Smart Mobility in a Secondary City: Insights from Food Delivery App Adoption Among Thai University Students

Abstract

Food delivery apps (FDAs) have emerged as transformative tools in the digital age, reshaping consumer behavior and urban mobility through their convenience and accessibility. This study explores the factors influencing the adoption of FDAs among university students in a secondary city in Thailand, framed within the broader context of smart mobility. This study employs an extended Unified Theory of Acceptance and Use of Technology (UTAUT) framework, incorporating key constructs including performance expectancy, effort expectancy, social influence, facilitating conditions, and environmental concerns. Data were collected from 396 students at Mahasarakham University through a structured questionnaire and analyzed using structural equation modeling. The results reveal that effort expectancy, social influence, and environmental concerns significantly impact behavioral intention, while behavioral intention and facilitating conditions drive actual usage behavior. Environmental concerns emerged as a critical determinant, reflecting the growing alignment between consumer preferences and sustainability goals. The findings underscore the role of FDAs as key enablers of smart mobility, optimizing urban logistics, reducing transportation inefficiencies, and supporting sustainable city systems. By integrating environmental concerns into the UTAUT model, this study contributes to understanding technology adoption dynamics in secondary cities. Practical implications include promoting eco-friendly practices, enhancing digital infrastructure, and leveraging FDAs to foster sustainable and inclusive mobility ecosystems.

1. Introduction

Food delivery apps (FDAs) have emerged as a critical component of this digital transformation, connecting consumers with restaurants and delivery services through convenient and user-friendly platforms. This transition was further expedited by the global COVID-19 pandemic, which mandated social distancing measures and heightened dependence on contactless services [1]. Globally, the food delivery app market has shown remarkable growth, with revenue projected to reach USD 323.30 billion in 2023 and an annual growth rate of 9.89%. In Southeast Asia, including Thailand, this trend is also evident, with the market volume expected to reach USD 1.01 billion in 2023 [2].

In Thailand, the food delivery market has seen significant growth, with major players such as GrabFood, Food Panda, and Line Man dominating the landscape. A report by Kasikorn Research Center [3] indicates that the food delivery business in Thailand was valued at approximately THB 78–80 billion in 2022, showing a slight contraction of 0.8% to 6.5% from the previous year but still significantly higher than pre-pandemic levels. This trend underscores the lasting impact of the pandemic on consumer behavior and the increasing integration of food delivery services into daily life. The adoption of FDAs is particularly notable among younger demographics, especially university students. This group, often referred to as digital natives, is typically more receptive to new technologies and plays a crucial role in shaping technological trends [4]. Understanding the factors that influence their acceptance and use of food delivery apps is not only crucial for businesses operating in this space but also provides valuable insights into broader technology adoption patterns among young adults.

FDAs are increasingly recognized as integral components of smart mobility systems, reshaping how urban transportation operates and enhancing the efficiency of city logistics [5,6,7]. By connecting consumers with food vendors through digital platforms, these apps reduce the need for individual trips to restaurants, thereby decreasing traffic congestion and minimizing the environmental impacts associated with urban travel. FDAs optimize resource allocation by leveraging real-time data to streamline delivery routes, improve fuel efficiency, and reduce idle time for delivery drivers [8,9]. This integration into urban transportation systems not only supports more sustainable consumption patterns but also contributes to the broader goals of smart mobility by aligning technology with urban efficiency and sustainability.

In secondary cities, where infrastructure and mobility systems often lag behind metropolitan counterparts [10], FDAs present unique opportunities to address local transportation challenges. These apps can bridge gaps in public transit availability and enhance access to essential services, particularly in areas with limited connectivity [11]. Their role in urban planning extends beyond convenience, offering insights into consumer behavior and mobility patterns that can inform data-driven decisions for sustainable urban development [12]. By embedding FDAs within the framework of smart mobility, this study highlights their potential to transform not only consumer habits but also the overall landscape of transportation and urban living in secondary cities.

The Unified Theory of Acceptance and Use of Technology (UTAUT), proposed by Venkatesh et al. [13], provides a robust framework for understanding technology adoption across diverse domains. Initially developed to unify multiple acceptance models, the UTAUT has been widely applied to technologies such as mobile banking [14], e-government services [15], and autonomous vehicles (AVs) [16,17,18,19], demonstrating its versatility in explaining user behavior. For instance, in the adoption of autonomous vehicles, [16] found that performance expectancy and social influence significantly shaped consumer intentions, reflecting the model’s relevance to mobility-related innovations. Similarly, in mobile banking, Alalwan et al. [14] highlighted effort expectancy as a key driver of adoption, underscoring the UTAUT’s applicability to consumer-oriented technologies. Building on this foundation, our study applies the UTAUT to food delivery applications (FDAs), where performance expectancy pertains to perceived gains in the convenience and efficiency of food ordering. However, the unique context of FDAs—coupled with growing environmental awareness among young consumers—requires extending the model. Environmental concerns, increasingly central to decision-making among younger generations [20], are particularly relevant to FDAs due to their reliance on packaging and transportation. For example, Arunan and Crawford [21] reported that packaging-related emissions for a single FDA order range from 0.15 to 0.29 kg of CO₂e, with over 50% attributed to raw material production. Thus, this study extends the UTAUT by incorporating environmental concerns, tailoring it to the specific dynamics of FDA adoption in a secondary city.

