Food Quality and Preference 41 (2015) 103–111

Contents lists available at ScienceDirect

Food Quality and Preference

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

Chinese consumers’ understanding and use of a food nutrition label

and their determinants
Rongduo Liu a,b, Christine Hoefkens a,⇑, Wim Verbeke a
a
Department of Agricultural Economics, Ghent University, Coupure Links 653, 9000 Ghent, Belgium
b
Department of Finance, China Women’s University, Beijing 100101, China

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

Article history: This paper investigated Chinese consumers’ understanding and use of the Chinese food nutrition label
Received 18 June 2014 and their determinants. Quantitative data were collected during March 2012 through a self-adminis-
Received in revised form 10 October 2014 trated structured questionnaire conducted in Beijing (n = 213) and Baoding city (n = 447). Questions
Accepted 9 November 2014
assessed Chinese consumers’ use and understanding (objective and subjective) of food nutrition labels,
Available online 20 November 2014
nutrition knowledge (objective and subjective), socio-demographic characteristics, diet status, diet-
health awareness, body mass index (BMI) and familiarity with food nutrition labels. A moderate degree
Keywords:
of subjective understanding and a low degree of objective understanding of food nutrition labels were
Food
Nutrition label
found among the participants, and 70% of the participants claimed to rarely or never use nutrition labels
Consumer when shopping for food. Nutrition knowledge (objective and subjective) positively affected participants’
China understanding (objective and subjective) of food nutrition labels. Familiarity with food nutrition labels
Label use had a strong positive effect on understanding of and use of food nutrition labels. Subjective nutrition
Understanding knowledge and subjective understanding also played a significant and positive role in Chinese
consumers’ label use. Age yielded a negative effect on both subjective and objective understanding, while
education only affected participants’ objective understanding. None of the socio-demographic character-
istics associated with self-reported use. Implications for future policies to improve Chinese consumers’
understanding and promote their use of food nutrition labels are discussed.
Ó 2014 Elsevier Ltd. All rights reserved.

Introduction 2012) and concerned about diet-related hazards (Liu, Pieniak, &
Verbeke, 2013, 2014).
The improvement of living standards has brought about signif- In 2011 the Chinese government has announced the ‘‘Healthy
icant changes in Chinese consumers’ dietary behavior, lifestyle as China 2020” program. The program’s primary goal is to reduce
well as their health and disease pattern (Chen & Zhao, 2012). the incidence of NCDs by promoting healthy eating (Hu, Liu, &
Non-communicable chronic diseases (NCDs) such as obesity, dia- Willett, 2011). In this regard, nutrition labeling has been consid-
betes, cardiovascular diseases, hypertension and stroke have ered to be an attractive and potentially effective policy instrument
become the major causes of death in China (Chen & Zhao, 2012; as it provides nutrition information to consumers while maintain-
NCCD, 2013; Xu et al., 2013; Yang, Yang, Zhu, & Qiu, 2011). Scien- ing freedom of choice and reducing information search costs
tific evidence strongly supports the relation between diet and (Capacci et al., 2012; Grunert & Wills, 2007). Nutrition information
health (Buckland et al., 2013; Yamamoto, 2013). In China the prev- on food labels may help consumers to appraise the nutritional con-
alence of NCDs has been associated with high-fat and high-salt tribution of foods to the overall diet and, as such, to make better
diets and other unhealthy dietary choices (e.g., Kang, Guan, Ning, informed and more healthful food choices (Campos, Doxey, &
Wu, & Guan, 2012; NCCD, 2013). Notwithstanding this nutrition Hammond, 2011; Wahlich, Gardner, & McGowan, 2013). To better
transition, Chinese consumers have become increasingly aware of communicate about food and nutrition to Chinese consumers,
the relationship between food and health (Cheng, Cao, & Xu, China published its first Chinese Food Nutrition Labelling Regula-
2007; Sakamaki, Toyama, Amanoto, Liu, & Shinfuku, 2005; Zhang, tion in 2008 (MOH, 2008). In 2011 China’s Ministry of Health
released the National Food Safety Standard for Nutrition Labelling
of Pre-packaged Foods (GB 28050-2011). This regulation requires
⇑ Corresponding author. Tel.: +32 9 264 59 25; fax: +32 9 264 62 46.
that the energy value and the amount of protein, fat, carbohydrate
E-mail address: [email protected] (C. Hoefkens).
and sodium as well as their percentages in relation to Nutrient

http://dx.doi.org/10.1016/j.foodqual.2014.11.007
0950-3293/Ó 2014 Elsevier Ltd. All rights reserved.
104 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111

