The influence of quality on satisfaction and customer loyalty with an importance-performance map analysis: Exploring the mediating role of trust

Rocío Carranza (University of Castilla-La Mancha, Ciudad Real, Spain)
Estrella Díaz (University of Castilla-La Mancha, Ciudad Real, Spain)
David Martín-Consuegra (University of Castilla-La Mancha, Ciudad Real, Spain, and Universidad Autónoma de Chile, Santiago, Chile)

Journal of Hospitality and Tourism Technology

ISSN: 1757-9880

Publication date: 1 October 2018

Abstract

Purpose

The purpose of this study is to verify the existence of loyalty among fast-food customers and its dependence on fast-food service quality, comprising service quality, food quality and store atmosphere. This study also examines the direct and mediating role of constructs such as satisfaction and trust in creating loyalty in fast-food restaurants (FFRs).

Design/methodology/approach

A sample of 456 fast-food customers was collected using a structured questionnaire. This paper uses partial least squares path modeling to test and validate the study’s research model and hypotheses.

Findings

The results suggest that fast-food service quality has a positive influence on satisfaction, trust and loyalty among fast-food customers. The findings also reveal a mediating effect of trust (partial mediation), increasing the effect of satisfaction on loyalty.

Practical implications

This study reinforces the importance of considering the attributes that influence customer loyalty. Specifically, food quality is considered key to increasing loyalty among FFR customers.

Originality/value

This study proposes an integrated model influenced by three factors that contribute to fast-food service quality (i.e. food quality, service quality, atmosphere) along with classical variables used in the marketing literature (i.e. satisfaction, trust) in the creation of FFR loyalty. This study also follows modern procedures in PLS-SEM by challenging conventional methods.

研究目的

本论文旨在验证快餐消费者的忠诚度及其与快餐质量之间的关系, 其中质量包括服务质量、食物质量、和饭店氛围。本论文还旨在检验满意度和信任度在快餐饭店(FFRs)的消费者忠诚中的直接和间接作用。

研究设计/方法/途径

本论文采用结构问卷采样形式, 456位快餐消费者为问卷样本。本论文采用偏最小二乘回归(PLS)的路径建模数据分析方法来分析和验证模型及假设。

研究结果

快餐服务质量对于快餐消费者的满意度、信任度、和忠诚度有着积极促进作用。此外, 本论文发现信任度对于增加满意度对忠诚度的影响有着中介调节作用(部分中介调节)。

研究实践意义

本论文重新提出了影响消费者忠诚度的各种因素。具体是, 食物质量是增加FFR消费者忠诚度的重中之重。

研究原创性/价值

本论文提出一体化模型包括三种快餐服务衡量因子(食物质量、服务质量、环境)以及营销文献中常用变量(满意度、信任度), 来检验FFR忠诚度。本论文还挑战传统研究方法, 采用PLS-SEM现代程序来进行分析研究。

关键词

质量、满意度、信任度、忠诚度、快餐饭店(FFRs)、PLS-SEM

Keywords

Citation

Carranza, R., Díaz, E. and Martín-Consuegra, D. (2018), "The influence of quality on satisfaction and customer loyalty with an importance-performance map analysis", Journal of Hospitality and Tourism Technology, Vol. 9 No. 3, pp. 380-396. https://doi.org/10.1108/JHTT-09-2017-0104

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


1. Introduction

The demand for fast-food services and fast-food restaurants (FFRs) has been growing steadily across the world. This growth has been driven by a series of factors that attract the attention of the customer, such as the food’s intense flavor, the ease and speed with which it is acquired and its low acquisition cost (Namin, 2017). However, the service provided by a fast-food restaurant chain is limited to counter service and requires the customer to participate in the creation of their own experience (Bufquin et al., 2017). Another concern for FFRs is that some of the world’s largest fast-food brands, including McDonald’s and Burger King, have struggled to maintain their sales numbers and customer counts. These companies have lost some of their appeal over the last few years, as the economy has improved and people now have more dining options(Oches, 2015). Moreover, new generations of consumers are looking for value and authenticity in their dining experience, so traditional FFRs are having a hard time keeping up. Thus, large multinational chains such as McDonald’s have devoted their marketing efforts to increasing loyalty strategies and improving their commitment to providing service quality(Ryu et al., 2012).

Several researchers have indicated that the FFR industry is closely associated with behavioral intentions and related constructs, such as service quality, perceived value, corporate image and customer satisfaction (Qin and Prybutok, 2009). Some studies have examined restaurant service quality, using different scales to measure this concept and its implications. However, they have been insufficient for the examination of fast-food service quality because restaurant service quality should be considered to be an essential strategy in studies related to satisfaction and loyalty in FFRs (Wu and Mohi, 2015). Customer trust in FFRs is another important element. This concept could be considered a precursor of loyalty in the fast-food industry (Stathopoulou and Balabanis, 2016) and must thus be considered a necessary mediating variable in the relationship between satisfaction and loyalty. To fill these gaps in the literature, this study makes a twofold contribution. First, it contributes to the marketing literature by providing a better understanding of the components of fast-food service quality and its effect on satisfaction and behavioral intentions such as loyalty. Second, this study conceptualizes and measures the mediating effect of consumer trust in the relationship between satisfaction and loyalty in the FFR industry.

Partial least squares structural equation modeling (PLS-SEM) approach is particularly useful when the study’s focus is on multidimensional constructs. This study provides insight regarding the use of composites and common factors in the same model, as well as the alternation of the use of PLS and consistent PLS. Specifically, fast-food service quality is proposed in this research as an operationalized multidimensional construct (composite model estimated in mode B). Furthermore, this research illustrates the use of the Importance-Performance Map Analysis (IPMA) to extend the standard results reporting on path coefficients of each of the elements in the multidimensional construct of fast-food service quality. The goal is to identify components that have relatively high and low importance for the target construct (fast-food service quality).

