extended TAM—consisting of PI, PU,
PEOU, attitude toward using (ATT)—to identify the level of consumers'
acceptance of the food traceability system and BI to measure those
structural relationships. Each construct was measured with multiple
items using a 5-point Likert scale (ranging from 1 = strongly dis-
agree/extremely unlikely to 5 = strongly agree/extremely likely)
(e.g., Davis et al., 1989; Muk & Chung, 2015; Rese et al., 2014;
Venkatesh & Davis, 1996).
This study uses a measure of the TAM drawn from eleven items to
assess consumers' acceptance of new technology related to the food
traceability system, as suggested by Muk and Chung (2015) and Rese
et al. (2014). The eleven items were divided into four categories: PU,
PEOU, ATT, and BI. Academic experts reviewed the appropriateness of
the measurement items and confirmed the reliability of the pre-
determined four categories. In order to measure the additional con-
struct PI three items were applied: 1) the QR code for the food traceabil-
ity system shows as much information as I expected; 2) the QR code for
the food traceability system provides detailed information about food;
and 3) the QR code for the food traceability system provides information
that helps me in my decisions (Bruner & Kumar, 2005; Hausman &
Siepke, 2009; Rese et al., 2014).
The twelve items from the FIS developed by Bell and Marshall
(2003) was employed to measure food involvement using 7-point
Likert scale, based on the suggestion of Bell and Marshall (2003). Two
items each were associated with acquisition, preparation, and cooking;
one item each was associated with eating and procurement; and three
items each were associated with eating and disposal (Goody, 1982)
(Table 1).
3.2. Data collection
Before the main survey was fielded, the initial questionnaire was
reviewed by experts in the business and hospitality areas for unclear ex-
pression and grammar. Then the questionnaire was pilot-tested with
fifty university students to refine the research instrument, and the
final questionnaire was scrutinized for content validity by experts in
the fields of management and consumer science.
The main survey was conducted in Seoul, South Korea, on May 1–7,
2015, with three well-trained research assistants administering the sur-
vey to a convenience sample of customers in supermarkets and near
food markets. Potential participants were asked whether they had an
 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272 269

Table 1 (AGFI), normalized fit index (NFI), root mean square error of approxi-
Items on the food involvement scale (FIS) and the result of descriptive analysis. mation (RMSEA), and root mean square residual (RMR)). Table 3 dis-
Items Mean (SD) plays the results of the CFA, and the fit indices were within accepted
I don't think much about food each day.⁎ 5.08(1.60)
thresholds: chi-square value (df) = 175.960(67); CFI = 0.967; GFI =
Cooking or barbequing is not much fun.⁎ 5.61(1.65) 0.944; AGFI = 0.912; NFI = 0.948; RMSEA = 0.062; and RMR = 0.021.
Talking about what I ate or am going to eat is something I like to do. 4.92(1.72) Construct validity was determined by assessing convergent validity
Compared with other daily decisions, my food choices are not very 4.77(1.73) and discriminant validity (Bagozzi & Yi, 1988; Ping, 2004). Construct va-
important. ⁎
lidity was established based on the factor loading estimates and con-
When I travel, one of the things I anticipate most is eating the food 5.82(1.49)
there. struct reliabilities (Hair et al., 2009). All loading estimates are
I do most or all of the clean up after eating. 4.96(1.60) significant (p b 0.00), with the lowest at 0.675 and the highest at
I enjoy cooking for others and myself. 4.55 (1.59) 0.914, and the construct reliability estimates are adequate, ranging
When I eat out, I don't think or talk much about how the food tastes.⁎ 5.21(1.50) from 0.811 to 0.943. Thus, the results support the construct validity of
I do not like to mix or chop food.⁎ 5.13(1.57)
I do most or all of my own food shopping. 5.03(1.64)
the measurement model. To ensure construct reliability, the squared
I do not wash dishes or clean the table. ⁎ 4.65(1.59) correlation coefficients between any pair of constructs should be
I care whether a table is nicely set. 5.03(1.56) lower than the AVE for each construct (Ping, 2004). Table 4 presents
⁎ Denotes items requiring reverse scoring. All items were measured on a 7-point Likert the correlation coefficients and square roots of AVEs, which indicate ac-
scale. ceptable discriminant validity.