Reference Values (NRV) are mandatory items to be labeled on a reported to be important in shaping the acceptability of different
food nutrition label. The content of trans fatty acids shall be listed nutrition label formats (Mejean, Macouillard, Péneau, Hercberg, &
on the nutrition label if the ingredients contain hydrogenated fat Castetbon, 2013).
and/or partial hydrogenated fat, or if these trans fatty acids are Both subjective knowledge (i.e., what individuals perceive that
used in the production process. From 1st of January 2013 this they know, also indicated as perceived or self-rated knowledge)
national standard for mandatory nutrition labeling went into force. and objective knowledge (i.e., what an individual actually knows)
This changing regulatory environment is expected to have long- have been shown to influence information search and processing,
term positive effects on Chinese consumers’ diets which may even- although probably in different ways (Brucks, 1985). A low level
tually reduce the incidence of NCDs. of subjective knowledge, resulting from a lack of confidence in cur-
Based on consumer behavior models, people who understand rent knowledge, may motivate the search for additional informa-
and use nutrition labels are more likely to make better food choices tion, whilst a high level of subjective knowledge increases
(Grunert & Wills, 2007; Hoefkens, Pieniak, Van Camp, & Verbeke, reliance on previously stored information (Brucks, 1985; Ruddell,
2012). Labels need to be used in order to affect both immediate 1979). Objective knowledge facilitates deliberation and the use of
and future decisions about purchasing the product. These labels newly acquired information (Ruddell, 1979; Selnes & Gronhaug,
can also change the overall pattern of shopping, e.g., by altering 1986). Both positive and no effects of objective nutrition knowl-
the perception of food categories that are subsequently considered edge on consumers’ use of nutrition labels have been reported in
more or less healthy than before. Nutrition information on food literature (positive effects in e.g., Fitzgerald, Damio, Segura-Pérez,
labels may hence affect consumers’ dietary intake. Provided that and Pérez-escamilla (2008), Grunert, Wills, and Fernández-
the Eastern society has a stronger tendency to show socially desir- Celemín (2010), Grunert, Fernández-Celemín, et al. (2010); no
able behavior (Middleton & Jones, 2000), food nutrition labels may effects in e.g., Drichoutis, Lazaridis, Nayga, Kapsokefalou, and
potentially be very successful in China not only in terms of use but Chryssochoidis (2008), Nayga, Lipinski, and Savur (1998) and
also with regard to their impact on food choices, dietary intake and Nayga (2000)). Subjective nutrition knowledge has been found to
health. While there is a growing body of literature on the use and positively affect consumers’ use of nutrition labels (Hess,
understanding of nutrition labels in Europe and North America Visschers, & Siegrist, 2011; Petrovici & Ritson, 2006). Raju, Lonial,
(Cowburn & Stockley, 2005; Grunert & Wills, 2007), in China nutri- and Mangold (1995) and Pieniak, Aertsens, and Verbeke (2010)
tion labels are a new source of nutrition information and relevant found that subjective knowledge was related to consumer decision
research is very scarce. Given that China represents quite a differ- making more strongly than objective knowledge. Specifically, con-
ent regulatory, industrial and socio-economical context compared sumers’ decoding of nutrition messages largely depends on subjec-
to Europe or North America (Hawkes, 2008), three questions are tive inferences (van Trijp, 2009).
raised. First, do Chinese consumers understand nutrition informa- Socio-demographic factors are important because unhealthy
tion on food nutrition labels? Second, do Chinese consumers (claim eating habits are usually unequally distributed across social sub-
to) use nutrition information on food nutrition labels? Third, what groups. In addition, accounting for socio-demographic differences
factors affect their understanding and use of food nutrition labels? helps to investigate whether a lower use of nutrition information
Understanding and use of nutrition labeling are influenced by on food nutrition labels in a particular social subgroup is due to
numerous factors. The theoretical framework of Grunert and a lower understanding, lower level of nutrition knowledge or other
Wills (2007) presents the most prominent factors that have been factors (Grunert, Fernández-Celemín, et al., 2010). Conflicting
discussed in the literature and/or are likely to play a role based effects of age, gender and education on the use of nutrition labels
on consumer behavior and food choice theory. In this framework have been described in previous studies (see reviews of
consumers’ use of nutrition labels is influenced by factors associ- Drichoutis, Lazarids, and Nayga (2006) and Hieke and Taylor
ated with their understanding of the nutrition information on food (2012)), probably due to the differences in samples, method and
nutrition labels (including subjective and objective understand- study time (Drichoutis, Lazaridis, & Nayga, 2005) or the different
ing), while understanding and use are both affected by socio- sets of variables used in the different studies (Hess et al., 2011).
demographic characteristics and nutrition knowledge. Other factors such as the diet status have been reported to posi-
Understanding has a central role in the processing of informa- tively influence label use, which is mainly due to the consumers’
tion in general, and nutrition label information in particular diet-health awareness (Drichoutis et al., 2005). Obese consumers
(Grunert, Fernández-Celemín, Wills, Storcksdieck genannt (Body Mass Index (BMI) > 30 kg/m2) have been found to be more
Bonsmann, & Nureeva, 2010; Hoefkens, Veettil, Van likely to use food nutrition labels than consumers with a normal
Huylenbroeck, Van Camp, & Verbeke, 2012). Understanding nutri- weight (BMI between 18.5 and 25 kg/m2) (Satia, Galanko, &
tion information requires a certain level of nutrition knowledge Neuhouser, 2005).
which is often lacking, especially detailed knowledge about daily Familiarity with the nutrition label is found to be a key factor in
dietary needs for specific nutrients (van Trijp, 2009). A lack of consumer information processing in general, and nutrition label
understanding is an important reason for consumers not to use use in particular (EUFIC, 2012; Moorman, 1990). Consumers evalu-
nutrition labels (Besler, Buyuktuncer, & Uyar, 2012; Chen & Niu, ate themselves more able to process nutrition information if they
2009; EUFIC, 2005; Gorton, Mhurchu, Chen, & Dixion, 2009). A dis- are familiar with the information (Moorman, 1990). Consumers
tinction is made between subjective (or perceived) and objective may search food labels more actively (Grunert & Wills, 2007) or
understanding of information on food nutrition labels which may read labels more accurately (Cowburn & Stockley, 2005) if they
be quite different for one and the same consumer (Grunert & are familiar with the label. By contrast, consumers’ perceived
Wills, 2007). Objective understanding is whether a person’s inter- understanding of nutrition labels may decrease if they are not
pretation of information is consistent with what is intended by familiar with the information cues on the label and their meaning
the information sender (Grunert & Wills, 2007). Subjective under- (Burton, Biswas, & Netemeyer, 1994).
standing refers to ‘‘feeling states associated with the event, an The first objective of this study was to investigate the effects of
appraisal of the purpose or intent of the event, and an appreciation nutrition knowledge (objective and subjective), socio-demographic
of the significance of the event” (Powers, Welsh, & Wright, 1994), i.e., characteristics, diet status, diet-health awareness, BMI and famil-
the meaning a person attaches to the information and the extent iarity with food nutrition labels on consumers’ understanding
to which a person believes he/she can understand it (Grunert & (objective, subjective) of food nutrition labels in China. The second
Wills, 2007). Subjective (or perceived) understanding has been objective was to gain insights into the determinants (i.e., the same
 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111 105