The research focuses on students at Mahasarakham University, located in Maha Sarakham Province, Thailand. This choice of location is significant as it represents a secondary city, offering insights into technology adoption patterns outside of major metropolitan areas. Mahasarakham University, with a student population of 40,932 as of 2023 [22], serves as an ideal setting for this study. The university’s diverse student body provides a representative sample of young adults from various backgrounds, allowing for a comprehensive analysis of food delivery app adoption in a university context. The food delivery landscape in Maha Sarakham, while not as developed as in major cities like Bangkok, has seen significant growth. Local platforms like MSU LIKE SHOP (Facebook group) have emerged alongside national players, catering specifically to the university community. This mix of local and national services creates a unique ecosystem for studying food delivery app adoption [23].

The UTAUT model has been extensively applied and validated in diverse technological contexts. For example, Alshehri et al. [16] utilized the UTAUT framework to investigate the adoption of e-government services in Saudi Arabia. Similarly, Alrawashdeh et al. [24] applied the UTAUT to web-based training systems, confirming the model’s applicability in educational technology contexts. In the specific context of food delivery apps, several studies have utilized the UTAUT or its extensions. For example, Zhao and Bacao [25] explored the factors influencing customer retention during the COVID-19 pandemic, emphasizing the critical roles of perceived usefulness and social influence. Similarly, Alalwan [26] utilized an extended UTAUT model to investigate the adoption of mobile food ordering apps in Jordan, identifying performance expectancy and hedonic motivation as significant predictors of behavioral intention. However, limited research has focused on food delivery app adoption in secondary cities, particularly among university students in Southeast Asia. Additionally, the integration of environmental concerns into the UTAUT model within this context remains underexplored. This study seeks to fill these gaps by offering a comprehensive analysis of food delivery app adoption among university students in a secondary city in Thailand. The specific objective of this study is to identify the key factors influencing the acceptance of food delivery applications among students at Mahasarakham University.

2. Literature Review

2.1. Environmental Concerns in Technology Adoption

The UTAUT model offers a comprehensive lens for analyzing technology adoption, with applications spanning diverse fields such as autonomous vehicles (AVs), smart home technologies, and food delivery apps. For instance, Wang et al. [16] found that environmental concerns indirectly shaped AV adoption intentions via attitudes and norms, while Farzin et al. [17] identified trust and trip context as key factors in mandatory and optional AV use. Jing et al.’s systematic review [18] highlighted performance expectancy and safety perceptions in AV acceptance, and Yuen et al. [19] noted facilitating conditions and sustainability benefits in shared AV adoption. Similarly, Ferreira et al. [27] integrated environmental awareness into the UTAUT2 to study smart home technologies, emphasizing its role in shaping consumer preferences. These studies illustrate the UTAUT’s flexibility in accommodating contextual factors like sustainability. In our study, we narrow this focus to food delivery apps (FDAs), where environmental concerns are increasingly salient, particularly among younger consumers [28,29]. Kanchanapibul et al. [20] found that ecological affect and knowledge significantly drive green involvement and purchase behaviors among young adults, suggesting that environmental awareness may influence FDA adoption. This is especially relevant given the industry’s environmental footprint—Heard et al. [30] noted that meal kits, akin to FDAs in packaging and delivery, exhibit lower emissions than grocery meals, yet packaging remains a key impact factor. Thus, extending the UTAUT to include environmental concerns enhances its applicability to FDAs in our secondary city context [31,32].

Several studies have explored the role of environmental concerns in technology adoption across various domains. Wang et al. [16] extended the Theory of Planned Behavior (TPB) to study the adoption of hybrid electric vehicles (HEVs) in China. They found that environmental concern indirectly affected adoption intention through its influence on attitudes, subjective norms, perceived behavioral control, and personal moral norms. In the context of smart home technologies, Ferreira et al. [27] incorporated environmental awareness into an extended UTAUT2 model. Their findings highlight the importance of considering environmental approaches when studying the adoption of such technologies, which could be applicable to food delivery apps given their potential environmental implications. Lampo et al. [33] challenged the assumption that environmental concern is a key determinant of behavioral intention in the adoption of battery electric vehicles (BEVs). Their study suggests that while environmental concern may be relevant, new variables such as “technology show-off” might better explain technology acceptance in some contexts. This highlights the need to consider a range of factors, beyond just environmental concerns, when studying technology adoption. In the realm of sustainable banking, Amrutha and Santhi [34] proposed an innovative UTAUT model incorporating environmental concerns along with other variables such as risk perception and trust perception. Their findings underscore the importance of understanding customer perceptions and identifying barriers to adoption in promoting sustainable practices. Wang et al. [35] examined ride-sharing services, which share some similarities with food delivery in terms of their technology-driven, on-demand nature. They found that environmental awareness positively influenced consumers’ intention to use ride-sharing services, along with personal innovativeness and perceived usefulness.

These studies collectively suggest that environmental concerns can play a significant role in technology adoption, but their influence may be complex and context dependent. In the case of food delivery apps, environmental concerns might relate to issues such as packaging waste, carbon emissions from delivery vehicles, or food waste. However, the exact nature of this influence, whether direct or indirect, and its interaction with other factors in the UTAUT model, remains an area requiring further investigation.

2.2. Food Delivery App Adoption: Empirical Findings

The adoption and continued use of food delivery apps (FDAs) have been the subject of numerous empirical studies, revealing a complex interplay of factors that influence user behavior. These studies have employed various theoretical frameworks and methodologies to understand the dynamics of FDA adoption in different contexts. Yeo et al. [36] explored consumer attitudes toward online food delivery (OFD) services in their study, employing the Contingency Framework and Extended Model of IT Continuance. The findings revealed significant relationships between various factors and behavioral intentions. Specifically, convenience motivation, post-usage usefulness, hedonic motivation, price-saving orientation, and time-saving orientation were identified as positive influences on consumer attitudes and behavioral intentions toward OFD services. Notably, prior online purchase experience did not significantly impact post-usage usefulness, indicating that the distinct features of OFD services may necessitate separate consideration from general online purchasing behaviors.

