Research Article

Received: 25 November 2014 Revised: 17 April 2015 Accepted article published: 23 April 2015 Published online in Wiley Online Library: 19 June 2015

(wileyonlinelibrary.com) DOI 10.1002/jsfa.7232

Age and gender differences in the influence of

extrinsic product information on acceptability
for RTD green tea beverages
Ye-Won Bae,† Soh Min Lee† and Kwang-Ok Kim*

Abstract
BACKGROUND: The introduction of ready-to-drink (RTD) green tea beverage has allowed diverse consumers to consume
green teas and related products. Green tea that has been traditionally consumed for its delicate flavor characteristics is also
widely consumed for its recognition as a healthy product. Because it is reported that age difference exists in consideration of
health-related information, the objective of the study was to investigate how sensory and non-sensory factors, in particular
health-related information, price and packaging, would affect the flavor acceptability of green tea beverages, depending on
consumers’ age and gender.

RESULTS: Regardless of the product information, old consumers preferred products that provided an indication of health
beneficial effect. On the other hand, young consumers tended to be influenced by extrinsic product information such as
packaging, brand/manufacturer and/or price, though these consumers were not so much influenced by health beneficial
information as were the old consumers.

CONCLUSION: The findings of the study implied that the influence of non-sensory information such as health beneficial
information in flavor liking differed depending mostly on consumers’ age, and little on gender, for RTD green tea beverages.
© 2015 Society of Chemical Industry

Keywords: RTD green tea beverages; consumer acceptability; age and gender difference; extrinsic product information; expectation in
health improvement

INTRODUCTION green tea and related items, in particular those who consider con-
It is well known that many non-sensory variables, including pack- venience when selecting a product. In addition, the introduction
aging, price and nutritional information etc.,1 – 4 may affect the of RTD green tea beverage has brought a significant change in the
sensory acceptability of a food product. Although the sensory green tea market. Although the green tea market is still dominated
attributes perceived by consumption are known to be the major by loose tea and tea bags, it is reported that the consumption of
determinant factors of food acceptability,5 – 7 various researchers loose tea and tea bags is decreasing, whereas the consumption
have also identified how consumer acceptability can be affected of the RTD type is increasing both in the Korean and in the US
by non-sensory factors, with examples of lamb sausages,8 beer,4 food markets.21 In spite of the growth of the RTD green tea bev-
pasta,9 juices,6,10 – 12 bread,13 black teas,14 commercial canned erage market, not much sensory and consumer research has been
tea,15 powdered drink preparations16 and various commercial reported. Previous studies of RTD green tea beverages have mainly
beverages.17 focused on investigating the change in quality properties of green
Green teas are particularly known to be consumed not only teas based on different processing methods.22,23 Considering the
for their delicate flavor characteristics, but it also for their recog- increasing interest in RTD green tea beverages and the new inflow
nition as a healthy product.15,18,19 In the study of Cho et al.15 of consumer pools, investigation on how sensory and non-sensory
and Kim et al.,17 older respondents showed a tendency to pre- factors, in particular health information, would affect the accept-
fer green tea beverages over other various types of beverages. ability of RTD green tea beverages is needed in order to better
Nowadays, the health food market is steadily growing,20 and prod- understand the green tea market.
ucts that are consumed for their health claims, such as green
tea, will be increasingly consumed for their health beneficial
effect. ∗ Correspondence to: Kwang-Ok Kim, Department of Food Science and Engi-
Green teas can be found not only in the form of loose tea or neering, Ewha Womans University, 52 Ewhayeodae-gil, Seodaemun-gu, Seoul
tea bags, but now can also be seen in the form of ready-to-drink 120–750, Republic of Korea. E-mail: kokim@ewha.ac.kr
(RTD). RTD green tea beverages are produced worldwide, and var-
† There are two first authors in the article: Ye Won Bae and Soh Min Lee.
ious types of RTD green tea drinks exist, with varying levels of
green tea extracts and added flavorings. The introduction of RTD Department of Food Science and Engineering, Ewha Womans University, Seoul,
1362

green tea beverage has allowed diverse consumers to consume 120-750, Republic of Korea

J Sci Food Agric 2016; 96: 1362–1372 www.soci.org © 2015 Society of Chemical Industry
Age and gender difference in green tea beverage acceptability www.soci.org

