Domestic confinement crisis and its un-intensifying effect on service quality expectations towards e-stores: the case of the COVID-19 pandemic in Mexico

ABSTRACT

Certain factors may exert un-intensifying effects on service performance expectations. In the pandemic crisis era of domestic confinement, as consumer shopping at physical stores is restricted, e-stores have become a useful source of goods. It is proposed that such a confinement crisis, as an extreme situational factor, can have a mitigating effect on the expectation level of service performance of e-stores. Hence, it is hypothesised that consumers may become more forgiving with e-suppliers, reporting a higher perceived performance under a confinement crisis than under normal social conditions. Two sets of data were compared for assessment, employing the same questionnaire to evaluate consumer opinions of service performance, brand perceptions, satisfaction, trust and intentions of e-stores. One set of data was collected under normal conditions, prior to domestic confinement being imposed due to the COVID-19 pandemic in Mexico. The second set of data was collected during a high point of contagion of the COVID-19 emergency in Mexico. Results suggest that there are significant differences in some variables related to perceived service performance, brand perceptions and behavioural intentions between the two sets of data. Apparently, consumers tend to perceive, in some respects, a higher level of performance of e-stores under confinement conditions than under normal social conditions. This work aims to shed light on how such extreme situational factors can affect consumer expectations and perceptions.

KEYWORDS:

• Customer expectations
• mitigating effect
• situational factor
• e-store service performance
• zone of tolerance

Introduction

The COVID-19 pandemic caused important changes in shopping behaviour and in the ways in which business is conducted (Mehta, Saxena, and Purohit 2020). The pandemic has led to physical social distancing in the form of domestic confinement taking place worldwide. Thus, some industries have been experimenting with the need to enhance online transactions. According to the Organisation for Economic Co-operation and Development (OECD (Organisation for Economic Co-operation and Development) 2020), the sales of essential goods have been particularly susceptible to this phenomenon during the pandemic.

Extreme situational factors such as bad weather, catastrophes and other possible transitory circumstances have been theoretically proposed as factors that can have a potential effect on consumers’ expectations of the quality of services (Zeithaml, Berry, and Parasuraman 1993). According to this proposition, this effect can work to potentially mitigate consumers’ expectations of quality from a service provider. Apparently, in such situations, consumers may readjust their perspectives on the level of performance that should be received from a service provider. It seems that consumers may become more forgiving and sensitive with their service providers when one of these extreme situational factors occurs. Consumers may understand that it is unlikely that a service provider can perform at the same level as in normal conditions during such situations. Hints of this forgiving effect can be seen in a study wherein confinement (isolation) due to the COVID-19 pandemic generated feelings of apprehension and loneliness in clients of hospitality services (Wong and Yang 2020). Under these conditions, good service engenders a sense of warmth and care, which helps to mitigate anxiety. These theorisations were originally proposed in a conceptual article by Zeithaml, Berry, and Parasuraman (1993); however, to the best of the authors’ knowledge, after a thorough literature review, there has been no further published research to empirically explore these claims. In fact, the literature addressing the effects of extreme (harmful) situational factors on the perception of service performance is surprisingly scarce. One of the goals of this research is to address this gap and offer some empirical evidence exploring this proposition. Although the extreme situational factor addressed by Zeithaml, Berry, and Parasuraman (1993) refers to catastrophic events prevalent over short periods of time and in specific geographical locations (e.g. earthquakes, floods and hurricanes) and a pandemic such as COVID-19 is an enduring global phenomenon, the latter can have an equal or even greater effect (than the former) on reducing customer expectations precisely because of the duration and geographical reach of the event. Attending this gap may be relevant as it can improve our knowledge about how emergencies such as pandemics affect consumer behaviour. The relevance of this knowledge relies on the idea that it can give us tools to be better prepared to face future extreme situations. Therefore, for the current research work, the COVID-19 pandemic offers a unique opportunity to analyse the effects.

Therefore, for the specific settings of this work, and taking advantage of the opportunity of having recently generated data under normal conditions (prior to the pandemic), it was decided to perform this research in the context of e-store service performance and other related variables, such as brand perceptions, satisfaction and behavioural intentions.

In the COVID-19 pandemic, electronic commerce has become an extremely important source of goods for consumers, as visiting physical stores constituted a potential cause of contagion (Statista 2020; Enrico 2020; Chávez 2020). To reduce the probability of contagion, authorities recommend reducing the frequency of going to public places, such as grocery stores (World Health Organization (WHO), Western Pacific 2020; Asperin 2020; Groth 2020).