The rest of this paper is organized as follows. First, it reviews the study’s conceptual framework and presents its research model and hypotheses. Second, it describes and justifies its empirical methods. Next, the paper reports the study’s findings and explains them. Finally, it offers a conclusion and examines the study’s implications.

2. Literature review and hypothesis development

2.1 Fast-food service quality and customer satisfaction

Fast-food service quality is an essential strategy for restaurant success (Wu and Mohi, 2015) and is used to measure customers’ perception of service experiences derived from intangible factors (i.e. service interaction) and tangible elements (i.e. food or atmosphere) (Ryu et al., 2012). Several researchers have developed scales to measure service quality in several contexts, considering a combination of tangible and intangible elements. Thus, the SERVQUAL instrument was developed, consisting of five service quality dimensions: tangibles, reliability, responsiveness, assurance and empathy. The DINESERV scale measures restaurant customers’ perceptions and expectations of service quality attributes. To expand upon the research on the tangible aspects of the restaurant industry, the TANGSERV scale was developed. Later, a new version of TANGSERV was developed, consisting of five dimensions: food and service, staff, ambiance and social aspects, cleanliness and accessibility. Finally, the DinEX scale was proposed (Bufquin et al., 2017). Consequently, elements such as service quality, food and atmosphere are considered vital to the study of fast-food service quality (Ryu et al., 2012). Additionally, previous research shows that the right combination of tangible and intangible elements affects the customer’s perception of fast-food service quality, as well as satisfaction and favorable behavioral intentions.

Customer satisfaction is defined as the assessment a customer makes of a characteristic of a product or service and that provides a positive result derived from consumption (Ali, 2016; Oliver, 1999). This concept has been examined in the restaurant industry as a consequence of service quality. Several studies have found important evidence in the relationship between several aspects of fast-food service quality (i.e. service quality, food quality, atmosphere) and consumer satisfaction with FFRs (Bufquin et al., 2017; Wu and Mohi, 2015). On one hand, perceived service quality is defined as a customer’s judgment of the overall excellence or superiority of a service (Ryu et al., 2012). In line with this definition, Liu and Jang (2009) examined service quality and found that friendly and helpful employees were positively related to customer satisfaction. On the other hand, a growing number of studies have examined food quality and its influence on customer satisfaction. Liu and Jang (2009) found that food taste was the most important attribute influencing customer satisfaction. In a study by Ryu et al. (2012) on the relationships between three dimensions of quality (i.e. food, service, physical environment), price, satisfaction and behavioral intention in quick-casual restaurants, the findings showed that the quality of food and service quality were significant determinants of customer satisfaction. Thus, food and service quality have been considered as some of the most significant aspects in customer assessments of fast-food service quality and a key factor in increasing customer satisfaction (Wu and Mohi, 2015).

Finally, many researchers have studied how restaurants’ perceived atmosphere affects their guests’ satisfaction. For example, Liu and Jang (2009) found that interior design, décor and aroma significantly influenced restaurant customer satisfaction. Some studies have claimed that an establishment’s atmosphere is important for attracting customers because they occasionally go to restaurants merely for the perceived atmosphere (Ryu et al., 2012). Therefore, fast-food service quality, comprising service quality, food quality and atmosphere, has a positive influence on FFR customer satisfaction(Clemente-Ricolfe, 2016). Based on the literature, the following hypothesis is proposed:

H1.

Fast-food service quality has a positive influence on customers’ satisfaction.

2.2 Customer satisfaction and loyalty in fast-food restaurants

The operational and financial success of companies largely depends on the extent to which favorable loyalty or return intentions are expressed by customers. This is why it is crucial to know which fast-food service quality components influence customers’ return intentions most strongly. The relationship between customer satisfaction and loyalty has been verified in the literature. For instance, when measuring the influence of food quality on customers’ satisfaction and behavioral intentions in five mid- to up-scale restaurants, Namkung and Jang (2007) found a significantly positive relationship between customers’ satisfaction and behavioral intentions.

In recent years, several researchers have emphasized and validated the relationship between customer satisfaction and loyalty (Bufquin et al., 2017; Wu and Mohi, 2015). Although other psychological processes may be associated with consumer loyalty, it has been shown that satisfaction has a significant and positive effect on repurchase intentions and on positive cognitive attitudes toward the product or service (Sahagun and Vasquez-Párraga, 2014). Based on these arguments and the previous literature, it is assumed that the greater the consumer’s satisfaction, the greater their loyalty will be; thus, the following hypothesis is established:

H2.

Customer satisfaction has a positive influence on customer loyalty to fast-food restaurants.

2.3 The role of trust as a mediating variable in fast-food restaurants

Trust has been defined primarily as an exchange relationship between the company and the customer, in which it is important to conserve the investments made and maintain belief among those involved (Garbarino and Johnson, 1999). Several studies have confirmed that trust is the result of a progressive generation of customer satisfaction with a product, service or restaurant (Aldás et al., 2010).

In addition, it has been found that trust is an important, basic precursor of loyalty: the customer becomes loyal to a brand as a result of a trust that has been generated. Thus, trust is prior to loyalty and contributes to the belief that those involved are not moved by opportunism (Garbarino and Johnson, 1999). In this sense, trust can be interpreted as a customer’s relative attitude towards the restaurant or brand, causing a latent loyalty or an attitudinal loyalty relationship (Stathopoulou and Balabanis, 2016). Several studies have confirmed that satisfaction and trust are precursors of loyalty among fast-food customers (Sahagun and Vasquez-Párraga, 2014). Such customers do not become loyal merely due to product satisfaction; trust is a necessary mediating variable in the relationship between satisfaction and loyalty. Therefore, the following hypothesis is proposed:

H3.

The relationship between customer satisfaction and loyalty in FFRs is mediated by trust.