Therefore, the purpose of this study was to investigate how sen- using the water provided between samples. After tasting the first
sory and non-sensory factors, in particular health-related informa- four samples, panelists were given a 5 min break before tasking the
tion, would affect flavor preferences of RTD green tea beverages. next four samples, in order to minimize fatigue. Flavor and mouth-
Furthermore, because it has been documented in several studies feel attributes were evaluated first, in a monadic sequence, and
that older consumers have a tendency to consider health-related then appearance attributes were examined at the end. All sensory
information more than the adolescent group,15,17,24 the study attributes were evaluated on a 16-point category scale, from ‘none
investigated how the influence of sensory and non-sensory factors (0)’ to ‘strong (15)’. The evaluation of flavor and mouth attributes
would differ based on different age and gender groups. was conducted in an isolated booth under a dim red light, so as
to avoid any biases that might occur from the color differences
of the samples. For appearance evaluation, samples were placed
EXPERIMENTAL in a light box (D65 Superlight-III; Boteck, Siheong, Korea) under
Materials daylight conditions. Each evaluation session took approximately
Various commercial RTD green tea beverages were selected that 40 min to complete, and the test was repeated four times.
differed in price, country of origin, brand/manufacturer and nutri-
tional information. In addition, green teas were selected to differ Consumer test
in their main ingredients in order to provide a diversity in sensory Selection of consumers
characteristics. Samples GT-I, GT-II, GT-O and GT-JP had higher pro- In total, 358 consumers who drink green teas or green tea bever-
portions of green tea (GT) extract or infusion in the product. FGT-L, ages were recruited as participants. In order to see how the influ-
FGT-A and FGT-US were flavored green teas (FGT) that included ence of extrinsic product information would differ based on differ-
flavorings and contained HFCS (high-fructose corn syrup) with a ent age/gender groups, consumers were recruited based on four
sugar content of 11%, 10% and 7%, respectively. GT-CS, a sweet- subdivided groups, with two age groups (young, 10–30s; and old,
ened version of GT-I sample, was made by adding HFCS (Samyang 40–60s) and with two gender groups (male and female). In order
Genex Corporation, Inchon, Korea) to GT-I in order to meet 4% of to represent the diverse range of consumers, recruitment was per-
sugar content, which is set to be in the middle of 0% and 7%. formed at various places including universities, government insti-
The manufacturers of GT-I, FGT-US and GT-JP are number one tutions, business offices, churches and temples, or local commu-
sellers in the RTD green tea market for Korea, the USA and Japan, nity service centers, all located in Seoul (South Korea). To sum up,
respectively. GT-CS used the same bottle as that for GT-I. The 358 consumers were recruited consisting of 87 young males (YM),
GT-O sample had an ‘organic green tea’ label on its bottle. A more 94 young females (YF), 90 old males (OM) and 87 old females (OF).
detailed description for each sample is given in Table 1.
Sample preparation and presentation
Descriptive analysis Aliquots (50 mL) of samples were poured into paper cups (6.5 oz,
Panel selection and training Easepack Co., Namyangju, Gyeonggi-do, Korea), which were coded
Descriptive analysis was performed using eight trained panelists with three-digit random numbers. The serving order of the sample
(aged 25–27, eight females). All the panelists were students of was based on an 8 × 8 Williams Latin square design.27 Filtered tap
Ewha Womans University (Seoul, South Korea) and had previous water was provided for palate cleansing.
experience in descriptive analysis. During the training, the panel
generated sensory descriptors for the green tea beverages used Evaluation procedure
in the study. They then defined and selected an appropriate Recruited people were randomly segmented into two groups:
reference sample for each sensory descriptor (Table 2). The training a blind condition group (n = 179) and an informed condition
continued until the panel met consensus in each term and were group (n = 179). Blind condition consumers were only provided
able to reproduce all the information. Training sessions were with the sample itself for tasting, whereas the informed condition
held three times per week for 2 months and each session lasted consumers were provided with the actual bottle of the product
approximately an hour. along with the sample for tasting. Note that there was a tag that
was attached on the bottle to highlight extrinsic information,
including the name of the product, manufacturer (with country
Sample preparation and presentation
of origin for imported products), price, volume and calories per
Fifty milliliters of each sample were poured into amber glass
100 mL.
vials (50 mL) that were coded with three-digit random numbers.
For each given green tea beverage, consumers were instructed
The samples were served and evaluated at room temperature
to provide the degree of flavor liking, expectation in health
(20 ± 2 ∘ C) and the presentation order of the samples was random-
improvement and purchase intent, using a 15-point category
ized for each subject. For appearance evaluation, a new set of sam-
scale, with labels from ‘dislike very much’ (1) to ‘like very much’ (15),
ples was poured into a transparent glass vial (50 mL) with new ran-
and ‘neither dislike nor like’ as a middle point (8); ‘not at all help-
dom numbers. Each panelist was provided with filtered tap water
ful’ (1) to ‘very helpful’ (15); and ‘no intent to purchase’ (1) to ‘have
(Ceramic Filter System, Supercape, Dalton, Fairey Industrial Ceram-
much intent to purchase’ (15), respectively. On the other hand, the
ics Ltd, London, UK) (20 ± 2 ∘ C) for palate cleansing. informed group consumers were additionally asked to provide the
degree of package liking and price satisfaction, with labels from
Tasting and rating procedure ‘dislike very much’ (1) to ‘like very much’ (15), and ‘neither dislike
A quantitative descriptive analysis (QDA®) procedure25 was used nor like’ as a middle point (8); and ‘not satisfied at all’ (1) to ‘very
in conjunction with partial adoption of the Spectrum descrip- satisfied’, respectively. Consumers evaluated the first four samples
tive analysis method.26 Panelists were asked to sip each sample first, and then took a 10 min break before evaluating the other four
(approximately 10 mL per sip), and to evaluate the intensity of each samples. Instruction was given to rinse one’s palate between the
1363

sensory attribute. Subjects were instructed to cleanse their palate tested products using the water provided.

J Sci Food Agric 2016; 96: 1362–1372 © 2015 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org Y-W Bae, SM Lee, K-O Kim

Table 1. Information on eight green tea beverage samples used in this study

Product information provided to consumers

Sugar
Sample Main content Price (won) Volume Calories Country
identification ingredients (w/v %) [won 100 mL−1 ] (mL) (kcal 100 mL−1 ) of origin Other key features

FGT-L Concentrated lemon 11 1000 [294] 340 42.5 Korea

juice, green tea,
green tea extract,
lemon flavor
FGT-A Concentrated apple 10 1500 [429] 350 40.0 Korea
juice, green tea,
green tea extract,
honey flavor
FGT-US Honey flavor, ginseng 7 2800 [591] 473 29.2 U.S.A US number 1 market share
extract, green tea product
GT-CS Green tea, vitamin C, 4 1000 [286] 350 16.0 Korea
high fructose corn
syrup
GT-I Green tea, vitamin C 0 1000 [286] 350 0 Korea Korean number 1 market
share product (same
bottle used for GT-CS)
GT-II Green tea, vitamin C 0 1200 [293] 410 0 Korea
GT-O Green tea, vitamin C 0 1000 [200] 500 0 Korea Labeled as ‘organic green
tea’
GT-JP Green tea, vitamin C 0 2300 [460] 500 0 Japan Japanese number 1 market
share product

Statistical analysis Based on the ANOVA and PCA results (Table 3 and Fig. 1), the
On the descriptive analysis data, analysis of variance (ANOVA) eight green tea beverages were mainly classified into two groups
was conducted with the main effects of ‘judge’, ‘sample’, ‘replica- along with the PC1 dimension. The negative side of PC1 was
tion’ and their two-way interactions, and the sample means were defined by strong ‘brownness’ appearance, ‘bitter’ taste, ‘green
compared for each attribute using Duncan’s multiple range test tea leaf’, ‘roasted cassia seed’, ‘roasted barley’, ‘beany’, ‘metallic’,
(𝛼 = 0.05). Principal component analysis (PCA) was also conducted ‘roasted Solomon’s seal tea’, ‘toasted rice’, ‘chestnut shell’, ‘root’,
on the descriptive data averaged across panelists and replications, ‘burnt’ flavors and ‘astringency’ mouthfeel, whereas the positive
in order to summarize the interrelationship among the sensory side of PC1 was defined by ‘sweet’ taste, ‘sweet’, ‘honey’ and ‘floral’
attributes and the eight tested samples. Flavor liking scores of the flavors.
four consumer groups in the blind condition (i.e. B-OM, B-OF, B-YM Along with PC1, samples were mainly divided into flavored green
and B-YF) were plotted on to the PCA plot as a supplementary vari- teas (FGT-L, FGT-A and FGT-US) and plain type green teas (GT-I,
able to identify sensory factors affecting flavor liking. GT-II, GT-O and GT-JP) (Fig. 1). Plain type green tea samples (GT-I,
On the consumer data, ANOVA was performed on the data of the GT-II, GT-O and GT-JP), containing relatively higher proportions
358 consumers to evaluate the effect of age, gender, product and of green tea extract, were shown to have strong characteristics
information on flavor liking and on expectation in health improve- that defined the negative side of PC1. On the other hand, green
ment ratings. Then, for each consumer group in each test condi- tea beverages that were flavored (FGT-L, FGT-A and FGT-US) were
tion, ANOVA with the main effects of ‘consumer (random effect)’ located on the positive side of PC1, and were characterized by
and ‘sample’ was performed. Sample means were compared using strong ‘sweet’ taste, and ‘sweet’, ‘honey’ and ‘floral’ flavors. Sample
Duncan’s multiple range test (𝛼 = 0.05). Correlation analysis was GT-CS was located in the middle of PC1 dimension, which had
performed on the data for each consumer group in order to see medium intensities of the sensory characteristics that defined the
how expectation in health improvement, package liking and price PC1 dimension.
satisfaction scores were related to flavor liking scores. All statis- The PC2 dimension, which explained 9.68% of the total variation,
tical analyses were conducted using XLSTAT for Windows 14.0 further separated flavored green tea beverages (FGT-L, FGT-A and
(Addinsoft, France) and SPSS for Windows 18.0 (SPSS Inc., Chicago, FGT-US) and GT-CS. FGT-L, the lemon-flavored green tea beverage,
IL, USA). was highly loaded on the positive PC2 dimension, and was further
defined by ‘sour’ and ‘salty’ tastes and ‘lemon’ flavor. On the other
hand, sample GT-CS was located on the negative PC2 dimension,
RESULTS AND DISCUSSION and was shown to have the strongest ‘sweet’ flavor intensity
Sensory characteristics of green tea beverages among the eight tested products. It should be noted that although
From descriptive analysis, 23 sensory descriptors were generated the sugar content of this sample was not the highest, nor was its
for the eight green tea beverages (Table 2). ANOVA results showed perceived sweet taste intensity, the product was found to have the
that the intensities of all sensory attributes significantly (P < 0.05) strongest sweet flavor instead. It is assumed that the overpowering
differed among the samples. The PCA result showed that the first flavor notes of the FGT drinks suppressed the sweet flavor, whereas
two principal components (PCs) explained 77.31% and 11.03% of sweet flavor was more noticeable in GT-CS because there was
1364

