International Journal of Gastronomy and Food Science 7 (2017) 32–40

Contents lists available at ScienceDirect

International Journal of Gastronomy and Food Science

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

When Chinese cuisine meets western wine MARK

Shu-Tai Wang
Department of Hospitality Management, Tunghai University, P.O. Box 891, Taichung 407, Taiwan

A R T I C L E I N F O A BS T RAC T

Keywords: This study explores the guiding principle for pairing common western table wines such as Chardonnay, Riesling,
Sensory evaluation Merlot and Cabernet Sauvignon with authentic Chinese cuisines. Sensory evaluation was carried out to measure
Chinese cuisine the affective level on the pairing of food and wine. A five by eight (4 different wines and no wine pairing with 8
Wine pairing map different cuisines) factorial experiment design was carried out to attain the sensory affection from the taste
MDS graph
panel. Hedonic rating was adopted to assess the affective response of the cuisine and wine pairings. The results
of affective test indicated that Riesling was the preferred wine to pair with most of the Chinese cuisines in
question. The interaction was significant between different cuisines and wines (p=0.000), indicating that the
hedonic sensory pattern of the cuisine can be influenced by the type of wine paired. In addition,
Multidimensional Scaling (MDS) graph was proven to be an effective tool for visualizing the guiding principle
of food and wine pairing.

Introduction
According to the statistics of per capita wine consumption, French
people consumed 43 liters of wine per capita; each American drank 11
liters of wine; whereas the Chinese only consumed 1.3 liters per capita
in 2013 (Wine Institute, 2015). Camillo (2012) stated that although
the demand of wine decreased in Europe, the demand continued to
grow in Asia. China, in particular, had the sales growth of wine
for more than 30% within the 10 years period since 2001. Due to this
high demand, the number of wine importers soared 73% in 2012
(Mercer, 2012). In addition, wine pairing with meal became trendy in
China and the major revenue was contributed from Chinese young
customers.
An experienced sommelier can easily recommend a good pairing
wine to western cuisine. However, there seems to be no certain
guidelines for pairing wines with Chinese cuisines. The connection
between the diverse flavors of Chinese cuisines and western wines is a
missing piece in the puzzle of the knowledge for the sommelier.
Exploration on sensory perception on the pairing between Chinese
cuisines and western wines will provide a whole new dining experience
and business opportunities. Dodd (1996) found that placing wine on
the food menu will have a significant impact on the sales of wine, which
greatly facilitated the revenue of a restaurant. The sales of wine in a
modern Chinese restaurant could easily account for 30–40% of the
revenue. It becomes increasingly important for the practitioners in the
hospitality industry to have the knowledge in order to lead their
customers for different dining pleasure. In order to extend the fun of
mix and match in dining, the art of pairing western wine with food will
have to be extended from the western cuisines to different types of
Chinese cuisines.
Wine was found to be strongly associated with food in three
dimensions: complementary, social meanings, and lubrication effect
(Pittigrew and Charters, 2006). Lubrication effect refers to the social
interaction enhancement in a gathering influenced by the relaxation
from the alcohol and the pleasant taste of wine. A good pairing is based
on the complementary role of wine on the three important factors of
food: components, texture, and flavors. The operational definition for
the role of wine was the complementary effect on the flavor of a cuisine
in this study. Though there were a few articles attempted to discuss the
concept of pairing some Chinese foods with wines, qualitative and
empirical studies are still in need to establish the appropriate matching
between western wine and regional Chinese cuisines. In order to
introduce Chinese cuisines pairing with wine in a more systematic
manner, a model integrated categorization, consumer affective test,
and summarized by the visual mapping was established to explore the
pairing rules in this study.
The objectives of this study were to:
1. Identify the best pairing between four popular wines with selected
authentic Chinese cuisines
2. Establish a graphical method to elucidate the relationship between
cuisine and wine

E-mail address: s2719w20@thu.edu.tw.