3. Methodology

3.1 Data collection

A self-administered survey was used to examine the proposed relationships between fast-food service quality as formed by several attributes (i.e. quality food, quality service, atmosphere), satisfaction, trust and loyalty among fast-food customers. The study focused on Spain due to the recent increase in fast-food restaurant franchises in that country. Some figures show that the number of Spain’s fast-food restaurants increased by 24 per cent from 2010 to 2016, reaching a total of 2,807. In addition, consumer behavior around FFR visits has changed in the country. A quarter of the Spanish population visited FFRs in 2016, far exceeding the 2015 figure (Statista, 2017). A questionnaire was distributed by researchers to customers of well-known fast-food restaurant chains in central Spain, located in a metropolitan area close to a university campus and train station. The restaurants where the survey was distributed have an average meal price of €5 to €10and a maximum seating capacity of 100. The data were collected over a period of one month during normal restaurant operating hours, at different times of the day (i.e. breakfast, lunch, afternoon snack, dinner), from Monday to Sunday to try to balance the sample and ensure its representativeness.

Upon the respondents’ arrival, the researchers asked them if they would take the survey when they had finished eating. They were also asked if they had done the survey earlier; if they had, they were not allowed to complete the survey again to ensure the independence of observations. To reduce single-respondent bias, the questionnaire was focused on individuals who visited FFRs more or less frequently, since their information is more reliable than that given by individuals who are not fast-food customers. In addition, the comprehensive data collection process and extensive sampling minimized the risk of single-respondent bias (Kumar, 2015). Following Podsakoff et al. (2012), to reduce common method bias risk related to a single respondent for dependent and independent constructs, the questionnaire was conducted in different groups and sequences. The independent variables were provided at the beginning of the interview and the dependent variables at the end. First, items related to service quality, food quality and atmosphere elements were presented to individuals. The second set of questions measure customer satisfaction, trust and loyalty through the statements related to each.

Once the survey had been finished, the researchers checked whether the questionnaires were complete and tried to determine if the respondents had any doubts when completing the form. Of a total of 500 customers, 456 surveys were completed and included in the analysis, for a response rate of 91.2 per cent. Table I presents the customer profile of the sample. The sample consists of 44 per cent men and 56 per cent women. The largest age range comprises individuals between 34 and 43, representing exactly 33 per cent of the sample. However, the cumulative percentage of respondents up to 33 years of age is 53.7 per cent; thus, the sample comprises mainly young and middle-aged individuals. In addition, 45.3 per cent of the sample is highly educated. Regarding the consumption characteristics,39.3 per cent of the respondents visited this type of establishment once a month, 25.4 per cent once a week, 3.5 per cent several days a week and 2.2 per cent every day; thus, much of the sample is made up of regular fast-food customers, with an average expenditure of €6 to €12 per person. Most restaurant visits (69.8 per cent) take place during the weekend. Most visits occur in groups, especially families (representing 51.5 per cent of the sample).

3.2 Measures

The instrument used to measure the variables of this study is a structured questionnaire. All the variables included in the study were measured with multi-item scales validated by other researchers, using a Likert scale ranging from 1 to 5. Variables for service quality, food quality, atmosphere, and the general assessment of the attributes were based on Wu and Mohi (2015). The satisfaction variable was measured using the instrument developed by Oliver (1999) and updated by Stathopoulou and Balabanis (2016). Eleven items adapted from Aldás et al. (2010) were used to measure trust. Loyalty was measured using a six-item instrument adapted from Dick and Basu’s (1994) scale, combining attitudinal and behavioral perspectives.

To adapt the study to the Spanish context, the scales of service quality, food quality, atmosphere, satisfaction and loyalty went through a process of translation and synthesis to ensure that there were no errors of interpretation. To that end, the following procedures were followed in accordance with Martín-Consuegra et al. (2015). First, a direct translation of the original scales was carried out with the collaboration of two independent bilingual translators. After this first translation, two other translators carefully compared two independent questionnaires to find translation errors. Finally, a bilingual translator performed the reverse translation into the original language (English). Thus, it is possible to compare and examine the equivalence and accuracy of the translated questionnaire.

3.3 Partial least squares structural equation modeling analysis

The model estimation was performed using PLS-SEM, structural equation techniques based on variances (Rigdon et al., 2017).PLS-SEM was selected primarily because:

  • This study uses composite models (Mode B; Henseler, 2017; Rigdon, 2016).

  • The research model is a complex one given the type of relationships hypothesized (direct and mediation) and the levels of dimensionality (first-order and second-order constructs; Ali et al., 2018).

According to Henseler (2017), the type of construct determines the measurement. If we have behavioral constructs, these variables are more likely to be measured using common factor models, while the design of the constructs (artifacts) is likely to be measured by composites. This study considers the fast-food service quality construct as a multidimensional construct, operationalized through a composite model. This theoretical construct consists of three different constructs (service quality, food quality, atmosphere) that are not correlated; therefore, the model’s composites are considered using mode B. Likewise, considering to their behavioral nature, satisfaction, loyalty and trust are operationalized as common factors(Rigdon, 2016, Sarstedt et al., 2016).

This study uses Smart PLS 3.2.7 software alternating between the use of PLS to estimate the part of the study established as a model composite mode B (fast-food service quality: service quality, food quality and atmosphere) and the consistent PLS for common factors (satisfaction, trust and loyalty; Sarstedt et al., 2016). Due to the multidimensional nature of the fast-food service quality variable, a two-step approach was used to estimate the model, consisting of a construction method using latent variable scores. In the first step, the aggregate scores of the first-order dimensions were estimated. Then, these aggregate scores were used to model the second-order construct (Wright et al., 2012). Finally, following the PLS-SEM analysis literature, a two-step approach was followed, in which the measurement model was developed and evaluated, and then, separately, the structural model was developed and evaluated (Hair et al., 2016).

4. Results

4.1 Measurement model

The first step carried out is the approximate assessment of the model fit. Thus, the standardized root mean square residual (SRMR) of the proposed model is calculated (see Figure 1) as the root mean square discrepancy between the correlations observed and the model’s implied correlations. The results of 0.078 (estimate model) and 0.045 (saturate model), indicate an appropriate fit, given the accepted 0.008 cutoff point (Henseler et al., 2016).

