CONSUMER PREFERENCES FOR INDONESIAN FOOD

ARTICLE INFO Introduction/Main Objectives: Food industries have been growing fast in Indonesia for recent decade, so it is necessary for food companies to understand Indonesian consumer attitudes and examine how Indonesian consumption behavior may change in consuming food. Background Problems: The development of society, with the concept of modernization at this time, encourages the interest people have for consuming food from other countries, so that Indonesian food is increasingly being displaced in its own country. Novelty: The results of this study provide a method for evaluating the combination of different attributes for food combinations, which can be used as a reference for selling food. Research Methods: This research used a conjoint analysis to explore consumers’ preferences for different cuisines, especially Indonesian cuisine. Finding/Results: Compared with previous studies, the origin of food is an important food attribute, whilst “western food” is the preferred type of food. Conclusion: Western food being the most preferred type of food, followed by Indonesian food, a crispy and salty taste, fresh food is preferred, as is food at a cheap price. Price was the most important attribute. Article history: Received 17 December 2019 Received in revised form 25 February 2020 Accepted 28 February 2020


INTRODUCTION
Indonesia has many traditional types of food which makes it a preferred food tourism destination. The role of food in Indonesian culture is an expressive activity that reaffirms people's social relations with life, and with trust, the economy, technology and its various differences. The food arts in Indonesia have several differences (Siti Marti'ah, 2013).
Diversity of food represents the food habits of people from different countries. Verbeke, W., & López, G. P. (2005). It found that European consumer's trade-off the relative expensiveness and time-consuming preparation of traditional food for the specific taste, quality, appearance, nutritional value, healthiness and safety they find in traditional food preferences. In this study, five attributes: the foods 'origin (which represents the diversity of food from different countries); its taste; its freshness (which represents the food's appearance and healthiness); its price; its presentation (which represents its preparation and service), were set for the consumers to state their preferences at different levels. Information about how consumers make these decisions can be useful to those making production and marketing decisions, including providing an indication of the value of different attributes and how to focus advertising for the greatest effect (Lin & Kuo, 2016).
Food has different attributes, such as its style of cooking, cooking course, i.e., fish course, nutrition, ingredients and flavors. The diversity of food has a strong effect on our social and personal life (Rozin et al., 2003). Each cuisine has its own particular style for preparing food, related to the geographic location. It plays a very important role in culture, which reflects its unique history, lifestyle, values, and beliefs, as well as people tend to identify themselves with their food (Kamal, Jabreel, & Rashwan, n.d.) A food's attributes can be defined as product features that are different from competing products (Lewis & Churchil 1983). The definition of these attributes can be applied to food attributes, because food attributes have features that distinguish between one type of food and another. Ma, Chow, Cheung, & Lee (Ma, Chow, Cheung, & Lee, 2018) proposes that certain special attributes are found in food, such as taste, health, social status, and cost.
The development of society, with the concept of modernization at this time, encourages the interest people have for consuming food from other countries, so that Indonesian food is increasingly being displaced in its own country. Because of concerns about the loss of food identity for Indonesian food, it is necessary to examine what the consumers' preferences for Indonesian cuisine are.
The theoretical framework of this research is rooted in the Lancastrian approach to the consumer theory. Breaking away from the traditional view that utility is derived from a product, Lancaster proposed that a product per se does not give utility to the consumer. Rather, a product possesses characteristics, and these characteristics give rise to utility. Furthermore, Lancaster generalized that products can possess multiple characteristics which can be shared by multiple products, and that products in aggregate can possess different characteristics from those pertaining to the products separately (Lancaster, 1966) Following Lancaster, a consumer with preferences for each of the aforementioned characteristics will choose the bundle of attributes of the product that maximizes his/her utility subject to budget constraints, in which case a consumer has to select a product from a set of options.
According to Frank (2011), preference is the process of ranking of all things that can be consumed, with the aim of obtaining a preference for a product or service. Consumer preference arises during the alternative evaluation stage of the purchasing decision process, wherein the consumer is faced with a variety of product choices, as well as services, with a variety of different attributes. Therefore, it can be concluded that preference is a choice taken and chosen by consumers, from a variety of available choices.
The preference stage that is owned by consumers towards a product is the beginning of consumer loyalty to the product. So companies must learn how to create a sense of consumer's preference for their products (Kotler, 2007).
In addition, we are concerned with the preference for food due to its origin, taste, price, and the way it is presented and served. How consumers 'preferences or interests in Indonesian food can be identified as well the market share, and the market's segmentation for it? We attempt to classify the food preferences from different countries with the flavors of their food. Thus, this study aims to: 1) Identify consumers 'preferences for Indonesian food. 2) Estimate the market share for Indonesian food. 3) Determine the most preferred type of food. 4) Determine the market's segmentation for Indonesian food.
The main contributions of this paper are that, through the results of this study, it is expected that consumers 'preferences or interests in Indonesian cuisine can be identified, as well the market share, and the market's segmentation for it, which can provide inputs for food business people in their products 'development, then decide the optimal way of presenting and serving Indonesia's cuisine in order to be able to compete with the cuisines from other countries.
The main contributions of this paper are that, through the results of this study, it is expected that consumers' preferences or interests in Indonesian cuisine can be identified, as well the market share, and the market's segmentation for it, which can provide inputs for food business people in their products 'development, then decide the optimal way of presenting and serving Indonesia's cuisine in order to be able to compete with the cuisines from other countries.