Zhao and Bacao [25] examined FDA adoption in China using an integrated model that combined the UTAUT, Expectation–Confirmation Model (ECM), and Task–Technology Fit (TTF) framework, incorporating trust as an additional factor. Their findings identified satisfaction as the most influential determinant of continuance intention. Furthermore, the perceived Task–Technology Fit, trust, performance expectancy, social influence, and confirmation were shown to have direct or indirect positive effects on users’ intentions to continue using FDAs during the pandemic. This study emphasized the critical role of both technological and psychological factors in understanding FDA adoption, particularly under extraordinary circumstances. In a similar vein, Muangmee et al. [37] investigated FDA adoption in Bangkok, Thailand, during the COVID-19 pandemic. These studies highlight the multifaceted nature of FDA adoption, integrating both technological and safety factors within the context of a global health crisis.

Recent studies have further elucidated specific factors driving FDA adoption across different regions. In India, convenience and aggressive discounts were identified as primary motivators for FDA adoption [38]. The same study also highlighted the importance of app service quality and fulfillment in influencing user decisions. In Malaysia, user interface design emerged as a significant factor affecting the adoption and usage of popular FDAs [39], underscoring the importance of user experience in technology acceptance. Pandey et al. [40] found that perceived usefulness, various food choices, and overall usefulness directly impacted perceived value and satisfaction, which in turn influenced continuous-use intention. This finding aligns with the work of Elgammal et al. [41], who identified trust, convenience, and food variety as significant factors in the continuous usage of FDAs. The quality of information provided through these apps also plays a crucial role in their adoption. Talwar et al. [42] demonstrated that both user-generated and firm-generated information, along with system quality, significantly affected perceived usefulness and ease of use. These factors, in turn, influenced users’ attitudes toward mobile app usage.

Collectively, these empirical findings paint a comprehensive picture of FDA adoption, highlighting the interplay between technological factors (such as app quality and user interface), psychological factors (including trust and satisfaction), and contextual factors (like the COVID-19 pandemic). The studies also reveal the importance of considering regional differences and specific user needs in understanding and promoting FDA adoption.

2.3. University Students and Food Delivery Apps

The adoption and usage of food delivery apps (FDAs) among university students have become significant areas of research, given the growing popularity of these services within this demographic. Several studies have explored the factors influencing students’ use of FDAs and the potential impacts on their eating habits and health. Okumus et al. [43] examined the psychological factors influencing restaurant customers’ intention to use smartphone diet apps. Their study, based on the UTAUT, found that expected performance, anticipated effort of usage, social influence, and user innovativeness were significant predictors of app usage intentions. These findings suggest that similar factors may influence university students’ adoption of FDAs.

In Indonesia, Prabowo and Nugroho [44] investigated the factors affecting attitudes and behavioral intentions toward online food delivery services, specifically focusing on the Go-Food app. Their research revealed that the perception of usefulness, influenced by external factors such as hedonic motivations and time-saving orientation, was a key determinant of attitudes and behavioral intentions. This highlights the importance of convenience and time-saving aspects for students using FDAs. Similarly, Lim et al. [45] explored the factors affecting Malaysian college and university students’ satisfaction with FDAs. Their study found that app content and user interface design were the most significant predictors of student satisfaction, followed by perceived e-service quality, payment methods, and perceived ease of use. This research emphasizes the importance of user experience in FDA adoption among students.

Buettner et al. [46] conducted an in-depth study on food delivery app usage among young adults in the United States, including college students. Their findings revealed that participants used FDAs approximately twice a week. Key factors associated with more frequent FDA use included higher perceived subjective social status, food insecurity, financial responsibility, and being a full-time student. This study offers valuable insights into the demographic and socio-economic characteristics of young adults who frequently utilize FDAs.

The potential health implications of FDA use among students have also been a focus of research. Stefani and Layalia [47] investigated the relationship between FDA use and the risk of obesity among students in the Jabodetabek area of Indonesia. While their study did not find a significant relationship between the frequency of FDA use and obesity risk, they noted that certain types of foods frequently purchased through these apps (such as fried chicken, baso aci, ice cream, and coffee) were associated with an increased risk of obesity. Osaili et al. [48] investigated FDA usage among Jordanian consumers, including students, and found that fast food was the most commonly ordered option (87.1%), with lunchtime being the preferred time for placing orders (67.3%). The study identified key determinants of food choice, including price, appearance, delivery time, nutritional information, and the availability of healthy options. Similarly, Homyamyen et al. [49] explored factors influencing college students’ selection of food delivery service providers. Their findings highlighted service quality as the most critical factor, followed by price, indicating that while students prioritize high-quality services, they remain cost sensitive. Additionally, menu diversity emerged as a significant factor, reflecting students’ preferences for a broad range of culinary options.

These studies collectively paint a complex picture of FDA use among university students. While these apps offer convenience and variety, they may also contribute to less healthy eating habits. The research highlights the need for FDA providers to consider factors such as app design, service quality, pricing, and menu diversity to appeal to student users. Additionally, there is a growing need to address the potential health implications of frequent FDA use among this demographic, possibly by increasing the availability and promotion of healthier food options on these platforms.

