3.1. Data and Study Area
This study investigates consumer preferences for fragrant rice in China and Thailand, with a specific focus on the factors that influence purchasing decisions. This research was conducted in two phases, ensuring a rigorous data collection process and thorough analysis aligned with the study’s objectives. The geographical focus included Guangzhou, China, and Bangkok, Thailand, two prominent urban centers representing significant rice consumer markets in their respective regions.
Data Types and Data Sources
Phase 1: Pre-testing and Questionnaire Development
The initial phase involved collecting preliminary data through field trials and interviews with both consumers and distributors. The primary goal was to identify critical attributes influencing consumer decisions about rice purchases, such as price, rice quality, consumer knowledge, and perceptions of Chinese and Thai rice. These findings informed the creation of a structured questionnaire, which was collaboratively designed by researchers and professors at South China Agricultural University, China, and Kasetsart University, Thailand.
Pre-testing Strategy and Study Areas
Pre-testing was conducted with 30 participants from each study location: Kasetsart University in Bangkok, Thailand, and South China Agricultural University in Guangzhou, China. These locations were chosen for their strategic importance to the rice industry:
Bangkok, Thailand, is known for its rich culinary diversity; Bangkok is a global culinary hub where international and traditional Thai cuisines coexist. This makes it a prime setting for exploring consumer preferences, including both domestically produced and imported rice. The city’s cultural and economic influence provides a nuanced understanding of Thai consumer behavior.
Guangzhou, China: As a major economic center in southern China, Guangzhou is pivotal in the rice trade, particularly as a distribution hub for imported rice varieties, including Thai fragrant rice. This position makes it ideal for investigating cross-cultural perceptions and preferences, especially in relation to imported products.
Questionnaire Refinement
The pre-testing phase offered valuable feedback on several aspects of the questionnaire, including clarity of attribute descriptions, cultural relevance, and overall question design. Revisions were made to ensure that the questionnaire accurately captured the nuances of consumer preferences in both contexts. This iterative refinement process enhanced the methodological rigor and cultural adaptability of the study.
Phase 2: Main Data Collection
The second phase of this study focused on a comprehensive examination of consumer preferences through sensory evaluation trials and face-to-face interviews conducted in Guangzhou, China, and Bangkok, Thailand, while incorporating additional insights from Vietnam and Cambodia. This holistic approach aimed to capture a nuanced understanding of preferences for aromatic rice varieties across four culturally distinct countries. The study emphasized critical sensory attributes such as aroma, texture, flavor, and appearance, which are widely recognized as pivotal determinants of consumer satisfaction and willingness to pay [
23].
Sensory Evaluation and Questionnaire Design
The final questionnaire was meticulously crafted to encompass a broad spectrum of factors influencing consumer preferences. These factors included rice consumption behavior, sensory evaluations, quality attributes, and purchasing decisions, which were further contextualized by social, economic, and environmental considerations [
24].
Sensory evaluation trials involved blind testing of representative aromatic rice varieties sourced from China, Thailand, Vietnam, and Cambodia. This methodological approach ensured that assessments of sensory attributes remained unbiased and were conducted under standardized conditions [
25]. Blind testing was particularly instrumental in isolating the intrinsic sensory qualities of each rice variety, enabling robust comparisons across regions.
Integration of Secondary Data
To complement the primary sensory data, secondary data were gathered from government bodies, the academic literature, and industry reports. These additional resources offered crucial insights on rice standard and certification adopted in the four countries as well as marketing trend and the implications for aromatic rice.
The role of sensory attributes is vital in shaping consumer perceptions and market positioning, particularly for aromatic rice, where aroma, taste, and texture drive preferences and brand loyalty. By integrating sensory evaluation and market data, this study developed a comparative framework to assess the competitiveness of aromatic rice varieties regionally and globally, highlighting the importance of aligning product attributes with consumer expectations to enhance market success [
26,
27].
Data Collection Method
Primary data were collected through an experimental survey approach, targeting a total of 1330 respondents (665 from each country). The sample was selected using an accidental sampling method, and the sample size was determined based on the following formula [
28]:
with
The result was a sample of 665 respondents per country, ensuring a robust dataset for the analysis of consumer preferences for fragrant rice, including organic and conventional varieties.
Data Examination and Verification
Prior to executing the analysis, a stringent data examination protocol was instituted to ascertain the precision and comprehensiveness of the dataset. Absent values and anomalies were methodically detected and addressed utilizing descriptive statistical evaluation. The examination process encompassed validating response coherence and juxtaposing flagged data against the original records to mitigate potential biases. Further elucidations regarding the data examination methodologies are delineated in
Appendix A.
3.2. Theoretical Models
This section presents an advanced theoretical framework for examining how sensory perception, country of origin (COO), and ethnocentrism shape consumer behavior in the rice markets of China and Thailand. Drawing upon seminal models from consumer economics, we integrated the Random Utility Model (RUM), Lancaster’s Characteristics Theory, and Ethnocentrism Theory into our approach while accounting for heterogeneity in preferences using advanced econometric techniques, particularly the Mixed Logit Model.
