Factors influencing consumers’ continuance purchase intention of local food via online food delivery services: the moderating role of gender

Abstract

In a post-pandemic era, digitalization is embedded in all service sectors, including the food industries. Today’s restaurant industry faces increased competition and the need to adapt to changing consumer preferences for competitive pricing, convenience, and reliable food delivery. In this study, we explore the factors that shape the continued purchase intention of local food via online food delivery (OFD) services. Two hundred valid responses were acquired from Malaysian customers with experience ordering local food through OFD services. Data analysis via SmartPLS revealed that perceived ease of use was the only insignificant predictor, while health aspects, food quality, service quality, and price value significantly influenced customer purchase intention. Notably, price value emerged as the most dominant determinant of purchase intention. Gender moderates the effects of perceived ease of use and service quality on continuance intention. The framework provides a more robust explanation for females, with service quality being their primary focus, while males prioritize price values. These findings offer valuable insights for local restaurants and OFD service providers striving to remain competitive in the post-pandemic epoch.

Keywords:

• Food quality
• health aspects
• online food delivery
• price value
• perceived ease of use
• service quality
• gender

REVIEWING EDITOR:

• Len Tiu Wright, De Montfort University Faculty of Business and Law, United Kingdom

SUBJECTS:

• Behavioral Sciences
• Consumer Psychology
• Economics, Finance, Business & Industry
• Entrepreneurship and Small Business Management
• ICT
• Marketing
• Social Sciences

1. Introduction

Online food delivery (OFD) platforms allow users to order food from restaurants and deliver it to their location through a website or mobile app. This process is streamlined with users browsing menus, choosing dishes, personalizing orders, paying electronically, and getting the dishes from a delivery person or courier (Suhartanto et al., 2019). OFD platforms offer convenience, choice, and flexibility, making them essential in today’s modern society. Thus, It is unsurprising that OFD services are ubiquitous globally, with platforms like Doordash, Deliveroo, Uber Eats, Zomato, and Swiggy leading the market (Keeble et al., 2022; Research, 2023).

The OFD service industry’s popularity continues to surge and is expected to grow from $128.32 billion in 2022 to $143.05 billion in 2023, projecting a compound annual growth rate (CAGR) of 11.5% (Research, 2023). The spread of smartphones, the rise of metropolitan centres, the hectic pace of modern life, and, most recently, the effects of the COVID-19 epidemic have all played a role in this expansion. The pandemic accentuated the significance of contactless food delivery and safety standards, increasing consumer reliance on these services. Sustainability, healthier food options, and enhanced delivery logistics are at the forefront of this market’s evolution, driving its rapid growth. This trend is expected to continue worldwide as more customers discover the ease and selection of online food delivery, reforming how people access, customize, and purchase their favourite meals.

Paralleling the advent of OFD services, research on the effects of OFD has emerged, with scholars examining consumer behaviour, technical advancements, and market dynamics (Chew et al., 2023; Wyse et al., 2021; Yeo et al., 2017). A recent meta-review reveals a diverse corpus of behavioural theories and models adopted in studies, primarily focusing on China and South Korea (Shankar et al., 2022). Scholars seek to understand how consumers perceive and engage with these platforms to shed light on the variables impacting their acceptance and adoption. Recent studies, such as Keeble et al. (2022)’s work, have explored aspects like the importance of takeaway food, convenience, and saving effort. Other studies encompass menu design, pricing methods, and the importance of reviews and ratings in influencing consumer decisions (Shankar et al., 2022; Yeo et al., 2017).

Jun et al. (2022) drew on the technology acceptance model (TAM) to assess the factors impacting customers’ intention to use OFD services. They discovered that perceived usefulness, directly and indirectly, impacts customer usage of OFD through customer attitude. At the same time, enjoyment, trust, and social influence are key characteristics influencing the technology’s acceptance. It was also shown that customer attitude positively influences behaviour intention of OFD services.

Pillai et al. (2022) investigated how perceived benefits (convenience, trust, order accuracy, variety of choices), perceived risks (psychological, financial, product), and online persuasion (central and peripheral routes) influence consumer attitudes and purchase intentions in the United States. Adopting the theories of planned behaviour, perceived risks, and the elaboration likelihood model, their findings indicated that perceived benefits, perceived risks, and online persuasion significantly influenced consumer attitudes and purchase intentions. A more recent study by Long (2023) identified factors impacting consumer intentions in online food purchasing in Phnom Penh. The results showed that advertising and technology eagerness significantly affected online purchase attitudes and, in turn, influenced purchase intention, while customer control and perceived convenience did not impact online purchase attitudes.

The global trend of OFD services extends to Malaysia, which has become the new norm for urban consumers (Lau & Ng, 2019). This shift transforms the dynamics and level of interaction between customers and food suppliers (Ibrahim et al., 2022). Before the COVID-19 pandemic, Malaysian small- and medium-sized restaurants faced intense competition against established food service brands (Ernst & Young, 2020). However, pandemic-induced lockdown measures and consumers’ hesitance and anxiety in visiting restaurants precipitate a significant situational factor that drives consumers to use OFD services. During the Malaysian Movement Control Order (MCO), the food and beverage industry income fell by 90% against the prior year (Bernama, 2020). Reports indicate that 98% have been using OFD services, with 78% of Malaysians utilizing these services regularly during the Recovery Movement Control Order (RMCO) implementation (Vase, 2020). Local food enterprises, including hawkers, realized the need for food delivery platforms to sustain profits. Should they not embrace digitalization, they will fail to interface with their usual pool of customers (Parmar, 2021).