Literature review

Situational factors influencing customer expectations

Evidence suggests that consumers have different expectation levels of service companies. These expectations can range from the minimum they are willing to receive (adequate service) to the real desired service (ideal service) (Zeithaml, Bitner, and Gremler 2009, 80). What lies between these two extremes is identified as the ‘zone of tolerance’ (Ho, Sharma, and Hosie 2015). According to Zeithaml, Berry, and Parasuraman (1993), customer expectations are formed according to different factors; they are (1) transitory service intensifiers, (2) perceived service alternatives, (3) customer self-perceived service roles, (4) predicted service and (5) situational factors. This article focuses on what these authors define as ‘service performance contingencies that customers perceive are beyond the control of the service provider’. For example, they argue that ‘catastrophes that affect numerous people at one time (earthquakes or hurricanes) would likely lower service expectations […] customers appear to recognise that these contingencies are not the fault of the service company and tend to accept lower levels of adequate service given the context’.

After an extensive literature review, some studies were found in which the effect of some factors on customer expectations were assessed; however, none of this literature considered extreme situational factors (e.g. Durvasula et al. 2006; Ramsaran-Fowder 2013; Nadiri and Hussain 2005; Nadiri 2011; Ho, Sharma, and Hosie 2015). While some articles do mention situational factors, they do not conceptualise situational factors in the same sense that Zeithaml, Berry, and Parasuraman (1993) do (as catastrophic events). These studies refer to situational factors as elements related to the context where the service meeting, store price, sense of urgency of the purchase, brand loyalty, emotional impact and sense of surprise occurs, as related to transactions and waiting, time and place, urgency and need of products and consumer location (Strombeck and Wakefield 2008; Dadzie and Winston 2007; Kazancoglu and Aydin 2018).

Service performance of e-stores and website design

Service performance of e-stores can be defined as consumers’ evaluation and judgment of excellence and e-service quality by features offered by the virtual marketplace (Santos 2003). The aspect of e-commerce service performance (or quality) is usually assessed by attributes of the company’s website, which are likely to hold an important role in improving customer satisfaction with the website (Amin 2016; Manasra, Zaid, and Taherqutaishat 2013; Shin et al. 2013; Kaur et al. 2012; Kim and Stoel 2004). According to this, e-store service performance is also likely to have a positive effect on customer e-loyalty, directly or through satisfaction and/or customer trust as mediating variables. In addition, there is evidence that consumers’ shopping intention can be favourably affected by convenience, perception of safety and product characteristics when these are appropriately addressed by e-retailers (Mahadin, Akroush, and Bata 2020). A good website is considered to be one that provides good information for consumer purchase decision-making; thus, website quality can be measured based on website content, website design and website structure (Winnie 2014). Hence, website design tends to be a relevant aspect of service quality performance (Lee and Lin 2005; Ziaullah, Yi, and Akhter 2014).

Consumer trust in e-stores

E-shoppers often question the security, return policies or quality of information provided on the internet (Kim and Park 2013). The most common issues that customers face when shopping online are related to trust, warranties, paying with card, delivery time and the risk of strangers hacking into personal or financial data (Makwana, Pathak, and Sharma 2017). Trust in terms of e-commerce has been defined as the expectation that others will behave in a reliable, ethical and appropriate manner and abide by their commitments (Gefen, Karahana, and Straub 2003). Trust serves as a foundation for initial consumer relationship and is a more important aspect of building and maintaining a long-term relationship for e-commerce success (Kim et al. 2014). Trust plays a key role in commercial transactions between customers and vendors, especially in the online environment where there are no face-to-face interactions (Shin et al. 2013). Accordingly, trust creates favourable feelings towards online vendors that are likely to increase customer intentions to purchase a product. Commitment towards the site is derived from a client who trusts in the genuineness of the data exhibited and is confident in the protection of individual data; such ‘e-trust’ may exert a noteworthy effect on repurchase intentions (Hsu and Vui 2019). Previous studies suggest that there is also a positive relationship between website attributes performance and trust (Manasra, Zaid, and Taherqutaishat 2013; Shin et al. 2013; Winnie 2014). In a study exploring the level of consumer satisfaction in online shopping, it was observed that despite numerous advantages, fraud and privacy concerns caused by web-hackers is a source of great concern to users, and website users are doing their best to ensure the use of secure portals (Giao, Vuong, and Quan 2020). Conversely, lack of trust tends to be a reason why people hesitate to make online purchases.