Then, before estimating the aggregate scores of the first-order dimensions, an assessment of the measurement model is conducted, differentiating between the variables treated as composites and those treated as common factors. The reliability presented by the measurement scales for constructs considered as common factors (satisfaction, trust and loyalty)are analyzed (Henseler, 2017).To assess the reliability of each construct, the composite reliability index (CRI)and Dijkstra–Henseler’s rho (ρA) are calculated. The CRI is higher than 0.7 for all components (Hair et al., 2016). Likewise, ρA exceeds 0.7 in all cases, confirming its reliability (Dijkstra and Henseler, 2015). For Likert-type scales with seven positions or fewer, Cronbach’s alpha underestimates reliability; consequently, its use is not recommended (Gadermann et al., 2012). Table II shows a generally high degree of internal consistency among the constructs. To evaluate convergent validity, three criteria are analyzed. First, the size of the loadings is examined, followed by the average variance extracted (AVE) and, finally, the significance of the indicator loadings. Although not all loadings are greater than 0.70, only the LOY1 indicator is removed because it is too low compared to the rest (close to 0.7). Additionally, eliminating this indicator increases the AVE value of the loyalty variable. In all cases, the AVE is higher than 0.5, establishing that more than 50 per cent of the construct’s variance is due to its indicators (Hair et al., 2016). Finally, the significance of the loadings is determined through the bootstrapping resampling procedure (5,000 subsamples of the original sample) to obtain the t-statistic values (Hair et al., 2016). In this case, all indicators are significant, with a 99.9 per cent confidence level.

The Fornell–Larcker criterion is used to contrast discriminant validity values (Hair et al., 2016). The results indicate a satisfactory level of discriminant validity (see Table III). Henseler et al. (2015) propose the evaluation of the heterotrait–monotrait ratio (HTMT). This criterion tends to be more demanding and entails a stricter evaluation than the previous criteria. This measure establishes the ratio of heterotrait–monotrait correlations, with discriminant validity confirmed when the values are less than 0.90 (Hair et al., 2016). This value is surpassed by the latent variables in the model estimated as common factors; hence, it is concluded that the measures of this study show sufficient evidence of reliability and of convergent and discriminant validity.

Finally, the validity of the measuring instrument is assessed for the constructs of general quality, the composite of service quality, food quality and atmosphere, estimated as composite model mode B. In this case, as they do not need to be correlated (see Table IV), the traditional evaluations of reliability and validity are considered to be inapplicable. Thus, an assessment of the possible existence of multicollinearity is conducted, as well as an assessment of the magnitude of the weights and their significance (Hair et al., 2016). First, the collinearity of indicators estimated in mode B is analyzed with the aim of observing the absence of correlation between items. This is evaluated using the variance inflation factor (VIF) and the significance levels of the weights. The VIF should be less than 3.3 (Henseler et al., 2016) to confirm that there are no collinearity problems. Afterwards, we study the weights and their significance through a bootstrapping process (see Table V). All weights are significant except FOQ3, ATM1, ATM3 and ATM7. Since there are no collinearity problems and the loadings are higher than 0.5, the indicators are maintained despite being non-significant (Hair et al., 2016).

4.2 Structural model

After analyzing the measuring model and verifying its validity and reliability, the proposed structural model was examined (see Table VI). Similarly, the significance of the estimated structural coefficients is verified using bootstrapping. The model’s explanatory capacity is evaluated using the R2 value, which reflects the explained variance of the dependent constructs. The model explains 46.4 per cent of the variance in satisfaction, 45.8 per cent in trust and 54.7 per cent in loyalty. Analyzing the decomposition of variance, of the 54.7 per cent of the explained variance of loyalty, 24.1 per cent is due to trust and 30.6 per cent to satisfaction. Both relationships are significant, but the contribution of satisfaction to loyalty acquires greater weight than its contribution to trust.

After evaluating and confirming the predictive relevance of the proposed model, we proceed to study the size of the effects (f2; Hair et al., 2016).The results show that, as stated, if the assessment of the overall quality of the restaurant attributes is eliminated, a large effect on satisfaction will occur (f 2 = 0.866), significant at 99.9 per cent. Similarly, if satisfaction is eliminated, a large effect on trust will emerge (f2 = 0.846) as well as a moderate effect on loyalty (f2 = 0.235), both with a significance level of 99.9 per cent. However, it is important to highlight that if confidence is eliminated, it produces a moderate effect on loyalty (f2 = 0.157) at a 95 per cent significance level, but this is lower than the effect produced by eliminating satisfaction.

To analyze the significance of the structural relationships, the path coefficients and their corresponding significance levels are calculated (see Table VI). To do this, it is necessary to verify significance through the t-values and the strength of the relationships. Overall quality appears to be positive and significant, at 99.9 per cent in FFR customer satisfaction (β = 0.681/t = 18.510). Thus, H1 is supported, as food quality is the attribute with the greatest weight in the generation of satisfaction. As proposed in H2, satisfaction is positively associated with loyalty (β = 0.444/t = 5.305). To reinforce the valuation of statistical significance by means of bootstrapping, the confidence intervals are reported together with the t-values (see Table VI). If a confidence interval does not include a zero value for an estimated path coefficient, the null hypothesis is rejected by accepting the proposed hypothesis (Henseler et al., 2016). Therefore, hypotheses H1 and H2 are accepted by the percentile method.