According to Statista (2023), Malaysia’s OFD business, encompassing groceries and meal delivery, is projected to witness a considerable boost, with an anticipated annual growth rate of 14.24%. The revenue for this sector is estimated to reach US$2.31 billion in 2023 and increase to US$3.94 billion by 2027. In 2023, it is expected that 30.5% of the population will be employing meal delivery services. This number is set to advance significantly by 2027, with about 13.04 million people expected to benefit from meal delivery services, demonstrating a notable surge in user penetration (Statista, 2023). The restaurant industry in Malaysia is already highly competitive, with a peaked saturation point (Gao et al., 2018). Thus, OFD platforms present a practical prospect for local restaurants to extend their business beyond their geographical boundaries. The capacity for success in this endeavour is closely linked to the availability and efficiency of delivery coverage (Ibrahim et al., 2022).

Nevertheless, local restaurant owners face challenges when venturing into OFD services. Consumer concerns about the safety and quality of food served by local restaurants impact their purchase intentions. Consumers believe that reputable chain restaurants practice higher sanitation standards and prioritize hygiene, cleanliness, and food quality (Yang et al., 2021). Despite no concrete evidence of COVID-19 transmission through food, consumer anxiety persists. Therefore, ensuring that restaurants and OFD companies adhere to proper operating procedures to mitigate the risk of virus transmission is crucial (Kayes, 2020). Maintaining a good reputation is vital for restaurants and OFD companies, as negative feedback on social media can significantly impact their sales.

The rise of OFD services has attracted research interests in Malaysia. Lau and Ng (2019) and Mohd Nor et al. (2022) focused on the intention to use OFD, whereas Perumal et al. (2021) concentrated on the quality (customer satisfaction) of OFD. Koay et al. (2022) linked OFD service quality with customer satisfaction and then customer satisfaction to customer loyalty, while Pitchay et al. (2022) looked at UTAUT variables to attitude and then attitude to the intention with age as a moderator. Lau and Ng (2019) revealed that convenience, time-saving, privacy, and security positively influence Klang Valley consumers’ choice to use OFD but not for perceived ease of use. Mohd Nor et al. (2022) reported that perceived ease of use, price value, performance expectancy, and food safety risk perception positively correlated with future intention to use OFD after COVID-19. Perumal et al. (2021) claimed that information quality, privacy and security, and time-saving orientation positively influence the quality of OFD services but not for price saving; however, they only portray results for demographic and normality tests for the quality of OFD services. Pitchay et al. (2022) revealed that attitude significantly influences the intention to use OFD, and age is not a significant moderator.

Lau and Ng (2019) studied urban residents in the Klang Valley before the COVID-19 pandemic, focusing on time-saving and convenience motivations. In contrast, Mohd Nor et al. (2022) conducted research during the pandemic, but their analysis was limited to descriptive and correlation studies without path analysis. Neither Lau and Ng (2019) nor Mohd Nor et al. (2022) validated respondent status as OFD users or included health and quality criteria in their studies.

The COVID-19 pandemic has caused a substantial shift in people’s food consumption habits, resulting in a notable surge in OFD services usage. This shift in behaviour is primarily driven by heightened anxieties stemming from perceived infection risks associated with dining out, prompting individuals to persist in selecting these services, even as COVID-19 restrictions are gradually relaxed (Francioni et al., 2022).

However, a recent report by Oppotus (2023) indicates a decline in the trend of using delivery apps, reaching an all-time low of 29% in Q1 2023 as the country steadily reopens and recovers from the pandemic’s impact. This situation underscores the importance of examining consumers’ intentions to persist in using OFD apps in the post-pandemic era. Recent studies conducted in Malaysia have explored the factors influencing consumers’ continuance intention to use OFD services in the context of the COVID-19 pandemic, with notable contributions from Yapp and Kataraian (2022) and Yap and Lee (2023). Both studies employed the Extended Unified Theory of Acceptance and Use of Technology (UTAUT2) in their models. Yapp and Kataraian (2022) identified that social influence, hedonic motivation, price value, habit, time-saving, and convenience as predictors of users’ intention to continue using OFD, and performance and effort expectance were not. Focusing on Malaysian Millennials, Yap and Lee (2023) replicated some of the findings of Yapp and Kataraian (2022) related to habit and hedonic motivation but diverged results regarding price value, performance, and effort expectancy. None of these studies considered the situational and quality factors associated with the pandemic or the moderation effects of gender.

Francioni et al. (2022) delved into the impact of various variables related to food safety concerns, the pandemic situation, app effectiveness, and the moderating influence of gender among Italian university students. They discovered that perceived healthiness, adherence to quarantine procedures, perceived hygiene, perceived ease of app use, and personal attitude significantly influence continuance intention. However, they did not find a significant gender difference. Notably, the study by Francioni et al. (2022) excluded the quality dimension and price value.

Notably, there needs to be more specific studies on the factors influencing the continuance purchase intention of local food through OFD services in Malaysia. This study aims to address this gap by focusing on three primary dimensions. The initial dimension involves the ‘health aspects’ predictor related to the pandemic situation, which has prompted the promotion of OFD services. It includes health worries, the desire to minimize exposure to COVID-19, attention to food hygiene, and adherence to social distancing measures. The second set of factors pertains to the effectiveness of OFD applications, encompassing app user-friendliness, discounts, search, and filtering functionalities. It is represented by the perceived ease of use and price value. The final cluster of predictors pertains to quality, comprising both food and service quality. In addition, this study explores the role of gender as a moderator to the proposed research framework.