2.4. Secondary Cities and Technology Adoption

While much research on food delivery app adoption has focused on major metropolitan areas, fewer studies have examined this phenomenon in secondary cities. Secondary cities are often characterized by smaller populations and less developed infrastructure compared to primary cities. These cities offer unique spatial and economic efficiencies that yield different dividends to their communities compared to primary urban centers [50,51,52]. As such, understanding the dynamics of technology adoption in these settings is of particular importance. The adoption of new technologies in secondary cities is influenced by a complex interplay of factors. External influences such as government policies and business strategies play a significant role, as do internal factors and the market-driven participation of customers [53]. This multifaceted nature of technology adoption reflects the unique position of secondary cities in responding to globalization and their varying capacities for planning within the global context [54].

Contrary to common perceptions, secondary cities are not limited to technological invention alone. They can specialize in various innovation activities beyond patents, challenging the notion that urban innovation is dominated by large cities [55]. This diversity in innovation potential highlights the unique role secondary cities can play in technological advancement and adoption. However, secondary cities, particularly in developing countries, face distinct challenges. While experiencing rapid growth, many lack the institutional capacity to manage urbanization effectively. This leads to unique obstacles related to infrastructure development, economic growth, and sustainability [56,57]. These challenges can significantly impact the pace and nature of technology adoption in these cities.

While the existing literature provides valuable insights into food delivery app adoption, several gaps remain. There is limited research on food delivery app adoption in secondary cities, particularly in Southeast Asia, where the dynamics of technology adoption may differ significantly from major metropolitan areas. Few studies have incorporated environmental concerns into models of food delivery app adoption, despite a growing awareness of the industry’s environmental impact, particularly among younger consumers. The interplay between technological factors (as captured by the UTAUT) and environmental considerations in shaping app adoption intentions has received limited exploration, leaving a significant gap in our understanding of how these factors interact. Additionally, there is insufficient research on the unique factors influencing food delivery app adoption among university students in non-metropolitan settings, a demographic that represents a significant market for these services. This study aims to address these gaps by applying an extended UTAUT model, incorporating environmental concerns, to examine food delivery app adoption among university students in a secondary city in Thailand. By doing so, it contributes to a more comprehensive understanding of technology adoption in this specific context, while also advancing the broader literature on technology acceptance and environmental consciousness in consumer behavior. This approach allows for a nuanced examination of how traditional technology acceptance factors interact with emerging environmental considerations in shaping the adoption of digital services in evolving urban environments.

3. Research Framework and Hypotheses Development

This study adopts the UTAUT as its theoretical foundation. The UTAUT combines elements from eight influential technology acceptance theories, providing a robust framework for examining the user adoption of new technologies. The model identifies four primary constructs—performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating conditions (FC)—that influence behavioral intention (BI) and use behavior (UB). To address the increasing significance of sustainability in consumer behavior, this study extends the original UTAUT model by incorporating environmental concerns as an additional construct [58]. Our research framework is illustrated in Figure 1.

3.1. Performance Expectancy (PE)

Performance expectancy refers to the extent to which an individual believes that using a specific system will improve their performance or productivity [13]. In the context of food delivery applications, it encompasses perceived advantages such as time savings, convenience, and access to a wide range of food options. Prior research has consistently demonstrated that performance expectancy significantly impacts behavioral intention to adopt mobile applications [14,59]. Based on these findings, we propose the following hypothesis:

H1: 

PE positively influences BI to use food delivery applications.

3.2. Effort Expectancy (EE)

Effort expectancy is defined as the extent to which an individual perceives a system to be easy to use [13]. In the context of food delivery applications, it pertains to the ease of navigating the app, placing orders, and completing transactions. Numerous studies have highlighted effort expectancy as a significant determinant of behavioral intention across various technology adoption scenarios [60,61]. Based on this, we hypothesize the following:

H2: 

EE positively influences BI to use food delivery applications.

3.3. Social Influence (SI)

Social influence refers to the degree to which an individual believes that significant others think they should adopt a new system [13]. In the context of food delivery applications, this may include peer pressure, recommendations from friends and family, or the perceived social status associated with using these apps. Prior research has consistently shown that social influence significantly affects behavioral intention in technology adoption [62,63]. Based on this, we propose the following:

H3: 

SI positively influences BI to use food delivery applications.

3.4. Facilitating Conditions (FC)

Facilitating conditions are defined as the extent to which an individual perceives that adequate organizational and technical infrastructure is available to support system use [13]. In the context of food delivery applications, this encompasses factors such as smartphone compatibility, reliable internet connectivity, and accessible customer support. While the original UTAUT model suggested that facilitating conditions primarily influence use behavior, later studies have indicated that they can also affect behavioral intention [63,64]. Based on these insights, we propose the following two hypotheses:

H4: 

FC positively influences BI to use food delivery applications.

H5: 

FC positively influences UB regarding food delivery applications.

3.5. Environmental Concerns (EC)

As an extension of the UTAUT model, we include environmental concerns to capture the growing awareness of sustainability issues among consumers. Environmental concerns refer to an individual’s consciousness about environmental problems and their willingness to contribute to their solution [58]. In the context of food delivery services, this could relate to concerns about packaging waste, carbon emissions from delivery vehicles, or food waste. Recent studies have shown that environmental concerns can significantly influence consumer behavior in various contexts, including food consumption [65,66]. Therefore, we hypothesize the following:

H6: 

EC positively influences BI to use food delivery applications.

3.6. Behavioral Intention (BI) and Use Behavior (UB)

Behavioral intention, a key construct in the UTAUT model, serves as a critical predictor of actual usage behavior. It reflects an individual’s intention to engage in a specific action, such as using food delivery applications. Extensive research has consistently demonstrated a strong correlation between behavioral intention and actual use behavior across diverse technology adoption contexts [64,67]. Accordingly, we propose the following:

H7: 

BI positively influences UB regarding food delivery applications.