Behavioural intentions toward e-stores

Repurchase intentions tend to grow when consumers have a qualitative relationship with vendors (Prashar, Vijay, and Parsad 2017). Purchase intention towards specific online shopping websites tends to predict the purchase decisions of customers and the actual behaviour of repurchase (Kim, Ferrin, and Rao 2008; Blut et al. 2015; Liu et al. 2015). Repurchase has been portrayed as a genuine activity of clients to purchase or use an item once more (Ibzan, Balarabe, and Jakada 2016). Thus, behavioural intentions can be considered as an indicator of a certain type of brand loyalty (Oliver 1999). Perceived e-service performance (quality) and value have also been confirmed as being related to loyalty in e-retailing (Zehir et al. 2014). Providing good service performance and building customer confidence result in enhancing customer trust, thereby generating loyal customers (Sarwar, Abbassi, and Pervaiz 2012). Subsequently, e-loyalty can be considered as customers’ favourable attitudes towards an online business that may result in a repurchase behaviour (Safa and Ismail 2013). Many e-business models stress the need to find and secure adequate numbers of clients prior to gaining the long-term benefits from those potentially loyal customers (Shafiee and Bazargan 2018).

Framework and hypotheses

Consumers tend to form expectations of performance from previous encounters and experiences with other products and services, personal interactions with known consumers, promotional content and communication channels (Cardozo 1965; Vasić, Kilibarda, and Kaurin 2019; Oliver 2014, 63). Several factors can have an effect on consumers’ level of expectation of service performance. Such elements can include enduring service intensifiers, personal needs, transitory service intensifiers, perceived service alternatives and (extreme) situational factors (Zeithaml, Berry, and Parasuraman 1993). Some of these elements can exert a positive effect on customers’ expectations regarding service performance; meaning that these factors can increase the level of expectation, reducing the zone of tolerance and causing the brand to struggle more to achieve consumers’ perception of acceptable performance (Parasuraman, Berry and Zeithaml, 1991). The cognitive mechanisms to explain this increase of expectation can be twofold. From one perspective, a previously unfavourable experience can cause the consumer to be more attentive and thorough when evaluating the potential for a new encounter with a provider, as a defence mechanism to avoid possible additional drawbacks (i.e. failures in previous delivery services, alternative service providers, negative word of mouth and unreliable service promises). Conversely, favourable previous experiences related to the service provider will increase the expected level of perceived quality as well as the level of expectation for future encounters (Rust and Oliver 2000).

Effects on perceived service performance

Additional factors can exert negative effects on service performance expectations, some of which have been termed situational factors. Zeithaml, Berry, and Parasuraman (1993) proposed that situational factors might temporarily lower the level of accepted service performance, extending the zone of tolerance regarding perceptions of adequate service. As these situational factors refer to massive destructive events, they could be called extreme situational factors. Such factors exert inverse effects compared to those noted previously, it may be appropriate to call them un-intensifiers of service performance expectations. In other words, they exert an un-intensifying (mitigating, reducing) effect on consumers’ quality expectations; presumably, making it easier for service providers to attain high consumer perceptions of performance. According to literature, these reducers or mitigating factors are contingencies beyond the control of the service provider, such as earthquakes, floods, hurricanes and other state emergency conditions (Zeithaml, Berry, and Parasuraman 1993; Sakurai et al. 2011). A domestic confinement crisis such as the one generated by the COVID-19 pandemic can certainly be considered an extreme situational factor that could reduce consumers’ expectations of quality from a service provider. Apparently, under extreme adverse social conditions, customers tend to be more forgiving with their service providers, understanding that these providers cannot perform at the same level as under normal conditions. Moreover, under a confinement crisis, an adequate service can provide a sense of solace that reduces the level of anxiety (Wong and Yang 2020). Therefore, customers may tend to give a less rigorous evaluation to their service providers under an extreme adverse social condition, probably resulting in a better evaluation of performance, due to the forgiving effect. Perhaps traditional service attributes become less important when experiencing an extreme situational condition from the customer’s perspective. However, these propositions have remained on a conceptual (theoretical) level. To the best of the authors’ knowledge, no literature has yet empirically tested this hypothesis regarding the un-intensifying effect on service performance expectations generated by extreme situational conditions. These effects may occur for a wide range of consumer services. In the current study, it was decided to consider the e-store sector, as it has had extreme relevance to the provision goods under the confinement crisis, as consumers have been recommended to reduce the frequency of in-person shopping in traditional stores during the COVID-19 pandemic. In fact, partial evidence points to the use of e-commerce increasing during the pandemic as well as a rise in the perceived service performance and customer satisfaction of online stores (Dirgantari et al. 2020). Therefore, as the need for such stores increased under confinement conditions, in addition to the proposed un-intensifying (reducer) effect on customer expectation, it is expected that customers may perceive better service performance from e-stores while under confinement conditions than under normal conditions, leading to the following hypotheses related to e-store’s service performance attributes:

H1: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perceived home page attractiveness of the e-store.