The theoretical underpinnings of this study stem from the Random Utility Model (RUM), which posits that consumer decisions are driven by their attempt to maximize utility under constraints [
29]. In the context of rice purchasing behavior in China and Thailand, consumers derive utility not merely from the product as a whole but from the specific attributes associated with the rice, such as aroma, texture, and country of origin. These attributes are essential in shaping their purchasing decisions, particularly in culturally distinct markets.
In this study, utility,
for consumer
choosing rice option
in a given choice set can be decomposed as follows:
where
represents the observable utility based on the characteristics of the rice, and
is a stochastic error term reflecting unobserved factors. The systematic component
is a linear function of the rice’s attributes:
The Mixed Logit Model is employed to estimate this utility function, allowing for variation in consumer preferences by accommodating unobserved heterogeneity in coefficients. This framework effectively captures the interaction between sensory perceptions and ethnocentric tendencies with COO effects, which are critical in cross-cultural studies of rice markets.
Lancaster’s (1966) Characteristics Theory posits that utility is derived from a product’s attributes rather than the product itself [
30]. This theory aligns with sensory marketing concepts, emphasizing the role of sensory attributes—such as aroma, flavor, and texture—in driving consumer preferences for aromatic rice.
In this study, sensory attributes such as aroma and texture are modeled as key determinants of utility. For example, aromatic rice varieties like Jasmine and Basmati hold cultural significance in China and Thailand, where sensory experiences are integral to consumer satisfaction. Southern Chinese consumers, for instance, highly value aroma due to cultural preferences for fragrant rice varieties, while Thai consumers often prioritize texture, such as softness, for meal satisfaction [
31]. The sensory utility component is expressed as follows:
In this framework, it is hypothesized that Chinese consumers will assign a higher weight (β1) to aroma, whereas Thai consumers will place greater emphasis on texture (β2). These cultural differences provide nuanced insights into sensory-driven preferences in both countries.
The country-of-origin (COO) effect suggests that consumers often use the origin of a product as a proxy for quality and safety [
32]. This phenomenon is particularly strong in food markets where food safety concerns are prevalent. In China, recurring food safety scandals have heightened consumers’ reliance on domestic products, perceived as safer or more reliable compared to imports [
33]. The COO effect can be particularly pronounced among ethnocentric consumers who prioritize products that symbolize national pride or self-reliance [
34]. The consumer ethnocentrism construct, measured by the CETSCALE, interacts with the COO effect to further influence consumer preferences. Ethnocentric consumers prefer domestic products, even if imports are perceived to be of higher quality or offer better value. The interaction term
in the utility function captures the additional utility that ethnocentric consumers derive from consuming domestic rice:
It is hypothesized that Chinese consumers exhibit stronger ethnocentric tendencies, driven by sociopolitical and cultural factors, resulting in a marked preference for domestic rice. Thai consumers, while valuing domestic products, may demonstrate relatively lower ethnocentric bias and greater openness to imports.
The Mixed Logit Model was employed to account for heterogeneity in consumer preferences across individuals. Unlike the standard logit model, which assumes homogeneous preferences, the Mixed Logit Model allows coefficients to vary across individuals, reflecting diverse consumer attitudes towards sensory attributes and COO. This is particularly relevant in cross-cultural studies, where heterogeneity in taste and preferences can be substantial.
The Mixed Logit Model assumes that individual-specific preferences are distributed across the population, allowing for a richer interpretation of consumer behavior. The choice probability for consumer
i selecting rice product
j from a choice set
C is given by
This model captures the complex interactions between sensory perceptions, ethnocentrism, and COO, providing robust insights into the drivers of consumer preferences in the rice markets of China and Thailand.
3.3. Structural Equation Modeling (SEM)
Purpose and application: SEM was employed to evaluate the relationships between sensory perceptions, COO, and consumer behavior. The technique enables the simultaneous analysis of multiple variables and their interdependencies.
Software: the analysis was conducted using AMOS (version 23.0).
Model validation:
Measurement model: convergent validity was confirmed through AVE > 0.50, and the CR ranged from 0.88 to 0.93.
Structural model: goodness-of-fit indices (CFI, TLI, RMSEA) confirmed the model’s adequacy for both Chinese and Thai datasets.
Cross-validation was performed to confirm the stability of path coefficients, goodness-of-fit indices (CFI, TLI, RMSEA), and latent variable relationships across subsets, ensuring robustness and reliability.
3.4. Choice Set Design
The selection of fragrant rice as the principal product for this investigation is motivated by its prominence as a dietary fundamental in both China and Thailand, wherein contemporary consumer behavior indicates a notable transition towards premium rice offerings. Fragrant rice, distinguished for its exceptional sensory properties, has increasingly garnered preference due to its superior aroma, texture, and aesthetic appeal. These sensory dimensions have been recognized as fundamental factors affecting consumer willingness to pay (WTP) for premium rice strains [
35]. Within this framework, the present study seeks to examine consumer preferences regarding fragrant rice alongside their corresponding WTP, employing the Engel–Kollat–Blackwell (EKB) model to systematically delineate consumer decision-making processes across organized stages [
36].
