Negative memorable experience: North American and British Airbnb guests’ perspectives

Purpose – This paper aims to explore the components of a negativememorable Airbnb experience. Design/methodology/approach – Two studies of North American and British nationals were conducted online using an open-ended survey questionnaire with photo-elicitation via Amazon Mechanical Turk (MTurk). The grounded theory was used to analyse the collected data. Findings – The findings of the current study are destination-specific and generalisation is limited. In addition, this study gathered data using an open-ended survey questionnaire with visual images (photoelicitation technique) inMTurk. Moreover, the study participants weremainlyWesterners. Research limitations/implications – Airbnb could provide hosts with a service quality checklist to warrant quality assurance across listings. Hosts must be informed, guided andmonitored so that service quality standards are fulfilled. In addition, hosts should be incentivised to write an honest and accurate description of their listing. Practical implications – Airbnb can provide hosts with a service quality checklist to ensure standardisation and quality assurance across listings. Hosts must be informed, guided andmonitored so that service quality standards are fulfilled. In addition, hosts might benefit from training or workshops on the role of hosting and service qualitymanagement. Originality/value – This is one of the first studies to explore the components of a negative memorable experience in the context of Airbnb.


Introduction
According to Kim et al. (2012), a memorable tourism experience (MTE) is one that is positively recalled after the event. However, because emotional stimuli with both positive and negative valence contribute to the memorability of an event (Kensigner and Corkin, 2003), negative experiences should also be considered potential MTE components. Considering both the positive and negative components of MTEs will provide researchers with a more comprehensive understanding of the essence of MTEs (Kim, 2014). Indeed, recent studies have argued that negative experiences are a critical component of MTEs (Coudounaris and Sthapit, 2017;Sthapit and Coudounaris, 2018).
The emergence of new forms of accommodation services, such as Airbnb and other shared, peer-to-peer (P2P) hosting services, has changed the consumer profile of today's tourist (Ert et al., 2016). These services are intangibly experienced goods, characterised by inseparability (they are produced and consumed simultaneously) and the fact that their quality cannot be verified before they are consumed (Zeithaml et al., 2006). In addition, service failures are inevitable in the hospitality industry and can occur in both the process and outcome of the service delivery (Lewis and McCann, 2004). These services have an inherently inconsistent nature; guests can always develop adverse feelings, such as anger and frustration, during tourism experiences (Kim, 2014). For these reasons, experiences can vary widely among customers (Sthapit and Bjö rk, 2019a). The concept of Airbnb exacerbates these inconsistencies, as hosts are renting rooms to strangers (Ert et al., 2016), and the quality of the accommodation service is highly dependent upon each host's competence in hospitality (Zhang et al., 2018). Consequently, unforeseen happenings are common, as neither guests nor hosts can determine one another's reliability in advance (Sthapit and Bjö rk, 2019a). For example, in a recent unfortunate incident, a 51-year-old woman from New Mexico was sexually assaulted by her Airbnb host during her stay (Levin, 2017). Such horrific experiences are always possible and generate negative memorable Airbnb experiences, which are the focus of this study.
A review of recently published studies on MTEs shows a focus on tourists' recollections of positive memorable experiences related to souvenir shopping (Sthapit and Bjö rk, 2019b), hotels (Sthapit, 2018a), food (Sthapit et al., 2019) and Airbnb stays (Sthapit and Jimé nez-Barreto, 2018a), including the relationship between MTE dimensions and other constructs . In addition, recent studies related to Airbnb have focussed on value co-destruction (Sthapit, 2018b), service quality attributes (Ju et al., 2019), the consumer experience (Pappas, 2019), distrust (Sthapit and Bjö rk, 2019a), unique experiences (Mao and Lyu, 2017), sharing (Sthapit and Jimé nez-Barreto, 2018c), continuance intention  and discontinuance (Huang et al., 2020). Studies on MTEs have proliferated in tourism research and, to date, have focussed overwhelmingly on positive or pleasant events or feelings. Far fewer studies have focussed on negative accounts of tourism experiences (Mackenzie and Kerr, 2013). In addition, few studies have examined guests' negative experiences with Airbnb in particular (Sthapit 2018b;Sthapit and Bjö rk, 2019a).
The current study builds on both types of MTEs. In particular, this study aims to explore the components of a negative memorable Airbnb experience. This topic was chosen for two reasons. First, this study focusses on the negative aspects of the Airbnb experience, while many previous studies have predominantly examined the Airbnb customer experience from a positive perspective (Li et al., 2019). Second, Airbnb offers an un-standardised experience (Birinci et al., 2018;Huang et al., 2020), which may generate different negative memorable elements than a standardised experience (such as in a hotel). In addition, recent studies suggest that negative dimensions can also influence the experience (Li et al., 2019). Moreover, studies indicate that there is a positivity bias in Airbnb users' comments (Cheng and Jin, 2019;Bjö rk, 2019a, 2019b) and that negative experiences are under-reported (Dann et al., 2019). In the context of this study, a negative experience refers to a tourist's recollection of an on-site Airbnb experience while at a destination and does not include the pre-booking experience with the platform.