H2: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perceived product catalogue width of the e-store.

H3: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perceived order form ease of filling of the e-store.

H4: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perceived customer support effectiveness of the e-store.

Effects on brand perceptions

In this research, it is also considered that if the extreme situational factor (confinement crisis) can have an effect on the perception of performance of service provider attributes (e-store service attributes, in this case), it would also exert a similar effect on the perception of other business-related consideration, such as brands related to the service provider. In case of e-stores, the extreme situational factor (the COVID-19 confinement crisis) may also impose a similar reducing effect (the forgiveness impulse) on the perception of brands in concordance with that proposed for the perception of the service performance. If the customer has reduced expectations regarding service performance, driving better perceptions of service attributes, it follows that a spill-over effect can occur concerning the perception of all the brands related to the e-store. If the perception of service performance may increase under such conditions, the perception regarding related brands may also improve. Here, we are considering two types of brands: the name of the store as the e-store brand (regardless of whether the same name can be used as a private label for some products) and the brands of the products. Hence, the situational factor may play a role in improving the perception toward the e-store brand (e.g. Amazon), as well as the perception of the brands of the products that the e-store offers. Hence, the following hypotheses are offered:

H5: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perception of the brands of the products in the e-store.

H6: The presence of an extreme situational factor (confinement crisis) has a positive effect on the perception of the e-store brand.

Effects on behavioural variables of business performance

If an extreme situational factor can have a reducer effect on customer expectations and thus generate higher values of perceived service performance and brand perceptions, it would also follow that this factor can have a similar effect on the expected outcomes of these constructs. The literature from which service performance has been established as an antecedent of expected outcomes, such as satisfaction, trust, intentions, brand equity, changing intentions, willingness to recommend and disposition to pay an overprice, is vast (e.g. Host and Knie-Andersen 2004; Kyle et al. 2010; Martins, Hor-Meyll, and Ferreira 2013; Ajmal, Raza, and Fatima 2018). In the context of e-store services, these relationships have been also widely attended and empirical evidence has been repeatedly published (e.g. Madu and Madu 2002; Ribbink et al. 2004; Cristobal, Flavián, and Guinalíu 2007; Yeh and Li 2009; Rose et al. 2012; Vos et al. 2014; Vásquez and Vera-Martínez 2020). If the customer perceived good performance from the service provider and has a favourable perception of the brand(s) involved in the pre-purchase process, this will tend to favour a positive overall cognitive evaluation (satisfaction and trust) toward the service provider (e-store in this case). This confident cognitive evaluation would increase the likelihood of the consumers’ expressing possible behaviour toward the service provider in the form of a purchase and likely even to accept paying a bit more, given this confidence.

Sakurai et al. (2011) tested the effects of a hurricane (situational factor) over a model of customer service regarding insurance companies. This study revealed that the hurricane had a mitigating effect on customer perceptions in such a way that an indirect positive effect between customer satisfaction and some of its antecedents, such as service performance attributes, were identified. Therefore, an extreme situational factor such as the confinement crisis of the COVID-19 pandemic in the current work can be expected to reduce or mitigate customer expectations regarding service provider performance, due to the forgiveness or understanding effect and a greater sense of satisfaction and trust. Hence, a higher level of behavioural intentions and an elevated level of disposition to pay an overprice to the service provider (e-store, in this case) would be expected as part of such effect. Hence, the following hypotheses are posited:

H7: The presence of an extreme situational factor (confinement crisis) has a positive effect on satisfaction toward an e-store.

H8: The presence of an extreme situational factor (confinement crisis) has a positive effect on trust toward an e-store.

H9: The presence of an extreme situational factor (confinement crisis) has a positive effect on behavioural intentions toward an e-store.

H10: The presence of an extreme situational factor (confinement crisis) has a positive effect on the disposition to pay overprice to the e-store.

Ultimately, higher elevated intentions to purchase and to pay overprice have been proposed as business performance antecedents (Ladhari 2009; Meyer, Shankar, and Berry 2018; Jaiswal and Lemmink 2017). This can be confirmed in exploring the specific case of the confinement crisis generated by the COVID-19 pandemic for the sector of e-stores, as online shopping has become an extremely relevant source for consumers to satisfy their needs given the public health recommendations to reduce the number of visits to traditional stores (World Health Organization (WHO), Western Pacific 2020; Asperin 2020; Groth 2020). Figure 1 shows a diagram summarising this framework.

Figure 1. Conceptual framework.