LITERATURE REVIEW
The demand for western-style convenience food is growing around the world, a likely result of the modernization in food consumption patterns. Proper targeting of consumers who exhibit preferences for western food will be essential for companies wishing to successfully enter the local market. A population base in excess of one billion combined with rapid and sustained economic growth has made Indonesia as an obvious target for western companies in search of new customers.
The entrance of Indonesia into the World Trade Organization (WTO) is another compelling argument for western companies to develop business strategies tailored to Indonesian people and markets. To be successful in Indonesia, it is necessary for western companies to understand Indonesian consumer attitudes and examine how Indonesian consumption behavior may change as Indonesia integrates into the global economy and faces increased exposure to industrialized countries, cultures, and product (Radhiah, Ab, Zakiah, & Nazirah, 2015).
Additionally, western food and culture are fashionable. The increased demand for westernstyle convenience foods has been associated with a higher frequency of dining out, increased patronage of grocery stores (compared to traditional wet markets), and the increased consumption of snack foods (Jussaume Jr, R. A., 2001).
Empathic and social concerns influence consumers' attitude toward, and preference (Jussaume Jr, R. A., 2001, Roininen et al., 2006. Local patriotism influences the preference for local food, even if such consumers evaluate it as being of lower quality and less desirable than other food products (Onozaka et al., 2010). Local food is an important part of the local culture and is held in high regard by local communities. Onozaka et al., (2010) define "local" as located within a county, whereas Darby et al. (2008) use state boundaries. While place-based definitions are frequently used, other criteria are also appliedfor example, the product's type (i.e. where local food is thought to be fresh produce), production techniques (expected to be traditional), farm size (allegedly small and family owned), and recipes (specific to the area).
The consumption of local food is one of the fastest-growing trends, especially in developed countries (Aprile et al., 2016;Bianchi and Mortimer, 2015;Penney and Prior, 2014). Governments also show an interest in assisting and promoting local food. Young, Choe, & Sam suggesting that the local food trend will continue to expand in the future (Young, Choe, & Sam, 2019). Local food, which is also called "the origin of the food," attract consumers because of the transparency of the food chain and their growing awareness of environmental and healthrelated concerns (Kühne, Vanhonacker, Gellynck, & Verbeke, 2010) showed that the country of origin is the most important cue attribute, followed by price (Moon & Han, 2018). A large body of research using a lottery as the focal subject has found that many people stated that they preferred one product (e.g., A) over another (e.g., B), yet made lower bid price for the preferred product (e.g., bid higher for B than for A) (Tversky& Thaler,1990).
Two general perspectives within the literature of economic sociology are useful for understanding why consumers may choose to buy local food: (1) embeddedness and (2) marketers/ instrumentalism. Embeddedness is a key concept in the theoretical construction of Alternative Food and Agriculture Networks (AFANs) within economic sociology (Hinrichs, 2003;Maye and Kirwan, 2010).
Folgado & Maria (Folgado-fern & María, 2019)found that intrinsic factors or the practical "self-gratifying" benefits of local food (e.g. taste, freshness, appearance, availability, and healthiness) are important drivers influencing purchase In this study, taste, freshness, and food that appeared to be fresh, were used to measure the attributes of food preference.

METHOD, DATA, AND ANALYSIS
This section introduces the research process and method. First, the design attributes and attribute levels of food were determined, followed by a combination of the attributes' levels. Next, the conjoint analysis was used to evaluate and measure customers' preferences for food design attributes in order to find the best design schemes for different groups of customers.