http://dx.doi.org/10.1016/j.ijgfs.2016.11.003
Received 7 March 2016; Accepted 16 November 2016
Available online 15 December 2016
1878-450X/ © 2016 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
S.-T. Wang
3. Explore the potential of pairing wine with Chinese cuisine for
younger diners
Chinese cuisine
With the growing popularity and consumption of Chinese cuisines,
various cuisines have been introduced to consumers across the world
both in casual fast food restaurants as well as in fine dining restaurants.
Regardless of the level of service, the regional cuisines can be defined
by their cooking styles, ingredients, and tastes (Newman, 2004). Rozin
(1982) categorized the authentic regional cuisines based on their
primary ingredients, cooking techniques, and unique flavoring princi-
ples. It is generally recognized that Chinese cuisine is a combination of
soy sauce, salt, sugar, chili sauce, garlic, green onion, and rice wine.
However, with the addition of local produce, the combination of
different ingredients and the cooking skills gave the regional cuisine
its unique taste. For example, if a fish steamed with ginger, green
onion, salt, rice wine and with local vinegar sauce, it will be a Zhejiang
(Eastern) style “West Lake Fish in Vinegar Gravy”. If this fish was deep
fried, then sautéed with vinegar, rock sugar and soy sauce, it became
the Cantonese style “Sweet and Sour Fish”.
There are eight popular categories of cuisines corresponding to
different regions in China; these eight regions from the west inland to
the northern region are: 1. Szechuan, 2. Hunan, 3. Fujian, 4.
Cantonese, 5. Zhejiang, 6. Jiangsu, 7. Anhui, and 8. Shandong
(Anderson, 1988; Newman, 2004) as shown in Fig. 1.
In general, despite the central region, Chinese regional cuisines
traditionally belong to four destinations. Inevitably, the characteristics
of the typical dishes of each region in Table 1 were distinctly different
due to the geological barrier. With the improvement of transportation,
the eight categories of cuisine merged into four general regions. The
northern region such as Beijing was known for the dishes made by
braising, baking, and stewing (Newman, 2004). People from the north
often have strong preference of pairing grain-made distilled liquor with
their food. The cuisines from the other three regions in Table 1 all
have potential to be paired with wine since the trend has been observed
in various restaurants. For instance, roast goose and salted goose
wing are the popular dishes in southern China. The marinated sauce for
the goose contains light-color soy sauce and rock sugar that can
eliminate the game odor and consequently promote the savory taste
of the goose.
8) Shandong cuisine:
Seafood with rich, umami flavor
Ex. Stewed dishes
1). Szechuan cuisine:
spicy, numbing hot China
sensation
Ex. Kongpao chicken
2) Hunan cuisine:
oily,piquant and intense
flavor
Ex. Smoked fish/ham
3)Fujian cuisine:
fresh ingredients with original
taste, seasoned with rock sugar
Ex. Emulsion fish ball
Fig. 1. Origins and the characteristics of the regional cuisines.
33
International Journal of Gastronomy and Food Science 7 (2017) 32–40
Eastern Chinese cuisines are often cooked with brown rock sugar
and a special black vinegar to form a sweet, decent sour and savory
taste (Newman, 2004; Kittler and Sucher, 2008). Chefs from Jiangsu or
Zhejiang use freshwater fishes or shrimps with light seasoning, or
seasoning with locally grown tea. It is characteristic to eastern cuisines
that the fresh and original taste of the food is highlighted rather than
being overwhelmed by strong flavorings.
Contrarily, a lot of spices were often applied in western cuisine to
add the strong flavor in their food and also to get rid of the so-called
“humidity” in their body. Cooking with red hot peppers or Szechuan
peppers is a popular way to season a wide spectrum of dishes. The well-
known dish in almost every Chinese restaurant, Kongpao chicken, is
from this region.
Western table wines
The U.S. has been the largest wine-consuming nation in the world
since 2010. California wine sales to both domestic and international
markets totaled 276 million cases in 2015. It reached a record high, up
2% since 2014, with an estimated retail value of $55.8 billion (Wine
Institute, 2016). According to the statistics of Wine Institute (2016),
the most popular table wines are in these eight categories: Cabernet
Sauvignon, Syrah, Zinfandel, Pinot Noir, Chardonnay, Sauvignon
Blanc, Pinot Gris, and Riesling.
Although the traits of wine might be slightly influenced by the
growing environment, Egli et al. (1998) indicated that the attributes of
wine were mainly determined by the grape variety, whereas the
fermentation method or inoculation starter culture would alter the
flavor yet not as much. German law requires that all Spätlese Riesling
wines to have a minimum Brix level of 20. Riesling keeps its acidity,
even when it’s very ripe and sugary. This makes Riesling not taste
overly cloying because it is balanced with a nice freshness. Clary et al.
(2006) indicated that Riesling processed intense attributes of citrus,
Sherry character, and length of aftertaste. Riesling is extremely
versatile, making wines that range from crisp and dry, to very sweet
dessert wines. In order to be distinguished with Chardonnay, the
Riesling selected for this study has good length of aftertaste with semi-
sweetness. Based on their sensory characteristics, Riesling was more
distinct for its acetic, caramelized, overall fruitiness, pineapple/tropical
flavor and stronger body compare to Chardonnay; Chardonnay was
characterized for the astringent/phenolic, floral, vegetative, pear flavor.
5). Zhejiang cuisine:
fresh water fish/shrimp
with light seasoning
Ex. Sweet &sour fish
4) Cantonese cuisine:
savory, sweet and intense
flavor
Ex. Roast goose, Salted goose
wing
S.-T. Wang International Journal of Gastronomy and Food Science 7 (2017) 32–40

Table 1
Typical regional food and the characteristics.

Food region Typical Cuisine Flavor Characteristics

Western Cuisine • Kongpao chicken Flavor:

• Poached fish with chili sauce – Pungent
• Hot and spicy pot – Spiciness
• Hot and spicy tofu – Numbing hot sensation
– Intense

Ex. Kongpao Chicken

Southern Cuisine • Roast suckling pig Flavor:
• Steamed crab – Sweet
• Sticky rice wrapped in lotus leaves – Fresh savory flavor
• Barbeque baby pork – Savory
• Dim sum

Ex. Dim Sum

Eastern Cuisine • Brine cooked goose Flavor:
• Steamed fish – Light flavor
• West Lake vinegary fish – Fresh and original taste
• Chicken with shredded green onion – Seafood
• Thousand-year-old egg – Umami
• Stinky tofu

Ex. Shrimp with Longjing Tea

Northern Cuisine • Peking roast duck Flavor:
• Sautéed clams – Rich body flavor
• Orange beef – Salty
• Stewed oxtail – Less greasy
• Shredded pig ears – Delicate snack
• Mu shu pork