Methodological approach and measurements

Hypotheses testing called for a contrast between service performance data gathered under ‘normal’ (habitual) conditions against data gathered during a high point of contagion (in Mexico) in the COVID-19 pandemic. This period of high contagion was the one in which the confinement measures for citizens were at the highest level of strictness. It can be argued that this was a major period of confinement. Thus, for this work, we sought the opportunity and took our data from a previous study in which variables related to service performance, brand perceptions and behavioural performance had been measured in the context of well-known e-stores in Mexico (Amazon, Liverpool, Sears, Palacio de Hierro, Sanborns, Best Buy, Fábricas de Francia and Walmart). To participate in this study, its subjects had to be autonomous purchase decision makers who were accustomed to internet shopping. They had to be shoppers between the ages of 25 and 50 years old that bought products in e-stores for themselves with their own monetary resources. Participants had to have made purchases at least five times in the last twelve months in at least one of the stores participating in the study. Each participant evaluated only one e-store – the one in which she/he made the most number of purchases in the past 12 months. Hence, via an internet survey, a sample of n₀ = 500 e-store shoppers was obtained. This sample, gathered a year before the COVID-19 pandemic, served as the contrast data for the data gathered during the highest period of contagion in Mexico. The confinement data was collected following the same exact procedures, specifications and participants’ characteristics of the original data (excepting the situation). However, she/he had to specifically evaluate the e-store regarding the purchases made under the confinement conditions. In this case, a total of n₁ = 312 valid observations were obtained; thus, n = 862 in total. Data from the normal condition was coded as condition zero (0 = absence of confinement) and data of the confinement condition was designated as condition one (1 = presence of confinement). For hypotheses testing, analysis of variance (ANOVA) analyses were implemented. The independent variable was the absence (0) or presence (1) of the confinement condition. The observable dependent variables were each of the measurements for the conceptual dependent variables declared in the hypotheses.

Table 1 shows the items used as measurement, along with reliability and validity indicators. For each dependent variable, four items were used (for behavioural loyalty, five items) in a structured questionnaire that was administered on an online survey platform. Cronbach’s alpha and the composite coefficient (omega) indicate an adequate level of reliability of the items. For each group of measurements, a factorial analysis was performed. These analyses were used to confirm the convergent validity of the items. In each case, a single component with an eigenvalue over 1.0 was extracted, showing an adequate convergence. High values in factorial loadings, AVE and KMO coefficients confirm the convergence and unidimensionality of the items of each dependent variable. Only the item of overprice disposition rendered a questionable loading.

Table 1. Measurements for dependent variables n = 862

Construct /dimension	Item code	Item declaration	Single factor loadings	C’s alpha	AVE	CR	KMO
Home page attractiveness	Homep1	The homepage of the e-store is very attractive.	0.87	0.89	0.77	0.93	0.83
	Homep2	The design of this e-store looks professional.	0.87
	Homep3	The homepage is of high-quality.	0.89
	Homep4	I like the appearance of the homepage of this e-store.	0.87
Product catalogue width	ProdC1	The e-store offers a wide line of products.	0.93	0.95	0.86	0.96	0.87
	ProdC2	The e-store offers a good selection of products.	0.94
	ProdC3	The e-store offers a complete range of products.	0.94
	ProdC4	In this e-store there is a great variety of products.	0.92
Order form ease of filling	OrdF1	Filling out the e-store purchase order form is simple.	0.93	0.95	0.86	0.96	0.82
	OrdF2	The e-store facilitates the process of filling out the purchase order.	0.92
	OrdF3	It is very easy to generate a purchase order in this e-store.	0.96
	OrdF4	I have no difficulty filling in the information required by the e-store.	0.89
Customer support efficiency	CustS1	Returns to the e-store are possible without problems.	0.88	0.94	0.84	0.95	0.84
	CustS2	When a problem arises, the e-store shows a sincere interest in solving it	0.94
	CustS3	Complaints are answered quickly by the e-store.	0.94
	CustS4	Customer support for this e-store is easily accessible.	0.91
Product brand perception	PBP1	This e-store offers high-quality branded products.	0.94	0.94	0.86	0.96	0.83
	PBP2	This e-store offers products from well-known brands.	0.94
	PBP3	This e-store offers products from leading brands.	0.94
	PBP4	This e-store offers reliable branded products.	0.92
E-store brand perception	BSP1	This e-store brand has a good reputation.	0.94	0.96	0.89	0.97	0.85
	BSP2	This e-store brand is recognised.	0.94
	BSP3	This e-store brand is reliable.	0.95
	BSP4	This e-store brand has good quality products.	0.92
Customer satisfaction	Satisf1	The service in the e-store is excellent.	0.92	0.95	0.87	0.97	0.87
	Satisf2	Buying from this e-store makes me happy.	0.91
	Satisf3	I am very satisfied with my shopping experience at this e-store.	0.96
	Satisf4	The e-store meets my expectations.	0.95
Trust on the e-store	Trust1	This e-store protects my credit card details.	0.92	0.93	0.84	0.95	0.80
	Trust2	I feel secure in the transactions I make in this e-store.	0.91
	Trust3	This e-store guarantees the confidentiality of customers’ information.	0.92
	Trust4	This e-store protects my privacy.	0.92
Behavioural intentions	BInt1	If I had to, I would buy again from this e-store.	0.95	0.97	0.89	0.98	0.90
	BInt2	I would use this e-store again to meet my needs.	0.95
	BInt3	I would recommend this e-store to friends and family.	0.943
	BInt4	For future purchases this e-store is recommendable.	0.94
	BInt5	I will buy again from this e-store.	0.95
Overprice disposition	PriceD1	I would still buy from this e-store even if it was a little more expensive.	0.91	0.87	0.73	0.91	0.75
	PriceD2	Even if the prices of this e-store went up, I would still buy here.	0.95
	PriceD3	I wouldn’t mind having to pay a little more at this e-store than I would pay at other.	0.91
	PriceD4	At the same price, I prefer to buy from this e-store than any other e-store.	0.58