Negative emotions, memory and their interrelationships
Some prior studies have indicated that travel is stressful and can be associated with negative emotional outcomes (Mackenzie and Kerr, 2013). If these emotions are sufficiently intense, they can result in the creation of negative memorable experiences (Kim, 2014). Negative emotions result from a mismatch between a person's preferred way of feeling (dictated by his or her current motivational state) and the experience that person is having (Mackenzie and Kerr, 2013).
Tourism experiences are complex psychological processes in which memory plays a paramount role (Larsen, 2007). Memory is "an alliance of systems that work together, allowing us to learn from the past and predict the future" (Baddeley, 1999, p. 1). Episodic memory, which includes individuals' long-term storage of factual memories concerning personal experiences, is considered the type of long-term memory most relevant to the study of tourists' experiences (Larsen, 2007). Episodic memories focus on events and permit tourists to travel backward in subjective time to re-experience previous events (Matlin, 2005). These memories are created from in situ experiences (Tung and Ritchie, 2011).
The most painful or joyous moments in one's life are often remembered in vivid detail. It is well known that emotion can affect both memory encoding and the retrieval processes (Dolcos et al., 2017). Although Piqueras-Fiszman and Jaeger (2015) used the term "memorable experience" to describe an experience with a positive connotation that is emotionally remembered, memory researchers believe that negative valence leads to the creation of a stronger memory than positive valence does (Kensigner and Schacter, 2006).

2.2
The sharing economy, Airbnb, its characteristics and negative incidents linked to Airbnb The sharing economy as an economic model enables individuals to share access to underused goods or services for monetary or nonmonetary benefits (Ferrell et al., 2017). The sharing economy may also require face-to-face interactions between providers and demanders, which creates greater uncertainty related to service quality and personal security (Ert et al., 2016). Sharing economy transactions that require face-to-face interactions involve higher risk, more uncertainty and greater interdependence between parties (Zhang et al., 2018).
Airbnb is a collection of private rooms, apartments and homes, each owned by an individual owner, located in different places and managed independently (Mody et al., 2019). According to Zervas et al. (2014), Airbnb can potentially expand supply wherever houses and apartment buildings already exist, and individual hosts solely determine the prices of their Airbnb listings. Airbnb has become a key competitor of not only other online travel agents (e.g. Expedia) but also traditional hotels. In addition, Airbnb enables hosts to profit by renting out their available accommodations while providing travellers with a wide range of accommodation options (Sthapit and Jimé nez-Barreto, 2018a).
While Airbnb creates value on both the supply and demand sides, there have been numerous reports of negative incidents linked to Airbnb stays. In addition, several scholarly articles have documented the risks associated with Airbnb stays Bjö rk, 2019a, 2019b). These incidents can be serious: in one recent incident, Airbnb guest Mike Silverman was attacked by his host's Rottweiler during his stay in Salta, Argentina and spent two nights in hospital as a result (Lieber, 2015). These examples illustrate that negative experiences are pivotal to the comprehensive understanding of memorable Airbnb experiences.