4.2. Structural equation model

interest in filling out a questionnaire, and the research assistants ex-
plained to those who were interested what the food traceability system The extended TAM (TAM with PI) showed an acceptable level of fit,
is and how to use it. A total of 470 questionnaires were distributed, of with a chi-square value (df) of 249.918 (70), CFI = 0.945, GFI = 0.922,
which 420 (89.3%) were usable. AGFI = 0.844, NFI = 0.926, RMSEA = 0.078, and RMR = 0.035
Of the 420 respondents, 247 (58.8%) were female. More than a third (Table 5).
of the respondents (n = 141, 33.6%) were 25–34 years old, while 26% As shown in Fig. 3, all relationships among the variables were statis-
were 35–44 years old (n = 109), 15% were 45–54 years old (n = 63), tically significant: PI → PU (β = 0.531, p b 0.001); PEOU → PU (β =
and 12.1% were younger than age 25 (n = 51). More than a third of 0.283, p b 0.001); PU → ATT (β = 0.474, p b 0.001); PEOU → ATT
the respondents (n = 159, 37.9%) held a university degree, while (β = 0.254, p b 0.001); and ATT → BI (β = 0.260, p b 0.001).
28.3% held a two-year college degree (n = 119), and around 17% had
done postgraduate work (n = 74). The annual household income of al- 4.3. Moderating effect of food involvement
most a third of the respondents was USD 40,000–49,999 (30.7%),
followed by USD 30,000–39,999 (27.1%), USD 50,000 or more (21.7%), 4.3.1. Grouping
USD 20,000–29,999 (13.3%), and less than USD 20,000 (7.1%) (Table 2). To test the moderating function of food involvement on the relation-
ship between PI and PU, respondents were divided into a high food-
4. Findings involved group and a low food-involved group. The number of cases
using the 420 responses for the score of FIS was split sequentially into
4.1. Measurement model two groups, and the cases were divided into a high food-involved
group (high-FIG; n = 228) and a low food-involved group (low-FIG;
Confirmatory Factor Analysis (CFA) using AMOS 18.0 was conducted n = 192), based on a moderator score (m = 5.07), recommended by
to test the measurement model before examining the relationships be- Chandrashekaran and Grewal (2003).
tween the latent constructs and their indicators. Given the sensitivity of
the chi-square statistics to sample size (Anderson & Gerbing, 1988; Hair, 4.3.2. Invariance test of the path for hypothesized moderation
Black, Babin, Anderson, & Tatham, 2009), the overall model fit was In the next step, the invariance of specific paths was tested. The par-
assessed using six goodness-of-fit indices (i.e., comparative fit index ticular parameters of paths (PI → PU) in the nested models were
(CFI), goodness of fit index (GFI), adjusted goodness of fit index constrained to equality across groups, and paths in the baseline models
can be freely estimated. A test of invariance was conducted by investi-
gating a model with no constrained paths between the unconstrained
model and the constrained model (Anderson & Gerbing, 1988). The
Table 2 model showed acceptable model fit: χ2 (df) = 347.544 (141), and the
Description of the respondents (n = 420). chi-square difference between the unconstrained model and the
Characteristics Category N % constrained model was significant at the 0.05 level: the moderating role
of food involvement had a statistically significant impact on the relation-
Sex Male 173 41.2
Female 247 58.8 ship between PI and PU (Δχ2 = 5.956, df = 1 N χ20.5 (1) = 3.84). The dif-
Age Under 25 51 12.1 ference between them was statistically significant in both levels of food
25–34 141 33.6 involvement. The path coefficient between PI and PU was 0.546
35–44 109 26.0 (p b 0.001) for the high-FIG, while the path coefficient was 0.456
45–54 63 15.0
Above 55 56 13.3
(p b 0.001) for the low-FIG. Therefore, this finding reveled that PI was
Education High school 58 13.8 more effective in enhancing PU in the high-FIG than in the low-FIG (Fig. 3).