Ex. Peking roast duck

The flavor of Chardonnay was described with the attributes of citrus,
buttery, toasted flavor, sourness, and bitterness by Cliff and Dever
(1996).
Merlot and Cabernet Sauvignon are generally known for their flavor
of berry, plum and tannin. Gürbüz et al. (2006) compared Merlot and
Cabernet Sauvignon from California and Australia with the analytical
instrument GC-MS. Both wines were characterized by high fruity,
caramel, green, and earthy aroma, yet there were 4–5 times of ethyl
octanoate in Merlot than Cabernet Sauvignon, which are accounted for
the smoother flavor in Merlot.
Food and wine pairing
Food and wine present a great combination at the dining table
throughout human history; people enjoyed food and wine together,
since wine can raise people’s spirits and assist in the digestion of food
(May and Sharpe, 1997). Beverages are often associated with different
types of meals in various situations, such as the common notion
that Sushi and Sake go well together, and the great complementary pair
for the Oyster and the Chablis. The decisions for wine selection can
also be influenced by consumption habits (Jaeger et al., 2010). Asian
consumers tend to like the wine with sweeter taste that contradicts
with the concept of western wine pairing. Westerners usually drink
drier wine to complement their food for not disturbing the salty taste of
the cuisine. Saura et al. (2008) stated that the menu and wine list
were equally important for revenue contribution and reputation
establishment of a restaurant. Manske and Cordua (2005) found that
the role of the sommelier has its strategic importance, as it may
lead to an increase of wine sales of 10–15% to 25%. Traditionally,
foods were served with local wines because of familiarity and the cost;
however, with the trend of globalization, there was a growing
interest for pairing food with a greater variety of wines (Wesson,
2007). Wine is commonly served in the designated wine glasses
accompanying the meal in Chinese restaurants. Though diners use
chopsticks for their food, there are no conflicts with the serving
pattern between food and wine. In the household, wine-drinking