The last step in the model assessment is to evaluate the influence of trust as a mediating variable between satisfaction and loyalty (see Figure 2). First, the indirect effect is calculated (ab). To test the indirect effects, following studies such as Cepeda et al. (2018), the percentile bootstrap and bias-corrected bootstrap is calculated. To determine the effect of mediation, it is necessary to evaluate the significance of the indirect effect. The valuation of significance is done through the confidence intervals (CI), where 0 should not be included to be significant (see Table VII). In this case, the indirect effect is significant, confirming the mediation of trust between satisfaction and loyalty in FFRs (H3 is accepted). After the indirect effect is determined, it is necessary to study the significance of the direct effect (c’) to know whether the mediation is full or partial. Thus, the significance of the direct effect is evaluated without including the mediating variable between satisfaction and loyalty; this is significant at 99.9 per cent (β = 0.444/t = 5.305), thus confirming H2. As both effects are significant, a partial mediation relationship is established. To confirm this result, we calculate the Variance Accounted For (VAF) that determines the size of the indirect effect on the total effect, obtaining a VAF of 0.36 and showing partial mediation according to the established mediation criteria (VAF > 0.8 = full mediation; 0.2 ≤ VAF ≤ 0.8 = partial mediation; VAF < 0.2 = no mediation; Hair et al., 2016). According to Cepeda et al. (2018), our results indicate complementary partial mediation because the result of the x b (ab) and c’ are significant and a x b x c’ is positive.

4.3 Impact-performance map analysis

The Importance-Performance Map Analysis (IPMA) (see Figure 3) extends the reported PLS-SEM results of the path coefficient estimates using an analysis dimension that considers the average values of the latent variables’ scores. Specifically, the IPMA verifies the total effects, representing its importance in constructing a construct with their average latent variable scores indicating their performance. The objective is to identify the elements that are more important in the construct and therefore have a strong overall effect on the construct but with a low yield; in other words, the average scores of the latent variables are low (Ringle and Sarstedt, 2016).

In this case, the IPMA shows that food quality performs the lowest of the three elements, with a value of 69.53. On the other hand, the importance of this element is high because it has a total effect of 0.42. A one-unity increase in food quality will increase the satisfaction yield up to 0.42 unities. Thus, to increase the performance of the satisfaction variable, aspects related to food quality should be given priority, since they have the greatest importance but display low performance. Service quality shows greater performance and minor importance, with values of 75.92 and 0.16, respectively. Likewise, atmosphere is located in the matrix with values of 69.5 and 0.18.

5. Discussion and conclusions

5.1 Conclusions

The main objective of this study was to analyze how fast-food service quality – composed of food quality, service quality and atmosphere – can affect customer satisfaction and loyalty and to verify the direct and mediating roles of constructs such as satisfaction and trust in creating loyalty to FFRs.

An important contribution of this research is the determination of service, food quality and restaurant atmosphere as critical elements in customer satisfaction and components of fast-food service quality. The empirical findings of this study confirm the existence of significant relationships between fast-food service quality (service quality, food quality and atmosphere) and satisfaction. The results also show food quality to be the most important element of overall quality. Satisfaction, in this type of establishment, significantly and positively influences loyalty and, in turn, customer trust in FFRs. Furthermore, this research incorporates the variable trust as a mediator between satisfaction and loyalty in the FRR industry. Findings have confirmed that trust exerts a partially mediating effect between customer satisfaction and loyalty, affirming that this is established prior to loyalty. Therefore, satisfaction and trust are key to developing loyalty in FFR customers.

Another contribution of this study is the use of the impact-performance map analysis. Given the importance of the dimensions of fast-food service quality considered in this study (i.e. service quality, food quality and atmosphere), it is important to analyze how these elements contribute to increasing satisfaction in FFRs and generating loyalty. The results obtained by the IPMA also indicate that service quality is one of the three most-valued attributes among those examined in FFR studies. This is logical given the assumption that customers expect quick and comfortable service because these are characteristic features of this sector. However, service quality is the variable that carries the least weight for fast-food service quality. By contrast, the variable that is least-valued by customers but has the greatest weight is the product itself (i.e. the food). The atmosphere becomes an average between the two previous elements. Therefore, by examining only three of the key elements of fast-food service quality, this study confirms the earlier finding that food quality has the greatest weight in fast-food service quality and contributes most to customer satisfaction and therefore to the generation of loyalty.

The findings of this study provide FFRs with a broader understanding of the restaurant service dimensions that may influence customer satisfaction and behavioral intentions, such as the mediating role of trust and loyalty. This study also contributes to the PLS-SEM approach, providing knowledge regarding the use of composites and common factors in the same model, as well as the alternation of the use of PLS and consistent PLS. Additionally, the IPMA presents a number of practical implications for FFR decision makers.

5.2 Theoretical implications

This study should be of great interest to those studying loyalty in the restaurant business, particularly FFRs. Previous studies on FFRs have examined customer intentions and satisfaction using various scales to measure the quality of service or related constructs. However, this study shows that it is possible to analyze the overall quality of the restaurant in terms of three uncorrelated vital attributes: food quality, service quality and atmosphere (Ryu et al., 2012).

For these reasons, this study represents a remarkable contribution to the literature related to traditional variables such as satisfaction, trust and loyalty, as well as to the field of FFRs and service quality. It uses a model that integrates multiple elements previously studied in the literature in a paired or individual way. This study will assist scholars in expanding their research on restaurant service attributes. Methodologically, this study adds to the PLS-SEM field, where new conceptualizations and updates have been provided by recent works such as Ali et al. (2018), Cepeda et al. (2018), Henseler (2017) and Rigdon (2016).

5.3 Practical implications

This study indicates that improving the quality of restaurant attributes such as atmosphere, food and service increases loyalty and customer satisfaction and therefore can be used as a guide for the managers of small and large FFRs. The results show that food quality is the most important element in the overall quality of the restaurant (compared to service and atmosphere) and yet is the least-valued by customers. Therefore, these findings offer practical recommendations for FFR managers. Analyzing the extent to which the product offered is high-quality is considered of utmost importance because it allows an increase in the customer’s perceived quality, thus enhancing customer satisfaction and loyalty. Future researchers are expected to make further progress on this issue.