College 119 28.3
Undergraduate 159 37.9 5. Discussion and conclusions
Postgraduate 74 17.6
Others 10 24
Annual Less than USD 20,000 30 7.1 5.1. Discussions and implications
household income USD 20,000–29,999 56 13.3
(USD) USD 30,000–39,999 114 27.1 The current study began at the current point of the research; al-
USD 40,000–49,999 129 30.7 though previous studies have applied the TAM to consumer behaviors
USD 50,000 or more 91 21.7
(e.g., Venkatesh & Davis, 1996), limited research in the food industry
270 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272

Table 3
Results of Confirmatory Factor Analysis for the measurement model.

Factors and items Std. loadings Critical ratio Construct reliability AVE⁎

Perceived information (PI) 0.949 0.861

QR code for food traceability system shows much information I expected. 0.889 Fixed
QR code for food traceability system provides detailed information about foods. 0.886 24.56
QR code for food traceability system provides much information that helps me in my decision. 0.824 21.82
Perceived usefulness (PU) 0.866 0.684
For me, QR code for food traceability system has great value. 0.808 Fixed
I find QR code for food traceability system useful in my life. 0.757 14.24
Using QR code for food traceability system saves me much time. 0.712 13.48
Perceived ease of use (PEOU) 0.811 0.588
It is easy to learn how to use QR code for food traceability system. 0.712 Fixed
QR code for food traceability system is easy to use. 0.706 11.49
QR code for food traceability system is easy to understand. 0.675 11.11
Attitude toward using (ATT) 0.943 0.846
I am positive about QR code for food traceability system. 0.914 Fixed
The use of QR code for food traceability system is a good idea. 0.862 24.28
It makes sense to use QR code for food traceability system. 0.830 21.25
Behavioral intention to use (BI) 0.821 0.697
I will use QR code for food traceability system in the future. 0.824 Fixed
I will recommend using QR code for food traceability system to my friends. 0.732 11.83

Chi-square value (df) = 175.960(67); CFI = 0.967; GFI = 0.944; AGFI = 0.912; NFI = 0.948; RMSEA = 0.062; RMR = 0.021
Note: All items were measured on a 5-point Likert scale.
⁎ AVE: average variance extracted.

has dealt with the importance of PI in consumers' decisions related to
the QR code for the food traceability system and the moderating influ-
ences on the relationship between PI and PU in the TAM. The study ap-
plied the TAM with PI to individuals' BI to use the QR code for the food
traceability system in the context of the food industry; and determined
the moderating effects of food involvement on the relationship between
PI and PU with respect to consumers' decision-making.
The study's findings demonstrated that the extended TAM has a sat-
isfactory fit with the data and that the underlying dimensions have a
significant effect on consumers' intention to use the QR code for the
food traceability system. The effect of PI was found to be significant in
the PU of the QR code, similar to the results of previous research
(e.g., Chen & Huang, 2013; Lee et al., 2016; Rese et al., 2014; Siegrist,
2008; Stolzenbach et al., 2013). These other studies found a positive re-
lationship between PI and information usefulness and considered PI an
important antecedent of PU in the TAM.
This study's finding showed significant relationships among the con-
structs. The PI is the most powerful influence on PU (β = 0.531,
p b 0.001) in the extended TAM, and the effect of PU on ATT (β =
0.474, p b 0.001) is the strongest in the original TAM. This finding is con-
sistent with the findings of previous studies (e.g., Rese et al., 2014;
Venkatesh & Davis, 1996), which suggested the important role of PU
in consumers' acceptance of new technology. More specifically, the
work of Rese et al. (2014) indicated a positive and strong relationship
between PU and ATT, similar to findings of the current study. The result
indicates that people understand the importance and usefulness of the
QR code in the agriculture and food industries, and that this consider-
ation has a positive effect on their attitude toward using the QR code
for the food traceability system in their decision-making.