the total variance, respectively (Fig. 1). no overpowering flavor note that interrupted the sweet flavor.

wileyonlinelibrary.com/jsfa © 2015 Society of Chemical Industry J Sci Food Agric 2016; 96: 1362–1372
Age and gender difference in green tea beverage acceptability www.soci.org

Table 2. Definitions of the descriptive attributes of green tea beverages

Sensory attribute Definition Reference sample

Brownness Intensity of brown color of green tea beverage Color guide (DIC Co., Tokyo)
(appearance)
Sweet taste Fundamental taste sensation of which sucrose is 5% sucrose solution (Sigma Chemical Co. Ltd, St Louis, MO)
typical
Sour taste Fundamental taste sensation of which citric acid is 0.05% citric acid solution (Duksan Pure Chemical Co. Ltd, Ansan
typical Gyeonggi-do)
Bitter taste Fundamental taste sensation of which caffeine is 0.03% caffeine solution (Sigma Chemical Co. Ltd, St Louis, MO)
typical
Salty taste Fundamental taste sensation of which sodium 0.3% NaCl solution (Duksan Pure Chemical Co. Ltd, Ansan,
chloride is typical Gyeonggi-do)
Sweet Aromatics associated with high-fructose corn syrup 4% high-fructose corn syrup solution (Samyang Genex, Incheon)
Green tea leaf Aromatics associated with green tea leaf 1 g green tea leaves (Sulloc Jungjak, Amorepacific Corp., infused with
200 g boiling water for 2 min) mixed with 60 g green tea)
Black tea Aromatics associated with Assam tea Assam tea (1 g Assam tea (Nokchawon, Seoul) extracted with 200 g
water at 95 ∘ C for 3 min)
Roasted cassia seed Aromatics associated with roasted cassia seed Cassia seed tea (10 g Cassia seed tea (Dong Suh Co. Ltd, Siheung,
Gyeonggi-do) extracted with 400 g water at 95 ∘ C for 10 min
Roasted barley Aromatics associated with roasted barley Barley tea (10 g Cassia seed tea (Dong Suh Co. Ltd, Siheung,
Gyeonggi-do) extracted with 400 g water at 95 ∘ C for 10 min
Beany Aromatics associated with steamed green peas Steamed green peas (local supermarket, Seoul)
Solomon’s seal tea Aromatics associated with Solomon’s seal tea Solomon’s seal tea (10 g Solomon’s seal tea seed tea (Dong Suh Co.
Ltd, Siheung, Gyeonggi-do) extracted with 400 g water at 95 ∘ C for
10 min
Toasted rice Aromatics associated with toasted rice Toasted rice (60 g toasted rice (Ottogi Co. Ltd, Hwasung,
Gyeonggi-do) extracted with 400 g water 95 ∘ C for 4 min
Honey Aromatics associated with honey Acacia honey (Ottogi Co. Ltd, Anyang, Gyeonggi-do)
Apple (artificial) Aromatics associated with apple essence 0.05% apple oil solution (Seoul Perfumery Co. Ltd, Seoul)
Lemon (artificial) Aromatics associated with lemon essence 0.05% lemon oil solution (Seoul Perfumery Co. Ltd, Seoul)
Red ginseng Aromatics associated with red ginseng Red ginseng extract (0.5 g Korean red ginseng extract (Korea
Ginseng Corporation, Daejeon) mixed with 150 g green tea)a
Floral Aromatics associated with jasmine tea 10 g jasmine tea (1 tea bag, Hyanggiropin jasmine, Amorepacific Co.
Ltd, Jincheon, Chungcheongbuk-do) infused with 200 g boiling
water for 2 min) mixed with 60 g green tea)a
Root Aromatics associated with dried arrowroot Arrowroot extract (10 g arrowroot extract (Birak Co. Ltd, Jincheon,
Chungcheongbuk-do) mixed with 100 g green tea)a
Chestnut shell Aromatics associated with boiled chestnut shell 30 g chestnut shell (local supermarket, Seoul) boiled with 500 mL
water at low heat for 10 min
Burnt Aromatics associated with burnt wood, burnt 1 g burnt arrowroot (local supermarket, Seoul)
arrowroot
Metallic Aromatics associated with metals, tinny or iron Stainless steel spoon
Astringency The feeling which shrivels the tongue with tannin and 0.1% tannic acid (Duksan Pure Chemical Co. Ltd, Ansan,
aluminum sulfate (mouthfeel) Gyeonggi-do) solution
a Green tea beverage (Oiocha green tea, Itoen, Tokyo, Japan).