The design of the choice experiment in this research emphasizes sensory perception, which is critical in influencing consumer purchasing choices. Recent scholarly discourse posits that sensory elements such as aroma, softness, and grain integrity have been widely acknowledged as fundamental indicators of rice quality, directly impacting consumer preferences and behaviors [
14,
37]. These sensory properties were utilized as essential variables in the formation of the choice sets, thereby enabling the study to ascertain how consumers negotiate trade-offs among various quality dimensions when selecting their preferred rice product.
In addition to sensory properties, this investigation also integrates variables such as price, brand identifiers, traceability information, and certification status. Price was conceptualized as a continuous variable, quantified in the domestic currencies of the two sampled nations, China and Thailand, and categorized into low, medium, and high brackets to assess consumer sensitivity to price variations [
13,
38,
39]. Notably, traceability—the capacity to monitor a product throughout its supply chain—was operationalized as a binary variable (Yes or No) to evaluate its influence on consumer trust and safety apprehensions [
40,
41,
42]. Certification labels issued by governmental authorities or independent organizations were incorporated to provide consumers with assurances regarding food safety and quality, further shaping their decision-making processes [
43,
44,
45].
This design not only facilitates a comprehensive analysis of how sensory attributes interact with other product characteristics but also underscores the intricacies of consumer decision making within cross-national contexts [
46]. Considering both sensory quality and supplementary factors such as price and traceability, the choice sets developed in this study offer an extensive understanding of the elements that impact consumer preferences for fragrant rice.
In recent years, the use of country-of-origin (COO) labeling has gained significant traction in the global food market, serving as a key determinant in consumer decision making. COO labels provide vital information about where a product was grown, processed, and packaged, allowing consumers to differentiate between domestic and imported products [
7,
47]. In the rice industry, COO labels play a critical role, as they not only indicate the geographic origin of the rice but also signal its quality and safety. Consumers often associate rice from certain countries with higher standards of production and superior sensory attributes, such as aroma and texture, which can influence their willingness to pay a premium for products from these regions [
7].
The significance of COO lies in its ability to act as an extrinsic quality cue. Studies have shown that COO information can enhance consumer trust, particularly in cases where the origin country has a positive reputation for food safety and high agricultural standards. This trust, in turn, increases the consumer’s willingness to pay (WTP) for rice from countries perceived as high-quality producers [
48,
49]. For example, consumers may prefer rice from countries like Thailand or Japan, which are known for their premium rice varieties, over rice from less well-known rice-producing regions.
3.5. Experiment Design
The experiment design for this study aimed to systematically evaluate consumer preferences for fragrant rice by incorporating key attributes and their respective levels. Initially, a full factorial design was considered, identifying five essential attributes influencing consumer choices: price, country of origin (COO), aroma, certification, and grain texture. Among these attributes, three had two levels, while two had three levels, resulting in a total of (3 × 3 × 2 × 2 × 2) = 72 possible combinations of choice sets.
Optimizing Choice Sets for Practicality
Due to the impracticality of presenting all 72 combinations to each respondent—given the potential for fatigue, reduced data quality, and higher costs—the study adopted a D-efficient design using STATA 17. This approach reduced the choice sets to 24 statistically optimized sets while maintaining rigor and reliability in the results [
40]. By employing this method, the design minimized cognitive load on respondents, ensuring ease of response while preserving the integrity of the data [
50,
51].
To streamline further, the twenty-four choice sets were divided into six versions of the questionnaire, each containing four choice sets. This subdivision mitigated the risks of fatigue and complexity for respondents while reducing survey administration costs. Each respondent was randomly assigned to one version, balancing the distribution of responses across all sets.
Choice Set Structure
Each choice set presented three options to respondents:
Option A: a combination of identified rice attributes at varying levels.
Option B: another combination of attributes differing from Option A.
Option C (Opt-out Option): this allowed respondents to abstain if neither Option A nor Option B met their preferences.
This structure reflects real-world decision-making scenarios, where consumers often have the flexibility to choose or abstain based on personal preferences.
Incorporating Sensory and Economic Attributes
Key sensory attributes, including sweetness, aromatic quality, softness, and grain integrity, were central to this study, alongside non-sensory attributes such as price, country of origin (COO), and certification status. The study examined aromatic rice varieties from China, Thailand, Vietnam, and Cambodia, each showcasing distinct sensory characteristics and market preferences (
Appendix C). Details of the character-istics and attribute levels used in the choice sets are presented in
Table 1 and
Table 2.
This carefully structured design allows for a robust analysis of consumer preferences, particularly the trade-offs they make between sensory and economic attributes. For example, the inclusion of price as a continuous variable enables an analysis of price sensitivity across different levels. Similarly, COO, certification, and sensory factors such as aroma and grain texture provide insights into how consumers value premium rice attributes.
By employing D-efficiency and optimizing the choice sets, this design ensures that respondents are neither overwhelmed nor fatigued, thus maintaining high-quality data collection that is representative of real-world purchasing behaviors.