The study found that food involvement plays a significant
moderating function in the relationship between PI and PU and
that PI is more effective in improving PU in the high-FIG than in the
low-FIG. As demonstrated in past studies (e.g., Chen & Huang,
2013; Karlsen et al., 2011; Verbeke & Vackier, 2004), this study
revealed a difference in the perception of the usefulness of using
the QR code based on the degree of food involvement; that is, the
degree of PU is based on the QR code's providing detailed
information about food. Therefore, the higher a person's degree of
involvement, the greater their intention to use the QR code for the
food traceability system.
The findings of this study can help marketers to develop effective
strategies, as the results suggest that it is important to emphasize
food information for consumers. The study findings demonstrate
that consumer considers QR code for food traceability system to
provide detailed information about foods and much information
that helps me in my decision. Additionally, even though there is a
difference of degree between the high-FIG (β = 0.546, p b 0.001)
and the low-FIG (β = 0.456, p b 0.001) on the path (PI → PU), this
result showed that most respondents perceive the importance of
information on food products.
As previous works pointed out, whenever there is incomplete
information in consumers, there emerges hesitation in their mind
before making buying decisions and the perception of incomplete
information negatively influences consumer purchase intention.
Table 5
Goodness-of-fit indexes for models.
Index Recommended value⁎ Observed statistics

TAM TAM + PI
Table 4 2
Model χ (df) N/A 112.791 (39) 249.918 (70)
Construct validity of the measurement model. Explanatory power (R2) PU 0.389 0.561
ATT 0.364 0.441
F1: PI F2: PU F3: PEOU F4: ATT F5: BI
BI 0.425 0.454
F1: PI 0.861⁎ p-Value ≥0.05 0.000 0.000
F2: PU 0.563 0.684⁎ CFI 0 ≤ 1, close to 1.0 0.965 0.945
F3: PEOU 0.525 0.491 0.588⁎ GFI 0 ≤ 1, close to 1.0 0.954 0.922
F4: ATT 0.625 0.496 0.436 0.846⁎ AGFI 0 ≤ 1, close to 1.0 0.922 0.884
F5: BI 0.519 0.419 0.346 0.473 0.697⁎ NFI 0 ≤ 1, close to 1.0 0.948 0.926
Mean(SD) 3.94(0.635) 3.74(0.671) 3.62(0.645) 3.97(0.683) 3.83(0.745) RMSEA ≤0.08 0.067 0.078
RMR ≤0.05 0.025 0.035
Note: ⁎AVE, The scores range from 1 to 5. All correlations are significant at the 0.01 level or
better; PI: perceived information, PU: perceived usefulness, PEOU: perceived ease of use, Note: ⁎Recommended values were based on past studies (Byrne, 2009; Hair et al., 2009);
ATT: attitude toward using, BI: behavioral intention to use. PU: perceived usefulness, ATT: attitude toward using, BI: behavioral intention to use.
 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272
Fig. 3. Structural parameter estimates for the proposed model. Note: Numbers in parentheses are critical ratio, and numbers outside of parentheses are the standardized path coefficients;
the path coefficient between PI and PU was 0.546 in the high-FIG, the path coefficient between PI and PU was 0.456; *p b 0.01.
Furthermore, Ajzen and Fishbein indicated that consumers tend to
gather information to decrease the incomplete information, existing
in the purchase decision making. Thus, it recommends that a retail
encounter should spend as much time as possible in providing
information about: basic cue such as brand and price; pre-purchase in-
formation including product descriptions, providers' addresses and
phone numbers if the goods cross state lines; and the safety and quality
of the whole process; etc. thru the QR code for food traceability system.