2. Theoretical framework

2.1. Health aspects

Covid-19 serves as a stark reminder of the importance of health and hygiene. Since the pandemic, elements linked with safety and health concerns have critically influenced customers’ attitudes regarding OFD services. OFD service consumers demand fresh and clean food (Yang et al., 2021)—the intention and actual purchase via OFD services are determined by food hygiene (Al Amin et al., 2021). Thus, OFD service providers must collaborate with restaurant operators to ensure safe food processing and packaging to eliminate contamination (Duda-Chodak et al., 2020; Kayes, 2020). Francioni et al. (2022) demonstrated a significant impact of perceived hygiene on continuance intentions. However, their study did not establish a clear connection between perceived risk associated with dining out, fear of COVID-19, and the likelihood of continued usage of Online Food Delivery (OFD) apps. Contactless deliveries and cashless transactions align with hygienic protocols, thereby augmenting consumers’ willingness to continue use OFD services for local food. Therefore, the first hypothesis is proposed as follows:

H1: Health aspects positively affect consumers’ continuance purchase intention of local food via OFD services.

2.2. Perceived ease of use

Perceived ease of use denotes the user’s assumption about how uncomplicated it is to use a technological application (Davis, 1989). The order tracking, order process, and interface filtering choices are essential features that reflect the perceived ease of use of the OFD applications (Ray et al., 2019). Roh and Park (2019)’s study in South Korea found that the higher customers’ perceived ease of use, the more inclined they are to employ OFD services. Troise et al. (2021) ‘s study on Italian consumers showed that perceived ease of use positively relates to consumer behavioural intention of OFD services. Francioni et al. (2022) asserted that perceived ease of use has a favourable influence on Italian university students’ continuance intention to use OFD service. Hong et al. (2023) found that effort expectancy (perceived ease of use) was not a predictor for purchase intention of OFD services in the United States. Mohd Nor et al. (2022) documented that perceived ease of use positively correlates with Malaysian’s behavioural intention to use OFD. Conversely, Lau and Ng (2019) research reported a nonsignificant relationship. This work extends the investigation into the relationship between perceived ease of use and consumer intention to re-order local food via OFD services. We put forward the second hypothesis:

H2: Perceived ease-of-use positively affects consumers’ continuance purchase intention of local food via OFD services.

2.3. Price value

Online food delivery providers that integrate other tools, such as discounts and coupons, with their marketing plan see a significant increase in their revenue (Gupta, 2019). Kumar and Shah (2021) recommended future studies in OFD services to incorporate elements of emotional influence, which can be evoked through offering discounts and coupons. In addition, Prasetyo et al. (2021) asserted that promotions positively impact the likelihood of utilizing OFD services. Besides, the Competition Commission of India (2020) found that one of the drivers of online purchases is enhanced pricing transparency that allows ad-hoc price comparisons. Furthermore, Alalwan (2020) found that price value impacts consumers’ intention to reuse OFD services. Lee (2019) has a similar finding; it was discovered that cost-effectiveness is an essential factor influencing consumers to adopt OFD services in China. Thus, we formulate the following hypothesis:

H3: Price value positively affects consumers’ continuance purchase intention of local food via OFD services.

2.4. Food quality

Food quality defines how well the food meets the demands of restaurants’ customers; it is a critical factor in determining consumer repurchase or continuous intention when placing orders through OFD services (Suhartanto et al., 2019). Culinary attributes, such as the degree of well-cooked quality, freshness, and food presentation, impact customer satisfaction and repurchase intention via OFD services (Kedah et al., 2015). Consumers favor OFD services that offer broad menus, healthy options, and food delivery, ensuring fresh and warm meals upon arrival (Sjahroeddin, 2018). The respective literature has documented food quality as the most crucial aspect in meeting consumers’ expectations that drive decision-making criteria for online food ordering (Saad, 2020; Sjahroeddin, 2018). Consumers’ willingness to use OFD services depends on their satisfaction with the food quality (Banerjee et al., 2019). Koay et al. (2022) also found that maintaining meal quality and hygiene affects customer satisfaction with OFD services. Hence, the fourth hypothesis is as follows:

H4: Food quality positively affects consumers’ continuance purchase intention of local food via OFD services.

2.5. Service quality

Online food delivery service encapsulates the process of service providers delivering meals from numerous restaurants when a consumer places an order via a website or mobile application (Pigatto et al., 2017). While OFD services offer convenience, several operational challenges need addressing. Order fulfilment is essential in satisfying customers on OFD service (Sjahroeddin, 2018). High traffic levels on online food delivery platforms, especially during peak mealtimes, lead to delays in the order processing (Chen et al., 2021). The perceived level of service offered by OFD providers significantly affects customer satisfaction (Kedah et al., 2015). Saad (2020) revealed that service quality influenced consumers’ decisions while purchasing meals online. Koay et al. (2022) examined the service quality dimension and further elaborated that reliability, assurance, security, and system operation are determinants of customer satisfaction with OFD services. Therefore, the fifth hypothesis is developed as follows:

H5: Service quality positively affects consumers’ continuance purchase intention of local food via OFD services.

2.6. Gender as moderator

Gender, as a social construct and demographic factor, has been extensively studied in marketing research, examining its impact on customer behaviour and consumption intentions. Recent studies have delved into gender moderation in diverse scenarios, such as online purchase intent during COVID-19 in Bangladesh (Awal et al., 2023), mobile wallet adoption in India (Chawla & Joshi, 2020), peer-to-peer mobile payment systems (Kalinić et al., 2020), mobile augmented reality in physical stores (Alesanco-Llorente et al., 2023) in Spain, Instagram health influencers in Morocco (Chetioui et al., 2023), contactless fitness services during COVID-19 in China (Zhu et al., 2023), and environmentally responsible purchases in Pacific Alliance Countries (García-Salirrosas et al., 2023).

Gender’s influence on customer behaviour and intentions in food purchase and consumption has also been explored by several researchers. These studies encompass OFD in Italy (Francioni et al., 2022) and India (Anbumathi et al., 2023), beverage food products in Pakistan (Sharif et al., 2023), cloud kitchens in the United States (Pookulangara et al., 2023), drone food delivery services in South Korea (Hwang et al., 2019), and online local food purchases in Indonesia (Suhartanto et al., 2023).