5.4 Limitations and future research

This study has several limitations. The first is the geographic location of the sample. Future research can incorporate customers located in a wider geographical environment. In addition, difficulties were experienced in distinguishing the two-dimensional concept of loyalty, since some FFRs are known worldwide, with customers acquiring loyal attitudes towards the brand rather than towards the restaurant itself. Further studies can analyze the differences between establishments with less well-known brands and establishments with globally recognized brands. Moreover, the moderating role of the socio-demographic variables of sex, age and education level was not considered. Future studies could examine the moderation of these variables in the relationships established in the proposed research model. Finally, the cross-sectional nature of the study is a limitation. Future studies can perform dynamic studies that analyze the variation in the relationships presented here.

Figures

Model proposed on the generation of loyalty in FFRs

Figure 1.

Model proposed on the generation of loyalty in FFRs

Structural model results

Figure 2.

Structural model results

IPMA of components of fast-food service quality

Figure 3.

IPMA of components of fast-food service quality

Characteristics of the survey sample

Customer profile Categories Sample (%)
Gender Men 44
Woman 56
Age ≥23s 24.2
24-33 29.5
34-43 33
44-53 11
+54 2.2
Education level Basic 26.9
Middle 27.8
Higher 45.3
Frequency of visit Everyday 2.2
Several days a week 3.5
Once a week 25.4
Once a month 39.3
Lower frequency 29.6
Time of day From Monday to Friday, during lunchtime 11.5
From Monday to Friday, during dinnertime 12.2
From Monday to Friday, other times of the day 1.8
Weekend, lunchtime 39.0
Weekend, dinnertime 30.8
Weekend, other times of the day 4.7
Expenditure (average) per person <6 € 20.9
6-12 € 64.7
12-18 € 6.4
18-24 € 3.9
>24 € 4.1
Who is accompanying you today? Alone 5.1
Myfamily 51.5
Myfriends 23.8
Mypartner 17.8
Mycoworkers 1.8

Measurement model evaluation

Construct/Indicator Loading Dijkstra–Henseler’s
rho (ρA)
CRI AVE
Satisfaction (Common factor)
SAT1. This fast-food restaurant meets my expectations 0.823*** 0.858 0.857 0.667
SAT2. My overall evaluation of this fast-food restaurant is good 0.837***
SAT3. I made a good choice when I decided to go to this fast-food restaurant 0.789***
Trust (Common factor)
TRU1.I think that this fast-food restaurant usually fulfills the commitments it assumes 0.804*** 0.931 0.929 0.545
TRU2. I certify that the information provided by this fast-food restaurant is true 0.746***
TRU3.I believe I can have confidence in the promises that this fast-food restaurant site makes 0.788***
TRU4.This fast-food restaurant is characterized by the frankness and clarity of the products and services that it offers to consumer 0.789***
TRU5. I consider that the information offered by this fast-food restaurant is sincere and honest 0.759***
TRU6. To me this fast-food restaurant has sufficient experience in the marketing of the products and services that it offers 0.696***
TRU7.I consider that this fast-food restaurant has the necessary resources to successfully carry out its activities 0.624***
TRU8.I believe that this fast-food restaurant has employees with necessary abilities to carry out their work 0.688***
TRU9.To me this fast-food restaurant is concerned with the present and future interests of its customers 0.772***
TRU10. I think that this fast-food restaurant would not do anything intentional that would prejudice the customer 0.694***
TRU11. I believe that this restaurant is receptive to the needs of its customers and gives them all the information they need 0.739***
Loyalty (Common factor)
LOY2. I would recommend this fast-food restaurant to someone who seeks my advice 0.686*** 0.840 0.832 0.555
LOY3. I consider this fast-food restaurant my first choice to eat away from home 0.657***
LOY4. I seldom consider switching to another fast-food restaurant 0.788***
LOY5.I am likely to continue eating in this fast-food restaurant in the next few years 0.835***
Notes:

n = 5,000 subsample:

***p < 0.001; ns: non-significant (one-tailed t Student) t(0.05; 4,999) = 1.645; t(0.01; 4,999) = 2.327; t(0.001; 4,999) = 3.092 CRI = composite reliability index; AVE = average variance extracted; SAT: satisfaction; TRU: trust; LOY: loyalty

Measurement model: Discriminant validity

Fornell–Larcker criterion analysis Heterotrait-Monotrait (HTMT)
SAT TRU LOY SAT TRU LOY
SAT 0.817 SAT
TRU 0.677 0.738 TRU 0.678
LOY 0.689 0.663 0.745 LOY 0.690 0.660
Notes:

SAT: satisfaction; TRU: trust; LOY: loyalty. Fornell–Larcker criterion: diagonal elements (bold) are the square root of the variance shared between the constructs and their measures (average variance extracted). Off-diagonal elements should be larger than off-diagonal elements. AVE: average variance extracted

First-order correlations of constructsa

Construct SER FOQ ATM SAT TRU LOY
Service quality
Food quality 0.493
Atmosphere 0.478 0.642
Satisfaction 0.447 0.597 0.525 0.817
Trust 0.578 0.621 0.646 0.676 0.739
Loyalty 0.373 0.550 0.545 0.689 0.662 0.745
Note:
a

The diagonal element is the square root of the average variance extracted. To be discriminant, the diagonal elements should be large tan all corresponding off-diagonal elements to show discriminant validity