factors used to explain label understanding and understanding Familiarity relates to the experience a consumer has about an
(objective, subjective) itself) of Chinese consumers’ use of food object (Bettman & Park, 1980; Park & Lessig, 1981). Familiarity
nutrition labels. Results from this study are valuable to govern- (unaided) with nutrition labels in this study measured visual famil-
mental and non-governmental organisations, food manufacturers iarity by means of a yes–no question without showing the partic-
and retailers to gain insights into the potential effectiveness of ipants an example of the food nutrition label (Aldridge, Dovey, &
nutrition labeling to provide nutrition information and to promote Halford, 2009): ‘‘Have you ever noticed food nutrition labels
healthier food choices and dietary patterns. Recommendations for before?” Participants’ awareness of the relation between diet and
how to promote nutrition label use in China will be presented. health was measured by five items (e.g., ‘‘My health is determined
by the food I eat”) using the 7-point Likert scale described by
Ragaert, Verbeke, Devlieghere, and Debevere (2004) (Cronbach’s
Methods alpha = 0.72).
Subjective knowledge is measured by asking participants to rate
Data collection how much they think they know about an object (Brucks, 1985;
Park, Mothersbaugh, & Feick, 1994). Subjective nutrition knowl-
Quantitative data were collected during March 2012 through a edge in this study was assessed by three items on a 7-point interval
self-administrated structured questionnaire conducted in Beijing scale ranging from ‘totally not know’ (=1) to ‘totally know’ (=7)
and Baoding, a medium-sized city close to Beijing. Baoding is a less (Cronbach’s alpha = 0.82; e.g., ‘‘To what extent do you think you
developed city compared with Beijing. In order to obtain a more know a lot about food and nutrition?”; ‘‘To what extent do you
diverse sample not only limited to metropolitan citizens, data col- think you know how to evaluate the nutritional value of a food?”;
lection was extended beyond Beijing alone. Beijing and Baoding ‘‘To what extent do you think you know which food has the nutri-
city have similar historical, cultural and policy environments. ents your body needs?”).
Although Beijing and Baoding city differ in terms of economic Objective nutrition knowledge was measured by three compo-
development, income, as an important indicator of economic nents. The first two components were based on the study of
development level, has previously been reported not to influence Grunert, Wills, et al. (2010). The first component assessed partici-
Chinese consumers’ use of food nutrition labels (Chen & Niu, 2009). pants’ knowledge on dietary recommendations and consisted of 12
Sample selection and contact procedures differed between the items measuring awareness of whether health experts recommend
two survey cities depending on cost efficiency and time effective- one should have more, about the same, less, try to avoid or ‘don’t
ness. The data collection in Beijing was performed at supermarkets, understand what it means’ for a series of nutrients, energy or
shopping malls, residence and public gardens in urban areas. Par- ingredients. Seven items measured awareness of whether experts
ticipants were selected in these areas based on convenience sam- recommend that one should have more, about the same, less, try
pling with the restriction that they had lived in Beijing for more to avoid or ‘don’t understand what it means’ for different food
than 1 year, considering the large floating population in Beijing. groups (i.e., referring to the food-based dietary guidelines for China
A total of 213 participants from Beijing completed the survey. (Ge, Jia, & Liu, 2007)). The second component including 36 items
The same questionnaire was delivered to about 700 students of a measured participants’ knowledge about the sources of nutrients.
middle school in Baoding city. A self-selection sampling method Ten food products (rice, wheat products, beans, red meat (e.g., beef,
was applied. The students were told to ask their parents to com- pork), skimmed milk, full-fat yoghurt, vegetable, soft drink, choco-
plete the questionnaire and return it. A total of 447 participants late, edible oil) that Chinese consumers are familiar with, were
from Baoding completed the survey. This yields a total valid sam- chosen as sources of nutrients. Participants were asked whether
ple of 660 adult participants. The sample from Beijing included these 10 food products were high or low (or ‘not sure’) in fat, cho-
younger (Chi-square test, v2 = 79.718, p < 0.001) and higher edu- lesterol and sugar, each, thus resulting in a total of 30 items. In
cated (Chi-square test, v2 = 122.463, p < 0.001) participants and addition, based on the study of Nayga (2000), participants were
participants with a lower familiarity with nutrition labels (Chi- asked which of two foods (egg yolks versus egg white, skimmed
square test, v2 = 13.547, p < 0.001), while participants from Bao- milk versus whole milk, pig liver versus hair tail) contained more
ding city were older, less educated and claimed to be more familiar cholesterol, and which of two foods (beans versus wheat, rice ver-
with nutrition labels. There were no differences between the city sus hair tail, and egg yolks versus egg whites) provided more pro-
samples in terms of gender (Chi-square test, v2 = 1.682, tein, resulting in another six objective nutrition knowledge items.
p = 0.195), diet status (Chi-square test, v2 = 0.857, p = 0.355) and The third component consisted of four items measuring partici-
BMI (Chi-square test, v2 = 5.119, p = 0.163). pants’ knowledge on salt and energy recommendations. Partici-
Compared to the census data for China, there was an overrepre- pants were asked the maximum amount of salt that an adult on
sentation of the middle-aged population group (40–60 years), as average should eat a day and whether they agree or disagree (or
well as female and higher-educated (college and above) partici- are not sure) with three statements about energy requirements
pants. The over-representation of female and middle-aged partici- (e.g., ‘An active man needs the same amount of energy as an active
pants is probably due to the self-selection method in Baoding city, woman’). The answer for each item was coded as correct or wrong.
i.e., participants were the parents of high schools students. The For analysis, an overall index of objective nutrition knowledge was
over-representation of higher educated participants is due to the calculated as:
selection of the survey city Beijing whose population has a higher
education compared to the national Chinese population. OBJ KNOW ¼ ðnumber of correct answers on dietary recommendations=19Þ
þ ðnumber of correct answers on sources of nutrients=36Þ
þ ðnumber of correct answers on salt and energy requirements=4Þ
Questionnaire
With regard to subjective understanding of nutrition labels,
The questionnaire included the following measures: use and participants were asked to indicate to what extent they believed
understanding (objective and subjective) of food nutrition labels, to understand the terms fat, saturated fat, sodium, carbohydrate,
nutrition knowledge (objective and subjective), socio-demographic energy and sugar on nutrition labels, and to what extent they
characteristics, diet status, diet-health awareness, BMI and famil- believed to understand the role that different nutrients on labels
iarity with food nutrition labels. play in the diet. These questions were adapted from Cowburn
106 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111

with post hoc Tukey comparison of means were used to test the dif-
ferences of understanding (subjective and objective) and use of
food nutrition label among socio-demographic groups and other
factors. Reliability tests with Cronbach’s alpha were performed to
test construct reliability of consumers’ awareness of the relation
between diet and health, subjective nutrition knowledge, and sub-
jective understanding of nutrition labels. Cronbach’s alpha coeffi-
cients above 0.70 were considered to denote acceptable internal
consistency reliability (Nunnally, 1978). Correlation analysis was
applied to test the relationship between diet-health awareness
and subjective nutrition knowledge. Six regression models were
estimated with Stata/SE 12.0 software (Statacorp, College Station,
TX, USA). Depending on the nature of the dependent variable, a lin-
Fig. 1. Sample of food nutrition label in China. ear regression model or an ordered logistic regression model was
applied. Results from ordered logistic regression model are
and Stockley (2005) and measured on a 7-point interval scale rang- reported as odds ratio (OR) with 95% confidence intervals (95%
ing from ‘not understand at all’ (=1) to ‘totally understand’ (=7) CI). An OR larger than 1 indicates that higher values on the indepen-
(Cronbach’s alpha = 0.90). dent variable make it more likely that participants will be in a
Objective understanding of nutrition labels was measured by higher category of the dependent variable while an OR lower than
three questions about the Nutrition Reference Values (NRV): the 1 indicates that a higher value of independent variable increases
meaning of ‘‘NRV%” on a nutrition label (Fig. 1), the percentage of the likelihood of participants to be in the current or a lower cate-
energy and fat per 100 g that the example food contributed to gory of the dependent variable. In case the assumption of parallel
the respective daily recommendation. The overall objective under- lines was violated in some independent variables of the ordered
standing of the nutrition label was calculated as the average num- logistic regression models, a generalized ordered logit model with
ber of correct answers to the three questions resulting in four partial proportional odds (PPO) was estimated using the STATA
values: 0, 0.33, 0.67 and 1, which were categorized and recoded gologit2 command with its autofit option (Williams, 2006).
as ‘not understand’ (=1), ‘understand slightly’ (=2), ‘understand The dependent variables in the first two regression models were
moderately’ (=3) and ‘totally understand’ (=4). The latter two subjective understanding and objective understanding of nutrition
response categories were further merged for regression analysis. labels, respectively. The dependent variable in the other four
Participants’ use of nutrition labels was measured on a 4-point regression models was the use of food nutrition label. The common
scale (ranging from never to always) by asking: ‘‘In general, how explanatory variables in the six regression models included socio-
often do you look for nutrition information on a food package demographic characteristics like age, gender, education and other
when you go shopping”1 with answering possibilities: ‘never’ (=1), factors: BMI, diet status, consumers’ familiarity with nutrition
‘rarely/not often’ (=2), ‘regularly’ (=3) and ‘always’ (=4). This formu- labels, and subjective and objective nutrition knowledge. To inves-
lation is based on a large-scale cross-European study on the use and tigate city difference on understanding and use of food nutrition
understanding of nutrition information on food labels (Grunert, labels, city was included as an additional explanatory variable in
Fernández-Celemín, et al., 2010). The same formulation has been the six regression models. To investigate the impacts of subjective
used in other studies as well (e.g., Satia et al., 2005; Kim, Nayga, & and/or objective understanding on use of food nutrition labels,
Capps, 2000; Nayga, 2000). subjective and/or objective understanding also served as explana-
The overall diet status was measured based on whether a par- tory variables in the models with use as the dependent variable.
ticipant followed any of the following diets: low-fat, low-sugar,
low-salt and low-energy diet. Participants on any of these diets
were coded as ‘1’, the others as ‘0’. Results
The questionnaire was developed in English and translated into
Chinese by a professional English–Chinese translator. The language Description of nutrition knowledge, understanding and use of food
in the Chinese version of questionnaire was further checked by a nutrition label
Chinese language teacher. Then back-translation was undertaken
to ensure linguistic equivalence. The first author of the study, Participants reported on average a moderate degree of both
who is a native Chinese speaker, was closely involved in the subjective knowledge (SUBJ_KNOW; Mean = 4.15 on 7-point scale)
back-translation process. A preliminary version of the question- and objective nutrition knowledge (OBJ_KNOW = 1.65 in a range
naire was pretested twice in a small sample of 20 Chinese adults from 0 to 3). Regarding the knowledge about expert recommenda-
for clarity of content, language/wording, overall understanding tions, more than half of the participants were familiar with the die-
and length of the survey. Based on their feedback, the question- tary recommendations for fat, salt, sugar, fiber, while only less than
naire was refined and finalized. 10% of the participants knew the recommendation for omega-3
fatty acids and polyunsaturated fat. The recommendations for
Data analysis monounsaturated fat, omega-3 fatty acids and trans fat were not
known among about 40% of the sample. Most of participants
Descriptive analysis including the calculation of mean values (75%) knew the recommendation of increasing the intake of whole
and standard deviations (SD) was conducted using SPSS 20.0 soft- grain. Regarding food-based dietary guidelines, a large majority of
ware (IBM SPSS, Armonk, NY, USA). Cross-tabulation and chi-square participants (88%) knew they should eat more fruits and vegeta-
analysis were conducted to test differences in demographic charac- bles. More than half of the participants knew that they should con-
teristics between the two city samples. One-way ANOVA F-tests sume more milk and dairy products, maintain the current level of
intake of starchy food and try to avoid food and drinks high in fat,
1
Translated to Chinese the statement probing for participants’ use of nutrition
sugar and salt. The recommendation of increasing the consumption
information on food labels reads as follows: ‘‘一般来说, 当您购物时您会经常寻找食品包 of food rich in protein was less known; only 38.1% could provide
装上的营养信息吗?”. the correct answer to this issue.
 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111 107