Previous research has produced varying findings regarding gender moderation. For example, Hwang et al. (2019) found that gender moderated the relationship between customers’ attitudes and word-of-mouth tendency, product innovativeness, and the intention to use drone food delivery services, with higher path coefficients for females than males in South Korea. Awal et al. (2023) identified significant gender moderation effects on customer satisfaction and online shopping experience towards online purchase intent in Bangladesh.

Conversely, Francioni et al. (2022) discovered that attitude and perceived healthiness play more significant roles in shaping females’ intention to continue using OFD, while perceived ease of use, hygiene, and quarantine procedures affected males more. Nevertheless, none of the differences between genders are significant. Anbumathi et al. (2023) found no gender moderation effects on path coefficients in their study of OFD in India, aligning with the results of Francioni et al. (2022).

Despite diverse findings, previous research underscores the importance of gender’s moderating role in consumer behaviour and perception. Based on this theoretical and empirical foundation, this study proposes the following hypotheses:

H6: Gender significantly moderates the impact of predictors (health aspect = H6a, perceived ease of use = H6b, price value = H6c, food quality = H6d, service quality = H6e) on continuance purchase intention for local food via OFD services.

The research framework developed for this study is depicted in Figure 1.

Figure 1. Research framework.

3. Research methodology

A self-administered questionnaire was employed to gather data. The survey starts with demographic background in section A, general questions in section B, and measurement constructs in section C. Measurement constructs from previous research were adapted for this study, i.e., food quality from Sjahroeddin (2018), perceived ease of use from Ray et al. (2019), price value from Agmeka et al. (2019) and Pitchay et al. (2022), continuance purchase intention from Lee et al. (2019), one construct of service quality adapted from Ghosh (2020). The new items developed for this study are health aspects and three service quality constructs. All measurement constructs were evaluated using a 5-Likert scale from strongly disagree being 1 to strongly agree being 5. G-power showed that the smallest sample size required is 138 for five predictors, medium effect size (f² =0.15), and 0.95 power. The online survey was created in Google form and distributed via social media. This study was conducted on a convenience and purposive sampling approach; the targeted respondents are people with experience ordering local food via OFD services. Two hundred valid respondents met the criteria, and their feedback is used for data analysis. SPSS version 29 is used for descriptive analysis and t-test for equality of means, and Smart-PLS 4 assesses measurement constructs and the structural model.

4. Data analysis

4.1. Descriptive analysis

Table 1 tabulates the respondents’ demographic background. Most of the respondents are students (71%), generation Z (77%), and single (85.5%) with tertiary education (84%). Besides, more than half of respondents are Chinese (67%), male (56.5%) with household income less than RM3170 (54.5%).

Table 1. Respondents’ demographic.

Demographics		Frequency	Percentage
Gender	Male	113	56.5
	Female	87	43.5
Age	18–25 years old (Generation Z)	154	77
	26–41 years old (Generation Y)	29	14.5
	42–57 years old (Generation X)	8	4
	58 and above	9	4.5
Race	Chinese	134	67
	Malay	40	20
	Indian	21	10.5
	Others	5	2.5
Occupation	Employed	44	22
	Housewife	4	2
	Retired	5	2.5
	Student	142	71
	Unemployed	5	2.5
Marital status	Single	171	85.5
	Married	23	11.5
	Divorced	4	2
	Widowed	2	1
Education level	Secondary (PMR/PT3/SPM)	12	6
	Post-secondary (STPM/Foundation/A-Levels)	12	6
	Tertiary (Diploma/bachelor’s degree)	168	84
	Postgraduate (Masters/PhD)	8	4
Household income	Less than RM3170	109	54.5
	RM3171- RM4,850	47	23.5
	RM4,851- RM7,100	32	16
	RM7,101- RM10,970	8	4
	RM10,911 and above	4	2

As illustrated in Table 2, nearly half (49%) of the respondents have been using OFD services for more than a year and most of them (61.5%) use OFD services 1–4 times per month. 97.5% of them preferred to order food via food delivery apps with Grab Food (56%) as the most frequently used apps, followed by Food Panda (39%) and the remaining 5% using other apps such as Quick Sent, Deliver Eat, Shopee Food etc.

Table 2. General information.

How long have you been using online food delivery services?	1–6 months	78	39
	7–12 months	24	12
	More than 12 months	98	49
How often do you use online food delivery services per month?	1–4 times	123	61.5
	5–8 times	43	21.5
	9–12 times	10	5
	More than 12 times	24	12
What is your preferred method of ordering?	Through food delivery apps	195	97.5
	Through restaurant websites	2	1
	Through phone	3	1.5
Which online food delivery service that you used the most?	Grab food	112	56
	Food Panda	78	39
	Others	10	5

Table 3 tabulated the result summary of the t-test for equality of means, indicating significant differences in health aspects and perceived ease of use among genders. Female respondents gave higher ratings on these two variables than male respondents. On the other hand, ratings by male respondents for price values, food quality, and service quality were slightly higher than females, but the differences are neglectable. Generally, all variables have mean scores above three and standard deviations below 0.7 in Table 4, indicating respondents tend to agree with most of the measurement constructs. For both genders, perceived ease of use had the highest means, while food quality had the lowest average scores. The ratings reflect that OFD apps are user-friendly and local food quality on OFD services might not meet all respondents’ expectations.

Table 3. Different in central tendency and dispersion of variables among gender.

