Factors Influencing Repurchase Intention in Drive-Through Fast Food: A Structural Equation Modeling Approach

The drive-through fast-food industry has been one of the fastest businesses growing over the past decades in developing countries, including the Philippines. The purpose of this study was to evaluate factors influencing costumers’ repurchase intention in a drive-through fast food in the Philippines by utilizing the structural equation modeling (SEM) approach. A total of 305 Filipinos answered the online questionnaire, which contained 38 questions. The results of SEM indicated that subjective appetite (SA) was found to have a significant direct effect on menu options (MO). Consequently, MO was found to have significant direct effects on imagery elaboration (IE), vividness (VV), and convenience (CO), and an indirect effect on order accuracy (OA). Finally, SA, MO, IE, VV, OA, and CO were found to have significant effects on satisfaction (S), which subsequently led to loyalty (L) and repurchase intention (RI). Interestingly, MO was found to have the highest indirect effect on RI, indicating that MO is an important consideration for RI. This is the first comprehensive study evaluating drive-through fast food in the Philippines. The causal relationships of the present study can be applied and extended to evaluate the repurchase intention of drive-through fast food in other countries.


Introduction
Drive through service has been an important part of foodservice business research. It is a common feature of fast-food restaurants (FFRs). The primary goal of these drive-through services is to provide fast and convenient service to the customer [1] while increasing the number of customers that can be served, as opposed to the traditional walk-in transactions. Usually, drive-through systems are located on the assigned drive-through lane, remotely from the building. Upon entering the drive-through lane, the customer stops his vehicle momentarily to choose an order from the food offered by the restaurant on the menu.
Thus, a menu acts as the first point of interaction for customers in the restaurants, providing cues about their impending encounter, while, simultaneously, establishing the restaurant's strategic marketing plan, personality, and brand identity [2,3]. A welldesignated menu can facilitate restaurants' sales by steering the attention of customers to a specific menu item offered. Thomas and Mills [4] mentioned that few studies empirically assessed how menu information gathered by restaurants influences the attitudes of the customers towards the restaurant, including their consequent behavioral intentions.
In the Philippines, the fast-food industry (FFI) has been one of the fastest-growing business sectors over the past decade. According to a report by the Philippine Statistics Authority (Table 1), FFI generated the highest income of PHP 203 billion, with 4411 establishments in 2016. Local brands have transformed FFRs as pioneers and expanded their franchises into other countries all around the world. [5]. Aside from Korea, the Philippines is the only country where McDonald's is not the leading and most successful company in the fast-food industry (FFI), as Filipino fast-food market (FFM) prioritized localization while retaining the prestige of foreign brands [5,6]. Jollibee is a local brand that operates similar to a Filipino version of McDonald's [6], providing both rice dishes, a daily meal for locals citizens, and fast food products typically sold in the United States. As of November 2019, Jollibee currently holds a dominating 58% share of the FFM in the Philippines, with a total of 1150 outlets nationwide, outnumbering its competitor, McDonald's, with more than 640 stores. Interestingly, part of Jollibee's success stems from imitating the innovative principles and techniques pioneered by McDonald's in the United States decades ago.
Despite the customers' familiarity with order selection, there is a significant lack of academic research in the Philippines that addresses the relative importance of attributes affecting repurchase intention in drive-through fast food. Bowen and Morris [7] mentioned that the representational quality of the products listed on a menu, as a unique type of marketing situation is, comparable to a professional speech delivery. In addition, McCall and Lynn [3] highlight the relative importance of the role of restaurant menus suggesting that the menu drives a process that might entice a diner to enter a restaurant. According to Namkung and Jang [7], increased competition in the global FFRs has made customer satisfaction more important in order to build customer loyalty and to enhance business performance. Further studies are required to verify the extent of what influences customer satisfaction and repurchase intentions in a drive-through setting.
This research aims to evaluate factors influencing costumers' repurchase intention in a drive-through fast food in the Philippines by utilizing the structural equation modeling (SEM) approach. Following some previous studies [8][9][10], SEM is used to analyze the causal Foods 2021, 10, 1205 3 of 14 relationships between measured variables and latent constructs. The causal relationships in this study can be applied and extended to evaluate the repurchase intention of drivethrough fast food in other countries. Figure 1 represents the theoretical research framework of the current study. The theoretical basis of the current model was derived mainly from Peters and Remaud's [11] and other several studies. The model is developed from the factors of vividness, subjective appetite, and imagery elaboration, as investigated by Lee and Kim [12], and from the customer perceptions analysis of service performance, carried out by Cronin and Taylor [13]. The current model also incorporates factors such as value for money [14], convenience [15], taste preference [16], introducing latent variables such as menu options and order accuracy as well.