34
S.-T. Wang
accompanying the food becomes more casual, similar to American
household habits but, not as elaborate as the fine dining settings.
Such a serving pattern seems to be in a good harmony across east and
west.
According to Jackson (2002), the main reason of drinking wine with
meals was due to the enjoyment created from its combination.
Different flavors in food may suppress or enhance certain attributes
in wine, which in turn will alter the perception of that wine, and vice
versa. Tasting food and wine together reveals an integrated perception
of flavor by the palate (Wesson, 2007). If the wine and food
components and texture elements have been properly matched, a
potentially enjoyable and synergistic pleasant sensation will be
achieved (Harrington, 2007). Harrington et al. (2010) highlighted
the value of wine layering flavors, adding contrasts, and to improve
the taste of food by the interaction of food and wine. Therefore, wine
could be more than just a drink in terms of altering the taste of the
food.
Central to most concepts of food and wine pairing is the harmonial
balance among different flavor intensities, besides habits and cultural
reflection (Jackson, 2009). The association between wine and food was
well established in ancient Greek and Roman time (Jackson, 2009).
The wine culture has long been cultivated in China as well. Chinese
started to drink wine made from grape since East Han Dynasty (25–
220 A.D.). Subsequently, the habit of drinking wine along with food
spread from the royal family to the local society in Tang Dynasty (618–
907 A.D.). General recommendations such as food and wine flavors
combine well when they were either complement or contrast to each
other (Harrington, 2007) could be linked with the theory that food and
wine pairing may relate to how wine can enhance a sense of elegance to
dining, even just a simple meal (Jackson, 2009). It has been reported
that the interactions between the intricate sensory and chemical
components within a food or wine were inherently complex (Bastian
et al., 2010). For example, the reaction between the acids and tannins
in wine and food protein will reduce the perception of the alcohol, thus
creating a smoothness mouth feel for the food (Jackson, 2009). The
acidity of white wine tends to refresh the mouth and reduce the
perception of oiliness from food, whereas the bitterness and astrin-
gency in red wine can enhance the flavor of most red meat (Jackson,
2002).
If food and wine pairing is looked into from a more scientific way,
wine can cleanse the palate or enhance saliva production by its phenols
and ethanol contents (Jackson, 2002). It was reported that alcoholic
beverage taken 30 min before a meal may have the stimulating effect on
food intake since the ad libitum intake was significantly higher than
others (Hetherington et al., 2001), implicating that alcohol may
increase subjective feelings of hunger (Caton et al., 2004). Other
research stated that the stimulation from alcohol on food intake was
short and alcohol consumption had no effect on appetite ratings;
however, it can affect the appetite system by modulating the orosensory
reward from food (Caton, Marks and Hetherington, 2005). Table wines
are usually served as the complementary role to the food. Wine is
appreciated as an alternative flavor component of a meal. In addition,
the social enhancement associated with alcohol consumption makes
wine a good complement for dining either with friends, families or after
work (Jackson, 2002; Bastian et al., 2010).
Since food and wine pairing becomes an apparently crucial
component of a meal, chefs, sommeliers, and restaurant practitioners
all want to have better comprehensions in order to understand and
meet the real needs of their clients (Jackson, 2009). Wansink et al.
(2006) found that food and wine pairing recommendation could
increase the sales of wine by 7.6 percents. Harrington (2005) used a
hierarchical approach to study the pairing between food and wine
based on the 3 key parameters: components, texture, and flavors and
discovered that flavor is usually the most important factor among these
three.
35
International Journal of Gastronomy and Food Science 7 (2017) 32–40
Table 2
Characteristics of the wines for pairing with the cuisines.
Wine Riesling Chardonnay Merlot (M) Cabernet
Sauvignon
(CS)
Maker Schmitt Palmira Gallo Family Canyon Road,
Year Söhne, 2007 2007 2008 2008
Origin Mosel- Chile California California
Germany
Flavor Fresh apples, Lemon, buttery, Blackberry; Black currant,
profile peach, pear grapefruit raspberry; black cherry,
olive blackberry,
Materials and methods
Samples
Two red wines, Cabernet Sauvignon (Canyon Road, 2008-
California) and Merlot (Gallo Family Vineyard, 2008 - California),
and two white wines, Chardonnay (Palmira, 2007- Chile) and Riesling
(Schmitt Söhne, 2007 Mosel- Germany) were used in the Affective test
for food and wine pairing. The characteristics and flavor profile of these
four wines are shown in Table 2.
The wines with single grape cultivar were selected for the research
to avoid the introduction of mixed characteristics from different grapes.
The producer of the wines was not designated to be the same for the
pairing since this approach could explore guiding principles applicable
for wines from different origins. The selected Riesling (residual sugar
66 g/L) was sweeter than the rest of the wines. Red wines were served
between 14 and 18 ℃, while the white wines were served between 7
and 12 ℃. The 20 cl ISO wine tasting glass was used to avoid any bias
caused by the shape of the wine glasses. Fifty milliliters of each wine
were served with 50 gm food sample in a small, white plate to the
panelists.
Eight different cuisines with distinct flavor characteristics were
selected to represent the regional food from China. Roast Goose and
Salted Goose Wing represent the Cantonese cuisine. Salty and Peppery
Fried Chicken is a popular delicacy in Taiwan and southern China.
Sweet and Sour Fish and Poached Shrimp are popular Zhejiang dishes.
Hot and Spicy Tofu and Kongpao Chicken are typical Szechuan dishes.
Pork Fried Rice is a typical staple food not belonging to any region, and
was selected as one of the food sample to test the reliability of the
cluster analysis. All the food samples were prepared by a local
restaurant specialized in various regional cuisines, and were cooked
and delivered in a styrofoam container 30 min before the test to keep
the temperature of the samples above 40 ℃.
Experimental design
Factorial treatment design was adopted in order to explore the
effect of the two main factors, namely wine and food, on the consumers’
affection on the pairing (Meilgaard et al., 2007). A matrix of the first