	Male		Female
Variable	Mean	Std	Mean	Std	t-Value
Health Aspects (H)	4.135	.63916	4.305	.52748	−2.005*
Perceived Ease of Use (PEOU)	4.270	.56413	4.434	.56863	−2.031*
Price Value (PV)	3.896	.59455	3.839	.59183	0.673
Food Quality (FQ)	3.816	.57478	3.805	.51634	0.150
Service Quality (SQ)	4.254	.47125	4.230	.49152	0.358
Continuance Purchase Intention (CPI)	4.108	.64855	4.108	.64855	−0.400

Note: Based on a 5-Likert scale, *p < 0.05.

Table 4. Convergent validity.

Constructs	Factor loading	CR	AVE
Health aspects (H)		0.911	0720
H1. I tend to buy food that are fresh and clean for health reasons during the Covid-19 pandemic.	0.835
H2. Aspects such as sanitation standards (hygiene), safety and cleanliness of food delivered are important for me when buying food via online food delivery services during the Covid-19 pandemic.	0.819
H3. Public health initiatives such as practicing social distancing or self-quarantining have changed my eating patterns.	0.889
H4. I buy local food via online food delivery services because I am afraid of catching Covid-19.	0.849
Perceived ease-of-use (PEOU)		0.931	0.773
PEOU1. In my opinion, food delivery apps are easy to use.	0.908
PEOU2. The order placement process via food delivery apps is easy for me.	0.919
PEOU3. I like the feature to track order progress (order accepted/prepared/pickup) via food delivery apps.	0.813
PEOU4. The filter options (type of cuisine, estimated delivery time) are helpful for me	0.873
Price value (PV)		0.909	0.714
PV1. I can save money by comparing prices from different online food delivery services.	0.799
PV2. I like to search for cheap food deals in different food delivery apps.	0.866
PV3. Online food delivery services offer better value for my money compared to physical store.	0.842
PV4. I use online food delivery services because I can save money with discounts and vouchers.	0.870
Food quality (FQ)		0.878	0.643
FQ1. Local restaurants offer a healthy option via online food delivery services.	0.755
FQ2. Local food delivered via online food delivery services are fresh and clean.	0.807
FQ3. Local food delivered via online food delivery services taste just like in the restaurant.	0.803
FQ4. Local food delivered via online food delivery services are still warm upon arrival
Service quality (SQ)	0.841	0.883	0.653
SQ1. My order is delivered as promised and there are no missing orders.	0.761
SQ2. My order is delivered on time.	0.853
SQ3.The delivery staffs are courteous and presentable, and they treated me well.	0.795
SQ4. For me, the delivery charges are worth it for the service offered.	0.821
Continuance purchase intention (CPI)		0.949	0.822
CPI1. I intend to continue buying local food via online food delivery services in the future.	0.898
CPI2. I will always try to use an online food delivery service to buy local food in my daily life.	0.871
CPI3. I plan to continue using online food delivery service frequently when buying local food.	0.935
CPI4. I have decided to use online food delivery services when I buy local food for the next time	0.921

4.2. Assessment of measurement models

The convergent and discriminant measurement constructs and the structural model were validated in this work using SmartPLS 4. The factor loading, composite reliability (CR), and average variance extracted (AVE) of the measurement constructs are shown in Table 4. Construct factor loading ranges from 0.755 to 0.935, with CR and AVE having minimum values of 0.878 and 0.643, respectively. Since Hair et al. (2017) suggested a minimal criterion of 0.7 for CR and 0.5 for AVE to establish the convergent validity, we may say that our measurement constructs fulfil the requirement of convergent validity. As shown in Table 5, the greatest Heterotrait-Monotrait (HTMT) value of 0.553, much lower than the 0.9 criterion and has interval values below 1, demonstrates that the discriminant validity is attained in this study.

Table 5. Discriminant validity by heterotrait-monotrait (HTMT) criterion.

	CPI	FQ	H	PEOU	PV
FQ	0.512
H	0.336	0.145
PEOU	0.247	0.117	0.231
PV	0.588	0.415	0.115	0.221
SQ	0.553	0.456	0.335	0.316	0.295

Note: The values in the brackets represent the 90% bias-corrected and accelerated confidence derived from bootstrapping with 5000 samples. FQ = Food Quality; H = Health Aspects; CPI = Continuance Purchase Intention; PV = Price Value; SQ = Service Quality.

4.3. Assessment of structural model

In the realm of PLS-SEM, common method bias is a phenomenon that arises due to the measurement approach employed in an SEM study rather than being influenced by the interconnected relationships and outcomes within the model under examination (Kock, 2015). Kock (2015) and Hair et al. (2017) suggested that a VIF value exceeding 3.3 indicates problematic collinearity and potential common method bias in a model. Hence, all our VIF values were less than 3.3 (as shown in Table 6), confirming the absence of common method bias.

Table 6. A summary of the assessment for the structural model.

HP	Path	Std. Beta	Std. Error	t-value	Decision	R^2	Q^2	f^2	VIF
H1	H→CPI	0.165	0.057	2.915**	S	0.474	0.426	0.047	1.109
H2	PEOU→CPI	0.045	0.06	0.758	NS			0.004	1.119
H3	PV→ CPI	0.374	0.074	5.042**	S			0.223	1.189
H4	FQ→CPI	0.189	0.073	2.579**	S			0.054	1.270
H5	SQ→ CPI	0.262	0.064	4.113**	S			0.099	1.320

Notes: **Supported at p < 0.01, S = supported, NS = not supported.

After validating the measurement constructs’ convergent and discriminant, we assess the structural model using the SmartPLS 4 bootstrapping function with 5000 sub-samples. Figure 2 portrays the PLS diagram with path coefficients and R-square value. According to the standard beta column in Table 6, all variables have positive relationships with customers’ continuance purchase intention of local food via OFD services. However, based on the t-value, perceived ease of use has insignificant influence, leaving health aspects, food quality, service quality, and price value as significant predictors for customers’ continuance purchase intention of local food via OFD services. The price value with the highest standard beta value of 0.374 with medium effect size (f² = 0.223) emerges as the most dominant predictor.