Sampling and data collection
A qualitative case study approach was used. More specifically, an online, open-ended survey questionnaire using a photo-elicitation technique was used for data collection. First, informants were presented with images concerning a particular phenomenon (in this case, a negative Airbnb experience) using the online platform Qualtrics. These images were intended to work as metaphorical projective stimuli to motivate and contextualise subjects' participation (Collier, 1967). After viewing the images, participants were asked to write a brief story, prompted by open-ended questions, to gather their narratives, beliefs, experiences and opinions.
Photo elicitation is a tool through which participants can share their knowledge and through which intense feelings and truths can be realised and shared (Collier, 1967). One of the advantages of the photo-elicitation technique is that it increases participants' interest in and engagement with the phenomena being analysed (Matteucci, 2013).
The actual images presented to participants were chosen to represent three negative Airbnb experience scenarios, each in a different stage of the Airbnb experience: pre-visit, on-site and post-visit. After the presentation of the images, participants were instructed to write a brief story about their recent negative experience with Airbnb, prompted by openended questions. These questions included "after looking at these pictures, please describe in detail your negative Airbnb experience. Write only negative things and not positive, based on your experience" and "please, describe in detail what made your negative Airbnb experience memorable. Write only negative things and not positive, based on your experience" The final version of the questionnaire was modified based on feedback collected during pilot testing. Informants were also asked to provide demographic information after answering the open-ended questions.
The online open-ended questionnaire containing images was pilot-tested on five American and five British nationals in Amazon Mechanical Turk (MTurk) to ensure that informants understood the questions correctly. The pilot testing was conducted during the first week of June 2019. The final version of the questionnaire was modified based on feedback collected during the pilot test. In the final study, the survey questionnaire link was posted on MTurk during the past two weeks of June 2019. Upon clicking the link, participants were directed to the visual images and open-ended questions. Only informants who had stayed in an Airbnb within the six months preceding the time of data collection (December 2018-May 2019) and who were nationals of and residing in either the United States or the UK were eligible to complete the survey. These two nationalities were selected because the USA and the UK are two of the five countries with the most Airbnb listings and because, out of the 150 million Airbnb users worldwide, 41.1 million are from the USA (Lock, 2019) and 11.1 million are from the UK (Luty, 2019). Each informant was paid US$0.90.

Data analysis
Grounded theory (Glaser and Strauss, 1967) was used to analyse the collected data. Grounded theory is based on a range of qualitative research methods that use a systematic set of procedures and simultaneous processes of data collection and analysis to develop an inductive-derived grounded theory about a phenomenon (Strauss and Corbin, 1998).
The data analysis procedure followed the standard grounded theory format (Strauss and Corbin, 1990). The first step involved open coding (Glaser and Strauss, 1967). The data were broken into smaller parts and analysed deeply. Second, the data were assembled using axial coding, which is the process of relating codes (categories and concepts) to each other that were developed during open coding (Strauss and Corbin, 1990). The axial coding process reduced the database to a small set of themes or categories that characterised the process under study (Creswell, 2007). Third, selective coding was conducted after axial coding; the core categories were identified, integrated and refined to form a conceptual framework and to elaborate substantive theoretical concepts. The framework was reviewed against prior literature for consistency and validation of the findings (Matteucci and Gnoth, 2017).

Overall demographic and trip characteristics of North American and British informants
Of the 45 North American participants, 22 were female and 23 were male. Their ages ranged from 21 to 57 years old. The majority of informants had a bachelor's degree (39). Half (23) of the informants identified themselves as medium-frequency users of Airbnb (3-5 stays per year). The majority reported using Airbnb while travelling within the USA (33), while nine recently stayed in an Airbnb during international travel. The duration of their stays ranged from one to seven nights. All the participants were American nationals (Appendix).
Of the 30 British participants, 24 were male and six were female. Their ages ranged from 20 to 60 years old. Half of the informants had a bachelor's degree (15). The majority were medium users of Airbnb (19). Slightly more than half had stayed in an Airbnb while travelling abroad (16); of those who used Airbnb for domestic travel within the UK (13), most stayed in London. The duration of participants' stays ranged from two to nine nights. All the participants were British nationals (Appendix).