Selection of the relevant attributes
The first step in conducting a conjoint analysis is categorizing the attributes, which indicates the prominent factors that affect customers' choices, namely the attributes, and determines the levels of each prominent factor, namely the attributes' levels. A lot of design elements determine the foods' originality, taste, its price, and how it is presented and served. We have gathered the key attributes of food design, based on a broad literature review, prior research, and discussions. We eventually selected the following aspects as the combination of food attributes to be evaluated in our conjoint analysis: the foods' origin (Indonesian, Western, Chinese, Japanese), taste (sweet, crisp, spicy/hot), price (cheap, average cost expensive), and how it is served (with wait-staff or self-service). The price was added into the attribute list for analysis, because it is also a factor affecting consumers' choice decisions (Moon et al., 2017). Table 1 summarizes all of the attributes and their respective levels.
The design of the combination of food attributes involves four main parts: the foods 'origin, its taste, its price and how it is presented and served. Thus, each categorized attribute in this study involved that combination. Price is the main factor influencing consumers 'purchase decisions (Moon et al., 2017).

Research design
After determining the attributes and attribute levels for food, the products' portfolios should be determined. According to the number of attributes and respective attribute levels, there were 144 (4x3x2x3x2) possible product portfolios in total, which far exceeds the rational judgment scope of those being tested. In order to guarantee the feasibility and reliability of this experiment, this paper adopts an orthogonal design to simplify the product portfolios. As one of the most common and effective experiment design methods, orthogonal design can give consideration to the distribution of each attribute and the attribute level of the products (Zikmund,W. G.,et.all.,201). (The number of attribute level portfolios was reduced to 16 by using an orthogonal design in SPSS.
The test diagram for the16 food attribute combinations obtained from the ortogonal design is shown in Figure.

Structure of offline questionnaire
The structure of the offline questionnaire is shown in Figure 3. In the first part of the offline questionnaire, participants were required to provide their personal information, including age, gender, nationality, and monthly income. Then, 16 sample images (as shown in Figure 2) are displayed, and the participants were required to express their purchase intention through scoring, by using a five point Likert scale to rank the importance of their food attributes combination.
By conducting a conjoint analysis on the participants' responses to the simulated products, the participants' preferences for, and possibilities of purchasing the simulated products, could be investigated by scoring, ranking and other methods.

Data Analysis
The conjoint function in the SPSS software was used for data analysis, and the conjoint analysis operation was conducted by programming. First, the relative importance scores of the four attributes were calculated, which quantified the relative importance of each attribute toward the product' selection and provided information about the importance of one attribute in the attribute' selection relative to all the other attributes. Second, the food preferences of tourists of different ages, gender, nationality, and monthly income, were classified and compared.

Importance analysis of attributes
The relative importance of the food design attributes was reported first (Figure2). The participants listed the "foods 'origin" as the most important factor in their purchase decision with a mean value (µ ) of2.36, followed by "taste" (µ = 2.28), "price" (µ = 1.52), whereas "freshness" (µ = 1.22) was the least important factor influencing the consumers' purchase decisions. Table 3 shows the utility analysis results of all the attributes' levels. If consumers prefer one level over another, this level will have a greater utility value. In terms of the "foods' origin," "western food" was the most loved by consumers -in other words "western food" had the highest utility value. In terms of the "taste," consumers loved "crispy & salty" the most, followed by "spicy" and "sweet."For "freshness," the "fresh food" option was the most favored by consumers. In terms of the way of "serving," consumers prefer "fresh food" more than "tinned food",. With respect to "price," people often assume that cheaper items will sell better; in this study the consumers also prefer the cheaper food.