Also the QR code could be used as a useful tool to protect consumers'
rights by providing more information to ensure food safety and quality,
and the QR code could be expected to mitigate uncertainty, and then
strengthen consumers' purchase intention. Therefore, clear, detailed in-
formation about food products, such as information about its natural-
ness, brand, origin, packaging, price, nutrition, ingredients, safety,
sustainability, and environmental effects, should be provided, and infor-
mation provided by the QR code should be more abundant and detailed
than consumers expect. These efforts are important mechanisms with
which to improve consumers' intention to use the QR code and pur-
chase food that provides the QR code.
5.2. Limitation and future research
Notwithstanding the contributions of this paper, certain limitations
remain that future research should seek to overcome. There may exist
selection bias regarding sampling methods. The advantages of conve-
nience sampling employed in this study, are the availability and the
quickness with which data can be gathered, however the disadvantages
are the risk that the sample may be biased by volunteers (Hair et al.,
2009). Therefore, some causation should be raised whenever there is
an intention to generalize the findings of this study. Even though selec-
tion bias is inherent in traditional data collection methods, further re-
search is needed to consider other sampling methods to minimize its
problem. Lastly, future research should consider the possible effect of
demographic factors such as marital status, occupation, education, etc.
on acceptance of the food traceability system. Even though this study
was to identify the relationship among the constructs, the TAM with
PI; and to determine individuals' behavioral intention to use the QR
code for the food traceability system, those demographic factors have
been recognized as crucial elements in food consumption
(Bakhotmah, 2012; Dickson-Spillman & Siergrist, 2010).
References
Degrees Counterfeit Prevention (2013). 6DCP in Bentham science FNA (food, nutrition,
agriculture). Retrieved 15 December 2015 from http://www.6dcp.com/
communicating-food-safety-authenticity-consumer-choice-field-experiences.
271
Ajzen, I., & Fishbein, M. (1980). Understanding attitude and predicting social behavior. En-
glewood Cliffs, NJ: Prentice-Hall.
Amoako-Gyampah, K. (2007). Perceived usefulness, user involvement and behavioral in-
tention: An empirical study of ERP implementation. Computers in Human Behavior,
23, 1232–1248.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103(3),
411–423.
Badia-Melis, R., Mishra, P., & Ruiz-García, L. (2015). Food traceability: New trends and re-
cent advances. Food Control, 57, 393–401.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of
the Academy of Marketing Science, 16(1), 74–94.
Bakhotmah, B. A. (2012). Nutritional knowledge and desire to change of food preferences
among Saudi women in Jeddah, Saudi Arabia. Ecology of Food and Nutrition, 51(4),
313–328.
Bruner, G., & Kumar, A. (2005). Explaining consumer acceptance of hand held Internet de-
vices. Journal of Business Research, 58(5), 553–558.
Bell, R., & Marshall, D. (2003). The construct of food involvement in behavioral research:
Scale development and validation. Appetite, 40, 235–244.
Byrne, B. M. (2009). Structural equation modeling with AMOS: Basic concepts, applications,
and programming (2nd ed.). New York, NY: Routledge.
Chandrashekaran, R., & Grewal, D. (2003). Assimilation of advertised reference prices:
The moderating role of involvement. Journal of Retailing, 79(1), 53–62.
Chen, M., & Huang, C. (2013). The impacts of the food traceability system and consumer
involvement on consumers' purchase intentions toward fast foods. Food Control, 33,
313–319.
Cox, D., & Evans, G. (2008). Construction and validation of a psychometric scale to mea-
sure consumers' fears of novel food technologies: The food technology neophobia
scale. Food Quality and Preference, 19, 704–710.
Cozzolino, D. (2014). An overview of the use of infrared spectroscopy and chemometrics
in authenticity and traceability of cereals. Food Research International, 60, 262–265.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of in-
formation technology. MIS Quarterly, 318–339.
Davis, F. D., Bagozzi, P. R., & Warshaw, P. R. (1989). User acceptance of computer technol-
ogy: A comparison of two theoretical models. Management Science, 35(8), 982–1002.
Dickson-Spillman, M., & Siergrist, M. (2010). Consumers' knowledge of healthy diets and
it's correlation with dietary behaviour. Journal of Human Nutrition and Dietetics, 24,
52–60.