Measurement model: Model composite mode B

Construct/Indicator Loadings Weights Weights bootstrapping Collinearity
statistic
T Significance 5% 95% VIF
Service Quality (Composite Mode B)
SER1.Fast-food restaurant’s service is quick 0.799 0.408*** 5.426 0.000 0.283 0.524 1.443
SER2. Fast-food restaurant’s service servers have a pleasant attitude 0.780 0.365*** 4.258 0.000 0.223 0.504 1.473
SER3.My order placed at this fast-food restaurant is exactly the right one 0.867 0.449*** 5.027 0.000 0.298 0.592 1.710
Fast-Food Quality (Composite Mode B)
FOQ1. The food presentation in this fast-food restaurant is attractive and tempting 0.778 0.346*** 5.315 0.000 0.234 0.450 1.646
FOQ2.The food served in this fast-food restaurant tastes good 0.668 0.154* 2.000 0.023 0.023 0.281 1.614
FOQ3.The food in this fast-food restaurant smells great 0.666 0.080(ns) 1.106 0.134 −0.043 0.198 1.829
FOQ4.The food temperature in this fast-food restaurant is correct 0.637 0.176** 2.843 0.002 0.071 0.280 1.477
FOQ5.Considering price, the amount of food is what I expected 0.540 0.171** 2.729 0.003 0.064 0.267 1.250
FOQ6.The food at this fast-food restaurant is fresh and cooked properly 0.860 0.430*** 7.073 0.000 0.330 0.532 1.775
Atmosphere (Composite Mode B)
ATM1.Background music of the fast-food restaurant is good 0.484 0.038(ns) 0.621 0.267 −0.070 0.135 1.330
ATM2. There are no unpleasant odors, and the fast-food restaurant is clean 0.709 0.276*** 4.351 0.000 0.169 0.382 1.413
ATM3.The temperature inside the fast-food restaurant is pleasant 0.531 0.077(ns) 1.213 0.113 −0.026 0.182 1.387
ATM4.The fast-food restaurant’s ambiance allows for conversation 0.818 0.344*** 4.410 0.000 0.213 0.468 1.868
ATM5.The interior décor of the fast-food restaurant is enjoyable 0.698 0.154** 2.482 0.007 0.044 0.249 1.666
ATM6.Seating arrangement of the fast-food restaurant provides with adequate space 0.617 0.156* 2.311 0.010 0.044 0.263 1.524
ATM7. The place play of the fast-food restaurant is pleasing 0.532 0.064(ns) 1.029 0.152 −0.041 0.165 1.422
ATM8.Lighting compliments my experience in the fast-food restaurant 0.762 0.296*** 4.254 0.000 0.179 0.409 1.624
Notes:

n = 5,000 subsample:

*p < 0.05;

**p < 0.01;

***p < 0.001; ns: non-significant (one-tailed t Student) t(0.05; 4,999) = 1.645; t(0.01; 4,999) = 2.327; t(0.001; 4,999) = 3.092; SER: service quality; FOQ: food quality; ATM: atmosphere

Structural model evaluation

Relationships β t-statistic (Bootstrap) Variance explained (R2) Percentile bootstrap
R2 Adjusted-R2 95% CI
SER → FFSQ 0.266*** 14.349 [0.603; 0.758] Sig.
FOQ → FFSQ 0.492*** 29.580 [0.819; 0.914] Sig.
ATM → FFSQ 0.453*** 29.551 [0.805; 0.901] Sig.
H1 FFSQ → SAT 0.681*** 18.510 0.464 0.463 [0.621; 0.743] Sig.
H2 SAT → LOY 0.444*** 5.305 0.547 0.545 [0.296; 0.563] Sig.
(a) SAT → TRU 0.677*** 15.859 0.458 0.457 [0.605; 0.743] Sig.
(b) TRU → LOY 0.363*** 4.567 [0.213; 0.470] Sig.
Notes:

n = 5000 subsample:

***p < 0.001; sig: significant (one-tailed t Student) t(0.05; 4,999) = 1.645; t(0.01; 4,999) = 2.327; t(0.001; 4,999) = 3.092 SAT: satisfaction; TRU: trust; LOY: loyalty; SER: service quality; FOQ: food quality; ATM: atmosphere; FFSQ: Fast-food service quality

Mediating effects test

Mediating effects Coefficient Bootstrap 95% CI
Percentile BC
Direct effects
H2: SAT → LOY(c′) 0.444sig 0.307 0.585 0.304 0.582
SAT → TRU(a) 0.677sig 0.604 0.744 0.604 0.744
TRU → LOY(b) 0.363sig 0.229 0.490 0.231 0.492
Indirect effects Point estimate Percentile BC VAF
H3: SAT → TRU → LOY (ab) 0.246sig 0.157 0.328 0.159 0.330 35.6%

References

Aldás, J., Currás, R., Ruiz, C. and Sanz, S. (2010), “Factores determinantes de la lealtad en el comercio electrónico b2c. Aplicación a la compra de billetes de avión”, Revista Española de Investigación de Marketing, Vol. 14 No. 2, pp. 113-142.

Ali, F. (2016), “Hotel website quality, perceived flow, customer satisfaction and purchase intention”, Journal of Hospitality and Tourism Technology, Vol. 7 No. 2, pp. 213-228.

Ali, F., Rasoolimanesh, S.M., Sarstedt, M., Ringle, C. and Ryu, K. (2018), “An assessment of the use of partial least squares structural equation modeling (PLS-SEM) in hospitality research”, International Journal of Contemporary Hospitality Management, Vol. 30 No. 1, pp. 514-538.

Bufquin, D., DiPietro, R. and Partlow, C. (2017), “The influence of the DinEX service quality dimensions on casual-dining restaurant customers’ satisfaction and behavioral intentions”, Journal of Foodservice Business Research, Vol. 20 No. 5, pp. 542-556.

Cepeda, G., Nitzl, C. and Roldán, J.L. (2018), “Mediation analyses in partial least squares structural equation modeling: guidelines and empirical examples”, in Latan, H. and Noonan, R. (Eds), Partial Least Squares Path Modeling: Basic Concepts, Methodological Issues and Applications, Springer, Heidelberg.

Clemente-Ricolfe, J.S. (2016), “Atributos relevantes de la calidad en el servicio y su influencia en el comportamiento postcompra: el caso de las hamburgueserías en España”, Revista Innovar Journal Revista De Ciencias Administrativas y Sociales, Vol. 26 No. 62, pp. 69-78.

Dick, A.S. and Basu, K. (1994), “Customer loyalty: toward an integrated conceptual framework”, Journal of the Academy of Marketing Science, Vol. 22 No. 2, pp. 99-113.