4. Methodology

4.1. Research Design and Data Collection

This study employed a quantitative research design to examine the factors influencing food delivery application adoption among university students. The population for this study consisted of 40,932 undergraduate students at Mahasarakham University, as recorded by the university’s Registration Office in 2023. Stratified random sampling was employed to ensure representativeness across faculties and gender. Using Taro Yamane’s formula [68], with a 95% confidence level and a 5% margin of error, the required sample size was calculated as 396 students.

The sample was proportionally allocated across the university’s 19 faculties based on their respective student populations, ensuring a balanced representation of male and female students. The final sample included students who both used and did not use food delivery apps, capturing diverse perspectives. Data were collected through structured questionnaires designed to align with the study’s objectives, ensuring reliability and validity for the subsequent analysis. The questionnaire comprised sections on demographic characteristics, usage patterns of food delivery applications, and constructs from the UTAUT, including performance expectancy, effort expectancy, social influence, facilitating conditions, and an additional construct for environmental concerns. Data collection was conducted between 12 February and 30 June 2024.

4.2. Measurement and Instrument Validation

Each UTAUT construct and environmental concern was measured using a 7-point Likert scale, ranging from “strongly disagree” to “strongly agree” (See

Table 1. Measurement items.

Variable	Observed Variables	Questions	Ref.
Performance Expectancy (PE)	PE1	- Do you think food delivery applications are essential for ordering food during busy study or work periods?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	PE2	- Do you think using food delivery applications helps you manage your time better?
	PE3	- Do you think using food delivery applications saves you from waiting in long lines at restaurants?
	PE4	- Do you believe food delivery applications improve the convenience of accessing food?
Effort Expectancy (EE)	EE1	- Do you find learning how to use food delivery applications difficult?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	EE2	- Are the steps for ordering food through these apps straightforward and easy to follow?
	EE3	- Do you find it easy to use the features of food delivery applications?
Social Influence (SI)	SI1	- Do your friends, family, and close acquaintances frequently recommend using food delivery applications?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	SI2	- Do the usage habits of your friends, family, and acquaintances influence your decision to use food delivery applications?
	SI3	- Is there a trend among Mahasarakham University students to use food delivery applications?
Facilitating Conditions (FC)	FC1	- How well does your smartphone support food delivery applications?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	FC2	- Do the apps present menus and food options in a clear and accessible way, making it easy for you to search and select?
	FC3	- Do you think food delivery applications are compatible with other technologies you use?
Environmental Concerns (EC)	EC1	- Does using food delivery services generate more packaging waste?	Arunan, I., and Crawford [21]; Heard et al. [30]; own development
	EC2	- Does food delivery via current vehicles contribute to increased pollution?
	EC3	- Does the increasing number of food delivery drivers cause traffic congestion?
Behavioral Intention (BI)	BI1	- Are you willing to learn and use food delivery applications in your daily life?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	BI2	- When you are hungry, do you prefer food delivery applications as your first choice for ordering food?
	BO3	- Will you recommend food delivery applications to others?
User Behavior (UB)	UB1	- Do you frequently use food delivery applications during study sessions, after exercise, or in urgent and busy situations?	Venkatesh et al. [13]; Zhao and Bacao [25]; Alalwan [26]
	UB2	- Has your use of food delivery applications increased since entering university or after participating in university activities?
	UB3	- Do you use food delivery applications when you have limited time or no time to prepare meals?

). The questionnaire was pre-tested with 30 students to ensure reliability and validity.

4.3. Data Analysis

Structural equation modeling (SEM) was employed to test the hypothesized relationships between the constructs. Descriptive statistics were used to summarize the demographic data and food delivery app usage patterns. Model fit indices, such as the Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA), were evaluated to ensure the adequacy of the SEM model.

5. Results

5.1. Demographic and Behavioral Characteristics of Respondents

The descriptive statistics reveal a comprehensive profile of the sample comprising 396 undergraduate students at Mahasarakham University (See

Table 2. Demographic and behavioral characteristics of respondents (N = 396).

Category	Subcategory/Details	Frequency (N)	Percentage (%)
Gender
	Male	95	24.0
	Female	270	68.2
	LGBTQ+	31	7.8
Age
	18 years old	1	0.3
	19 years old	69	17.4
	20 years old	159	40.2
	21 years old	110	27.8
	22 years old	55	13.9
	23 years old	2	0.5
Academic Year
	1st Year	68	17.2
	2nd Year	157	39.6
	3rd Year	114	28.8
	4th Year	55	13.9
	5th Year	2	0.5
Monthly Income (THB)
	3001–5000	4	1.0
	5001–7000	52	13.1
	7001–10,000	146	36.9
	10,001–13,000	137	34.6
	>13,000	57	14.4
Duration of App Use
	>1 year	379	95.7
	6–12 months	12	3.0
	Less than 6 months	5	1.3
Reasons for Initial App Adoption
	Distant restaurants	59	14.9
	Lack of public transportation	13	3.3
	No parking space or difficult parking	16	4.0
	Traffic congestion	15	3.8
	Affordable prices	24	6.1
	Encouraged by friends	18	4.5
	Avoiding long queues	101	25.5
	Availability of popular restaurants	52	13.1
	Fear of disease outbreaks	69	17.4
	Influenced by advertisements	9	2.3
	No nearby restaurants	20	5.1
Impact on Dining Frequency
	Decreased significantly	58	14.6
	Decreased slightly	168	42.4
	No change	131	33.1
	Increased slightly	32	8.1
	Increased significantly	7	1.8