Negative Airbnb experiences of North American and British informants
In response to the prompt, "please describe in detail your negative Airbnb experience", the poor condition of the room emerged as a common and pivotal attribute in North American informants' experiences (mentioned by 38 of them). Interpretive codes such as "poorly ventilated", "dirty", "terrible", "horrible", "awful", "not clean like seen in the pictures", "dusty and grimy", "worst room", and "extremely dirty" are all indicative of the significance of the poor condition of the room in informants' negative Airbnb experiences. In addition, some of the negative Airbnb experiences were linked to the unpleasant behaviour of the host (mentioned by seven). Interpretive codes such as "host was still cleaning the house", "host had left a mess in the rooms", "had to wait two hours for the host" and "host was playing loud music" are all indicative of the significance of unpleasant behaviour from the host in these guests' negative Airbnb experiences. This is emphasised by the detailed responses of two informants: The majority of the British informants' negative Airbnb experiences were also linked to the poor condition of the room (mentioned by 21 of them). Interpretive codes such as "dirty rooms", "bedding was stained", "covered in dust", "dirty, smelly and did not look like it did in pictures", "more weathered than shown in pictures", "fully dirty rooms" and "not cleaned" are all indicative of the significance of the poor condition of the room to guests' negative Airbnb experiences. In addition, some negative Airbnb experiences were linked to the rude behaviour of the host (mentioned by nine). Interpretive codes such as "host family was annoying", "very rude about what I can do and cannot do", "getting shouted at for no reason" and "host did not greet us at all" are indicative of the significance of rude behaviour from the host in guests' negative Airbnb experiences. This is further exemplified by the detailed replies given by two informants:

Components of a memorable negative Airbnb experience (North American and British informants)
A dirty room (24), bad or deceptive host behaviour (16) and poor customer service (5) emerged as the top contributors to North American informants' memorable negative Airbnb experiences. Table I shows the codes indicating the importance of dirty rooms, bad hosts and poor customer service on the participants' negative memorable Airbnb experiences.
The following two responses detail this further. The fact that it was just dirty. The pictures made it seem clean and normal, but those photos must have been taken some time ago [. . .] I am glad that I did not catch anything while I was there since the place was just so dirty.
Among the British informants, the findings indicate that poorly maintained rooms (27) and rude behaviour from hosts (3) were the primary contributing factors to the forming of memorable negative Airbnb experiences. Table II shows the codes that exemplify the importance of poorly maintained rooms and unpleasant host behaviour in the informants' negative memorable Airbnb experiences.
The two responses below highlight this in further detail.
Participant UK9 (male, 21, visiting Sheffield) reported: The negative experience that made the Airbnb memorable was the dirty room. The room was unbearable and it was fairly obvious and certain to me that there had not been a thorough and intense cleaning in preparation for my arrival. The stains on the carpet too also made it obvious that the cleaning had not been done properly. The doors had stains on them and the handle did not look pleasant to touch.
Participant UK13 (male, 24, visiting Amsterdam) reported: "mould in room", "just so dirty", "bugs fly into my face", "bugs", "one cockroach per day", "ants were endless", "unclean", "dishes with food still on them scattered in the room", "flies and bugs in the room", "room was full of dust", "not cleaned", "needed to be cleaned up", "unclean", and 'infested with bed bugs "host did not care and did not do anything about it", "poor host service", "host did not help us and respect us as customers", "no response", "room much smaller than advertised", "pictures look nothing like what was advertised" "customer service did not really work to resolve the issue", "awful customer service", "the service agents were largely unhelpful about this", and "did not seem that they were very concerned" The components of a dirty and poorly maintained room and bad, deceptive and rude host behaviour are attributes of poor service quality. A dirty and poorly maintained room has a negative impact on service quality attributes such as facility condition (Ert et al., 2016) and accommodation experience (Liang et al., 2018). The accommodation facility provided by the host is considered a key service quality dimension in Airbnb as a main product for an overnight stay (Ju et al., 2019). A dirty room also negatively affects other service quality attributes, such as appearance, cleanliness (Parasuraman et al., 1988), reliability and trustworthiness (Grö nroos, 1990) and functionality (Johnston, 1995).
Many study participants' encountered bad, deceptive and rude behaviour from hosts. This behaviour runs directly contrary to the positive service-related attributes associated with hosts in the P2P marketplace, namely, understanding and caring (Priporas et al., 2017) and hospitality hosting behaviour (Lalicic and Weismayer, 2018). In addition, unpleasant behaviours such as "no answer again", "no response" and "never really heard from him", highlight the lack of communication on the part of multiple hosts. Certain other keywords can be linked to low benevolence or the belief that the trustee (the Airbnb host) wants to do well by the trustor (the Airbnb guest) for reasons other than an egocentric profit motive (Jarvenpaa et al., 1998). Such keywords include, for example, "host did not care and did not do anything about it" and "they do not deliver on their promises". Some narratives also clearly indicated that the Airbnb host acted dishonestly and cheated the guest out of the promised services, leading to distrust toward the Airbnb host. These include: "host is not reliable", "misleading host", "host had clearly lied", "misleading pictures", "room much smaller than advertised" and "pictures look nothing like what was advertised". This finding supports studies indicating that poor behaviour from hosts is a common cause of guests' negative Airbnb experiences (Sthapit, 2018b).
Some guests contacted Airbnb's customer service to remedy their negative Airbnb experience. This strategy has been categorised as a "complaining behaviour" and requires the consumer to directly act on service failure (Menon and Dubé , 2004). Some informants mentioned feeling helpless because of the lack of responsiveness and empathy from customer service personnel at Airbnb. This made their negative Airbnb experiences even worse. "awful smell in the room", "each bed was covered in dust", "hairs littered the place", "dusty", "bed bugs and spiders", "damp/ mould on the walls", "unkept", uncomfortable', "run down", "apartment was very old", "needed a lot of work", "beat up looking' and "nothing like the pictures" 'Tried to ring the host on the mobile several times, "but no answer again", "host is not reliable", "misleading host", "host did not care", and "no view as promised by the host" and "host made me feel uncertain" Poorly maintained rooms Rude behaviour from hosts Poorly maintained rooms and rude behaviour from hosts contributed to a memorable negative Airbnb experience The findings support studies indicating that the service quality of Airbnb is less predictable for the guest, leading to negative experiences (Cheng and Jin, 2019). There are multiple reasons for poor service quality from Airbnb. One is the shift in the service delivery process from proficient providers to individual hosts (Tussyadiah and Zach, 2017). Some studies indicate that these hosts are cheap rental accommodation providers who profit by renting out their accommodations, usually at lower rates than comparable hotels (Sthapit and Jimé nez-Barreto, 2018b). Another is the fact that transaction partners on Airbnb (the host and guest) are unable to inspect and evaluate the service before purchase or consumption (Yoon and Occeñ a, 2015). Third, there are no hospitality standards (Finley, 2013) and the listings are unregulated (Ert et al., 2016), including no health and safety regulations that hosts must follow (Bonnington, 2015). The findings also support studies indicating that risk perceptions in the context of Airbnb are related to concerns about receiving poor quality service (Sthapit, 2018b) and that guests are more likely to experience unpredictable and inconsistent service quality with Airbnb than in traditional accommodations (Huang et al., 2020). In addition, the findings tell a different story from studies indicating that Airbnb tourists enjoy a more personalised service quality (Mao and Lyu, 2017) and that household amenities are not of high significance when formulating Airbnb service quality perceptions (Priporas et al., 2017).