FGT-US was further characterized by strong ‘red ginseng’ and point), and GT-II was the least liked sample among the eight green
‘black tea’ flavors (Table 3). FGT-A, the apple-flavored green tea tea beverages tested by the blind group consumers.
beverage, was shown to have a strong ‘apple’ flavor characteristic Although consumers generally liked samples that were ‘flavored
(Table 3). and sweetened’, it was observed that consumers disliked a product
that was ‘sweetened’ without any flavoring. In Korea, it is difficult to
Influence of sensory characteristics on consumer find green teas or green tea beverages that are sweetened without
acceptability (blind condition) any flavorings. Hence familiarity may have played a role when the
Flavor liking of green tea beverages test participants evaluated the products and resulted in disliking
The blind group consumers’ (n = 179) mean flavor liking scores of sweetened plain green tea sample (GT-CS), whereas the flavored
for the eight green tea beverages are shown in Table 4. It was green teas were well accepted. Prescott et al.28 stated that certain
observed that in the blind group consumers generally preferred flavors may be regarded as inappropriate for specific foods in
flavored green tea beverages (FGT-L, FGT-A and FGT-US) over some cultures, while it can be accepted in others. Hence the result
other samples. Among the rest of the five green tea beverages, GT-I implied that when green teas are in a beverage form sweet taste
was liked the most, followed by GT-JP, which received a liking score and fruit flavors, such as lemon or apple flavors together would
that was close to the ‘neither like nor dislike’ point. GT-CS and GT-O have a positive effect on Korean consumers’ acceptability, whereas
1365

had lower liking scores than GT-JP (and thus below the neutral sweet taste/flavor on its own would not have the same positive

J Sci Food Agric 2016; 96: 1362–1372 © 2015 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org Y-W Bae, SM Lee, K-O Kim

(a) 1

F2 (11.03 %)
B-OF
Lemon
Salty taste
Metallic Sour taste
0.5
Astringency B-YF
Bitter taste B-YM
Root
Burnt
Green tea leaf
Roasted cassia seed B-OM
Chestnut shell
Sweet taste
Solomon's seal tea0 Apple
Roasted barley Floral
Beany Honey
Black tea
Toasted rice
Red ginseng

Brownness Sweet flavor

–0.5

F1 (77.31 %)
–1
–1 –0.5 0 0.5 1

Active variables Supplementary variables

(b) 15
F2 (11.03 %)

10

FGT-L

GT-JP
GT-O GT-I
0
GT-II FGT-A

FGT-US
–5

GT-CS

–10

F1 (77.31 %)
–15
–15 –10 –5 0 5 10 15

Figure 1. PC loadings of the sensory attributes, flavor likings of consumers differing in age and gender, (a) and eight green tea beverages (b). B, blind
condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young female.

influence in consumer acceptance. Further research is needed to observed that ‘salty’, ‘sour’ and ‘sweet’ tastes and ‘lemon’, ‘floral’
see how these sensory attributes are related to each other in terms and ‘honey’ flavors were the major factors that influenced liking
of consumer liking for green tea beverages. directions of all four groups of consumers (B-OM, B-OF, B-YM and
B-YF) for green tea beverages. The result is consistent with the find-
Sensory drivers of flavor liking ings of Cho et al.15 and Kim et al.17 Cho et al.15 reported a prefer-
In order to reveal sensory drivers of flavor liking of green tea bever- ence of Korean consumers toward lemon-flavored black tea bever-
ages for each consumer group, the mean score of flavor liking was age that had a sweetener added over other commercial beverages,
plotted as a supplementary variable on to the sensory space that including green tea, oolong tea and black tea with milk. Kim et al.17
1366

was derived from PCA on descriptive analysis data (Fig. 1). It was reported that sweetened and fruit-flavored product was strongly

wileyonlinelibrary.com/jsfa © 2015 Society of Chemical Industry J Sci Food Agric 2016; 96: 1362–1372
Age and gender difference in green tea beverage acceptability www.soci.org

Table 3. Mean intensity scores for sensory attributes of eight green tea beverages

Sensory attributes FGT-L FGT-A FGT-US GT-CS GT-I GT-II GT-O GT-JP

Appearance Brownness 1.53f 3.22e 4.41d 8.72b 7.16c 11.44a 8.94b 7.06c
Taste Sweet taste 9.47a 8.34b 6.00c 5.63c 0.97d 0.38d 0.63d 0.50d
Sour taste 9.25a 6.47b 3.56c 0.78de 1.34de 0.63e 1.53d 1.03de
Bitter taste 1.47d 0.97d 1.06d 1.13d 4.25c 5.22b 5.47b 7.00a
Salty taste 4.53a 2.72b 1.91c 1.31c 1.13c 1.53c 1.25c 1.31c
Flavor Sweet 5.34b 5.34b 5.00b 7.53a 0.63c 0.16c 0.44c 0.16c
Green tea leaf 1.81d 1.75d 1.94d 3.16c 6.53b 6.78b 7.06b 8.09a
Red ginseng 1.53b 1.72b 4.34a 1.13b 1.19b 1.56b 1.34b 1.25b
Honey 4.25a 4.59a 4.28a 2.53b 0.22c 0.13c 0.09c 0.06c
Floral 2.47c 3.81a 3.22b 1.13d 0.47e 0.34e 0.53e 0.28e
Black tea 2.91b 2.06c 6.53a 1.41cd 0.94d 1.16d 1.09d 0.69d
Roasted cassia seed 0.19d 0.34d 0.78cd 1.31c 4.72b 5.06ab 6.00a 5.72a
Roasted barley 0.97c 0.75c 1.97b 2.44b 4.34a 3.63a 4.62a 3.69a
Beany 0.41c 0.31c 0.88c 1.97b 3.03a 2.78ab 3.41a 3.25a
Metallic 2.28c 1.31cd 0.81d 1.16d 3.53b 3.50b 5.56a 4.41b
Roasted Solomon’s seal tea 0.31c 0.28c 0.63c 1.78b 3.13a 3.53a 2.69a 3.41a
Toasted rice 0.28d 0.28d 0.81cd 1.47bc 2.59a 2.38a 2.00ab 1.94ab
Chestnut shell 0.25c 0.63c 0.47c 1.19c 4.19b 6.44a 4.56b 5.94a
Root 0.22c 0.28c 0.34c 1.06c 3.63b 4.69a 4.69a 5.47a
Burnt 0.00c 0.13c 0.09c 0.78c 2.59b 3.66a 3.66a 4.47a
Lemon (artificial) 9.16a 1.06c 2.00b 0.13d 0.00d 0.00d 0.00d 0.00d
Apple (artificial) 1.38b 8.06a 1.13b 0.22c 0.00c 0.00c 0.00c 0.00c
Mouthfeel Astringency 2.28c 1.50cd 1.16d 0.97d 4.13b 4.66b 5.63a 6.44a

Values within a row not sharing a letter are significantly different (P < 0.05, Duncan’s multiple range test). 16-point category scale (0 = ‘none’,
15 = ‘strong’).