). Most respondents were female (68.2%), with male (24.0%) and LGBTQ+ (7.8%) groups also represented. The majority of participants were aged 20 (40.2%), followed by 21 (27.8%) and 19 (17.4%). A significant proportion were second-year students (39.6%), with first- and third-year students comprising 17.2% and 28.8%, respectively. Monthly family incomes clustered around THB 7001–10,000 (36.9%) and THB 10,001–13,000 (34.6%). Food delivery app usage was widespread, with 95.7% of students using these apps for over a year. Line Man, GrabFood, and Food Panda dominated preferences, often used in varying sequences. Initial adoption reasons included convenience factors such as avoiding long queues (25.5%) and distant restaurants (14.9%), alongside pandemic-driven concerns (17.4%). Food delivery habits influenced dining patterns, with 57.0% reporting a reduced frequency of dining out and 86.1% indicating a decrease in cooking at home. These findings underscore the prominence of food delivery applications in students’ daily lives, driven by ease of access, promotional offers, and societal shifts, particularly during and after the COVID-19 pandemic.

5.2. Model Measurement Assessment

Table 3. Convergent validity.

Construct	Item	Standardized Factor Loadings (>0.5)	AVE (>0.5)	Composite Reliability (>0.7)	Cronbach’s Alpha (>0.7)
Performance Expectancy (PE)	PE1	0.66	0.54	0.823	0.827
	PE2	0.7
	PE3	0.8
	PE4	0.77
Effort Expectancy (EE)	EE1	0.67	0.511	0.728	0.715
	EE2	0.68
	EE3	0.7
Social Influence (SI)	SI1	0.86	0.776	0.912	0.913
	SI2	0.91
	SI3	0.87
Facilitating Conditions (FC)	FC1	0.68	0.619	0.828	0.804
	FC2	0.86
	FC3	0.81
Environmental Concerns (EC)	EC1	0.88	0.767	0.907	0.893
	EC2	0.95
	EC3	0.78
Behavioral Intention (BI)	BI1	0.81	0.654	0.85	0.842
	BI2	0.86
	BI3	0.76
User Behavior (UB)	UB1	0.67	0.686	0.866	0.849
	UB2	0.89
	UB3	0.9

demonstrates that the measurement model meets the quality criteria. Indicator loadings range from 0.66 to 0.95, exceeding the threshold of 0.5, indicating convergent validity. AVE values (0.511–0.776) confirm that latent variables explain over 50% of the variance. CR values (0.728–0.912) and Cronbach’s alpha (0.715–0.913) surpass the recommended thresholds, establishing composite reliability and internal consistency [69,70]. These findings validate all latent constructs.

5.3. Results of the Structural Equation Model Analysis

The structural equation model analysis explored the factors influencing food delivery app adoption among students at Mahasarakham University (see Figure 2 and

Table 4. SEM analysis estimations.

Hypothesis	Variable (Path)	Std. Est.	t-Value	p-Value	R2	Result
H1	PE → BI	0.069	1.824	0.068	0.978	Not supported
H2	EE → BI	0.311	3.818	<0.001 ***	0.978	Supported
H3	SI → BI	0.326	2.903	0.004 **	0.978	Supported
H4	FC → BI	0.111	1.073	0.283	0.978	Not supported
H5	EC → BI	0.262	2.213	0.027 *	0.978	Supported
H6	BI → UB	0.359	2.889	0.004 **	0.415	Supported
H7	FC → UB	0.291	2.363	0.018 *	0.415	Supported

Note: Std. Est. refers to standardized estimate. * p < 0.05, ** p < 0.01, and *** p < 0.001.

). The model explained 97.8% of the variance in behavioral intention (BI) (R² = 0.978), a notably high value that reflects the strong predictive power of the extended UTAUT constructs in this context, likely driven by the homogeneity of our sample (95.7% with over one year of FDA use; Table 2) and the relevance of effort expectancy (EE), social influence (SI), and environmental concerns (EC) to these students.

Performance expectancy (PE) showed a positive but non-significant direct effect on behavioral intention (BI) (β = 0.069, t = 1.824, p = 0.068), explaining 97.8% of the variance. Effort expectancy (EE) significantly influenced BI positively (β = 0.311, t = 3.818, p < 0.001), indicating its critical role, with 97.8% of the variance explained. Similarly, social influence (SI) demonstrated a significant positive effect on BI (β = 0.326, t = 2.903, p = 0.004), also contributing to 97.8% of the variance. Conversely, facilitating conditions (FC) had a non-significant effect on BI (β = 0.111, t = 1.073, p = 0.283) despite explaining 97.8% of the variance. Environmental Concerns (EC) positively and significantly impacted BI (β = 0.262, t = 2.213, p = 0.027), highlighting their influence, with 97.8% of the variance explained.

Behavioral intention (BI) had a significant positive effect on Use Behavior (UB) (β = 0.359, t = 2.889, p = 0.004), explaining 41.5% of the variance. Additionally, facilitating conditions (FC) significantly influenced UB positively (β = 0.291, t = 2.363, p = 0.018), accounting for 41.5% of the variance. These findings underscore the importance of effort expectancy, social influence, and environmental concerns in shaping BI, while BI and FC are critical drivers of actual app usage behavior.

Additionally, the structural equation model assessing factors influencing food delivery app adoption among Mahasarakham University students demonstrated a strong alignment with empirical data. Seven latent variables, comprising 22 observed variables, were analyzed. The model’s fit indices were evaluated: Chi-Square = 285.959, df = 153 (acceptable), Relative Chi-Square (χ²/df) = 1.869, CFI = 0.979, IFI = 0.979, and TLI = 0.968, all exceeding the threshold of >0.90. The RMR = 0.035 (<0.05) and RMSEA = 0.047 (<0.08) met their criteria. These results confirm the model’s consistency with empirical data and its validity for interpretation.

