CC BY
0Volume 8 Issue 2, 2024, pp. 76-93
doi: 10.22363/2521-442X-2024-8-2-76-93
Original Research
Framing negative eWOM via conjunctions: The language of customer communication in the luxury hotel sector on social media
by Yau Ni Wan
Yau Ni Wan
ORCID 0000-0002-7585-1745 H ynwan@hksyu.edu Hong Kong Shue Yan University, Hong Kong
Article history Received March 29, 2024 | Revised April 30, 2024 | Accepted June 5, 2024 Conflicts of interest The author declared no conflicts of interest Research funding No funding was reported for this research doi 10.22363/2521-442X-2024-8-2-76-93
For citation Wan, Y. N. (2024). Framing negative eWOM via conjunctions: The language of customer communication in the luxury hotel sector on social media. Training, Language and Culture, 8(2), 76-93.
In the digital age, social media has a significant impact on consumer perceptions. This business communication study examines the linguistic strategies used to express and repair negative electronic word of mouth (eWOM) in Hong Kong's luxury hotel sector. The data, which were collected from travel forums, consist of 17,344 words from online hotel reviews (11,647 words) and responses (5,697 words) made by the visitors and senior hotel management of 18 five-star hotels. Using systemic functional linguistics as a framework, this study analyses the logical relations that exist between expectant and concessive conjunctions, as well as the evaluative judgment resources present in an appraisal system. The analysis focuses on the poor service standards framed in the complaints and how the hotel management responds to such situations differently, particularly with respect to repairing customer expectations. This study also examines how loyal fans and frequent travellers present their customer identity when dealing with service-related issues. By developing an understanding of how luxury hotels manage customer complaints, repair customer relationships, and protect their brand reputation through dynamic interactions, this study can provide valuable communication insights for the hotel sector.
KEYWORDS: negative eWOM, customer expectation, external conjunctions, judgement, relationship repair, luxury hotels
This is an open access article distributed under a Creative Commons Attribution-Noncommercial 4.0 International License (CC BY-NC 4.0), which allows its unrestricted use for non-commercial purposes, subject to attribution. The material can be shared/adapted for non-commercial purposes if you give appropriate credit, provide a link to the license, and indicate if changes were made.
Hong Kong, an international city for tourism and business travel, has a well-developed hotel industry due to its strategic location, financial district and cultural diversity (Kucukusta et al., 2013; Law et al., 2015; Wong & Kwan, 2001). However, the industry faces a number of challenges, including keen competition from surrounding cities, rising operating costs, changing consumer preferences and managing cultural differences (Dávid et al., 2024; Deng et al., 2023; Kulakhmetova et al., 2022). Within a short hotel stay, hotels must cater to a diverse customer population, including international tourists, mainland visitors and regional travellers with different purposes to their maximum satisfaction. To remain competitive, the hotel management must improve service quality, develop strong customer interactions and relationships, and build brand loyalty with their customers.
1. INTRODUCTION
The rise of social media has transformed the way businesses interact with their customers, with the hotel industry being no exception (Appel et al., 2020). Online travel forums and review systems have become influential in shaping customer purchasing decisions (Gupta, 2019; Lai et al., 2021; Varkaris & Neuhofer, 2017). While social media allows hotels to engage with customers more frequently, it also exposes them to public criticism such as negative word of mouth (WOM), which can damage their brand image and deter potential guests (Colmekcioglu et al., 2022; Wan & Forey, 2024). Electronic word of mouth (eWOM) on digital platforms is emerging as a new customer marketing tool (Verma & Yadav, 2021) that influences consumer decisions in the service industry (Yuan et al., 2020). Negative eWOM refers to the digital spread of negative customer comments, which poses a major challenge to brands as it can be intrusive
© Yau Ni Wan 2024
Licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
by Yau Ni Wan
'This study aims to investigate the underlying discourse patterns and attitudes of writers towards critical comments in relation to the reviews of luxury hotels on social media. External conjunctions, such as 'and', 'but', 'then', and 'because', are conjunctions that link clauses and indicate logical relationships between ideas. These conjunctions frame logical relations in texts and express the writer's ideas and expectations about a situation. Judgment resources in the appraisal system refer to language elements that express the writer's evaluation of others' behaviour, such as 'helpful', 'experienced'or 'rude' hotel staff
and negatively impact online brand presence (Dwivedi et al., 2021). Negative eWOM can influence consumer attitudes in ecommerce (Kietzmann & Canhoto, 2013), so managing it is important for hotel brands.