Notes: C’s alpha (Cronbach’s alpha); AVE average variance extracted; CR composite reliability; KMO Kaiser-Meyer-Olkin test coefficient.

Results

Table 2 presents the results of ANOVA analyses, along with the decision for each hypothesis. The decision rule for accepting or rejecting a hypothesis was agreed as: In order to accept a hypothesis, in 3 of the 4 items there had to be a significant result of the ANOVA showing a higher average value for the data under the confinement crisis condition than under the ‘normal’ condition (${\bar{X}}_1>{\bar{X}}_0$). To test significance, two coefficients were implemented (F and Welch). F coefficient serves to test inequality of means when variance between groups is homogeneous. Welch coefficient is for the existence of heterogeneous variances among groups. These two coefficients concurred in all cases. Thus, for the hypotheses related to e-store’s service performance, two of the four items were accepted (product catalogue width and order form ease of use). For the hypotheses associated with brand perceptions, one of the two was accepted (the e-store brand perception). Finally, regarding customers’ perceived performance hypotheses, one was rejected (satisfaction), one was accepted (behavioural intentions), and two were argued as a mild acceptance (trust and overprice disposition). Why the mild acceptance? In the case of the hypothesis regarding customer trust, for the four items, the means of the confinement group were superior to the means of the normal condition group (${\bar{X}}_1>{\bar{X}}_0$, confirming the direction of the hypothesis). Additionally, two of the four items were significant in the ANOVA. In terms of the hypothesis regarding overprice disposition, in three of the four items, the means were superior for the confinement group in comparison to the normal condition group (${\bar{X}}_1>{\bar{X}}_0$), and two of the four items were significant. In case of the evidence for these two hypotheses, it did not fully comply with the established decision rule; however, it was quite close.

Table 2. ANOVA analyses to test whether mean of group 1 (confinement) is superior to group 0 (normal)

Hypothesised relationship and measurements		F	Welch	${\bar{X}}_1>{\bar{X}}_0$	Decision
H1	Confinement→home-page-attractiveness-1	1.67	1.43	No	Reject
	Confinement→home-page-attractiveness-2	1.34	1.27	Yes
	Confinement→home-page attractiveness-3	.16	.15	Yes
	Confinement→home-page attractiveness-4	7.94**	7.07**	No
H2	Confinement→product-catalogue-width-1	15.76**	15.01**	Yes	Accept
	Confinement→product-catalogue-width-2	12.65**	11.85**	Yes
	Confinement→product catalogue width-3	11.77**	11.60**	Yes
	Confinement→product catalogue width-4	15.63**	14.83**	Yes
H3	Confinement→order-form-ease-of-filling-1	13.77**	13.13**	Yes	Accept
	Confinement→order-form-ease-of-filling-2	13.79**	13.15**	Yes
	Confinement→order-form-ease-of-filling-3	23.80**	23.16**	Yes
	Confinement→order-form-ease-of-filling-4	35.28**	37.22**	Yes
H4	Confinement→customer-support-effectiveness-1	8.26**	7.50**	Yes	Reject
	Confinement→customer-support-effectiveness-2	.36	.32	Yes
	Confinement→customer-support-effectiveness-3	.05	.05	Yes
	Confinement→customer-support-effectiveness-4	.35	.30	No
H5	Confinement→product-brand-perception-1	.76	.68	No	Reject
	Confinement→product-brand-perception-2	.03	.03	No
	Confinement→product-brand-perception-3	.01	.01	No
	Confinement→product-brand-perception-4	1.23	1.13	Yes
H6	Confinement→e-store-brand-perception-1	5.98*	5.66*	Yes	Accept
	Confinement→e-store-brand-perception-2	5.06*	4.58*	Yes
	Confinement→e-store-brand-perception-3	8.86**	8.28**	Yes
	Confinement→e-store-brand-perception-4	2.08	1.88	Yes
H7	Confinement→satisfaction-1	2.46	2.09	Yes	Reject
	Confinement→satisfaction-2	.48	.42	Yes
	Confinement→satisfaction-3	.02	4.64	Yes
	Confinement→satisfaction-4	1.48	1.25	Yes
H8	Confinement→trust-1	8.13**	7.33**	Yes	Mild accept
	Confinement→trust-2	14.55**	14.30**	Yes
	Confinement→trust-3	2.58	2.33	Yes
	Confinement→trust-4	.30	.27	Yes
H9	Confinement→behavioural-intentions-1	5.66*	5.14*	Yes	Accept
	Confinement→behavioural-intentions-2	6.50*	5.77*	Yes
	Confinement→behavioural-intentions-3	6.08*	5.36*	Yes
	Confinement→behavioural-intentions-4	2.64	2.32	Yes
	Confinement→behavioural-intentions-5	11.99**	11.06**	Yes
H10	Confinement→overprice-1	4.73*	4.53*	Yes	Mild accept
	Confinement→overprice-2	6.15*	6.02*	Yes
	Confinement→overprice-3	.81	.78	Yes
	Confinement→overprice-4	1.09	1.01	No