6. Discussion

6.1. Summary of the Main Findings

This study provides a nuanced understanding of the factors influencing food delivery app (FDA) adoption among university students in a secondary city in Thailand, using an extended UTAUT framework. By incorporating environmental concerns alongside traditional constructs of performance expectancy, effort expectancy, social influence, and facilitating conditions, this study contributes to the literature on technology adoption while addressing critical sustainability considerations. The findings offer important insights into the local context, where cultural norms, infrastructural limitations, and emerging environmental consciousness intersect to shape consumer behavior.

The significant positive effect of effort expectancy (EE) on behavioral intention (BI) underscores the pivotal role of ease of use in technology adoption [60,61]. For university students in a secondary city like Maha Sarakham, the simplicity of navigating FDAs and completing transactions may be particularly appealing given their busy schedules and limited time to visit physical establishments. Moreover, the prominence of mobile technology in their daily lives—often as the primary means of accessing digital services—further amplifies the importance of user-friendly interfaces. In the context of secondary cities, where students may have fewer dining options and rely more heavily on FDAs for convenience, intuitive app design becomes a decisive factor in adoption.

Similarly, the significant impact of social influence (SI) on BI highlights the role of peer and family networks in shaping attitudes toward FDAs. In the communal culture prevalent in secondary cities, recommendations from trusted social circles are likely to carry greater weight than in more individualistic urban environments [71,72]. The relatively close-knit nature of university communities in these settings reinforces the importance of social norms, as students are more likely to adopt behaviors perceived as popular or beneficial within their peer groups [73,74]. This finding aligns with prior research indicating the heightened influence of communal relationships in driving technology adoption [63,64].

Interestingly, performance expectancy (PE), a central construct of the UTAUT model, did not significantly influence BI in this study. This could reflect the baseline expectations of students for FDAs, where perceived usefulness is already assumed given the widespread availability and functionality of such apps. In the local context, students may view FDAs as standard tools rather than as innovative solutions, diminishing the weight of performance-related considerations. This aligns with Zhao and Bacao’s [25] observation that psychological factors like trust and satisfaction can sometimes override traditional utility-based evaluations, especially in environments shaped by extraordinary circumstances, such as the post-COVID-19 period.

The inclusion of environmental concerns (EC) as a construct provides a meaningful extension to the UTAUT model and highlights the growing environmental awareness among younger consumers in Thailand. The positive and significant effect of ECs on BI highlights the growing environmental consciousness among younger consumers [20,28]. In secondary cities like Maha Sarakham, where environmental challenges such as waste management and air pollution are more visible, students may be particularly attuned to the environmental implications of their consumption choices. Packaging waste and delivery vehicle emissions emerge as pertinent concerns, consistent with Heard et al. [30] and Wang et al. [31]. This finding underscores the need for FDA providers to adopt greener practices, such as using biodegradable packaging or deploying electric delivery vehicles, to align with consumer preferences and remain competitive. Facilitating conditions (FC), while not significantly affecting BI, had a direct influence on use behavior (UB). This finding emphasizes the practical infrastructure necessary to enable app usage, including smartphone compatibility, app reliability, and internet connectivity. In secondary cities, where digital infrastructure may be less robust than in metropolitan areas, these factors play a critical role in sustaining usage [75,76]. Smartphone compatibility, app reliability, and internet connectivity are indispensable, aligning with Dwivedi et al. [64] and emphasizing the importance of technical support for sustained usage.

The unique dynamics of secondary cities further shape the interplay of these factors. Unlike metropolitan areas, secondary cities often face infrastructural and institutional challenges, such as limited public transportation options and fragmented urban planning [53,54]. These constraints make FDAs an attractive alternative for accessing meals without navigating logistical difficulties. The significant role of social influence and environmental concerns in this context highlights how local cultural values and global sustainability trends converge in shaping adoption behaviors. This intersection provides valuable insights for policymakers and service providers seeking to enhance the integration of FDAs into smart mobility systems in similar urban environments. By situating these findings within the broader context of smart mobility, this study underscores the transformative potential of FDAs in secondary cities. Beyond their convenience, FDAs can serve as enablers of sustainable urban logistics, reducing unnecessary travel and optimizing resource allocation. Future research could explore how these dynamics evolve over time, particularly as secondary cities continue to develop and adapt to changing consumer and environmental demands.

6.2. Policy Implications

The findings of this study provide valuable insights for policymakers aiming to enhance the integration of FDAs into sustainable urban mobility systems, particularly in secondary cities. The significant impact of environmental concerns on behavioral intention underscores the importance of promoting eco-friendly practices in the food delivery sector. Policymakers can incentivize the adoption of green delivery vehicles, such as electric or hybrid bikes, through subsidies or tax benefits, and implement regulations mandating the use of biodegradable or reusable packaging materials. To effectively encourage sustainable practices in secondary cities, policymakers can implement concrete policies, such as subsidies or tax incentives for local restaurants and delivery providers adopting electric scooters or bicycles for deliveries. Also, municipalities could facilitate public–private partnerships to distribute affordable, biodegradable, or reusable packaging to local vendors, thus reducing the environmental footprint while simultaneously supporting local businesses. Requiring delivery platforms to disclose their environmental impact and adopt measurable sustainability goals can align industry practices with broader environmental policies. However, policymakers must also carefully consider trade-offs between sustainability initiatives and user convenience or acceptance. For instance, introducing biodegradable packaging may increase costs for restaurants, which could subsequently affect consumer prices and potentially reduce the attractiveness of delivery services. Therefore, it is crucial to conduct feasibility studies or pilot projects in secondary cities to assess consumer acceptance, willingness to pay, and the broader economic impacts of these green initiatives.