Table 1
Description of the study sample (n = 660); Mean (SD), use, objective understanding and subjective understanding of a nutrition label.

Sample characteristics Frequency (%) Use Objective understanding Subjective understanding

Mean (SD)
City Beijing 32.3 2.04a (0.65) 0.18a (0.26) 4.38a (1.29)
Baoding 67.7 2.17b (0.62) 0.24b (0.31) 4.33a (1.26)
Gender Female 64.8 2.18b (0.65) 0.22a (0.30) 4.36a (1.23)
Male 35.2 2.04a (0.68) 0.22a (0.28) 4.28a (1.32)
Age Below 25 15.0 2.14a (0.66) 0.24a (0.27) 4.76b (1.29)
26–40 31.4 2.04a (0.68) 0.24a (0.31) 4.27a (1.25)
41+ 53.6 2.17a (0.34) 0.20a (0.29) 4.27a (1.25)
Education Middle school and below 58.1 2.15a (0.63) 0.23a (0.30) 4.19a (1.23)
College and above 41.9 2.10a (0.69) 0.20a (0.28) 4.53b (1.27)
Diet status Yes 81.2 2.13a (0.66) 0.21a (0.29) 4.35a (1.26)
No 18.8 2.06a (0.63) 0.29b (0.31) 4.31a (1.34)
Familiarity Yes 71.6 2.26b (0.60) 0.24b (0.30) 4.47b (1.24)
No 28.4 1.79a (0.67) 0.17a (0.29) 4.01a (1.29)
BMI (kg/m2) Underweight (BMI < 18.5)* 9.0 2.30a (0.69) 0.20a (0.31) 4.55a (1.27)
Normal weight (18.5 6 BMI < 24) 62.1 2.11a (0.63) 0.24a (0.30) 4.37a (1.25)
Overweight (24 6 BMI < 28) 22.2 2.08a (0.67) 0.22a (0.30) 4.15a (1.23)
Obese (BMI P 28) 6.7 2.07a (0.75) 0.15a (0.22) 4.30a (1.44)

Note: (a, b) In superscripts indicate significant differences by one-way ANOVA with post hoc Tukey comparison of means.
*
According to China-defined cut-off points (China Obesity Group, 2002).

The average number of correct answers for sources of nutrients variable objective understanding were merged. Similarly, the cate-
was 19.7 out of 36. More than half of the participants got fat and gories ‘regularly’ and ‘always’ of the variable nutrition label use
sugar items right, except that only about 30% of them knew that were combined. The description of the different dependent and
rice and full-fat yoghurt were low in sugar and that cola was low independent variables is presented in Table 2.
in fat. For cholesterol, a common mistake was that participants Results of the regression Models 1 and 2 with subjective and
estimated that chocolate and edible oils were high in cholesterol. objective understanding as the dependent variables, respectively,
However, more than 60% of the participants answered correctly are shown in Table 3. City did not yield a significant effect on par-
to the questions about which of two foods has more cholesterol. ticipants’ subjective and objective understanding in food nutrition
Less than half of the participants knew the maximum amount of labels. A higher subjective understanding of nutrition labels was
salt that an adult on average should eat a day. With regard to cal- found with decreasing age (Model 1). Both subjective and objective
ories, a large majority of participants were aware of differences in nutrition knowledge positively affected participants’ subjective
energy requirement between subpopulations. understanding. Participants with a higher educational level or a
Participants reported to have a moderate level of subjective special diet status were more likely to have a lower objective
understanding (SUBJ_UND) of the nutrient terms presented on understanding of nutrition labels (Model 2). Younger and more
nutrition labels. Very few (about 20%) participants had a good knowledgeable participants were more likely to (objectively)
objective understanding of ‘‘NRV” on the example nutrition label. understand the nutrition label while participants with a lower
About two thirds of the participants reported that they rarely or BMI were more likely to totally understand the food nutrition label.
never used a nutrition label during food shopping, while only 7% The coefficient estimates of familiarity in Models 1 and 2 indicated
indicated to always use nutrition labels. Young participants with that the participants were more likely to understand, both subjec-
a higher education or who were familiar with a food nutrition label tively and objectively, nutrition labels if they were more familiar
reported a significantly higher subjective understanding compared with them.
with their counterparts. A higher level of objective understanding Results of the ordered logistic regression Models 3–6 with use
(OBJ_UND) of a food nutrition label was found among participants as the dependent variable are shown in Table 4. There was no sig-
who were familiar with food nutrition labels or participants with- nificant city effect on participants’ use of food nutrition labels. Par-
out a special diet status. Females, as well as participants who were ticipants with a lower BMI, a higher familiarity with nutrition
familiar with food nutrition labels used food nutrition labels more labels and participants who perceived themselves to be more
frequently when they went shopping (Table 1). knowledgeable about nutrition labels were more likely to use
nutrition labels when shopping for food. Based on the coefficient
Regression results estimates, the effect of familiarity on label use appeared to be
stronger than that of subjective nutrition knowledge.
To avoid multicollinearity, consumers’ awareness of the relation Models 4–6 in Table 4 included subjective understanding and/
between diet and health (Cronbach’s alpha = 0.72; Mean = 4.98; or objective understanding as explanatory variables. A higher
SD = 1.10) was excluded from the regression models because it Pseudo R2 for these models compared to Model 3 indicated that
was moderately correlated with subjective nutrition knowledge the entry of label understanding, especially subjective understand-
(correlation = 0.528; p value <0.001). To reduce the number of ing resulted in a better model fit. Subjective understanding played
empty cells in running the ordered logistic model, two approaches a significant and positive role in Chinese consumers’ use of nutri-
were adopted: keeping less categorical explanatory variables and tion labels, while objective understanding did not. The effect of
reducing the number of categories of a variable. Accordingly, first, familiarity with the nutrition label was consistently the strongest
only five categorical variables, i.e., city, gender, education, diet sta- among all studied explanatory variables. This significant positive
tus and familiarity served as explanatory variables. Second, the cat- effect remained even when subjective and objective understanding
egories ‘understand moderately’ and ‘totally understand’ of the were entered into the models as additional explanatory variables.
108 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111

Table 2
Variable description for regression analyses, frequency and Mean (SD).