Theoretical Research Framework
suggesting that the menu drives a process that might entice a diner to enter a restaurant. According to Namkung and Jang [7], increased competition in the global FFRs has made customer satisfaction more important in order to build customer loyalty and to enhance business performance. Further studies are required to verify the extent of what influences customer satisfaction and repurchase intentions in a drive-through setting.
This research aims to evaluate factors influencing costumers' repurchase intention in a drive-through fast food in the Philippines by utilizing the structural equation modeling (SEM) approach. Following some previous studies [8][9][10], SEM is used to analyze the causal relationships between measured variables and latent constructs. The causal relationships in this study can be applied and extended to evaluate the repurchase intention of drive-through fast food in other countries. Figure 1 represents the theoretical research framework of the current study. The theoretical basis of the current model was derived mainly from Peters and Remaud's [11] and other several studies. The model is developed from the factors of vividness, subjective appetite, and imagery elaboration, as investigated by Lee and Kim [12], and from the customer perceptions analysis of service performance, carried out by Cronin and Taylor [13]. The current model also incorporates factors such as value for money [14], convenience [15], taste preference [16], introducing latent variables such as menu options and order accuracy as well.  Meanwhile, the entire area of food intake, choice, motivation, and preference is covered by subjective appetite [17]. The sight of preferred or less preferred is sufficient enough to influence the customer's desire to eat [18]. Consequently, a study by Schulte-Mecklenbeck et al. [19] stated that customers who are aware of their food options commonly ignore complex information on the menu when ordering. Thus, we hypothesized that: Hypothesis 1 (H1). Subjective appetite of customers had a significant direct effect on the menu options offered by the drive-through. Menu options can be presented in many ways and are crucial for restaurants' branding because it is an internal advertisement component for informing customers about their impending dining experience [20]. Mitchell et al. [21], suggested that mental imagery and the manifestation of mental imagery have a positive correlation through cognitive style. In addition, the past study also stated that mental imagery exhibits a significant correlation with menu presentation [22]. Hence, customers can easily associate menu items with appetite if menus elicit a sensory response. Consequently, we hypothesize that: Hypothesis 2 (H2). Menu options had a significant direct effect on the imagery elaboration of the menu in drive-through.

Theoretical Research Framework
Menu designers claim that the elements of design (e.g., copy, color, paper, typeface layout, and so on) can draw the customers' attention to the items that restaurants want to sell [7]. Laying out the assigned item with a box, putting it in striking print, employing a larger sort estimate, and including a color photo are a few regularly prescribed ways to draw attention to the menu items, increasing sales, as a result [23]. If utilized wisely, specialists claim that these procedures can viably offer targeted items. Thus, we hypothesized that:

Hypothesis 3 (H3).
Menu options had a significant direct effect on the convenience of menus in drive-through.
The degree to which pictures have been reflected in one's intellect influences the conciseness of the stimuli [24]. Mental imagery has characterized as a mechanism by which tactile data are interpreted in one's intellect, commonly conveyed by vividness, imagery amount, and imagery elaboration [22]. Vividness is the accuracy with which the person encounters a picture [25], whereas imagery elaboration is the actuation of info within the generation of imaginative illustrations above what the stimulus provides [26]. The customer can interface prior encounters to the comparing information through picture handling and tactile experiences (such as scent, taste, sight, and other sensations) in working memory [24]. Thus, we hypothesized that: Hypothesis 4 (H4). Menu options had an influential direct impact on the vividness of menus in the drive-through.
Lee and Kim [12] initially attempted to study the effect of pictures and video menu presentation on the imagery perspective of the customers. Their study encompasses the relationship between vividness, imagery elaboration, imagery quantity, and subjective appetite. Hence, we proposed that:

Hypothesis 5 (H5). Vividness had a significant direct effect on the order accuracy in drive-through.
Convenience is defined as the decreased time and effort customers must devote to purchasing or utilizing items and services, with the help of innovation [27]. Coyle and Thorson [28] showed a high degree of vividness, contributing to a positive disposition towards the website. Correlating this information to our current study, we hypothesized that: Hypothesis 6 (H6). Vividness had a significant direct effect on the convenience of menu in drive-through.
Hodes [29] stated that, in the context of learning instructions, imagery instructions appear to be effective in absorbing the information. Being related to order selection, Beldona et al. [30] suggested that an elegant menu design can cause customers to draw their attention to a certain item in the menu. Imagery elaboration enables the conveyance of information in the development of visual representations beyond what the stimulus offers [25]. The relationship between imagery elaboration and subjective appetite has been studied based upon the elaboration-intrusion (EI) theory of desire [31]. The study of Schumacher et al. [32], which stated that craving triggers of the elaboration-intrusion theory (e.g., picturing myself having the food) were significantly positive in showing craving intensity. FFRs present an image of their product to capture their customers' attention [31]. The sensory stimuli (e.g., taste, look, and smell) are the primary thoughts of the customers while ordering the food that they want, which also become the basis for confirming accuracy after receiving the food [32]. Hence, we hypothesized the following: Hodes [29] stated that, in the context of learning instructions, imagery instructions appear to be effective in absorbing the information. Being related to order selection, Beldona et al. [30] suggested that an elegant menu design can cause customers to draw their attention to a certain item in the menu. Imagery elaboration enables information in the development of visual representations beyond what the stimulus offers [25]. The relationship between imagery elaboration and subjective appetite has been based on the elaborated intrusion (EI) theory of desire [31]. The study of Schumacher et al. [32], which stated that craving triggers the elaboration-intrusion theory (e.g., picturing myself having the food) was significantly positive in showing craving intensity. FFRs present an image of their product to capture their customers' attention [31]. The sensory stimuli (e.g., taste, look, and smell) are the primary thoughts of the customers while ordering the food that they want, which also become the basis for confirming accuracy after receiving the food [32]. Hence, we hypothesized the following: Hypothesis 7 (H7). Imagery elaboration had a significant direct effect on order accuracy.
Handling customer satisfaction in terms of order fulfillment is dependent on how the organization meets customer requirements. [33]. Order accuracy involves all operations starting with the customer's purchase decision until the customer receives its order and the quality of the products are fully satisfied [34]. Thus, we hypothesized the following: Hypothesis 8 (H8). Order accuracy had a significant direct effect on satisfaction.
According to Yale and Venkatesh [35], convenience has been broken down into time efficiency, ease of access, applicability, and avoidance of dissatisfaction. The positive views of customers toward the services and the relevance of a food company provide an increase in their satisfaction [36]. The study of Cheng et al. [37] has mentioned that convenience enhances customer satisfaction. Therefore, we hypothesized that: Hypothesis 9 (H9). Convenience had a significant direct effect on satisfaction.
Namin [10] suggests that having a high level of service quality will gain satisfied customers, leading to having more loyal customers. According to Oliver [38], loyalty is described as the willingness of customers to continuously re-buy and re-patronize a chosen product or service. The previous study by Belaid and Behi [39] also mentioned that a powerful brand-loyalty predictor is satisfaction. Therefore, we proposed that: Hypothesis 10 (H10). Satisfaction had a significant direct effect on loyalty.
The previous study by Yi and La [40] considered the characteristics of loyal customers by evaluating loyalty as a high proportion of the same choice of brand, a high motive of positive word-of-mouth, and a high intention to repurchase. Behavioral loyalty is defined purely as paying back the goods or services [41]. The study by Curtis et al. [41] has indicated that loyalty and repurchase intention have a strong positive relationship. Thus, we proposed that: Hypothesis 11 (H11). Loyalty had a significant direct effect on repurchase intention.

Participants
Following Akram et al. [9], the research instrument of this study is in the form of an online questionnaire. The chosen population consisted of those people who have had a recent experience of service in FFRs drive-through in the Philippines. A total of 305 participants willingly answered the survey questionnaire, which contains 38 questions.

Questionnaire
A self-administered questionnaire was developed to measure the perceptions of the customers about their recent drive-through experience in fast food. A convenience sampling method was utilized to distribute the questionnaire. The questionnaire consists of 11 sections and is supported by some previous studies: (1) demographic information (gender, age, time of visit, and so on); (2) value for money; (3) subjective appetite; (4) menu options; (5) vividness; (6) imagery elaboration; (7) convenience; (8) order accuracy; (9) satisfaction; (10) loyalty; and (11) repurchase intention ( Table 3). All latent constructs were measured using 5-point Likert. The imagery shown in the menu board was detailed. V3 The imagery shown in the menu board was vivid.

CO1
The menu board was easy to read.