Table 4. Flavor liking scores in blind and in informed conditions of consumers groups differing in age and gender for eight green tea beverages

Consumer group FGT-L FGT-A FGT-US GT-CS GT-I GT-II GT-O GT-JP

Blind group B-Total (N = 179) 10.68a 10.09a 9.26b 7.31d 8.68bc 6.85d 7.34d 8.07c
B-OM (N = 45) 9.87a 9.38ab 10.09a 8.38bc 8.27bc 7.53c 8.07bc 8.02bc
B-OF (N = 44) 10.14a 10.23a 8.84ab 6.55c 9.36ab 8.05b 8.30b 8.39b
B-YM (N = 43) 12.53a 11.70a 9.21b 7.33cd 8.14bc 5.09e 6.19de 7.49cd
B-YF (N = 47) 10.26a 9.19ab 8.91ab 6.98cd 8.94ab 6.70d 6.79d 8.36bc
Informed group I-Total (N = 179) 10.63ab 10.80a 10.08bc 7.05f 9.39c 7.74ef 8.66d 7.84e
I-OM (N = 45) 9.60a 9.98a 9.51ab 7.98bcd 8.84abc 7.09d 8.67abc 7.87cd
I-OF (N = 43) 9.84a 10.19a 9.67a 6.09b 8.98a 8.65a 8.67a 8.77a
I-YM (N = 44) 11.61a 11.61a 10.75a 6.84d 9.48b 7.64cd 8.32bc 7.43cd
I-YF (N = 47) 11.43a 11.40a 10.36a 7.23c 10.21ab 7.62c 8.96b 7.34c

Values within a row not sharing a letter are significantly different (P < 0.05, Duncan’s multiple range test).
B, blind condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young female.

preferred by Korean consumers among other various types of hot water.29 Hence, because old consumers are more exposed to
beverages, including ion drink, black tea drink, mixed tea drink etc. traditionally prepared green teas, older consumers may have rela-
tively preferred plain green tea samples than the young group. The
Age and gender differences in flavor liking of green tea beverages observation can be partially supported by the study of Lee et al.,30
Age and gender differences were observed in flavor liking ratings which found that consumers who only drink green teas that are
for the eight green tea beverages (Table 4 and Fig. 2). Old con- prepared in a traditional way tended to like green teas with more
sumers (B-OM and B-OF) seemed to like plain green tea beverages complex original flavors and less preferred other types of green
relatively more than the young, because there was no plain green teas that were lacking in tea characteristics due to the decaffeina-
tea sample that was strongly rejected by these consumers in tion process. Another explanation of the observation could be the
general. It is known that while younger people in Korea tend to health concerns of the old consumers. As reported by Ares et al.,31
drink green teas more in a teabag or in a PET form, which lack older respondents tend to be more concerned about health than
the complex original green tea flavors, older people tend to drink younger respondents. Because green teas are generally perceived
1367

green teas in a more conventional way, infusing loose tea leaves in as a healthy product,15,18,19 plain green teas may have had an

J Sci Food Agric 2016; 96: 1362–1372 © 2015 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org Y-W Bae, SM Lee, K-O Kim

(a) (b)

(c) (d)

Figure 2. Flavor liking scores of the green tea beverages. B, blind condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young
female.

Table 5. Expectation in health improvement scores in blind and in informed conditions of consumer groups differing in age and gender for eight
green tea beverages

Consumer group FGT-L FGT-A FGT-US GT-CS GT-I GT-II GT-O GT-JP

Blind group B-Total (N = 179) 8.36b 7.72b 8.22b 7.79b 9.96a 9.56a 9.59a 9.66a
B-OM (N = 45) 8.36bc 7.47c 8.73abc 8.36bc 9.49ab 9.09ab 9.84a 9.16ab
B-OF (N = 44) 8.00c 8.59bc 7.70cd 6.52d 10.20a 9.43ab 9.36ab 9.50ab
B-YM (N = 43) 9.53a 9.12a 9.12a 9.19a 9.49a 10.12a 9.58a 10.00a
B-YF (N = 47) 7.62b 5.85c 7.40b 7.17b 10.62a 9.64a 9.55a 9.98a
Informed group I-Total (N = 179) 7.92de 7.25f 8.46d 7.85e 10.09ab 9.74bc 10.44a 9.20c
I-OM (N = 45) 8.87cd 7.84d 8.18cd 8.67cd 10.64a 9.22bc 10.18ab 9.47cd
I-OF (N = 43) 7.07c 7.20c 9.05b 8.05c 9.64b 9.91ab 10.75a 9.70b
I-YM (N = 44) 8.98ab 7.70b 8.72ab 7.65b 9.51a 9.74a 9.86a 8.49ab
I-YF (N = 47) 6.83c 6.32c 7.94b 7.09bc 10.51a 10.06a 10.94a 10.09a

Values within a row not sharing a letter are significantly different (P < 0.05, Duncan’s multiple range test).
B, blind condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young female.

advantage of tasting like a healthy drink. As expected, all four con- r = 0.492; B-OF, r = 0.637) tended to have a higher correlation
sumer groups rated ‘expectation of health improvement’ scores coefficient in relation to health improvement score to the flavor
higher for plain green teas than the other sweetened samples liking score, compared to the young (B-YM, r = 0.272; B-YF,
(Table 5). Although most consumers agreed on the ‘healthier taste’ r = 0.358).
of the plain green tea products, it was found that older consumers On the other hand, gender difference in acceptability for green
were more highly affected by the health aspect of the product in tea beverages was also observed (Table 4 and Fig. 2). It was found
their flavor acceptability than the young. According to the corre- that, unlike the two male consumer groups (B-OM and B-YM) con-
1368

lation analysis results shown in Table 6, older consumers (B-OM, sumers, the two female groups (B-OF and B-YF) strongly rejected

wileyonlinelibrary.com/jsfa © 2015 Society of Chemical Industry J Sci Food Agric 2016; 96: 1362–1372
Age and gender difference in green tea beverage acceptability www.soci.org

Table 6. Correlation coefficients of expectation in health improvement, package liking, and price satisfaction to flavor liking rating

Blind group Informed group

Variable B-OM B-OF B-YM B-YF I-OM I-OF I-YM I-YF

Health improvement 0.492 0.637 0.272 0.358 0.504 0.484 0.088 0.056
Package liking – – – – 0.503 0.439 0.374 0.449
Price satisfaction – – – – 0.372 0.227 0.235 0.289

Bold entries indicate significance at P < 0.05.

B, blind condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young female.

GT-CS, a product that was ‘sweetened’ without any flavoring.