Figure 2. The finalised PLS diagram.

Note: Outer-path coefficient, Inner-R square.

The model successfully explains a variation in customers’ continuance purchase intentions for local food by 47.4%, according to the R square value of 0.474. Given that the Q² value is higher than 0, the model has predictive power (Hair et al., 2017). Effect sizes of 0.02 are considered a little effect, 0.15 are a moderate impact, and 0.35 are a considerable effect, according to Cohen (1988). While health aspects, food quality, and service quality have minor significant effects in producing R square for the endogenous variable, price value has a moderate effect. Additionally, as all VIF measurements are below 5, the model is free of multicollinearity problems.

4.4. Moderation analysis on genders

4.4.1. Assessment of measurement invariance

For assessing the moderation effects of genders, we conducted multiple group analysis (MGA) in SmartPLS 4. MGA efficiently examines moderation across multiple relationships by evaluating the impact of the moderator on all modelled relationships rather than just one specific relationship (Hair et al., 2017). MGA assumes group heterogeneity, emphasizing the need to ensure sufficient statistical power, meaning an adequate sample size, in subgroups to detect the moderating effect (Becker et al., 2013; Cheah et al., 2020; Hair et al., 2017). According to Hair et al. (2014) power table for least squares regression, a minimum sample size of 70 per group is required for an R-squared of 0.25 and a power of 0.80 with five predictors. Therefore, our sample sizes for males (n = 113) and females (n = 87) were adequate for conducting multigroup analysis on moderation.

Before examining the moderation effect of gender using MGA, it’s crucial to establish measurement invariance, specifically configural and compositional invariance. In SmartPLS, the measurement invariance of composite models (MICOM) procedure is to verify measurement invariance (using moderation- permutation). When using MICOM in SmartPLS 4, configural invariance is automatically confirmed (Cheah et al., 2020). As shown in Table 7, the permutation p-values for compositional invariance were above 0.05 for all measurement constructs, indicating support for compositional invariance in our measurement models. However, composite equality in mean values in Table 8 reveals significant differences in mean values (permutation p-values less than 0.05) for the health aspect between genders. Consequently, partial measurement invariance is confirmed. Therefore, we can present the standardized coefficients of the structural model separately for males and females for comparison.

Table 7. Summary of MICOM assessment.

Measurement construct	Compositional invariance				Composite equality of mean values
	R	5%	Perm p-value	Partial MI	Mean Diff	Confidence interval	Perm p-value	Full MI
Continuance Purchase intention	1	0.999	0.185	Yes	−0.055	(-0.219,0.226)	0.339	Yes
Food quality	0.997	0.987	0.549	Yes	0.023	(-0.24,0.227)	0.432	Yes
Health aspect	0.998	0.977	0.824	Yes	−0.282	(-0.235,0.233)	0.017	No
Perceived ease of use	0.985	0.920	0.24	Yes	−0.276	(-0.247,0.250)	0.034	No
Price value	0.999	0.995	0.499	Yes	0.093	(-0.235,0.232)	0.239	Yes
Service quality	0.994	0.991	0.163	Yes	0.058	(-0.228,0.242)	0.357	Yes

Notes: R = correlation; MI = measurement invariance; Diff = difference; Perm = permutation.

Table 8. Path coefficient of moderation–permutation.

HP	Path	Std Beta (male)	Std Beta (female)	Difference	Confidence interval	Permutation p-value	Decision
H6a	H→CPI	0.167	0.145	0.023	(-0.199, 0.187)	0.438	NS
H6b	PEOU→CPI	0.130	−0.120	0.249	(-0.203, 0.200)	0.022*	S
H6c	PV→ CPI	0.390	0.381	0.009	(-0.262, 0.264)	0.508	NS
H6d	FQ→CPI	0.147	0.227	−0.079	(-0.257, 0.261)	0.309	NS
H6e	SQ→ CPI	0.159	0.438	−0.279	(-0.202,0.214)	0.011*	S

Note:

* p-Value < 0.05, NS = hypothesis not supported, S = hypothesis supported.

4.4.2. Moderation analysis

Table 8 depicts the moderation analysis based on permutation results. In overview, males have higher beta values on health aspects, perceived ease of use, and price value than females; females have better beta values on food and service quality. Path coefficient differences on perceived ease of use and service quality fell outside the 95% confidence interval with a p-value less than 0.05. Hence, gender significantly moderates the influence of perceived ease of use and service quality on continuance purchase intention on local food via OFD services but not for the other variables. Therefore, H6b and H6e were supported, but H6a, H6c, and H6d were not.

4.4.3. Multiple group analysis based on genders

The MGA analysis tabulated in Table 9 revealed that the R square value of females is 0.691, higher than males (0.359), indicating the model has substantial explanation power (69.1%) for females and moderate explanation power (35.9%) on males’ continuance purchase intention on local food via OFD services. Besides, Q square values for both genders were higher than 0, showing the model has predictive power on genders (Hair et al., 2017).

Table 9. Summary of structured model based on genders.