Rydell et al. [15] CO2
The food items were easy to locate. CO3 The menu board is located within my eye level.
Order Accuracy (OA) OA1 I received the meal I ordered correctly. OA2 The staff repeats my order for recap. -OA3 I received the meal I ordered completely.
Satisfaction (S) S1 The quality of food was excellent. Mcneil

Structural Equation Modeling
Structural equation modeling (SEM) is a progressive multivariate procedure that focuses on the interrelationships among constructs [49]. We utilized AMOS 22 with a maximum likelihood estimation approach to derive the model. Figure 1 demonstrates that our SEM construct had nine latent variable constructs, including one exogenous latent variable (SA) and seven endogenous latent variables (MO, IE, VV, OA, CO, S, L, RI).
The model fit of the SEM was measured by six measures: goodness of fit index (GFI), adjusted goodness of fit index (AGFI), root mean square error of approximation (RMSEA), incremental fit index (IFI), Tucker-Lewis index (TLI), and comparative fit index (CFI). A value greater than 0.80 for GFI and AGFI is the minimum indication of good model fit [50]. In addition, a value lower than 0.07 is a good indication of an acceptable model for RMSEA [49]. Finally, a value greater than 0.90 for IFI, TLI, and CFI is an indication of a good model fit [49]. Figure 2 represents the SEM for evaluating factors influencing customers' repurchase intention in drive-through fast food, in the Philippines. In addition, Table 4 represents the construct validity and reliability of the model, which were calculated based on Cronbach's α, average variance extracted (AVE), and composite reliability (CR). Based on this figure, all hypotheses were significant, indicating that our model construct was acceptable. Hair [49] mentioned that all of the factor loadings should be statistically significant. An ideal standardized factor loading should be higher than 0.5, and ideally 0.7 or higher [49]. If the factor loadings are lower than 0.7, they can still be considered significant; however, more variances are found in the measure than that of the explained variance [49], such as SA1 (λ:0.61), SA2 (λ:0.68), MO1 (λ:0.61), MO2 (λ:0.54), MO3 (λ:0.58), and IE3 (λ:0.66). Subsequently, the results showed that VV has a significant effect on OA (β: 0.28, p = 0.001) but an insignificant effect on CO (β: 0.10, p = 0.46). An interpretation of this result is that customers would use the menu presentation as a basis for checking the accuracy of their orders by checking the correctness and completeness of their food choices. Managers who wish to enhance their order accuracy should ensure that their staff is accurately deliv-  As presented in Table 5, the GFI and AGFI values were 0.879 and 0.851, respectively, indicating that the model passed the minimum cutoff of the model fit [50]. In addition, the RMSEA value was 0.055, indicating that it was lower than the suggested cut-off of 0.07. IFI, TLI, and CFI values were greater than the suggested cutoff of 0.90, indicating that the hypothesized construct of the specified model reflected the observed data very well. Finally, the direct effect, the indirect effect, and the total effect are presented in Table 6 as the basis for the discussion particularly related to the causal relationships among latent variables.

Discussion
In the Philippines, the fast-food industry (FFI) has been one of the fastest-growing business sectors, over the past decade. The present study applied the structural equation modeling (SEM) approach to examine the causal relationships between subjective appetite (SA), menu options (MO), vividness (VV), imagery elaboration (IE), convenience (CO), order accuracy (OA), satisfaction (S), loyalty (L), and repurchase intention (RI) of drive through fast food in the Philippines. An online survey was conducted during this study, and a total of 305 respondents were participated. SEM indicated that SA (β: 0.58, p = 0.001) had a significant direct effect on MO, which makes several implications in the customers' choices when ordering in a drive-through. While visiting a drive-through restaurant, hunger and a strong desire to eat, positively affect customers' choice of food from the menu offered by the restaurant. Hence, management should take note that a well-designed menu should cater to all customers' food preferences.
Regarding the impact of menu options, the current model indicated that MO had a significant effect on VV (β: 0.64, p = 0.001), CO (β: 0.74, p = 0.001), and IE (β: 0.79, p = 0.001), which can be interpreted in several ways. First, customers can easily find a suitable food option when the menu offered by the restaurant is clear and easy to read. Second, customers find it easier to make their food choices when the menu is located within their eye level. Third, customers' food choices can be based on the menu's representational quality, as it can attract customers to order food in drive-through by eliciting a sensory response. These findings were supported by the study of Bowen and Morris [7], who mentioned that the customers' attention towards food items can be attracted by the menu presentation. Thus, management should develop a menu that can capture the customers' attention, leading them to visit and order in their drive-throughs. Despite the results from most recent studies [10,13], MO was found to have the highest indirect effect on RI (β: 0.42, p = 0.001), indicating that MO is one of the most important considerations for RI. Based on the results, customers who find drive-throughs that offer more options suited for them are most likely to continue visiting them in the future.
Subsequently, the results showed that VV has a significant effect on OA (β: 0.28, p = 0.001) but an insignificant effect on CO (β: 0.10, p = 0.46). An interpretation of this result is that customers would use the menu presentation as a basis for checking the accuracy of their orders by checking the correctness and completeness of their food choices. Managers who wish to enhance their order accuracy should ensure that their staff is accurately delivering the customers' orders based on the quality of their menu's presentation.
Regarding satisfaction, SEM indicated that S was significantly affected by OA (β:0.45; p = 0.001) and CO (β:0.56; p = 0.001). The result implied that customers are most satisfied when the image shown on the menu board is clear and when customers receive their orders completely. Moreover, S was found to have a significant direct effect on L (β:0.77; p = 0.001). According to Carpenter and Fairhurst [51], the success of the company's technique depends on its capacity to fulfill its promises to customers, which leads to building a long-term beneficial relationship. Based on the result, management should provide food with excellent quality in order to make customers satisfied with their visit to the drive-through, saying positive things about the restaurant, as a result. Thus, when customers are satisfied with their recent experience in drive-through services, they intend to encourage other people to visit the drive-through as well.
Lastly, the SEM proved that there was a positive interrelation between L and RI (β:0.81; p = 0.001), which favors the previous studies' theoretical research framework. The desire to purchase again [52] and the desire to indulge in good referral are two ways of ensuring consumers' intention to buy again. Excellent service would develop better word-of-mouth and continuously improve the customers' experience. Based on the results of the current study, customers who are willing to say positive things and recommend drive-through to other people are most likely to continue visiting the drive-through in the future, regardless of how often they visit.