Table 7. ANOVA F-ratios of each source of variation for flavor liking
Females in Korea are generally known to be sensitive to ‘calories’ and expectation in health improvement ratings using total consumers’
because of the social pressure on their body shape.32 Hence data (n = 358)
the strong ‘sweet taste’ and medium intensity of ‘sweet flavor’
of GT-CS may have led to a rejection of this sample. It should Expectation in
be noted that although FGT samples had a higher sugar content Flavor health
(7–11%) compared to GT-CS (4%), these samples were not disliked Source of variation d.f. liking improvement
by female consumers. The reason seems to be associated with
Product 7 54.370*** 32.906***
the discouragement of consumers’ expectation on ‘plain’ green
Age 1 0.271 0.413
tea beverages. It is assumed that health expectation for flavored
Gender 1 0.045 9.740**
green teas may not be as high as one expects from a plain green
Condition 1 13.246*** 0.0003
tea beverage. Therefore, the sweet taste and sweet flavor of GT-CS
Age × Gender 7 0.282 3.345
probably seemed unhealthy for a ‘plain’ green tea product that
was expected to taste ‘healthy’, and have resulted in a low flavor Product × Age 7 6.186*** 3.124**
liking rating for GT-CS among the female consumers. The observa- Product × Gender 7 3.230** 3.204**
tion implied that unhealthiness information was only affective for Product × Condition 1 2.495* 1.680
plain green tea beverages and not for the flavored and sweetened Product × Age × Gender 7 2.452* 2.632*
green tea beverages. Further investigation is needed to reveal Product × Age × Condition 8 2.264* 2.242*
how expectation can influence the perception of healthiness of a Product × Gender × Condition 8 0.604 1.204
product. Product × Age × Gender × Condition 8 1.474 1.190

Bold entries indicate significance at P < 0.05.

Influence of extrinsic product information (informed Asterisks indicate significance at * P < 0.05 ** P < 0.01 *** P < 0.001.
condition)
Expectation in health improvement, package liking and price
satisfaction hand, old male and female consumers (I-OM and I-OF) showed
As shown in Table 5, when product information was not given, con- differences in patterns of package liking ratings compared to the
sumers provided lower scores for expectation in health improve- young consumers. The young and old particularly disagreed about
ment when the product tasted sweet and higher when the product the package liking in the FGT-US and GT-O products. FGT-US, an
tasted unsweet. That is, consumers in general reported that plain imported product from the USA (Table 1), can be easily found
green tea samples GT-I, GT-II, GT-O and GT-JP were expected to be in coffee shops such as Starbucks that are mainly visited by the
more helpful in improving health compared to FGT samples and younger generation in Korea. Hence differences in the exposure to
GT-CS. On the other hand, when product information was given, the packaging of FGT-US may have induced the difference in pack-
consumers still rated plain green tea samples higher in expectation age liking score between the young and the old. In addition, older
in health improvement score. The ANOVA result also supported the consumers showed a relatively higher package liking score for
observation by showing no significant (P < 0.05) effect of ‘prod- GT-O, the organic labelled product, compared to the young con-
uct × condition’ source of variation (Table 7). Among the plain sumers. Because organic product is acknowledged by consumers
green tea samples, GT-O received the highest health improvement to be beneficial to health,33 old respondents seemed to prefer the
score when the product was given with a package informing that packaging of GT-O because older respondents have a tendency
it is an ‘organic’ product. According to Michaelidou and Hassan,33 to be more concerned about health than younger respondents.31
one of the consumers’ motives for choosing the organic product is Price satisfaction of the products was in agreement among the
consumers’ perception of organic produce as a healthier alterna- four consumer groups (Table 8). Satisfaction was low when the
tive to conventional foods. product had a higher price per same serving size and vice versa
In the informed condition, consumers were also asked to rate (Tables 1 and 8). It is noteworthy that upper side of the scale used
package liking and price satisfaction for each product (Table 8). For for ‘price satisfaction’ by the old female group was lower than the
package liking, FGT samples and GT-II had relatively higher pack- scores of other consumer groups (Table 8). In Korea, young con-
age liking scores. Interestingly enough, the two young consumers sumers or old males are the main RTD beverage buyers, while old
(I-YM and I-YF) showed similar patterns in package liking ratings females, especially housewives, tend to buy green teas in a form
of the products, preferring the package of FGT-US the most, sec- of tea bags instead of RTD form. Because generally RTD green tea
1369

ondly FGT-A, followed by samples FGT-L and GT-II. On the other beverages are much more expensive than teabags, old females

J Sci Food Agric 2016; 96: 1362–1372 © 2015 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org Y-W Bae, SM Lee, K-O Kim

Table 8. Package liking and price satisfaction scores in informed conditions of consumer groups differing in age and gender for eight green tea
beverages

Consumer group FGT-L FGT-A FGT-US GT-CS GT-I GT-II GT-O GT-JP

Package liking I-Total (N = 179) 10.10a 10.09a 9.84a 8.87bc 9.02bc 9.44ab 8.75bc 8.55c
I-OM (N = 45) 10.40a 8.96abc 8.78bc 9.60ab 9.89ab 8.44bc 9.40abc 8.04c
I-OF (N = 43) 10.95a 10.28ab 8.40c 9.23bc 9.42abc 10.05ab 9.51abc 9.00bc
I-YM (N = 44) 9.50bc 10.36ab 10.75a 8.91c 8.52c 9.64abc 8.39c 8.68c
I-YF (N = 47) 9.60bc 10.74ab 11.32a 7.79d 8.28d 9.66bc 7.77d 8.49cd
Price satisfaction I-Total (N = 179) 10.58b 8.06d 5.73e 9.96b 10.04b 9.35c 11.68a 4.12f
I-OM (N = 45) 10.58a 8.69c 5.69d 10.44ab 10.80a 9.13bc 11.64a 4.16e
I-OF (N = 43) 8.88b 6.67c 5.40c 8.65b 8.79b 9.12b 10.67a 3.98d
I-YM (N = 44) 11.25ab 8.25d 5.50e 10.55bc 9.80c 9.45c 12.11a 4.00f
I-YF (N = 47) 11.49ab 8.55d 6.30e 10.15c 10.68bc 9.68c 12.21a 4.34f

Values within a row not sharing a letter are significantly different (P < 0.05, Duncan’s multiple range test).
B, blind condition; I, informed condition; OM, old male; OF, old female; YM, young male; YF, young female.