Utility analysis of attribute levels
From the results of the conjoint analysis presented in Table 1, the most preferred combination of food items was the sixth combination (western food, salty & crispy, fresh, cheap, selfservice), which was chosen by 26.92% of the consumers. None of the consumers placed the sixth combination of food items as their least preferred combination. The characteristics of the consumers were: 70% in the age range from 25 to40 years old, male (64%), European (100%), with a monthly income ranging from US$ 1000 to US$ 5000 (50%).
The combination of food attributes that consumers most disliked (chosen by 13.46% of them) was the combination of Chinese food, spicy, tinned food, expensive, self-service. None of the consumers placed it as their most preferred combination. The characteristics of these consumers who don't like the food combination of Chinese food, spicy, tinned food, and expensive, were: in the age range from 25 to40 years old, women (71%) European, in the middle income range (between US$ 1000 and US$ 5000 per month) (52%).
The most preferred Indonesian food was the 14 th combination (salty & crispy, fresh, expensive, served by wait-staff), which was rated as the best combination by 9.62% of the consumers. None of the consumers rated that combination as their least preferred combination. The characteristics of the consumers were: in the age range from 25 to40 years old (80%), women (60%), European (80%) and African (20%), in the middle income range (US$ 1000 to US$ 5000).
The combination of Indonesian food that consumers mostly disliked was the 15 th combination (sweet, not fresh, cheap, self-service), which was rated as the least favored by 7.69% of consumers. None of the consumers rated it as their most preferred. The characteristics of these consumers were: in the age range from 25 to40 years old, had a balanced proportion for gender, European (75%) and African (25%), 50% of them in the high income range (US$ 5000 and up).
Thus, the market share for Indonesian food, as viewed by the consumers who placed it as their most preferred was 34%, 17 respondents from 50 total respondents mostly like Indonesian food.
The highest market share of the different cuisines was "western food" (36.54%), as chosen by the consumers who ranked it their most preferred. The market share for Japanese food was 19.23%, followed by Chinese food which was 9.6%.Thus Indonesian food occupied the second rank below western food.

Preference analysis by age
The result of the one-way ANOVA's analysis showed that the relative importance of the attributes of food design was not significantly different for respondents of different ages. The one-way variance analysis showed that, with regard to the "foods' origin" ( F = 0.53, þ = 0.59), taste ( F= 2.33, þ = 0.10), freshness ( F =0.24, þ = 0.78), price ( F = 0.95, þ = 0.39), method of service (F = 0.59, þ = 0.55), there is no significant difference among the three groups (under 25 years old, in the range from25 to 40 years old, or above 40 years old). But there was a significant difference in the relative important for the "foods 'origin" attribute between consumers who are below 25 years old and consumer over 40 years old (þ = 0.039); the mean difference was significant at the 0.05 level.
By the result shown in Table 2, Indonesian food was the most loved by consumers between 25 and 40 years old, more so than by those in the other two age groups,who preferred western food (38% of the total consumers), followed by Indonesian food (34% of the total consumers). They state "crispy & salty" food has the best taste (32%), followed by "spicy/hot" (22%) and "sweet", which was loved by consumers under 25 years old (8%).
Most consumers chose the cheapest food (62%), they were between 25 years old and 40 years old (34%). They also liked wait-staff to serve their food (30%), but for the other consumers outside that age group it did not matter if they served themselves or were waited on as this had the same score for their preference.

Preference Analysis By Gender
The result of the one-way ANOVA's analysis showed that the relative importance of the attributes of food design was not significantly different for participants of different gender. The one-way variance analysis showed that with regard to "foods' origin" (F = 0.000, þ = 0.984), taste ( F= 0.239 , þ=0.627), freshness ( F = 0.619, þ = 0.435 ), price (F = 0.892 , þ = 0.350), serving method (F =1.280, þ =0.264 ), there is no significant difference among the; two groups (male, and female)The most favorite cuisine was western food, which was chosen by males(63%), followed by Indonesian food, which also was chosen by males(58%).
Both women and men liked the "crispy & salty" taste the most, followed by the "spicy/ hot" taste. They prefer "fresh" food rather than "tinned unfresh" food. As predicted, price was an important attribute for them, they prefer cheaper food. Woman preferred "wait-staff service" but men did not care.
Europeans mostly loved "western food," Americans mostly loved Japanese food, Africans mostly loved Indonesian food, while Turkish and Asian participants mostly loved Chinese food.