Variable description Frequency (%)

Dependent variables Use of nutrition label Ordinal (1–3) Never (=1) 16.0
Rarely (=2) 55.5
Regularly or always (=3) 28.5
Objective understanding (OBJ_UND) Ordinal (1–3) Not understand (=1) 57.1
Understand slightly (=2) 23.6
Moderately or totally understand (=3) 19.2
Mean SD
Subjective understanding (SUBJ_UND) Metric 4.35 1.27
Independent variables City Baoding = 0; Beijing = 1 0.32 –
Gender Female = 0; Male = 1 0.35 –
Age Metric 37.98 10.04
Education Middle school & below = 0; College & above = 1 0.42 –
BMI Metric 22.75 3.39
Diet status No special diet = 0; Special diet = 1 0.81 –
Familiarity No = 0; Yes = 1 0.72 –
Subjective nutrition knowledge (SUBJ_KNOW) Metric 4.15 1.25
Objective nutrition knowledge (OBJ_KNOW) Metric 1.65 0.47

affected by age, familiarity with food nutrition labels, subjective

Table 3
Determinants of subjective understanding (Model 1: linear regression): coefficient and objective nutrition knowledge, while objective understanding
estimates, and objective understanding (Model 2: generalized ordered logit regres- was determined by age, education, BMI, diet status, familiarity
sion): odds ratios, OR (95% CI). with food nutrition labels, subjective and objective nutrition
Variables a
SUBJ_UND (Model 1) b
OBJ_UND (Model 2) knowledge. Nutrition labels are not commonly used yet in China.
70% of the participants reported that they rarely or never used
City 0.041 0.762 (0.512–1.134)
Gender 0.071 1.216 (0.863–1.713) nutrition labels when shopping for food. The empirical findings
Age 0.017*** c
0.977* (0.961–0.994) of this study indicated that familiarity with the nutrition label,
d
0.996 (0.976–1.016) subjective nutrition knowledge and subjective understanding were
Education 0.175 0.721* (0.495–1.052) significant and positive determinants of Chinese consumers’ nutri-
c
BMI 0.012 0.989 (0.939–1.042)
d
0.932* (0.872–0.997)
tion label use.
Diet status 0.006 0.643* (0.429–0.962) Age negatively affected the subjective and objective under-
Familiarity 0.308** 1.541* (1.047–2.267) standing of nutrition labels, indicating that younger consumers
SUBJ_KNOW 0.373*** c
1.184* (1.028–1.363) have a better understanding of nutrition labels. This result corrob-
d
1.000 (0.840–1.190)
orates the study of Moorman (1996) which found that aging wors-
OBJ_KNOW 0.403*** c
1.469* (1.006–2.144)
d
0.910 (0.570–1.453) ens consumers’ comprehension level although it may increase
Pseudo R2 0.039 consumers’ perception of their abilities to process nutrition infor-
R2 adjusted 0.217 mation on food nutrition labels. Although age yielded significant
a
p Values were calculated based on robust standard errors because heteroske- effects on understanding, socio-demographic factors such as age,
dasticity was not satisfied (Breusch–Pagan test: chi2(1) = 19.82; p < 0.001). gender and education as well as the diet status did not yield signif-
b
Independent variables age, BMI, subjective nutrition knowledge and objective icant direct effects on participants’ use of nutrition labels, which is
nutrition knowledge in this model did not fill the parallel assumption. Coefficients consistent with previous studies (e.g., Drichoutis et al., 2005;
from general ordered logit model (gologit2) were reported.
c Fitzgerald et al., 2008; Grunert, Fernández-Celemín, et al., 2010).
Category ‘‘not understand” vs. categories ‘‘understand slightly” and ‘‘moder-
ately & totally understand”. Both education and diet status were negatively associated with
d
Categories ‘‘not understand” and ‘‘understand slightly” vs. category ‘‘moder- objective understanding of nutrition labels. However, education
ately & totally understand”. and diet status did not explain the use of nutrition labels, which
*
p < 0.05.
**
is consistent with the study of Nayga (2000). Consumers’ under-
p < 0.01.
***
p < 0.001.
standing of nutrition information depends on motivation
(Cownburn & Stockley, 2005; Miller, Gibson, & Applegate, 2010).
Subjective nutrition knowledge was also an important determi- For education, a possible explanation could be that higher edu-
nant of label use. Its effect remained significant and positive in cated individuals are less likely to believe that nutrition labels con-
Models 3 and 5 while it became not significant when subjective tribute to healthier food choices (Nayga, 1999). Moreover,
understanding entered Models 4 and 6. The negative effect of education has been shown to negatively influence consumers’
BMI on the use of nutrition labels remained significant with sub- motivation to process nutrition information (Moorman, 1990).
jective understanding as explanatory variable but was not signifi- For diet status, no proper explanation was found for our finding
cant when objective understanding entered Models 5 and 6. City, that participants who were not in a special diet status had a higher
gender, age, education, diet status, objective nutrition knowledge objective understanding. Future study focusing on the motivation
and objective understanding did not significantly affect the use of this subgroup might help to get insight into their objective
of food nutrition labels. understanding of nutrition labels.
BMI significantly affected nutrition label use. The negative
effect of BMI on label use is not in line with previous research
Discussion (Drichoutis et al., 2008). A possible reason could be that individuals
with higher BMIs are less likely to agree that nutrition information
A moderate degree of subjective understanding and low degree on food labels is useful to them (Nayga, 1999). This result suggests
of objective understanding of nutrition labels were found among that normal and underweight consumers in China are more likely
the participants. Subjective understanding of nutrition labels was to use nutrition labels on food.
 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111 109

Table 4
Determinants of use of food nutrition label (ordered logistic regression models): odds ratios, OR (95% CI).