seemed to provide lower satisfaction scores than other consumer from two viewpoints. Firstly, as mentioned earlier, old consumers
groups. may have changed their diet to one that tastes ‘healthy’, and
have been familiar with products with less sugar or fat, etc.,
Age and gender difference in the influence of extrinsic information because old consumers are more concerned about health. When
on flavor liking the product information was given to consumers, there was not
The ANOVA results of the 358 consumers’ flavor liking scores much ‘new’ information that consumers could have related to
ratings are shown in Table 7. A significant (P < 0.05) effect healthiness other than what they already knew about healthiness
was observed on ‘product × age × condition’ on flavor liking information, judged based on the taste of the product. In other
rating, whereas there was no significant effect on ‘prod- words, the sweet-tasting products also had higher sugar content
uct × gender × condition’ or ‘product × gender × age × condition’. and calories, which did not conflict with taste quality and the
The result implied that the influence of information on product information given. This can be supported by the ‘expectation of
rating differed mainly based on age groups, and not so much on health improvement’ rating of consumers given to the product in
gender groups. blind and information conditions. Samples GT-I, GT-II and GT-O
In order to investigate how each consumer group is affected by and GT-JP were all rated relatively higher than other samples in
the product information, the flavor liking scores between the blind the expectation of health improvement rating for both OM and
and informed conditions were compared side by side within each OF groups in two test conditions (Table 5). Hence, within the
specific age and gender groups (B-OM vs. I-OM; B-OF vs. I-OF; B-YM design of the study, it was unclear how old consumers made the
vs. I-YM; and B-YF vs. I-YF). It was found that old consumers’ liking trade-off of taste quality and information, but it can be concluded
scores for each product did not change much and were almost that older consumers had a tendency to being affected by the
the same with or without the product information, whereas much health-related aspect of the product.
more dramatic change was found in the young consumers’ flavor Young consumers showed a much more dramatic difference in
liking scores (Table 4 and Fig. 2). Gender differences seemed to their flavor liking scores with or without the extrinsic product
exist, but it was not as dramatic a separator as age differences, as information (Table 4 and Fig. 2). For the young female group, all the
can be shown in the ANOVA results (Table 7). samples except GT-JP rated higher liking scores with the product
Although old consumers were particularly expected to be information than without. It has been reported that providing
affected by information on sugar content and calories, since product information results in a general increase in consumers’
this information can relate to the healthiness of the product, acceptability.15,35 For the young male group, the influence of
it was observed that flavor liking scores of the old consumers extrinsic product information seemed to differ based on products
were little influenced by extrinsic product information. The result showing no particular pattern.
seemed to be different from the findings of previous studies In both young consumer groups (YM and YF), flavor liking scores
which reported that older people, in particular for females, have a were higher when the product information was given for sample
tendency to make a trade-off of taste quality for health beneficial GT-O (2.13 points increase for YM and 2.17 points increase for
effects.15,17,34,35 However, it should be noted that having little influ- YF on a 15-point scale) (Table 4). The big difference in flavor
ence of extrinsic information does not necessarily mean that the liking of GT-O sample can be seen in two points. Firstly, including
old group consumers were not health conscious. By comparing ‘organic’ in its product name for green tea beverages may have
the correlation coefficient of the old groups and the young groups, a positive effect on young consumers’ acceptability. Klihlberg
it was found that older groups had relatively higher coefficients et al.13 reported that information about the organic nature of a
in health improvement to flavor liking score (I-OM, r = 0.504; and bread may increase consumer acceptability. In particular, in the
I-OF, r = 0.484) than the young groups (I-YM, r = 0.088; and I-YF, study of Magnusson et al.36 it was reported that young consumers
r = 0.056) (Table 6). Hence health aspect seemed to influence the were observed to hold more positive attitudes toward organically
liking of the product more for the old compared to the young. The grown food compared to the old. Meanwhile, the price factor also
reason that there was little change in old consumers’ flavor liking seemed to induce higher flavor liking of sample GT-O in I-YM and
1370

score with or without the product information can be explained I-YF groups. In the present study, GT-O sample not only received

wileyonlinelibrary.com/jsfa © 2015 Society of Chemical Industry J Sci Food Agric 2016; 96: 1362–1372
Age and gender difference in green tea beverage acceptability www.soci.org