*p-value ≤ 0.05

**p-value ≤ 0.01

${\bar{X}}_1$: group 1 (confinement crisis) mean

${\bar{X}}_0$: group 0 (normal condition) mean

Discussion and conclusions

This work offers some viable contributions to literature. Such contributions will be considered in the following lines. A term that has been newly introduced by the current work is ‘un-intensifier’ that is related to the notion of the mitigating or reducing effect that certain factors can have on consumers’ expectations of quality regarding a service. This term was inspired by use of the word ‘intensifier’ by Zeithaml, Berry, and Parasuraman (1993) as related to factors that can exert an effect of escalating consumers’ expectations of quality. Thus, in the current study, it is proposed (with some empirical evidence) that an extreme situational factor (e.g. a pandemic, earthquake and flood) can exert a mitigating or un-intensifier effect on consumer expectations of quality from a service provider. In addition, an explanation of this effect related to the forgiveness sentiment by consumers is offered, as consumers would be empathetic towards the circumstances of extreme catastrophic conditions and understand that a service provider cannot perform at the same level as in normal conditions in this situation.

Another possible contribution of the current work is the tentative empirical evidence of the effect that an extreme situational factor (as the confinement crisis due to the COVID-19 pandemic) can have on the level of perceived service performance. This effect is consistent with the reduction in consumer expectations generated by this same factor that is addressed above and was originally proposed by Zeithaml, Berry, and Parasuraman (1993). To the best of the authors’ knowledge and based on an extensive literature review, no previous work has addressed this kind of effect in either a conceptual or in an empirical fashion. This contribution could be considered relevant as it provides knowledge regarding how extreme situational factors can affect the perceptions of consumers toward service providers, and it could be relevant for businesses when planning for such extreme (harmful) situational factor that may arise. Thus, the findings related to the possible positive effect of an extreme situational factor (a confinement crisis, in this case) on brand perception is another novelty of this work as it presents supporting empirical evidence, wherein no antecedents of this were found in the literature review. The same can be said for the effect of the extreme situational factor on the expected outcome of service performance, such as purchase intentions toward the e-store.

As noted above, very few studies have addressed the effect of extreme situational factors on consumer perceptions. Those extreme situational factors treated in literature include phenomena such as earthquakes and floods; however, none of them cites a pandemic as an extreme situational factor affecting consumer perception of service performance. Thus, this work offers a small contribution to the current work exemplified in the literature. Moreover, regardless of whether the COVID-19 pandemic is a situational factor relatable with the phenomena that have been associated with this term, this study also offers some specific data about how the pandemic itself has affected the perception of consumers toward e-store services. The study has further exemplified the fact that e-stores play an important role in the provision of goods to consumers, considering the recommended restrictions in visiting physical stores during the pandemic (Asperin 2020; Groth 2020). Hence, in some cases, an increase in the level of usage of e-stores during this pandemic has been observed (Dirgantari et al. 2020).