Goody, J. (1982). Cooking, cuisine and class: A study in comparative sociology. Cambridge:
Cambridge University Press.
Hair, J., Black, B., Babin, R., Anderson, R., & Tatham, R. (2009). Multivariate data analysis
(7th ed.). New York, NY: Prentice-Hall.
Hausman, A., & Siepke, J. (2009). The effect of web interface features on consumer online
purchase intentions. Journal of Business Venturing, 62(1), 5–13.
Karlsen, K., Sorensen, C., Foras, F., & Olsen, P. (2011). Critical criteria when implementing
electronic chain traceability in a fish supply chain. Food Control, 22, 1339e1347.
Lee, P., Lusk, K., Mirosa, M., & Oey, I. (2016). Effect of information on Chinese consumers'
acceptance of thermal and non-thermal treated apple juices: A study of young Chi-
nese immigrants in New Zealand. Food Quality and Preference, 48, 118–129.
Legris, P., Ingham, J., & Collerette, P. (2003). Why do people use information technology?:
A critical review of the technology acceptance model. Information Management, 40
(3), 191–204.
Lin, Q., Mark, X., Zetian, F., Trebar, M., & Xiaoshuan, Z. (2014). C2SLDS: A WSNbased per-
ishable food shelf-life prediction and LSFO strategy decision support system in cold
chain logistics. Food Control, 38, 19–29.
Melo, J., Andrew, P. W., & Faleiro, M. L. (2015). Listeria monocytogenes in cheese and the
dairy environment remains a food safety challenge: The role of stress responses. Food
Research International, 67, 75–90.
Mollet, B., & Rowland, I. (2002). Functional foods: At the frontier between food and
pharma. Current Opinion in Biotechnology, 13, 483–485.
272 Y.G. Kim, E. Woo / Food Research International 85 (2016) 266–272
Muk, A., & Chung, C. (2015). Applying the technology acceptance model in a two-country
study of SMS advertising. Journal of Business Research, 68, 1–6.
Rese, A., Screnie, S., & Baier, D. (2014). Technology acceptance modeling of augmented re-
ality at the point of sale. Journal of Retailing and Consumer Services, 21, 869–876.
Pavlou, P., & Gefen, D. (2004). Building effective online marketplaces with institution-
based trust. Information Systems Research, 15, 37–59.
Ping, R. (2004). On assuring valid measures for theoretical models using survey data.
Journal of Business Research, 57, 125–141.
Shin, D., Jung, J., & Chang, B. (2012). The psychology behind QR codes: User experience
perspective. Computers in Human Behavior, 28, 1417–1426.
Siegrist, M. (2008). Factors influencing public acceptance of innovative food technologies
and products. Food Innovation Management, 19(11), 603–608.
Stolzenbach, S., Bredie, W. L., Christensen, R. H., & Byrne, D. V. (2013). Impact of product
information and repeated exposure on consumer liking, sensory perception and con-
cept associations of local apple juice. Food Research International, 52(1), 91–98.
Tarjan, L., Senk, I., Kovac, R., Horvat, S., Ostojic, G., & Stankovski, S. (2011). Automatic iden-
tification based on 2D barcodes. Journal of Industrial Engineering and Management, 2,
151–157.
Tarjan, L., Senk, I., Srdjan, Tegeltija, S., Stankovski, S., & Ostojic, G. (2015). A readability
analysis for QR code application in a traceability system. Computers and Electronics
in Agriculture, 109, 1–11.
US FSMA. (2011). H.R. 2751 FDA food safety modernization act. Retrieved 22 March 2012
from http://thomas.loc.gov/cgi-bin/bdquery/z?d111:H.R.2751
Venkatesh, V., & Davis, F. D. (1996). A model of the antecedents of perceived ease of use:
Development and test. Decision Sciences, 27(3), 451–458.
Verbeke, W., & Vackier, I. (2004). Profile and effects of consumer involvement in fresh
meat. Meat Science, 67, 159–168.