Managerial Implications
The results of the research would help Philippine fast-food restaurants to assess the factors influencing customers' repurchase from their drive-throughs. First, the researchers discussed the significant effect of SA on MO in a drive-through fast food. The results showed the willingness of the customers to buy food from the fast-food since it is inexpensive, and the food quality is acceptable. Hence, different drive-through fast food restaurants must ensure that the menu is attractive to the customers. Providing a well-designed menu will entice customers and excite them to order. Secondly, the importance of MO can be recognized by the positive effect on VV, CO, and IE for drive-through fast-food customers. The finding provides the insightful implication of imagery and vividness to improve the design of the menu. Customers much prefer ordering in a drive-through fast food because it is more convenient and inexpensive, and it tastes good. Companies who are not yet familiar with enhancing their menu should invest time in upgrading their menu. Developing menu designs is one of the most critical strategic campaigns of a restaurant to keeps the brand fresh in the customer's mind and to increase profitability while also maintaining a competitive edge.

Limitations and Future Research Direction
Although this research provides important evidence of the factors influencing customers' repurchase intention in drive-through fast food in the Philippines, the authors would like to acknowledge several limitations on the SEM of the current study. First, the data from the current study were collected from Filipino customers who had a recent experience in drive-through. Thus, it may not be generalized for all nationalities because of their different cultures. In other words, customer motives can differ based on their taste preference, and other countries offer different menus to cater to their local palates. Future work should cross-validate the existing paradigm for new ideas into how cultural influences regulate hierarchical interactions in different ways and across countries. Second, the results from the derived SEM could have different results for electronic, digital menus with varying interactive features, whereas the current study only focused on traditional menus [12]. Future studies should also incorporate food quality and customers' complaints as one of the key constructs in the current SEM model to reveal more information regarding their interrelationship with loyalty and satisfaction. Furthermore, when evaluating the role of intentions as a connection between satisfaction and repurchase of actions, future studies are recommended to consider two specific intention constructs, such as intentions as expectations and intentions as preferences to the current SEM. Finally, there are several additional factors that could probably enhance satisfaction such as Wi-Fi services [53], cleanliness, and even payment options (cash, debit card, and e-money). Future research can add these additional factors in the model.

Conclusions
Over the past decades, drive-through fast-food industry has been one of the fastest growing businesses in the Philippines. According to the Philippines Statistics Authority, the fast-food industry has generated the highest income of PHP 203 billion with 4411 establishments in 2016. The purpose of this study was to evaluate factors influencing costumers repurchase intention in a drive-through fast food in the Philippines by utilizing the structural equation modeling approach [54][55][56][57][58][59]. A total of 305 Filipinos answered the online questionnaire, which contained 38 questions. The results of SEM indicated that subjective appetite (SA) was found to have significant direct effects on menu options (MO). Consequently, MO was found to have significant direct effects on imagery elaboration (IE), vividness (VV), and convenience (CO), and an indirect effect on order accuracy (OA). Finally, SA, MO, IE, VV, OA, and CO were found to have significant effects on satisfaction (S), which subsequently lead to loyalty (L) and repurchase intention (RI). Interestingly, MO was found to have the highest indirect effect on RI, indicating that MO is one important consideration for RI. The results of the present study can be applied and extended to evaluate the repurchase intention of drive-through fast food in other countries.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author.