factor (4 different wines plus without wine) interacting with the second
factor (8 different cuisines) resulted in 40 distinct treatment combina-
tions that forms a single replication of a 5×8 factorial design in the
consumer affective test. Due to the physiological limitation of the
tasting number for flavor perception, the panel evaluated 2 cuisines
pairing with a series of wines in each session: 1. no wine; 2.
Chardonnay (C); 3. Riesling (R); 4. Merlot(M); and 5. Cabernet
Sauvignon (CS). Therefore, the complete study consisted of 5 sessions
of sensory tests held on Friday for 5 consecutive weeks; each and every
one of the 30 panelists tasted 10 pairs of food and wine within a test.
S.-T. Wang
Part 1 Descriptive analysis
Selection of 12 panelists
Threshold and focus group training
Rate the intensity of the attributes of the eight cuisines
Group the cuisines by cluster analysis
Part 2 Consumer affective test
Invitation for consumer test
Affective test: 10 pairs of foods and wines *5 weeks
ANOVA to explore the suitable wine pairing
MDS graph for the suitable wine for each cluster of cuisines
Fig. 2. The procedure of the sensory evaluation.
All 30 panelists tasted and evaluated a total of 40 different combina-
tions for the full factorial design and another 10 replicates to examine
the consistency of the panelists throughout the testing period (five
weeks). The procedure of the sensory evaluation tests in this study is
shown in Fig. 2.
Sensory evaluation
Sensory evaluation is a technique using human subjects as the
measuring instruments to gauge attribute intensity (Meilgaard et al.,
2007). Sidel and Stone (1979) and Moskowitz (1983) provided some
guidelines for in-house consumer test that executes the test on the site
of sensory test provider, ex. in a test lab. The advantage of an in-house
test is the preparation and presentation of the product can be fully
controlled. There are two parts of a sensory evaluation; descriptive
analysis as the qualitative analysis was applied first to categorize the
cuisines, followed by the consumer affective test to quantitatively
elucidate the acceptance level for each food and wine pairing (Fig. 1).
More than 50% of the young panelists were recruited for these two
stages because young people are more sensitive than their counterparts
due to the greater number of taste buds (Mojet et al., 2001; Nordin
et al., 2007).
36
International Journal of Gastronomy and Food Science 7 (2017) 32–40
Descriptive analysis
Twelve panelists were selected from a pool of candidates, the 120
students who took the sensory evaluation class, according to their
ability to discriminate the intensity of basic taste in a low threshold
for saltiness, sourness, sweetness. This group of 12 trained panelists
was applied in descriptive analysis to explore the attributes while
identifying the corresponding intensity of the attributes for each
food sample. Before the descriptive analysis test, three training
sessions were held for these selected panelists. Different concentrations
from the threshold to higher concentration of salt (0.08%, 0.16%,
0.24%, and 0.32%), sugar (0.32%, 0.64%, 0.96%, and 1.28%), and
vinegar (0.005%, 0.01%, 0.015%, 0.02%) solutions were provided for
the first two trainings on the basic taste. The panelists had to answer
the correct intensity of the taste on the 7-point Likert scale. The panel
passed the performance evaluation with correction rate more than
90%. These 12 panelists were further trained in a third session in a
round table setting with a selection of reference food samples repre-
senting certain taste attributes. The discussion was led by a moderator,
and the samples were evaluated one at a time, to develop the
descriptors that can precisely describe the flavor attributes with
consensus. These agreed descriptors were used to develop a ques-
tionnaire that was later applied for rating the intensity of each dish.
With the followed up cluster analysis, the results of descriptive analysis
categorized the cuisines into different groups based on the intensity of
the attributes.
Consumer affective test
The Consumer Affective Test is employed to access the personal
response (preference or acceptance) of current or potential customers
to a product, a product idea or specific product characteristics
(Meilgaard et al., 2007). In another word, the affective test was to
measure the level of “liking” from the consumers that reflected the
immediate experience or anticipation of pleasure from the orosensory
stimulation of eating a food often expressed by the hedonic value or
palatability (Mela, 2006).
A group of thirty consumers aged between 20 and 60 who had
drunk wine in the past month and were willing to try the pairing of
western wine and Chinese cuisine were invited from the campus and
nearby neighborhood. The residents from the nearby neighborhood
consist of engineers from a high tech science park, employees of a
national hospital, and members from the university, representing the
diversity of potential customers. The panelists were not remunerated
for their time but were given a bottle of wine at the end of the 5
sessions. They were invited to the sensory lab in the Department of
Hospitality Management at Tunghai University rather than the central
location. de Graaf et al., (2005) suggested that the field condition would
be more suitable for the consumer test of a meal. Therefore, the sensory
lab was decorated as the dining environment of the restaurant with
tablecloth, silver utensil, and uniformed waitresses. All the tests were
held at 3 pm when the panelists were not too hungry or too full to
influence the sensory outcomes.
The temperature of the lab was controlled at 24–26 ℃, without
odor, and brightly lighted. Upon arrival, an oral explanation was
carried out alongside a brief instruction sheet to the panelists prior
to the sensory test. Thirty panelists were asked to finish the test without
discussion and/or showing their facial expression. After the brief
introduction, each panelist was presented a set of sensory samples
including two dishes, a wine sheet with the three-digit coded glass
placement for the 4 glasses of different wines to avoid any bias. In order
to ensure the consistent perception of the combination of cuisine and
wine, the participants (panelists) were asked to take a sip of the wine
(5 s), taste the food (10 s), then take another sip of the wine (5 s).
Drinking water and having a small piece of toast in between samples
was recommended to cleanse the palate. The panelists were asked to
rate each cuisine alone as well as the cuisine and wine pairs based on
the hedonic level of the Likert scale from 1 to 7, 1=“Dislike extremely”,
S.-T. Wang International Journal of Gastronomy and Food Science 7 (2017) 32–40