Dijkstra, T.K. and Henseler, J. (2015), “Consistent and asymptotically normal PLS estimators for linear structural equations”, Computational Statistics & Data Analysis, Vol. 81, pp. 10-23.

Gadermann, A.M., Guhn, M. and Zumbo, B.D. (2012), “Estimating ordinal reliability, for likert-type and ordinal item response data: a conceptual, empirical, and practical guide”, Practical Assessment, Research and Evaluation, Vol. 17 No. 3, pp. 1-13.

Garbarino, E. and Johnson, M.S. (1999), “The different roles of satisfaction, trust, and commitment in customer relationships”, Journal of Marketing, Vol. 63 No. 2, pp. 70-87.

Hair, J.F. Jr, Hult, G.T.M., Ringle, C. and Sarstedt, M. (2016), A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), SAGE Publications, Thousand Oaks, CA.

Henseler, J. (2017), “Bridging design and behavioral research with variance-based structural equation modeling”, Journal of Advertising, Vol. 46 No. 1, pp. 178-192.

Henseler, J., Hubona, G. and Ray, P.A. (2016), “Using PLS path modeling in new technology research: updated guidelines”, Industrial Management & Data Systems, Vol. 116 No. 1, pp. 2-20.

Henseler, J., Ringle, C.M. and Sarstedt, M. (2015), “A new criterion for assessing discriminant validity in variance-based structural equation modeling”, Journal of the Academy of Marketing Science, Vol. 43 No. 1, pp. 115-135.

Kumar, A. (2015), “Managing knowledge at tourism destinations: conceptual foundations”, in Kumar, S., Dhiman, M.C. and Dahiya, A. (Eds), International Tourism and Hospitality in the Digital Age, IGI Global, Hershey PA, pp. 72-87.

Liu, Y. and Jang, S.S. (2009), “Perceptions of Chinese restaurants in the US: what affects customer satisfaction and behavioral intentions?”, International Journal of Hospitality Management, Vol. 28 No. 3, pp. 338-348.

Martín-Consuegra, D., Gómez, M. and Molina, A. (2015), “Consumer sensitivity analysis in mobile commerce advertising”, Social Behavior and Personality: An International Journal, Vol. 43 No. 6, pp. 883-897.

Namin, A. (2017), “Revisiting customers’ perception of service quality in fast food restaurants”, Journal of Retailing and Consumer Services, Vol. 34, pp. 70-81.

Namkung, Y. and Jang, S. (2007), “Does food quality really matter in restaurants? Its impact on customer satisfaction and behavioral intentions”, Journal of Hospitality & Tourism Research, Vol. 31 No. 3, pp. 387-409.

Oches, S. (2015), “Tale of two brands: QSR Magazine”, available at: www.qsrmagazine.com/fast-casual/tale-two-brands (accessed 2 December 2017).

Oliver, R.L. (1999), “Whence consumer loyalty?”, Journal of Marketing, Vol. 63, pp. 33-44.

Podsakoff, P.M., MacKenzie, S.B. and Podsakoff, N.P. (2012), “Sources of method bias in social science research and recommendations on how to control it”, Annual Review of Psychology, Vol. 63 No. 1, pp. 539-569.

Qin, H. and Prybutok, V.R. (2009), “Service quality, customer satisfaction, and behavioral intentions in fast-food restaurants”, International Journal of Quality and Service Sciences, Vol. 1 No. 1, pp. 78-95.

Rigdon, E.E. (2016), “Choosing PLS path modeling as analytical method in European management research: a realist perspective”, European Management Journal, Vol. 34 No. 6, pp. 598-605.

Rigdon, E.E., Sarstedt, M. and Ringle, C.M. (2017), “On comparing results from CB-SEM and PLS-SEM: five perspectives and five recommendations”, Marketing ZFP, Vol. 39 No. 3, pp. 4-16.

Ringle, C.M. and Sarstedt, M. (2016), “Gain more insight from your PLS-SEM results: the importance-performance map analysis”, Industrial Management & Data Systems, Vol. 116 No. 9, pp. 1865-1886.

Ryu, K., Lee, H.R. and Gon Kim, W. (2012), “The influence of the quality of the physical environment, food, and service on restaurant image, customer perceived value, customer satisfaction, and behavioral intentions”, International Journal of Contemporary Hospitality Management, Vol. 24 No. 2, pp. 200-223.

Sahagun, M.A. and Vasquez-Párraga, A.Z. (2014), “Can fast-food consumers be loyal customers, if so how? Theory, method and findings”, Journal of Retailing and Consumer Services, Vol. 21 No. 2, pp. 168-174.

Sarstedt, M., Hair, J.F., Ringle, C.M., Thiele, K.O. and Gudergan, S.P. (2016), “Estimation issues with PLS and CBSEM: where the bias lies!”, Journal of Business Research, Vol. 69 No. 10, pp. 3998-4010.

Stathopoulou, A. and Balabanis, G. (2016), “The effects of loyalty programs on customer satisfaction, trust, and loyalty toward high-and low-end fashion retailers”, Journal of Business Research, Vol. 69 No. 12, pp. 5801-5808.

Statista (2017), “Number of restaurants of fast-food franchises in Spain from 2010 to 2016”, available at: www.statista.com/statistics/761964/fast-food-franchises-number-of-restaurants-in-spain/ (accessed 1 December 2017).

Wright, R.T., Campbell, D.E., Thatcher, J.B. and Roberts, N.H. (2012), “Operationalizing multidimensional constructs in structural equation modeling: recommendations for IS research”, Communications of the Association for Information Systems, Vol. 30, pp. 367-412.

Wu, H.C. and Mohi, Z. (2015), “Assessment of service quality in the fast-food restaurant”, Journal of Foodservice Business Research, Vol. 18 No. 4, pp. 358-388.

Corresponding author

David Martín-Consuegra can be contacted at: david.martin@uclm.es