Conclusion
The main contributions of the present work include the three components of negative memorable Airbnb experiences, namely, dirty and poor room conditions; bad, deceptive and rude host behaviour; and poor customer service. The findings indicate that when an Airbnb guest is provided with poor accommodation; experiences rude, deceitful or unpleasant behaviour from the host; or is offered poor customer service by the company itself, he or she is more likely to have a negative memorable experience. The findings support studies indicating that the physical environment, in this context, the room conditions (Cheng and Jin, 2019;Huang et al., 2020) and favourable hosting behaviour (Cheng and Jin, 2019;Lalicic and Weismayer, 2018;Sthapit, 2018b;Jimé nez-Barreto, 2018a, 2018b) are considered crucial for guests while staying in Airbnb rental properties. The findings also share similarities with studies indicating poor customer service experienced by guests after a service failure in the context of Airbnb (Huang et al., 2020;Sthapit, 2018b;Sthapit and Bjö rk, 2019b).
The findings contribute to existing service quality research by highlighting the significance of service quality to guests' on-site Airbnb experiences (Cheng and Jin, 2019;Huang et al., 2020) and during the service recovery process (Huang et al., 2020;Sthapit and Bjö rk, 2019b). In addition, the findings support previously developed service quality models (Grö nroos, 1982); the identified components of negative memorable Airbnb experiences can, indeed, be categorised as technical (dirty and poor room conditions) or functional (bad, deceptive and rude host behaviour; poor customer service). The findings also support studies indicating that negative experiences should be discussed as a component of MTEs (Li et al., 2019; Sthapit and Jimé nez-Barreto, 2018a).
In terms of the managerial implications, first, given the dirty and poor room conditions of Airbnb rentals as reported in the informants' narratives, Airbnb could provide hosts with a service quality checklist to warrant quality assurance across listings. Hosts must be informed, guided and monitored so that service quality standards are fulfilled. In addition, hosts should be incentivised to write an honest and accurate description of their listing. Strict action against deceitful, dishonest hosts should be taken when guests report inconsistencies between advertised accommodations and reality, such as the misleading pictures and smaller-than-advertised rooms mentioned in numerous narratives.
Second, to address bad, deceptive and rude host behaviour towards guests, Airbnb should instruct hosts to remain well mannered, conscientious and responsive when hosting. Hosts must be required to take responsibility for service failures if guests complain about the poor quality of the listing and/or unresponsive, unhelpful, unprofessional, unpleasant host behaviour or a lack of hospitality from the host. The study suggests that the Airbnb host's role should transform from a low-priced rental property manager to a sincere lodging service provider who helps to make their guests' stays worthwhile. In addition, hosts who are repeatedly perceived as unresponsive, unhelpful, unprofessional and lacking in hospitality toward guests should not be allowed to host on Airbnb.
Third, customer support agents should be further trained to provide responsive, caring and professional service. These trainings should focus on upgrading skills for handling complaints and on effective service recovery efforts after a failure. In addition, appropriate, efficient, well-enacted and timely compensation procedures could help to respond to service failures.
The findings of the current study are destination-specific and generalisation potential is limited. In addition, this study gathered data using an open-ended survey questionnaire with visual images (photo-elicitation technique) in MTurk. Moreover, the study participants were mainly Westerners. Lastly, this study was restricted to recollections of on-site Airbnb experiences while at a destination and did not include the pre-booking experience with the platform. As P2P accommodation services such as Airbnb use internet technology as a platform, on which guests are influenced by descriptions and presentations of properties (Sthapit and Bjö rk, 2019c), future studies should examine how technology influences guests' memorable Airbnb experiences.

Corresponding author
Erose Sthapit can be contacted at: erose.sthapit@haaga-helia.fi For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com Notes: Ã Very high (more than 10 stays in a year). High (between 6-10 stays in a year). Medium (between 3-5 stays in a year). Low (Less than 3 stays in a year). None (zero stays in a year). NA Ã = Not Available