the highest score in expectation of health improvement, but also 6 Luckow T and Delahunty C, Consumer acceptance of orange juice
had the highest price satisfaction score in I-YM and I-YF consumers containing functional ingredients. Food Res Int 37:805–814
(2004).
(Tables 6 and 7). Several studies have reported that adolescent 7 Haddad Y, Haddad J, Olabi A, Shuayto N, Haddad T and Toufeili I,
consumers tend to be price sensitive.4,37–39 However, considering Mapping determinants of purchase intent of concentrated yogurt
the low correlation coefficient between flavor liking and health (Labneh) by conjoint analysis. Food Qual Prefer 18:795–802 (2007).
improvement score for I-YM and I-YF consumers (I-YM, r = 0.088; 8 Helgesen H, Solheim R and Næs T, Consumer purchase probability of
and I-YF, r = 0.056) compared to price satisfaction (I-YM, r = 0.235; dry fermented lamb sausages. Food Qual Prefer 9:295–301 (1998).
9 Di Monaco R, Cavella S, Di Marzo S and Masi P, The effect of expec-
and I-YF, r = 0.289), it can be assumed that the young consumers tations generated by brand name on the acceptability of dried
showed a higher score for GT-O in the informed condition, not semolina pasta. Food Qual Prefer 15:409–415 (2004).
mainly because of the product’s health-related implication, but 10 Deliza R, MacFie H and Hedderley D, The consumer sensory percep-
because of other factors including the sample’s relatively cheap tion of passion-fruit juice using free-choice profiling. J Sens Stud
price that gave strong satisfaction for these consumers. Further 20:17–27 (2005).
11 Sabbe S, Verbeke W, Deliza R, Matta VM and Van Damme P, Consumer
investigation is needed to better reveal the effect of organic label liking of fruit juices with different açaí (Euterpe oleracea Mart.)
on adolescent consumers’ acceptability for green tea beverages. concentrations. J Food Sci 74:S171–S176 (2009).
12 Gadioli IL, Pineli LLO, Rodrigues JDSQ, Campos AB, Gerolim IQ and
Chiarello MD, ‘Evaluation of packing attributes of orange juice on
consumers’ intention to purchase by conjoint analysis and consumer
CONCLUSIONS attitudes expectation. J Sens Stud 28:57–65 (2013).
The findings of the current study showed that, depending on 13 Kihlberg I, Johansson L, Langsrud Ø and Risvik E, Effects of information
consumers’ age, the ways that consumers are influenced by on liking of bread. Food Qual Prefer 16:25–35 (2005).
14 Schuh C and Schieberle P, Characterization of the key aroma com-
extrinsic product information differ. Acceptability scores of the
pounds in the beverage prepared from Darjeeling black tea: quanti-
young consumer groups for green tea beverages were influ- tative differences between tea leaves and infusion. J Agric Food Chem
enced by the extrinsic product information such as packaging, 54:916–924 (2006).
brand/manufacturer and/or price, but not so much by health ben- 15 Cho HY, Chung SJ, Kim HS and Kim KO, Effect of sensory characteristics
eficial information. On the other hand, old consumers’ flavor liking and non-sensory factors on consumer liking of various canned tea
products. J Food Sci 70:s532–s536 (2005).
score did not change much with or without the product informa- 16 Varela P, Ares G, Giménez A and Gámbaro A, Influence of brand
tion. That is, providing health-related product information had lit- information on consumers’ expectations and liking of powdered
tle influence on the consumers’ flavor acceptability. However, it drinks in central location tests. Food Qual Prefer 21:873–880 (2010).
should be noted that, even without the product information, the 17 Kim HK, Jeon SY and Kim KO, The Effects of demographics and brand
old consumers preferred plain green tea products (i.e. no sweet- information on acceptability of commercial beverage products.
Food Sci Biotechnol 22:1315–1324 (2013).
ener added), which provided a high expectation on the ‘health 18 Khokhar S and Magnusdottir S, Total phenol, catechin, and caffeine
improvement’ rating. When the extrinsic product information was contents of teas commonly consumed in the United Kingdom. J
given, not much change was observed because the same sam- Agric Food Chem 50:565–570 (2002).
ples provided similar health claims. Therefore, the flavor accept- 19 Lee OH, Lee HS, Sung YE, Lee SM, Kim YK and Kim KO, Sensory
characteristics and consumer acceptability of various green teas.
ability of the old consumers seemed to be affected by an indica-
Food Sci Biotechnol 17:349–356 (2008).
tion of ‘health benefit’ derived from green tea beverage. Overall, 20 Gould D, 2013 Food trends get technical, sustainable and healthy.
the findings of this study implied that green tea beverages are [Online]. Available: http://www.forbes.com/sites/daniellegould/
consumed for different reasons depending on different age and 2012/12/28/2013-food-trends-get-technical-sustainable-healthy
gender groups. Therefore, it could be important to promote differ- [20 January 2014].
21 Beverage world, Healthy and Functional Beverage Report. [Online]. Avail-
ent strategies depending on consumers’ profile in order to fulfill able: http://www.beverageworld.com/articles/full/15976/healthy-
the acceptability of diverse green tea beverage consumers. functional-beverage-report [20 January 2014].
22 Wang LF, Kim DM and Lee CY, Effects of heat processing and storage on
flavanols and sensory qualities of green tea beverage. J Agric Food
Chem 48:4027–4032 (2000).
ACKNOWLEDGEMENT 23 Navindra PS, Michael A, Yanjun Z, Susanne MH, Lydia F, Mark D
This research was supported by the Basic Science Research Pro- and David H, Comparison of antioxidant potency of commonly
gram through the National Research Foundation of Korea (NRF) consumed polyphenol-rich beverages in the United States. J Agric
funded by the Ministry of Education, Science and Technology Food Chem 56:1395–1402 (2008).
24 Luckow T and Delahunty C, Which juice is ‘healthier’? A consumer study
(20110005883). of probiotic non-dairy juice drinks. Food Qual Prefer 15:751–759
(2004).
25 Stone H and Sidel JL, Sensory Evaluation Practices. Academic Press,
REFERENCES Orlando, FL (1985).
1 Vickers Z, Incorporating tasting into a conjoint analysis of taste, health 26 Munoz A and Civile GV, The spectrum descriptive analysis method.
claim, price and brand for purchasing strawberry yogurt. J Sens Stud ASTM Manual 3:22–34 (1992).
8:339–352 (1993). 27 Williams EJ, Experimental designs balanced for the estimation of resid-
2 Solheim R and Lawless H, Consumer purchase probability affected by ual effects of treatments. Aust J Sci Res 2:149–168 (1949).
attitude towards low-fat foods: liking, private body consciousness 28 Prescott J, Young O, Zhang S and Cummings T, Effects of added
and information on fat and price. Food Qual Prefer 7:135–141 (1996). ‘flavour principles’ on liking and familiarity of a sheepmeat product:
3 Westcombe A and Wardle J, Influence of relative fat content informa- a comparison of Singaporean and New Zealand consumers. Food
tion on responses to three foods. Appetite 28:49–62 (1997). Qual Prefer 15:187–194 (2004).
4 Guinard JX, Uotani B and Schlich P, Internal and external mapping 29 Choi BY, Kwon TY, Park JH and Kim JH, A study of consumption
of preferences for commercial lager beers: comparison of hedonic characteristics of middle-aged women and female college students.
ratings by consumers blind versus with knowledge of brand and J Kor Tea Soc 16:46–56 (2010).
price. Food Qual Prefer 12:241–255 (2001). 30 Lee SM, Lee HS, Kim KH and Kim KO, Sensory characteristics and
5 Tepper BJ and Trail AC, Taste or health: a study on consumer acceptance consumer acceptability of decaffeinated green teas. J Food Sci
1371

of corn chips. Food Qual Prefer 9:267–272 (1998). 74:S135–S141 (2009).

J Sci Food Agric 2016; 96: 1362–1372 © 2015 Society of Chemical Industry wileyonlinelibrary.com/jsfa
 www.soci.org Y-W Bae, SM Lee, K-O Kim

31 Ares G, Gimenez A and Gambaro A, Consumer perceived healthiness pay more for a fat spread with a proven health benefit. Food Qual
and willingness to try functional milk desserts: influence of ingredi- Prefer 14:65–74 (2003).
ent, ingredient name and health claim. Food Qual Prefer 20:50–56 36 Magnusson MK, Arvola A, Hursti U, Aberg L and Sjoden P, Attitudes
(2009). towards organic foods among Swedish consumers. Br Food J 103:
32 Jung J and Lee S-H, Cross-cultural comparisons of appearance self- 209–227 (2001).
schema, body image, self-esteem, and dieting behavior between 37 French SA, Story M, Hannan P, Breitlow KK, Jefferey RW, Baxter JS et al.,
Korean and U.S. women. Fam Consum Sci Res J 34:350–365 (2006). Cognitive and demographic correlates of low fat vending snack
33 Michaelidou N and Hassan LM, The role of health consciousness, food choices among adolescents and adults. J Am Diet Assoc 99:471–475
safety concern and ethical identity on attitudes and intentions (1999).
towards organic food. Int J Consum Stud 32:163–170 (2008). 38 French SA. Pricing effects on food choices. J Nutr 133:841S–843S
34 Guinard JX and Marty C, Acceptability of fat-modified foods to children, (2003).
adolescents and their parents: effect of sensory properties, nutri- 39 Neumark-Sztainer D, Story M, Perry C and Casey MA, Factors influ-
tional information and price. Food Qual Prefer 8:223–231 (1997). encing food choices of adolescents: findings from focus-group
35 Bower JA, Saadat MA and Whitten C, Effect of liking, information and discussions with adolescents. J Am Diet Assoc 99:929–937
consumer characteristics on purchase intention and willingness to (1999).
1372

wileyonlinelibrary.com/jsfa © 2015 Society of Chemical Industry J Sci Food Agric 2016; 96: 1362–1372