Table 3 Table 4
Cluster analysis on the results from descriptive analysis on the attributes. Demographic profile and wine drinking habit of consumer panelists.

Cuisine Cluster Distance Demographic Information Frequency (%)

Roast Goose 1 2.201 Gender

Sweet & Sour Fish 2 2.422 Male 12(40%)
Poached Shrimp 2 2.422 Female 18(60%)
Salted Goose Wing 1 2.147
Pork Fried Rice 4 0.000 Age
Salty and Peppery Fried Chicken 1 2.993 20–30 17(56.7%)
Hot and Spicy Tofu 3 1.663 31–40 2(6.7%)
Kongpao Chicken 3 1.663 41–50 5(16.7%)
51–60 5(16.7%)
61–70 1(3.3%)

Frequency of drinking wine

About once per month 20(69%)
2=“Dislike”, 3=“Somewhat dislike”, 4=“Neutral”, 5=“Somewhat like”,
About once every two weeks 3(10.3%)
6=“Like”, and 7=“Like extremely”. The test measured the hedonic About once per week 4 (13.8%)
rating without any complicated evaluation terms that might introduce About twice per week 2 (6.9%)
the “Halo effect” of any bias (Meilgaard et al., 2007).
Habit of having wine with meal (now)
Yes 8(26.7%)
Statistic analysis No 22(73.3%)

Will try to pair wine with meal (future)

Cluster Analysis was well recognized as a systematic tool that could
Yes 24(80%)
categorize foods based on their characteristics (Godwin et al., 1978; No 6 (20%)
Jacobsen and Gunderson, 1986). The first part of sensory evaluation,
namely the results of the descriptive analysis for each cuisine, was
grouped by cluster analysis. The results from consumer affective test
were analyzed by one-way ANOVA to compare the hedonic level of each
pairing. The interaction between the foods and the wines was further
analyzed by two-way ANOVA analysis. Lastly, consumer sensory data Table 5
was also converted to multidimensional scaling with a configuration of The influence of pairing wine on the affective level of different cuisines.
points in a space where each point represents either the cuisine or the
wine. The distance between each two points indicated the appropriate- Food Food with wine pairing

ness of the match; the shorter the distance the better the pairing. The
Cluster No wine R C M CS F value
graphical relationship between cuisine and wine was well explained by
the perceptual map based on the ratings of the dishes and the “liking” 1 Roast Goose 6.25a 5.50a b
4.7b 4.7b 5.10a b 5.27***
level of each pairs. 1 Salted Goose 5.00a 5.39a b
4.12b 3.97b 4.16b 7.09***
Wing
1 Salty and 5.77a 5.65a 4.51b 3.94b 3.97b 13.37***
Results Peppery Fried
Chicken
2 Sweet & Sour 5.54a 5.47a 4.60a b
4.00b 3.83b 8.84***
Cuisine categorization from descriptive analysis
Fish
2 Poached 5.56a 6a 5.21a 3.97b 3.97b 17.09***
The dishes investigated were grouped by the trained sensory panel Shrimp
based on their respective intensity score of the attributes, including: 3 Hot and Spicy 4.94a 5.24a 4.27a b
3.41b 3.62b 10.80***
Tofu
saltiness, sweetness, sourness, spiciness, savory taste, umami taste
3 Kongpao 4.91a 4.78a 3.62b 3.57b 3.59b 7.30***
(one of the basic taste, which usually refers to the delicious flavor of Chicken
meat or seafood), and soy sauce taste. Though the sensory panel was 4 Pork Fried Rice 5.35a 5.35a 4.71a 3.61b 3.58b 16.76***
trained with the standard solutions of known saltiness, sweetness and
acidity, the perceived tastes were usually the consequence of the Note:
1. Symbols of food paring with wine (R: Riesling; C: Chardonnay; M:Merlot; CS:
comprehensive flavor interaction, which is why there is no instru-
Cabernet Sauvignon)
mental measurement that can completely replace human sensory taste. 2. within the same row, mean values with the same superscript alphabetic letter indicate
The results from descriptive analysis were applied for K-Mean cluster no significant difference from Scheffe post hoc comparison at the level of Alpha=0.05
analysis, as shown in Table 3. “Roast goose”, “Salted goose wing” and *p < 0.05
“Salty and peppery fried chicken” were grouped together in group 1, **p < 0.01
which reconfirmed the categorization as the Cantonese style cuisine. ***p < 0.001