Variables Use (Model 3) Use (Model 4) Use (Model 5) Use (Model 6)

City 0.907 (0.615–1.338) 0.917 (0.620–1.356) 0.906 (0.613–1.340) 0.916 (0.619–1.357)
Gender 0.764 (0.542–1.078) 0.785 (0.556–1.110) 0.752 (0.533–1.062) 0.775 (0.548–1.096)
Age 1.003 (0.987–1.020) 1.009 (0.992–1.026) 1.005 (0.988–1.022) 1.010 (0.993–1.027)
Education 0.811 (0.562–1.169) 0.760 (0.525–1.099) 0.827 (0.573–1.195) 0.771 (0.532–1.118)
BMI 0.949* (0.902–0.998) 0.949* (0.902–0.999) 0.949* (0.902–0.998) 0.949* (0.902–0.999)
Diet status 1.294 (0.850–1.971) 1.279 (0.838–1.952) 1.356 (0.887–2.075) 1.328 (0.867–2.036)
Familiarity 3.934*** (2.634–5.878) 3.663*** (2.440–5.498) 3.814*** (2.549–5.707) 3.579*** (2.381–5.379)
SUBJ_KNOW 1.291*** (1.123–1.484) 1.159 (0.998–1.346) 1.277** (1.110–1.469) 1.152 (0.992–1.338)
OBJ_KNOW 1.065 (0.743–1.524) 0.942 (0.654–1.357) 1.038 (0.723–1.490) 0.926 (0.641–1.336)
SUBJ_UND 1.350*** (1.159–1.572) 1.339*** (1.149–1.561)
#
OBJ_UND_2 1.375 (0.913–2.071) 1.298 (0.862–1.956)
$
OBJ_UND_3 1.242 (0.813–1.898) 1.173 (0.765–1.799)
Pseudo R2 0.073 0.086 0.075 0.088
#
OBJ_UND_2 is 1 if OBJ_UND is equal to 2 and zero otherwise.
$
OBJ_UND_3 is 1 if OBJ_UND is equal to 3 and zero otherwise.
*
p < 0.05.
**
p < 0.01.
***
p < 0.001.

Familiarity played a predominant role in participants’ under- Objective understanding did not yield a significant effect on the
standing and use of nutrition labels. Familiarity, as a kind of expe- use of food nutrition labels. A possible explanation is twofold. First,
rience could increase the perceived validity of information (Park food nutrition labels have been introduced in China since 2008,
et al., 1994). Nutrition labels have only been introduced on a vol- though only on a voluntary basis mainly on dairy products and
untary basis in China in 2008. Most of the Chinese consumers indi- food products from large-scale producers (Jiang, 2012). Although
cate that they have noticed nutrition labels on food packages and Chinese consumers were exposed to (voluntary) nutrition labels
that they trust this new kind of information (Ye, Feng, & Wu, in the period 2008–2013, there was limited communicative sup-
2010; Zhao, Xia, Yu, & Wu, 2009). However, because Chinese con- port introducing the new label standard, explaining its content
sumers are not used to reading nutrition labels (Chen & Niu, 2009), and motivating its use in the period preceding the implementation
familiarity with nutrition labeling indicates that they are aware of of the national standard for mandatory nutrition labeling in China
it but probably without knowing much about it (Ye et al., 2010; (January 1st, 2013). Another possible reason is that the present
Zhang, 2012). Future communication should make more efforts study measured conceptual understanding and not the practical
to increase Chinese consumers’ familiarity with the nutrition infor- understanding on how to apply the label information for relative
mation on labels, e.g., explaining its content, and stressing the judgments of products. Previous research has shown that these
advantages and different uses of nutrition labels. two types of understanding may differ significantly as such that
This study’s empirical results showed that participants’ subjec- consumers often experience more difficulties in understanding
tive nutrition knowledge was also an important and positive deter- the concepts than in making intra-category comparisons between
minant of label use. However, objective nutrition knowledge did food products (Grunert & Wills, 2007; Grunert, Fernández-
not determine label use, which is consistent with the study of Celemín, et al., 2010; Malam et al., 2009). The measurement of
Nayga (2000) suggesting that consumers with more objective practical understanding or a combination of both types of objective
nutrition knowledge do not necessarily use nutrition labels. Sub- understanding beside subjective understanding are recommended
jective knowledge relates to motivational factors such as self-con- for future studies.
fidence (Brucks, 1985). Subjective knowledge differs from objective The findings of this study are subject to some limitations. First,
knowledge when there is a gap in consumers’ confidence about we focused on samples of self-selected urban consumers, which
their actual knowledge level. Chinese consumers indicate not to limits the potential to generalize our findings to the Chinese pop-
have much knowledge about the hazards caused by unhealthy ulation at large. Future studies, using a more representative con-
diets (Liu et al., 2014), implying a low confidence in food nutrition sumer samples and extending to less urbanized areas in China,
issues which may hamper their actual use of nutrition labels. would provide further insight and contribute to a better under-
Future educational campaigns might focus more on how to standing of the findings reported here. Second, the present study
increase Chinese consumers’ confidence and motivation to use just like most research in this field depended on self-reported
nutrition labels. Results with respect to the role of objective nutri- behavior. Although such self-reported and subjective opinions pro-
tion knowledge suggest that the Chinese government should put vide valuable insights into consumer behavior, they likely suffer
more efforts to recommend consumers to increase the intake of from so-called social desirability bias and hence may deviate from
foods rich in protein, polyunsaturated fat and omega-3 fatty acids. actual behavior (Fisher, 1993). As according to Middleton and Jones
In addition, efforts to increase consumers’ understanding of NRV (2000), especially people from Eastern cultures are likely to pro-
and trans fat which are compulsory information items on the food vide more socially desirable responses consistent with the cultural
nutrition label, are also needed. dimensions predominant in their country of origin as compared to
Another cognitive variable, subjective understanding, also people from Western cultures. In this respect, participants’ actual
yielded a positive effect on the use of nutrition labels. Chinese con- use of food nutrition labels may be lower than their reported
sumers indicated that they perceive nutrition labels as too difficult use. In addition, self-reported measurements are based on partici-
to understand (Chen & Niu, 2009; Zhao et al., 2009). Reading nutri- pants’ interpretation of their past behavior and may be of limited
tion labels appears to be too time consuming or irrelevant (Zhang, predictive power for their future label use. Therefore, more exper-
2012). Reducing Chinese consumers’ perceived difficulty in their imental and observational studies on Chinese consumers’ use of
understanding of nutrition labels – and herewith also improving food nutrition labels are recommended. Third, objective under-
their subjective understanding – is a crucial factor to stimulate standing of a food nutrition label as measured in this study is lim-
nutrition label use in China. ited to the understanding of the concept of nutrition reference
110 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111