HP	Path	Std. Beta	Std. Error	t-value	Decision	R^2	Q^2	f^2	VIF
Male
H1m	H→CPI	0.167	0.078	2.143*	S	0.359	0.249	0.042	1.042
H2m	PEOU→CPI	0.130	0.101	1.288^#	NS			0.026	1.028
H3m	PV→ CPI	0.390	0.108	3.620**	S			0.208	1.146
H4m	FQ→CPI	0.147	0.100	1.482^#	NS			0.028	1.200
H5m	SQ→ CPI	0.159	0.081	1.963*	S			0.033	1.179
Female
H1f	H→CPI	0.145	0.074	1.958*	S	0.691	0.636	0.051	1.337
H2f	PEOU→CPI	−0.120	0.065	1.844*	NS			0.033	1.383
H3f	PV→ CPI	0.381	0.095	3.994**	S			0.357	1.319
H4f	FQ→CPI	0.227	0.077	2.949**	S			0.114	1.454
H5f	SQ→ CPI	0.438	0.081	5.417**	S			0.365	1.697

Notes: ** m = male, f = female, p < 0.01, *p < 0.05, #p < 0.10. S = supported, NS = not supported.

For males, perceived ease of use and food quality are insignificant at alpha = 5%, leaving health aspects, price value, and service quality as significant factors that affect continuance intention. Hence, H1_m, H3_m, and H5_m were supported and H2_m and H4_m were rejected. Price value with the highest beta value of 0.390 and moderate effect (f² of 0.208 > 0.15) was the most decisive antecedent for male respondents.

All the variables significantly influence females’ continuance intention to use OFD. However, the perceived ease of use shows a significant negative relationship with continuance intention, which does not align with our H2_f that assumed a positive relationship; hence, we rejected H2_f, and H1_f, H3_f, H4_f, and H5_f were supported. Both price value (f² = 0.357) and service quality (f² = 0.365) have f² values greater than 0.35, hence possess large effects on generating R² in females (Cohen, 1988). Nevertheless, service quality (beta = 0.438) emerged as the most important predictor, followed by price value (beta = 0.382) for female respondents.

The multigroup analysis (MGA) also revealed that there were significant differences in gender on the following factors: perceived ease of use and service quality. Perceived ease of use has an insignificant positive influence on males and a significant negative impact on females’ continuance intention to use OFD. On the other hand, service quality positively affects the continuance intention of both genders but with different magnitudes. The impacts of service quality play a dominant role in females but less for males.

Figure 3 illustrates the overview of MGA results with beta and R square values based on gender.

Figure 3. Overview of MGA results.

Notes: M = Male, F = Female, **p value < 0.01, *p-value <0.05, #p-value < 0.10.

5. Discussion

Both companies and governments must comprehend the elements influencing customer intention to utilize OFD applications in Malaysia. This study advances knowledge of customer behaviour in Malaysian OFD services that emphasize local foods. Given that the recent pandemic has drastically transformed the food service industry landscape, insights from this research can guide decision-making and crisis management strategies for local businesses in Malaysia.

Almost all the respondents (97.5%) in this study preferred to use food delivery apps when engaged with OFD services, in which Grab Food emerged as the top mind brand. This study asserts that health aspects, food quality, service quality, and price value have significant positive relationships with customers’ purchase intention of local food via OFD services in Malaysia. Our findings coincided with the 2021 Food Delivery Industry Overview that revealed food variety options (45%), availability of promotions (41%), and the speed of delivery (51%) are the three main variables influencing customers’ satisfaction with food delivery services (Grab, 2021). By focusing on health aspects, food quality, service quality, and competitive pricing, local restaurants can enhance their offerings and attract more customers through OFD services.

Our data show that perceived ease of use is an insignificant predictor, which aligns with Lau and Ng (2019) ‘s and Jun et al. (2022) ‘s findings. This can be attributed to most of this study’s respondents being tech-savvy Gen-Zers who likely find the OFD platforms effortless to use. In this regard, perceived ease of use (PEOU) has little impact on the purchase intention of local food via OFD services. Francioni et al. (2022) found that PEOU positively influenced the continuance intention of OFD in Italy, but no significant difference between genders. Our studies found a moderation effect of gender on PEOU, where PEOU has a positive insignificant influence on males and a negative significant influence on females. Jun et al. (2022) proposed that PEOU might be a crucial factor affecting customers’ intentions, mainly those less experienced, to make food purchases through online food delivery platforms.

Price value is the most dominant exogenous variable that positively influences customer purchase intention of local food via OFD services. Most individuals prioritize price value when buying local food via online food delivery services; they compare prices between different food delivery apps to find the best deal. Besides avoiding social interaction, one of the main motivating factors for new users to start using OFD services is the advantage of special discounts exclusively for online delivery (Grab, 2021). Dayak Daily (2021) reported that the small business campaign enticed 70% of new users ordering foods from small businesses to enjoy a 30% discount with a minimum spend of RM20 with the "PENIAGAKECIL" promotion code. The Ministry of Domestic Trade and Costs of Living Government could continue to work with OFD providers to have such a campaign from time to time to empower the sustainability of small local food restaurants in this digital era.

Our study reveals that gender influences the continuance purchase intention of local food through OFD services concerning service quality. Specifically, Service quality significantly impacts female respondents’ decisions to continue using OFD services. A similar trend was observed in Ma et al. (2014) research on customer loyalty in fine dining, where female customers placed slightly greater emphasis on staff service than male customers. This finding diverges from Zhong and Moon (2020) study on fast-food dining experiences in China, which indicated that service quality had a more pronounced effect on male customer satisfaction than female customers. Apart from regional disparities, this discrepancy may be attributed to Zhong and Moon’s focus on service quality provided by staff members in fast-food restaurants. Our finding suggests that female’s value ascribing a deeper emotional connection when engaging with OFD services. They may be more inclined to link the service experience’s emotional aspects, such as the warmth and friendliness of delivery personnel and the order accuracy, to their perception of service quality.

6. Research implications

6.1. Theoretical implications

This research contributed knowledge by exploring factors affecting continuance purchase intentions of local food via OFD services and moderation effects of gender. This research focused on three primary dimensions, namely pandemic situational factor (health aspect), technological advantages (perceived ease of use and price value), and quality dimensions (food and service quality). In addition, this study investigating the diversity or similarity among gender perspectives in the antecedents in continuance purchase intention in Malaysia, a culturally diverse country, provides new insight into the current literature of OFD.