Apparently, service performance expectations of retail stores tend to have a narrow zone of tolerance (Nadiri 2011). The same may possibly be occurring for retail e-stores. However, as customers might provide better evaluations of a service provider under an extreme situational factor, at least with regard to certain aspects (as found in the current work), it may suggest that customers are readjusting their expectations of performance and widening their zone of tolerance. Nonetheless, current findings also suggest that what customers are readjusting are their expectations as a prediction standard (or perceptions of what constitutes adequate service), rather than their expectations as an ideal standard. This would be consistent with previous propositions reported in literature regarding these two types of expectations (Zeithaml, Bitner, and Gremler 2009, 80; Nadiri 2011).

The zone of customer tolerance in terms of service performance may vary according to the type of service attribute, wherein attributes that are considered more important will tend to have higher zones of tolerance (Walker and Baker 2000). These different levels of zones of tolerance toward service performance by service dimension may explain why the situational factor in the current study had significant effects on the perceived performance of some, but not all, service attributes. Hence, it may be expected that if the zone of tolerance regarding the importance of each service attribute differs, then the situational factor may not have the same effect on each service attribute. Analysing the moderating effect of the importance of a service dimension on the relationship between an extreme situational factor and the perceived performance of this service dimension may constitute an interesting subject for future research.

In considering the results of the current study, it may be possible to make some sense of the hypotheses that were rejected and those that were accepted. Based on the results, the rejected hypotheses are those related to aesthetic, but non-critical, aspects of e-stores’ offerings (page attractiveness, support effectiveness, products and brand perception, satisfaction). Alternatively, it is possible that the elements related to those hypotheses that were accepted (catalogue width, ease of filling, e-store brand, trust, overprice) may refer to more critical aspects of the transaction per se and to the goods that the customer is demanding. Therefore, during a pandemic crisis, it is important that e-stores focus more on that aspects related to the offering and the delivery of goods more than the aesthetic aspects of e-store websites.

Disclosure statement

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

Additional information

Notes on contributors

Jorge Vera-Martínez

Jorge Vera-Martínez Jorge Vera holds a Ph.D. degree in Management Sciences, with honors, from the National University of Mexico (UNAM). Previously, he obtained a MBA degree at the Technological Institute of Mexico (ITAM). He also holds a degree in experimental psychology from the National University of Mexico (UNAM). Before his academic vocation, he developed a career as a professional market research consultant. Currently, Jorge Vera is a research-professor in the Technological of Monterrey Business School in Mexico City Campus. He has taught, at graduate and undergraduate levels, subjects such as marketing research, empirical research methodology, instruments design, strategic marketing, brand management and multivariate methods for marketing decisions. At EGADE Business School, currently, he holds the position as leader-coordinator of the Research Group (GIEE) Consumer Behavior and Conscious Marketing. As a researcher, he is a member of the Researchers National System (SNI) of Mexico’s National Science and Technology Council (CONACyT) since 2006. He has published several research articles in scientific reviews related to topics such as consumer behavior, service/product performance, consumer involvement, perceived value, brand perceptions and marketing strategies.

Andrea Trujillo-León

Andrea Trujillo-León Passionate about marketing, she has combined her studies with the practice acquired as a manager and consultant. In more than 25 years of professional life, she has published academic articles and books about service quality and green marketing. She has trained several generations of professionals obtaining awards and has being recognized for her teaching work. She has developed marketing projects for different companies and organizations. In addition, she is a micro-entrepreneur and coordinator of the book collection ‘Motives to Love Mexico’. She is a researcher at the Retail Research Group and a professor at the Business School of the ‘Tecnológico de Monterrey’ in the area of Marketing. Among her research topics are service quality, satisfaction and loyalty, as well as green marketing, on which she has published research articles and books. She is a member of the Researchers National System (SNI) of Mexico’s National Science and Technology Council (CONACyT) She holds Ph.D. degree in Management Sciences from a joint program between University of North Carolina at Chapel Hill and ‘Tecnológico de Monterrey’.

Fabiola Vásquez-García

Fabiola Vásquez-García Fabiola Vasquez holds a Ph.D. degree in Management Science (Marketing Concentration) from the ‘Tecnológico de Monterrey’ (2016). She holds a Master’s Degree in Logistics and Supply Chain Management from the Popular Autonomous University of the State of Puebla (2011) and a Bachelor in Industrial and Systems Engineer ‘Tecnológciod de Monterrey’ (2007). Fabiola has several years of experience as a business consulter. Formerly, she was a Manager of Products at a prestigious consulting firm in Mexico. Here, she was responsible for developing e-commerce and commercial strategy projects for Mexican and international companies, mainly from the Retail and Automotive industries. She also performed as a Retail Hub leader for Accenture México. Currently, she is a full time professor at the Business School of ‘Tecnológico de Monterrey’ in México City.