values. It is recommended to investigate consumers’ objective References

understanding of compulsory nutrient items such as protein, fat,
carbohydrate and sodium on a food nutrition label in order to Aldridge, V., Dovey, T. M., & Halford, J. C. G. (2009). The role of familiarity in dietary
development. Developmental Review, 29(1), 32–44.
improve educational strategies and programmes to promote Chi- Besler, H. T., Buyuktuncer, Z., & Uyar, M. F. (2012). Consumer understanding and use
nese consumers’ use of food nutrition label. of food and nutrition labelling in Turkey. Journal of Nutrition Education and
Finally, the pseudo-R2-values in the present study were low, Behavior, 44(6), 584–591.
Bettman, J. R., & Park, C. W. (1980). Effects of prior knowledge and experience and
which implies that there is not a lot of variance left to explain phase of the choice process on consumer decision processes: A protocol
and much of the variance that is left is probably due to unobserved analysis. Journal of Consumer Research, 7(3), 234–248.
causes. This means that other cultural, situational, behavioral, psy- Brucks, M. (1985). The effects of product class knowledge on information search
behavior. Journal of Consumer Research, 12(1), 1–16.
chological factors (among others) may influence nutrition label use
Buckland, G., Travier, N., Cotte, V., González, C. A., Luján-Barroso, L., Agudo, A., et al.
among Chinese consumer, which is not surprising given the quite (2013). Adherence to the Mediterranean diet and risk of breast cancer in the
different context of China compared to Western countries. Further European prospective investigation into cancer and nutrition cohort study.
research, such as qualitative focus group discussions or depth International Journal of Cancer, 132(12), 2918–2927.
Burton, S., Biswas, A., & Netemeyer, R. (1994). Effects of alternative nutrition label
interviews may be necessary to identify other potential determi- formats and nutrition reference information on consumer perceptions,
nants of label use among Chinese. comprehension and product evaluation. Journal of Public Policy & Marketing,
13(1), 36–47.
Campos, S., Doxey, J., & Hammond, D. (2011). Nutrition labels on pre-packaged
foods: A systematic review. Public Health Nutrition, 14(8), 1496–1506.
Conclusions Capacci, S., Mazzocchi, M., Brambila-Macia, J., Shankar, B., Verbeke, W., Pérez-Cueto,
F., et al. (2012). Policies to promote healthy eating in Europe: A structured review
NCDs have become major causes of death in China due to the of instruments and their effectiveness. Nutrition Reviews, 70(3), 188–200.
Chen, W. P., & Niu, M. C. (2009). Consumers’ use of food nutrition label and its
nutrition transition encouraged by industrial growth, alongside influential factors. Journal of Renmin University of China, 4, 105–113.
related socioeconomic changes and government policies. Chinese Chen, J. S., & Zhao, W. H. (2012). Diet, nutrition and chronic disease in Mainland
consumers are increasingly aware of the relationship between diet China. Journal of Food & Drug Analysis, 20, 222–225.
Cheng, D. L., Cao, H., & Xu, S. Y. (2007). Knowledge about food nutrition and hygiene
and health. This evolution has urged the Chinese government to
of residences in Luyang district, Hefei city, Anhui. Journal of Preventive Medicine,
bring the mandatory food nutrition labeling policy into force in 13, 32–33.
January 2013. This study investigated the factors affecting Chinese China Obesity Group (2002). The prediction of body mass index and waistline of
consumers’ understanding and use of food nutrition labels. The adult to risk factors of relevant disease: Study on proper body mass index and
waistline. Chinese Journal of Epidemiology, 1, 5–10.
results suggest that familiarity with food nutrition labels has a Cowburn, G., & Stockley, L. (2005). Consumer understanding and use of nutrition
strong positive effect on label understanding and use. Nutrition labelling: A systematic review. Public Health Nutrition, 8(1), 21–28.
knowledge (subjective and objective) positively affects partici- Drichoutis, A. C., Lazaridis, P., & Nayga, R. M. (2005). Nutrition knowledge and
consumer use of nutritional food labels. European Review of Agricultural
pants’ understanding (subjective and objective) of food nutrition Economics, 32(1), 93–118.
labels. Subjective nutrition knowledge and subjective understand- Drichoutis, A. C., Lazaridis, P., & Nayga, R. M. (2006). Consumers’ use of nutrition
ing play a significant and positive role in Chinese consumers’ label labels: A review of research studies and issues. Academy of Marketing Science
Review, 9, 93–118.
use. Age yielded a negative effect on both subjective and objective Drichoutis, A. C., Lazaridis, P., Nayga, R. M., Kapsokefalou, M., & Chryssochoidis, G.
understanding, while education only affected participants’ objec- (2008). A theoretical and empirical investigation of nutritional label use. The
tive understanding. None of the socio-demographic characteristics European Journal of Health Economics, 9(3), 293–304.
European Food Information Council (EUFIC). New insights into nutrition labelling in
associated with self-reported use. Europe. (2012). <http://www.eufic.org/article/en/artid/New-insights-into-
Our study has important implications for future educational nutrition-labelling-in-Europe/> Accessed December 2013.
campaign or policies to promote nutrition label use in China. There European Food Information Council (EUFIC). Consumer attitudes to nutrition
information and food labelling. (2005). <http://www.eufic.org/article/en/
is no national or regional educational campaign to promote nutri-
nutrition/food-labelling-claims/ expid/forum-2-consumer-attitudes-information-
tion labeling in China until now, except for some scattered efforts food-labelling/> Accessed December 2013.
through programs on television and articles in newspapers or on Fisher, R. J. (1993). Social desirability bias and the validity of indirect questioning.
the internet. These programs and articles are often limited to Journal of Consumer Research, 20(2), 303–315.
Fitzgerald, N., Damio, G., Segura-Pérez, S., & Pérez-escamilla, R. (2008). Nutrition
explaining the meaning of nutrient items on nutrition labels, which knowledge, food label use, and food intake patterns among Latinas with and
might increase consumers’ objective knowledge and objective without type 2 diabetes. Journal of the American Dietetic Association, 108(6),
understanding but may not yield the desired effect on nutrition 960–967.
Ge, K. Y., Jia, J. B., & Liu, H. (2007). Food-based dietary guidelines in China – Practices
label use unless Chinese consumers would also feel or believe they and problems. Annals of Nutrition and Metabolism, 51, 26–31.
are knowledgeable and understanding. Promotion of nutrition Gorton, D., Mhurchu, C. N., Chen, M. H., & Dixon, R. (2009). Nutrition labels: A
labeling among Chinese consumers is recommended to focus more survey of use, understanding and preferences among ethnically diverse
shoppers in New Zealand. Public Health Nutrition, 12(9), 1359–1365.
on how to increase consumers’ familiarity with the advantages and Grunert, K. G., Fernández-Celemín, L., Wills, J. M., Storcksdieck genannt Bonsmann,
possible uses of nutrition labels, as well as on reinforcing consum- S., & Nureeva, L. (2010). Use and understanding of nutrition knowledge on food
ers’ subjective beliefs in this respect. The findings suggest that, labels in six European countries. Journal of Public Health, 18(3), 261–277.
Grunert, K. G., & Wills, J. M. (2007). A review of European research on consumer
apart from educating consumers with respect to nutrition labels, response to nutrition information on food labels. Journal of Public Health, 15(5),
additional efforts will be needed to foster also their subjective 385–399.
knowledge and understanding, i.e., apart from making consumers Grunert, K. G., Wills, J. M., & Fernández-Celemín, L. (2010). Nutrition knowledge,
and use and understanding of nutrition information on food labels among
capable of understanding nutrition labels, providing them with
consumers in the UK. Appetite, 55(2), 177–189.
the subjective feeling that they can do so indeed equally matters. Hawkes, C. (2008). Agro-food industry growth and obesity in China: What role for
regulating food advertising and promotion and food nutrition labelling. Obesity
Reviews, 9, 151–161.
Acknowledgments Hess, R., Visschers, V. H. M., & Siegrist, M. (2011). The role of health-related,
motivational and socio-demographic aspects in predicting food label use: A
comprehensive study. Public Health Nutrition, 15(3), 407–414.
The authors gratefully acknowledge the PhD-scholarship from Hieke, S., & Taylor, C. R. (2012). A critical review on the literature on nutritional
the LiSUM program of the European Commission, the BOF (Special labelling. The Journal of Consumer Affairs, 46(1), 120–156.
Research Fund) PhD-scholarship program of Ghent University, and Hoefkens, C., Pieniak, Z., Van Camp, J., & Verbeke, W. (2012). Explaining the effects
of a point-of-purchase nutrition-information intervention in university
the Ghent University Faculty of Bioscience Engineering’s Commit- canteens: A structural equation modelling analysis. The International Journal of
tee for Scientific Research (CWO). Behavioral Nutrition and Physical Activity, 9, 111.
 R. Liu et al. / Food Quality and Preference 41 (2015) 103–111 111