The research affirms that gender plays a significant role in elucidating customers’ continuance intentions to buy local food via OFD services. This finding underscores the importance of implementing marketing segmentation. Specifically, establishing a customer database categorized by gender can assist restaurant businesses in effectively managing their prospective clients, as suggested by Hwang et al. (2019). Consequently, restaurants can tailor distinct marketing strategies for different genders based on their preferences, ultimately enhancing their patron’s overall satisfaction (Francioni et al., 2022).

6.2. Managerial implications

Some managerial implications can be derived from this study’s findings. Given that health aspects positively impact the purchase intention of local food through OFD services, food operators and OFD platforms must highlight hygiene and safety commitments in their advertisements, menus, and policies. Francioni et al. (2022) proposed that it is advisable to convey all sanitation measures to customers, such as sharing information on food preparation guidelines and specifics regarding the safety of packaging and delivery procedures. One way to evoke user confidence is to collaborate with health inspection agencies to obtain endorsements and badges, assuring hygienic standards for food ingredients, meal preparation protocols, and labour. OFD service providers can spotlight contactless transactions as their primary benefit in providing a novel health-centric approach to having meals without physical contact and exposure risks.

Food quality drives purchase intention of local food via OFD services. Implicatively, food operators and OFD platforms must leverage tactics to accentuate food quality to consumers. They must showcase high-quality and visually appealing food pictures that allude to the food qualities on OFD platforms. Also, OFD platforms can feature food details accentuating fresh and premium ingredients, cooking methods, chef expertise, and utensils to prime consumers’ positive food quality expectations. Food quality entails superior and aesthetically pleasing presentation comprising portions, textures, colours, food positions, and plates and utensils. Hence, food operators must synergize with OFD riders to prepare and maintain excellent packaging and delivery of meals to users in that the arrived meals’ presentation is intact and warm. Ma et al. (2014) discovered that food quality plays a more significant role in female customer satisfaction, indicating their higher expectations for various food quality attributes. Our results align with Ma et al. (2014), confirming that food quality significantly affects female customers’ intention to continue using OFD services.

Because users attend to service quality when intending to purchase local food via OFD platforms, OFD service providers must prioritize prompt and timely delivery processes. Managerial strategies to enhance deliveries include leveraging real-time location tracking technology, route optimization algorithms, and user notifications. Ongoing training provisions can ensure that food delivery personnel showcase professionalism in appearance, food handling, customer interaction, and etiquette. These measures cohesively exhibit service quality branding to consumers, boosting the intention to purchase local food via OFD platforms. The preceding is particularly imperative for female users, given that service quality is a primary factor affecting female decisions to purchase local food through OFD services. OFD services should know these elevated expectations of order accuracy, on-time delivery, and the overall experience among female customers and consistently deliver personalized, high-quality service. In addition, delivery personnel and customer assistance representatives can pay extra attention to providing service quality for female consumers, encompassing personalized, empathetic, warm, and friendly OFD service.

This study places price value as a dominant factor in driving the purchase intention of local food via OFD services. Thus, OFD platforms must feature transparent pricing to showcase meal costs, taxes, delivery fees, and surcharges. Price transparency establishes appropriate consumer fair price expectations and trust while avoiding scepticism about the restaurants’ and the OFD platforms’ profiteering intentions. OFD platforms can offer diverse packages to cater to varying budgets. Offering affordable menus, value-added deals, and promotions appeal to budget-conscious consumers. On the other hand, consumers looking for higher-end meals can benefit from OFD platforms availing premium menus and selections. The accessibility of diverse price-point menus necessitates designers and programmers to deploy effective and user-friendly search and comparison tools in the OFD interfaces. Such measures appeal to a broader range of OFD platform users with varying budgets in evoking positive price value perceptions, thereby boosting the purchase intention of local food via the technology.

7. Conclusion, limitations, and future works

This study underscores the multifaceted impact of digitalization within the food service industry, particularly in a post-pandemic era. The tenet of this study hinges upon the transformative digitization of the food service industry, an adaptation that the recent pandemic has notably expedited. Our research documents health aspects, food quality, service quality, and price value as factors shaping consumers’ intentions to purchase local food via OFD platforms.

These findings offer practical insights for various stakeholders, including restaurants, food operators, and OFD service providers. These factors can effectively connect with consumers and encourage local food purchase intentions through these technology-mediated platforms by emphasizing these salient attributes of health considerations, culinary excellence, service excellence, and value for money.

However, it is vital to acknowledge the limitations inherent in our study. The research scope was confined to the perspectives of Malaysian respondents, thereby constraining the generalizability of our findings to other cultural and national contexts. Future investigations can extend to diverse countries and cultures to enhance the robustness of this research stream. A more comprehensive understanding of the interplay between these factors and local food purchase intention in various settings can be attained through such cross-cultural extensions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Additional information

Notes on contributors

Siew Chein Teo

Siew Chein Teo is a Senior Lecturer at the Faculty of Business, Multimedia University, Melaka, Malaysia. Her research focus includes operation research, consumer behaviour, acceptance models and theories, education management.

Tze Wei Liew

Tze Wei Liew is an Associate Professor at the Faculty of Business Multimedia University, Melaka, Malaysia. His research interests and contributions intersect cognitive psychology, learning sciences, media psychology, cyberpsychology, human-computer interaction, and human-agent interaction, focusing on the experimental research approach.

Hong Yik Lim

Hong Yik Lim is an undergraduate student under supervision of Dr Siew Chein Teo in the Faculty of Business, Multimedia University, Melaka, Malaysia.