“Sweet and sour fish” and “Poached shrimp” in group 2 were cuisines
from Jiangsu and Zhejiang region with light or sweet and sour
seasoning. “Hot and spicy tofu” and “Kongpao chicken” were clustered
in group 3, which represent Sichuan style cuisine with hot and piquant
taste.
The results from descriptive analysis and cluster analysis confirmed The Pork Fried Rice used in this study was seasoned with salt, black
the generally agreed categorization of the cuisine. Therefore, this and white pepper, and garnished with chopped green onion. The
approach is an effective way to facilitate positioning a new dish to a mellow taste of rice stir-fried with shredded pork and scrambled egg
certain category for further wine matching. The only odd one was the made it a popular dish in any Chinese restaurant. Similar to pasta with
Pork Fired Rice that was shown in a solo manner. Rice is the staple red sauce in Italy, it is served as a part of the cuisine to fulfill the
food in Chinese cuisine that functions like a supporting actor in a play. integrated sensation of a meal.

37
S.-T. Wang
Consumer demographic profile
The demographic profile of the consumer sensory panel is shown in
Table 4. The age distribution of the participants for the consumer test
appeared young due to sampling from the neighborhood near Tunghai
University. Such an age distribution corresponded well to the purpose
of this study as we intended to explore the acceptability of pairing
Chinese cuisine and western wine within young potential drinkers. In
addition, the younger panelists have sharper senses for discrimination.
Sixty nine percent of the panelists drink wine once per month. Most of
them (73%) did not have a habit of having wine with their meal, which
could be attributed to never having the chance for the first trial. This
argument is in line with the findings that 80% of the participants were
willing to pair wine with their meal after this series of sensory tests.
Consumer affective test
One-way ANOVA analysis was conducted to verify the preference
for each cuisine when pairing with different wines (Table 5). Except for
Roast Goose, dishes including Sweet and Sour Fish, Salty and Peppery
Fried Chicken, and Kong Pao Chicken pairing with Riesling received
higher hedonic ratings than those paired with the rest of the wines. It
suggested that if one were to choose a wine to pair with the
aforementioned dishes, Riesling was the most preferred, followed by
Chardonnay, Cabernet Sauvignon, and Merlot. There appeared to be an
obvious and significant trend of white wines being the preferred match
than red wines when pairing with these Chinese cuisines.
According to Table 5, the acceptance level of each cuisine pairing
different wines were significantly different (p < 0.001), indicating that
wine pairing changed the “liking” level of the cuisine. In each row,
cuisines pairing with Riesling received the higher acceptance levels
indicated by superscript “a”. Both red wines showed low ratings on
pairing with fried rice. One might suspect that the tendency of affection
to either food or wine would influence the preference level of the
pairing. Blake (2004) suggested the role of habituation in directing
food preferences. The acceptance of food and wine is a complicated
issue and the preference of the pairing might not follow the hedonic
trend of either the cuisine or the wine. A two way ANOVA analysis was
therefore conducted to further explore the possible interaction between
western wine and Chinese food on consumer hedonic rating. There
were significant interactions between wine and cuisine, suggesting that
Fig. 3. Multidimensional scaling graph of relationship of affection on pairing cuisines and Riesling wine.
38
International Journal of Gastronomy and Food Science 7 (2017) 32–40
the preference of food could be altered by the wine paired. Therefore, a
proper matching of wine will benefit the enjoyment of the cuisine.
The inter-relationship between cuisines and Riesling
Since Riesling was rated as the favorite match with most of the food
studied, its role of interacting with the cuisine was further explored by
visual mapping. The inter-relationships between cuisines and Riesling
were plotted by Multidimensional Scaling (MDS) in Fig. 3. The distance
between the cuisine and Riesling represents their respective hedonic
rating. The closer distance indicates the better the match. The best
match with Riesling was Poached Shrimp and Sweet and Sour Fish,
both belonged to group 2 (eastern region).
The results attained in the present study suggest that dishes from
the same geographical region are well categorized in the same group
from cluster analysis based on the sensory characteristics. Such a
finding is reconfirmed in the MDS graph as the dishes from the same
region showed up within the same circle. Plotting MDS graph is a
convenient way to identify dishes with the suitable pairing wine for
customers. For instance, the dishes with intense saltiness such as
“Salted Goose Wing” and “Salty and Peppery Fried Chicken” paired
with Riesling gained pronounced preference.
Discussion
After participating in this study as their initial experience of pairing
Chinese cuisine with western wine, eighty percent of the panelists
indicated that they would try to pair wine that complements their food
in the future. In another word, their experience from the sensory
evaluation, though unintended initially, served as an experiential
marketing similar to promoting the sales of wine with the meal in a
restaurant. Restaurant practitioners can definitely use the knowledge
gained from this study to employ table-side tasting to promote the
enjoyment and subsequently generating new revenues from wine
ordering.
It is noteworthy that consumers showed similar preference on wine
pairing pattern among the dishes from the same cluster. For example,
“Roast Goose”, “Salted Goose Wing” and “Salty and Peppery Fried
Chicken” all belonged to cluster 1 (representing Cantonese and south-
ern cuisine) matched better with Riesling, followed by Cabernet
Sauvignon. Roast Goose is a typical Cantonese cuisine with distinct
S.-T. Wang
savory flavor and fatty crisp layer of skin. From two replicates, not only
the panelists’ consistency on evaluation was confirmed, Riesling and
Cabernet Sauvignon were preferred over the other two wines for
pairing with “Roast Goose”. Cabernet Sauvignon was the second choice
for better pairing with Roast Goose, the good body of the wine structure
with tannin can complement “Roast Goose” and achieve a good
harmony. While the tannin and acidity of the wine might suppress
the perception of fat, the layer of fat from the roasted goose skin would
lessen the perception of astringent sensation from tannin. The high
protein content in Roast Goose is another important factor. Proteins in
food have already been mentioned as minimizing the sensory impact of
tannins and acids, making the wine taste smoother, less sour, and more
balanced (Jackson, 2002). The sweet note from the roasting sauce was
a hint for welcoming the sweeter wine, in agreement with Jackson’s
(2009) suggestion that the use of sweet sauce and some characteristics
of Chinese cuisines may herald a renewed appreciation of semi-sweet
wine with food.
The distance between Szechuan style (hot and spicy) food and
Riesling wine was farther than other dishes (Fig. 2). This observation
contradicted with the common knowledge that sugar added in the
spicy dishes tends to reduce the pungent mouth feel by increasing
the threshold of capsaicin (Sizer and Harris, 1985), since the
Hedonic rating of Riesling pairing with spicy food did not support
the common idea. Rather, our finding is more in agreement with the
conclusion from Koone et al., (2014) that the best match of Cabernet
Sauvignon is the spicy Italian salami. It is intriguing that both findings
are breaking the common perception of matching sweeter wine with
spicy food.
Great matching between wine and food could only be reached based
on either emphasizing the similarities or the complementary in the
sensory perceptions for both the wine and food (Wesson, 2007). Only
with systematic knowledge on the sensory traits one can better pair
wine and food so that they make each other tastier (May and Sharpe,
1997). While most Chinese prefer red wine based on its healthy image,
the results from our sensory tests indicated that Riesling was the most
suitable wine out of the 2 reds and 2 whites for Chinese cuisines in this
study. Nevertheless, the popular red wine Cabernet Sauvignon showed