Preference Analysis By Income
The result of the one-way ANOVA's analysis showed that the relative importance of the attributes of food design was significantly different for participants of different income groups for the "taste" (F = 4.361, þ = 0.018 ), and "price" (F =3.667 þ = 0.033) attributes.
From the result of a post hoc one-way ANOVA analysis, consumers whose income level was below US$ 1000 had different preferences to consumers whose income level was within the US$ 1000 to US$ 5000 range (µ difference = -0.47863, þ = 0.017), and with consumers whose income was over US$ 5000 per month (µ difference = -0.72222, þ = 0.019), at the 0.05 significance level.
From the results shown in Table 2, consumers who preferred Indonesian food (58.82%) were in the middle income range (US$ 1000 to US$ 5000).
The attribute of "taste" was scored differently by people of different ages. There was a difference between "sweet" and "crispy & salty" (þ = 0.045), and between "sweet"and "spicy/ hot" (þ = 0.017). Consumers under 25 years old preferred "sweet," whereas consumers between 25 and 40 years old preferred "spicy/hot" and "crispy & salty," but they mostly liked "crispy & salty." There was a difference in preference for "taste" between consumers under 25 years old and consumer's between25 and 40 years old.
As well as consumers in the different income groups having different preferences for the "taste" attribute between "crispy & salty" and "spicy/hot," and between "spicy/hot" and "sweet."Consumers in the low income bracket liked "sweet" food. Whereas consumers in the middle income level liked "crispy & salty" and "spicy/hot" food, but the "spicy/hot" taste scored slightly higher than "crispy & salty." There was no significant difference among the four groups of consumers' characteristics for the "freshness" of food. For the "price" there was a significant difference among the "age" (F=3.466, þ= 0.039) and "monthly income" groups (F = 4.89,þ= 0.012).The consumers had no different preferences about how the food was presented and served.

Relevant Criteria for the Evaluation of Food Design
Compared with previous studies, the origin of food is an important food attribute, whilst "western food" is the preferred type of food. The study found that the best combination of attributes for a cuisine were "salty & crispy, fresh, expensive, and served by wait-staff," which was the preferred combination of consumers aged between 25 years old and 40 years old, who are males, originally from Europe, whose monthly income is in the range from US$ 1000 to US$ 5000 (middle income level).
The most preferred Indonesian food is the combination of "salty & crispy, fresh, expensive, and served by wait-staff," which was the preferred combination of consumers aged between 25 years old and 40 years old, who are female, originally from Europe, whose monthly income is in the medium income level (US$ 1000 to US$ 5000). Most consumers prefer "fresh food" over "tinned food". With respect to the "price," consumers prefer the cheapest cuisine.
The market share for Indonesian food, as seen by consumers who placed it as their most preferred, is 34%, this places it second, behind western food.

Recommendations for Food Design
Based on the results and discussion of the conjoint analysis, we recommend some guidelines for food designs. These guidelines can be divided into general guidelines and consumers' characteristics classification guidelines. General guidelines refer to the preference tendency of consumers for certain attributes in food design, such as western food being the most preferred type of food, followed by Indonesian food, a crispy and salty taste, fresh food is preferred, as is food at a cheap price. Price was the most important attribute. Chefs or people in the restaurant trade can take note of and use these four preferred attributes.
Therefore, the specific guidelines should refer to the consumers' characteristics. According to the preferences of consumers (tourist) of different "nationalities" for the choice of their "foods' origin," and their "income level" for the choice of the "foods' origin," and "price," there was a significant difference in the relative importance of the "foods' origin" attribute. Europeans mostly preferred "western food," Americans mostly preferred Japanese food, Africans mostly preferred Indonesian food, and Turkish and Asian people mostly loved Chinese food. Indonesian food was mostly loved by Europeans, although most of them preferred western food over Indonesian food.
The relative importance of the attributes of food design was significantly different for participants of different "income levels" for the "taste" and "price" attributes. Consumers who preferred Indonesian food were in the middle income bracket (US$ 1000 to US$ 5000). Consumers with a lower income liked "sweet" food. Consumers in the middle income level liked the "crispy & salty" and "spicy/hot" taste, but the "spicy/hot" taste scored slightly higher than "crispy & salty". Consumers in the upper income bracket preferred the "spicy/hot" taste.
There was slightly difference in preferences for "price" between consumers in the middle income level and those in the higher level.
Thus, businessmen in the restaurant/food retail sector for Indonesian food should be segmenting their target customers as follows: Consumers who comes from Europe and African, in the middle income level and up. The food should be crispy and salty, spicy might be permitted. Serving the food with wait-staff, providing fresh food, and selling it at a lower price, are better choices although there was no difference in the preferences for these. The customers should be aged between 25 and 40 years old, since this is the age group that preferred Indonesian food, although there was no real preference seen.

Limitations and further research
The results provide a method for evaluating the combination of different attributes for food combinations, which can be used as a reference for selling food. However, there are still some limitations that should be considered by future studies.
Future research should improve the number of attributes and their levels which were limited in this study. Attribute levels of the food can be further subdivided in a more comprehensive way to explore the influence of more features on the design's evaluation.
In this paper, we selected five important attributes. Another limitation of the current research is the chosen subject. In this paper, we discussed food as a general product, but not its uniqueness, which is created by its origin.