Hoefkens, C., Veettil, P. C., Van Huylenbroeck, G., Van Camp, J., & Verbeke, W. (2012). Park, C. W., Mothersbaugh, D., & Feick, L. (1994). Consumer knowledge assessment.
What nutrition label to use in a catering environment? A discrete choice Journal of Consumer Research, 21(1), 71–82.
experiment. Food Policy, 37(6), 741–750. Petrovici, D. A., & Ritson, C. (2006). Factors influencing consumer dietary health
Hu, F. B., Liu, Y., & Willett, W. C. (2011). Preventing chronic diseases by promoting preventative behaviours. BMC Public Health, 6, 222.
healthy diet and lifestyle: Public policy implication for China. Obesity Reviews, Pieniak, Z., Aertsens, J., & Verbeke, W. (2010). Subjective and objective knowledge as
12(7), 552–559. determinants of organic vegetables consumption. Food Quality and Preference,
Jiang, X. J. (2012). Belated 10 years food nutrition labels. Southern Weekly. <http:// 21(6), 581–588.
www.infzm.com/content/80229> Accessed September 2014 Powers, S. I., Welsh, D. P., & Wright, V. (1994). Adolescents’ affective experience of
Kang, J. H., Guan, T., Ning, G., Wu, J. R., & Guan, Y. F. (2012). Diabetes research in family behaviours: The role of subjective understanding. Journal of Research on
China: Current status and future challenges. Translational Medicine Research, 2, Adolescence, 4(4), 585–600.
1–24. Ragaert, P., Verbeke, W., Devlieghere, F., & Debevere, J. (2004). Consumer perception
Kim, S., Nayga, R. M., & Capps, O. Jr., (2000). The effect of new food labeling on and choice of minimally processed vegetables and packaged fruits. Food Quality
nutrient intakes: An endogenous switching regression analysis. Journal of and Preference, 15(3), 259–270.
Agricultural and Resource Economics, 25(1), 215–231. Raju, P. S., Lonial, S. C., & Mangold, W. G. (1995). Differential effects of subjective
Liu, R. D., Pieniak, Z., & Verbeke, W. (2013). Consumers’ attitudes and behaviour knowledge, objective knowledge, and usage experience on decision marketing:
towards safe food in China: A review. Food Control, 33(1), 93–104. An exploratory investigation. Journal of Consumer Psychology, 4(2), 153–180.
Liu, R. D., Pieniak, Z., & Verbeke, W. (2014). Food-related hazards in China: Ruddell, F. (1979). Consumer food selection and nutrition information. New York, NY:
Consumers’ perceptions of risk and trust in information sources. Food Control, Praeger Publishers.
46, 291–298. Sakamaki, R., Toyama, K., Amanoto, R., Liu, C. J., & Shinfuku, N. (2005). Nutrition
Malam, S., Clegg, S., Kirwan, S., McGinigal, S., Raats, M., Barnett, J., et al. (2009). knowledge, food habits and health attitude of Chinese university students – A
Comprehension and use of UK nutrition signpost labelling schemes. British Market cross sectional study. Nutrition Journal, 4, 4.
Research Bureau. Satia, J. A., Galanko, J. A., & Neuhouser, M. L. (2005). Food nutrition label use is
Mejean, C., Macouillard, P., Péneau, S., Hercberg, S., & Castetbon, K. (2013). associated with demographic, behavioural, and psychosocial factors and dietary
Consumer acceptability and understanding of front-of-pack nutrition labels. intake among African–Americans in North Carolina. Journal of the American
Journal of Human Nutrition and Dietetics, 26(5), 494–503. Dietetic Association, 105(3), 392–402.
Middleton, K. L., & Jones, J. L. (2000). Socially desirable response set: The impact of Selnes, F., & Gronhaug, K. (1986). Subjective and objective measures of product
country culture. Psychology & Marketing, 17(2), 149–163. knowledge contrasted. Advances in Consumer Research, 13, 67–71.
Miller, L. M. S., Gibson, T. N., & Applegate, E. A. (2010). Predictor of nutrition SPSS Software (2011). SPSS software, version 20.0. Chicago, IL, USA: SPSS Inc.
information comprehension in adulthood. Patient Education and Counseling, 80 StataCorp, Stata 18.0/SE. College Station, TX, USA.
(1), 107–112. van Trijp, H. C. M. (2009). Consumer understanding and nutritional communication:
Ministry of Health of the People’s Republic of China (MOH). The regulation of food Key issues in the context of the new EU legislation. European Journal of Nutrition,
nutrition labelling. (2008). <http://www.gov.cn/gzdt/2008-01/11/content_856260. 48(1), 41–48.
htm> Accessed December 2013. Wahlich, C., Gardner, B., & McGowan, L. (2013). How, when and why do young
Moorman, C. (1990). The effects of stimulus and consumer characteristics on the women use nutrition information on food labels? A qualitative analysis.
utilization of nutrition information. Journal of Consumer Research, 17(3), Psychology & Health, 28(2), 202–216.
362–374. Williams, R. (2006). Generalized ordered logit/partial proportional odds models for
Moorman, C. (1996). A quasi-experiment to assess the consumer and informational ordinal dependent variables. The Stata Journal, 6(1), 58–82.
determinants of nutrition information processing activities: The case of the Xu, Y., Wang, L., He, J., Bi, Y. F., Li, M., Wang, T., et al. (2013). Prevalence and control
nutrition labelling and education act. Journal of Public Policy & Marketing, 15(1), of diabetes in Chinese adults. The Journal of the American Medical Association, 310
28–44. (9), 948–959.
National Centre for Cardiovascular Disease (NCCD). Report of China Cardiovascular Yamamoto, T. (2013). Nutrition and diet in inflammatory bowel disease. Current
Disease in 2012. (2013). China Centre for Disease Control and Prevention. Opinion in Gastroenterology, 29(2), 216–221.
<http://www.chinacdc.cn/mtdx/mxfcrxjbxx/201308/t20130813_86592.htm> Yang, Z. Y., Yang, Z., Zhu, L. F., & Qiu, C. X. (2011). Human behaviors determine
Accessed November 2013. health: Strategic thoughts on the prevention of chronic non-communicable
Nayga, R. M. (1999). Toward an understanding of consumers’ perceptions of food disease in China. International Journal of Behavioral Medicine, 18(4), 295–301.
labels. The International Food and Agribusiness Management Review, 2(1), 30–44. Ye, W. Y., Feng, Z. G., & Wu, W. B. (2010). Cognition and use of food nutrition label of
Nayga, R. M. (2000). Nutrition knowledge, gender and food label use. The Journal of consumers in Guangzhou city. Modern Preventive Medicine, 37(10), 1850–1852.
Consumer Affairs, 34(1), 97–112. Zhang, J. (2012). Consumers’ cognition of food nutrition label and compliance of food
Nayga, R. M., Lipinski, D., & Savur, N. (1998). Consumers’ use of nutritional labels producers in Taiyuan city (Master thesis). Shanxi Medical University.
while food shopping and at home. The Journal of Consumer Affairs, 32(1), Zhao, Y., Xia, W., Yu, B. W., & Wu, K. (2009). Consumer knowledge, attitude and
106–120. behavior toward food nutrition label in HaerBin city. Chinese Journal of Public
Nunnally, J. G. (1978). Psychometric theory. New York, NY: McGraw-Hill. Health, 25(5), 603–604.
Park, C. W., & Lessig, V. P. (1981). Familiarity and its impacts on consumer decision
biases and heuristics. Journal of Consumer Research, 8(2), 223–231.