the potential of matching the category of cuisine with fatty and smoke
flavored food such as Roast Goose.
The limited number of items in the study restricted the generation
of findings to the broader category of all Chinese cuisine. We did not
consider the option of increasing the number of items for testing,
because gustatory fatigue usually appears after tasting more than 6–8
samples. Therefore, the tests were designed, so that the panelists
evaluated only a maximum of 10 sets of paired samples at any one
setting.
Conclusion
The descriptive analysis from the trained panel led to the success
of clustering the cuisine into the distinct categories. Additionally,
the preference pattern among the different regional cuisine was
distinguishable among each other. The sensory affective test in
combination with the Multidimensional Scaling was proven as an
effective tool to explore consumer preferences for pairing Chinese
cuisine with western wines. The outcomes of this study can be applied
to Chinese restaurant practitioners to enhance the dining experience of
their guests by complementing their food with wine. Riesling is the
most preferred wine, since it paired well with most of the Chinese
cuisine employed in this study. The significant interaction between
cuisine and wine suggests that consumption of wine will influence the
food preference.
Most importantly, the results of continuous sensory test suggest
that experience is a crucial factor for accepting wine. A changing
attitude on food and wine pairing among the panelists toward the end
of the tests is noteworthy. After the young drinkers explored the beauty
39
International Journal of Gastronomy and Food Science 7 (2017) 32–40
of food and wine pairing, they had further temptation of trying similar
pairings in the future. This result suggests that use of small portions of
wine for consumer tasting at the table might be a useful strategy for
wine promotion in a restaurant. Providing free wine tasting is also a
good way of utilizing the wine from an opened bottle. Further
exploration on the pairing of wine with Chinese cuisine may involve
a comparison of the responses among panelists from different coun-
tries.
References
Anderson, E.N., 1988. The Food of China. Yale University Press, New Haven, London.
Bastian, S.E.P., Collins, C., Johnson, T.E., 2010. Understanding consumer preferences for
Shiraz wine and Cheddar cheese pairings. Food Qual. Prefer. 21 (7), 668–678.
Blake, A.A., 2004. Flavor perception and the learning of food preferences. In: Taylor, J.,
Roberts, D.D. (Eds.), Flavor Perception. Blackwell Publishing, Ltd, Oxford, UK, 172–202.
Camillo, A.A., 2012. A strategic investigation of the determinants of wine consumption in
China. Int. J. Wine Bus. Res. 24 (1), 68–92.
Caton, S.J., Ball, M., Ahern, A., Hetherington, M.M., 2004. Dose-dependent effects of alcohol
on appetite and food intake. Physiol. Behav. 81 (1), 51–58.
Caton, S.J., Marks, J.E., Hetherington, M.M., 2005. Pleasure and alcohol: manipulating
pleasantness and the acute effects of alcohol on food intake. Physiol. Behav. 84, 371–377.
Clary, C., Gamache, A., Cliff, M., Fellman, J., Edwards, C., 2006. Flavor and aroma attributes of
Riesling wines produced by freeze concentration and microwave vacuum dehydration. J.
Food Process. Preserv. 30 (4), 393–406.
Cliff, M.A., Dever, M.C., 1996. Sensory and compositional profiles of British Columbia
Chardonnay and Pinot noir wines. Food Res. Int. 29 (3), 317–323.
Dodd, T.H., 1996. Techniques to increase impulse wine purchases in a restaurant setting. J.
Restaur. Foodserv. Mark. 2 (1), 63–73.
Egli, C.M., Edinger, W.D., Mitrakul, C.M., Henick‐Kling, T., 1998. Dynamics of indigenous and
inoculated yeast populations and their effect on the sensory character of Riesling and
Chardonnay wines. J. Appl. Microbiol. 85 (5), 779–789.
de Graaf, C., Cardello, A.V., Kramer, F.M., Lesher, L.L., Meiselman, H.L., Schutz, H.G., 2005. A
comparison between liking ratings obtained under laboratory and field conditions: the role
of choice. Appetite 44 (1), 15–22.
Godwin, D.R., Barmann, R.E., Powers, J.J., 1978. Use of cluster analysis to evaluate sensory
objective relations of processed green beans. J. Food Sci. 43, 1229–1230.
Gürbüz, O., Rouseff, J.M., Rouseff, R.L., 2006. Comparison of aroma volatiles in commercial
Merlot and Cabernet Sauvignon wines using gas chromatography-olfactometry and gas
chromatography-mass spectrometry. J. Agric. Food Chem. 54 (11), 3990–3996.
Harrington, R.J., McCarthy, M., Gozzi, M., 2010. Perceived match of wine and cheese and the
impact of additional food elements: a preliminary study. J. Foodserv. Bus. Res. 13 (4),
311–330.
Harrington, R.J., 2005. The wine and food pairing process-using culinary and sensory
perspectives. J. Culin. Sci. Technol. 4 (1), 101–112.
Harrington, R.J., 2007. Food and Wine Pairing: a Sensory Experience. John Wiley & Sons,
Inc., NJ, USA.
Hetherington, M.M., Cameron, F., Wallis, D.J., Pirie, L.M., 2001. Stimulation of appetite by
alcohol. Physiol. Behav. 74, 283–289.
Koone, R., Harrington, R.J., Gozzi, M., McCarthy, M., 2014. The role of acidity, sweetness,
tannin and consumer knowledge on wine and food match perceptions. J. Wine Res. 25 (3),
158–174.
Jackson, R.S., 2002. Wine Tasting. Academic Press, MA, USA.
Jackson, R.S., 2009. Wine Tasting: A Professional Handbook 2nd ed. Academic Press, MA,
USA.
Jacobsen, T., Gunderson, R.W., 1986. Applied cluster analysis. In: Piggott, J.R. (Ed.),
Statistical Procedure in Food Research. Elsevier Science, Essex, UK, 361–408.
Jaeger, S.R., Danaher, P.J., Brodie, R.J., 2010. Consumption decisions made in restaurants:
the case of wine selection. Food Qual. Prefer. 21, 439–442.
Kittler, P.G., Sucher, K.P., 2008. Food and Culture 5th ed. Thomson & Wadsworth, CA, USA.
May, D., Sharpe, A., 1997. The Everything Wine Book. Adams Media, Avon, MA, USA.
Meilgaard, M.C., Civille, G.V., Carr, B.T., 2007. Sensory Evaluation Techniques. CRC Press
Taylor & Francis, Avon, MA, USA.
Mela, D.J., 2006. Eating for pleasure or just wanting to eat? Reconsidering sensory hedonic
responses as a driver of obesity. Appetite 47 (1), 10–17.
Mercer, C., 2012. Global: wine consumption to rise faster, but not inEurope. Just-Drinks. Aug
28, 2016 Retrieved from〈http://www.just-drinks.com/news/wine-consumption-to-rise-
faster-but-not-in-europe_id105982.aspx〉.
Mojet, J., Christ-Hazelhof, E., Heidema, J., 2001. Taste perception with age: generic or specific
losses in threshold sensitivity to the five basic tastes? Chem. Senses 26 (7), 845–860.
Moskowitz, H.R., 1983. Product Testing and Sensory Evaluation of Foods: marketing and R &
D Approaches. Food & Nutrition Press, West Port, CT, USA.
Newman, J.M., 2004. Food Culture in China. Greenwood Press, London, UK.
Nordin, S., Brämerson, A., Bringlöv, E., Kobal, G., Hummel, T., Bende, M., 2007. Substance
and tongue-region specific loss in basic taste-quality identification in elderly adults. Eur.
Arch. Oto-Rhino-Laryngol. 264 (3), 285–289.
Pittigrew, S., Charters, S., 2006. Consumers‫ ׳‬expectations of food and alcohol pairing. Br. Food
J. 108 (3), 169–180.
Rozin, E., 1982. The structure of cuisine. In: Barker, L.M. (Ed.), The Psychobiology of Human
Food Selection. AVI Publish Co., Westport, CT, USA, 189–203.
Saura, I.G., Molina, M.E.R., Contri, G.B., 2008. Qualitative and quantitative engineering
criteria of restaurant wine lists. J. Wine Res. 19, 19–31.
Sidel, J.L., Stone, H., 1979. Consumer Testing Considerations. In: Johnson, M.R. (Ed.),
Sensory Evaluation Methods for the Practicing Food Technologist. Institute of Food
Technologists, Chicago, IL, USA, 10–11.
S.-T. Wang
Sizer, F., Harris, N., 1985. The influence of common food additives and temperature on
threshold perception of capsaicin. Chem. Senses 10, 279–286.
Wansink, B., Cordua, G., Blair, E., Payne, C., Geiger, S., 2006. Wine Promotions in
Restaurants: Do Beverage Sales Contribute or Cannibalize? Cornell Hosp. Q. 47,
327–336.
Wesson, J., 2007. Wine & Food. Free Press. A Division of Simon & Schuster, Inc., New York.
40
International Journal of Gastronomy and Food Science 7 (2017) 32–40
Wine Institute, 2015. Per Capita Wine Consumption by Country. Sept 9, 2015 Retrieved from:
〈http://www.wineinstitute.org/files/Per%20Capita%20Wine%20Consumption%20_(c)
TradeDataAndAnalysis.pdf〉.
Wine Institute, 2016. 2015 California Wine Sales in U.S. Hit $31.9 Billion Retail Value. July
23, 2016 Retrieved from: 〈http://www.wineinstitute.org/resources/pressroom/
07082016〉.