The critical role of facilitating conditions in driving actual app usage highlights the need for robust digital infrastructure. Expanding high-speed internet access in secondary cities and fostering partnerships between the government and private sectors to improve smartphone affordability can enhance digital accessibility. Furthermore, digital literacy programs targeting less tech-savvy populations can support the equitable adoption and use of FDAs.

Social dynamics also play a pivotal role in FDA adoption, particularly in the communal culture of secondary cities. Community-based campaigns can be designed to promote FDAs by emphasizing their convenience, safety, and environmental benefits. Policymakers should also support local entrepreneurship within the FDA ecosystem, encouraging collaborations between delivery platforms and small restaurants or local delivery services to strengthen community ties and boost economic participation.

The unique dynamics of secondary cities necessitate targeted strategies to address infrastructure and consumer behavior gaps. Policymakers should invest in data-driven urban planning, leveraging FDA usage patterns to identify underserved areas and guide infrastructure development. Integrating FDAs with public transportation systems can further optimize urban mobility and reduce inefficiencies. Establishing local innovation hubs to support technology adoption and adaptation can also foster sustainable growth in these regions.

Finally, fostering competition and innovation within the food delivery sector is essential for its long-term sustainability. Policymakers should introduce regulatory frameworks that ensure fair pricing, consumer protection, and quality assurance. Additionally, funding and support for startups in the smart mobility ecosystem can drive technological advancements and enhance service offerings. By addressing these policy priorities, FDAs can be effectively integrated into the broader framework of smart mobility, supporting sustainability, inclusivity, and efficiency in urban transportation systems.

7. Conclusions

This study advances the literature on FDA adoption by applying an extended UTAUT model, incorporating environmental concerns, to university students in a secondary city in Thailand. The findings demonstrate that effort expectancy, social influence, and environmental concerns are significant predictors of behavioral intention, while facilitating conditions and behavioral intention influence actual usage behavior. These results underscore the multifaceted nature of technology adoption, where ease of use, social dynamics, and sustainability considerations converge to shape consumer behavior. The theoretical contribution of this research lies in extending the UTAUT model to include environmental concerns, which are increasingly critical in consumer decision-making. Practically, the findings suggest that FDA providers should prioritize user-friendly interfaces, leverage social networks, and adopt environmentally conscious practices to appeal to young, tech-savvy consumers. Policymakers and service providers in secondary cities should also address infrastructural gaps to enhance access and support broader technology adoption [77,78].

Despite its contributions, this study has several limitations that should be acknowledged clearly. First, the research focuses exclusively on university students in a single secondary city, limiting the generalizability of findings to broader demographic groups or different urban contexts. Future studies should replicate this research across multiple universities in Thailand and extend the investigation to other Southeast Asian countries to capture diverse cultural, economic, and infrastructural contexts. Additionally, incorporating non-student populations such as working professionals, families, and older adults will provide a more comprehensive understanding of the universal applicability of the factors identified, particularly the roles of environmental concerns and effort expectancy. Second, the cross-sectional research design restricts the ability to capture behavioral changes over time. Future longitudinal studies could provide deeper insights into how environmental concerns and other influencing factors evolve in shaping behavioral intentions and actual usage behaviors related to FDAs. Third, although environmental concerns were incorporated into the UTAUT model, the specific mechanisms by which these concerns influence consumer decisions were not examined. Further research could investigate underlying mechanisms, including eco-conscious branding, consumer trust in sustainability claims, and the effectiveness of green delivery initiatives.

Fourth, the environmental concerns (EC) construct was narrowly operationalized, potentially introducing response bias through item wording that implied negative environmental impacts. Future studies should adopt more comprehensive and neutrally worded measures, addressing additional dimensions such as food waste, energy consumption, sustainable packaging preferences, and willingness to pay for eco-friendly delivery options, thus capturing environmental concerns more effectively. Fifth, the notably high R² value (0.978) for behavioral intention (BI) may indicate overfitting or limited variance within the sample, requiring cautious interpretation. Future research should validate these findings through alternative methods or model specifications and explore mediation and moderation effects—for instance, examining how EC moderates relationships between other constructs such as social influence (SI) and BI—to provide deeper insights into these dynamics. Sixth, cultural factors unique to Thailand, such as communal dining traditions and street food culture, were not explicitly analyzed. Additionally, differences between local (e.g., MSU LIKE SHOP) and global platforms (e.g., GrabFood) were underexplored. Given the quantitative and cross-sectional design constraints, future research incorporating qualitative approaches, such as interviews or comparative case studies, would offer richer insights into cultural drivers and platform-specific factors influencing FDA adoption.

Seventh, detailed cost–benefit analyses of proposed sustainability policies were beyond the current study’s scope. Future research should empirically assess the economic and behavioral impacts of sustainability initiatives in secondary cities to provide robust, evidence-based recommendations for policymakers. Eighth, although the UTAUT framework effectively addressed our study objectives, utilizing extended models such as the UTAUT2, which includes hedonic motivation, price value, and habit, could yield deeper insights into consumer behavior related to FDAs.

Finally, integrating FDAs into the broader smart mobility framework warrants additional investigation. Future studies should explore interactions between FDAs, public transportation, and urban logistics systems in various urban contexts, providing valuable insights for policymakers and urban planners aiming to foster sustainable urban mobility. By addressing these identified limitations, future research can significantly expand upon the foundations laid by this study.