Heskett et al. (1994) investigated the relationship between customer satisfaction and discovered that customers who reported low satisfaction and low ratings were more likely to complain about poor service whenever possible. To maintain customer satisfaction, hotel management must effectively respond to negative WOM and repair customer relationships (Olson & Ro, 2020). Understanding the communication strategies and linguistic features used by guests when expressing dissatisfaction with their hotel experience is crucial for hotels to effectively mitigate these new challenges.
This study aims to investigate the underlying discourse patterns and attitudes of writers towards critical comments in relation to the reviews of luxury hotels on social media. External conjunctions, such as and, but, then, and because, are conjunctions that link clauses and indicate logical relationships between ideas. These conjunctions frame logical relations in texts and express the writer's ideas and expectations about a situation (Martin & Rose, 2007). Judgment resources in the appraisal system refer to language elements that express the writer's evaluation of others' behaviour (Martin & White, 2005), such as helpful, experienced or rude hotel staff.
This study first examines the use of external conjunctions in online complaint interactions, the logical relations developed, and the customer expectations that they convey. In particular, this study investigates how visitors use external conjunctions to frame their dissatisfying experiences in complaints. These conjunctions are classified into the following four types, which were proposed by Martin and Rose (2007) and Halliday and Hasan (1976): addition, comparison, time, and consequences.
The study then compares the statistical similarities and differences between the conjunctions used by guests and hotel management, as well as the flow of conjunctions used in both guest complaints and management responses, to identify the negotiation patterns in complex and serious complaints. Finally, the research also explores the appraisal system, including judgment
resources, to understand how customers use their special customer identities, such as loyal customers and frequent travellers, in the context of negative eWOM. To achieve these goals, this study addresses the following research questions (RQ).
RQ1. What are the external conjunctions that frame online hotel complaints and responses in terms of addition, comparison, time, and consequences, as suggested by Martin and Rose (2007)?
RQ2. How do external conjunctions and appraisal resources, especially judgment resources, shape emotional labour and customer identity in negative eWOM?
2. THEORETICAL BACKGROUND
2.1. Negative eWOM studies of hotel reviews on social media and SFL
Social networking sites have emerged as an effective means of disseminating information and shaping consumer perceptions (Baker & Hashimoto, 2024; Xu & Wu, 2000). Previous research has highlighted the importance of linguistic analysis in understanding consumer behaviour and expectations through online comments, while effective communication strategies are essential for developing strong customer-brand relationships (Li et al., 2023; Van Herck et al., 2023; Grishechko, 2021). The key studies on hotel eWOM on social media platforms that are relevant to the present study are reviewed in Table 1.
Levy et al. (2013) and Min et al. (2015) conducted hotel communication studies on the linguistic strategies used to express complaints in reviews and gratitude, empathy, and apology in hotel responses. Levy et al. (2013) found that the most common complaints were related to front desk staff issues, bathroom issues, dirty rooms, and noisy guest rooms. They also found that highly rated hotels often expressed gratitude, offered apologies, and provided explanations without discussing compensation adjustments. Min et al. (2015) further investigated empathy statements, paraphrasing statements, and response time by conducting an analysis of variance (ANOVA) test on the reviews of 176 students who underwent a hotel experience at a Mid-Atlantic university. Their results suggested that hotels that respond to negative reviews in a more personalised and less generic way have the potential to improve their ratings. In another hotel language study, Perez-Aranda et al. (2018) found that competence and commitment significantly influenced positive electronic word of mouth (eWOM), suggesting that identifying these predictors can increase eWOM favourability while maximising hotel resources and efforts. Mariani et al. (2019) further supported the importance of language proficiency among hotel service providers and social media reviewers. They conducted a comprehensive analysis of 500,000 online hotel reviews for Rome and Moscow.
More recently, studies by Sánchez-Franco and Alonso-Dos-Santos (2021) and Fastrich (2024) explored the role of gender in online reviews, focusing on small lexical-grammatical features such as vocabulary terms and the first-person perspective (FPP). Sánchez-Franco et al.'s (2021) study discovered that
Volume 8 Issue 2, 2024, pp. 76-93
doi: 10.22363/2521-442X-2024-8-2-76-93
Table 1
Summaries of the studies of hotel eWOM on social media platforms
AUTHOR (S) /YEAR
COUNTRY
SAMPLE EXAMINED
RESEARCH FOCUS
MAJOR FINDINGS
Levy et al. (2013)
USA 1,946 one-star reviews from 10
well-known internet review sites and 225 management comments from 86 hotels in Washington, D.C.
Content and statistical analysis
Most complaints are about front desk staff, bathroom problems, dirty rooms, and noisy guest rooms. Highly rated hotels often express gratitude, apologise, and explain the problem without discussing compensation adjustments.
Min et al. (2015) USA
176 student reviews at a Mid-Atlantic university with a hotel experience (108 females and 68 males with an average age of 21)
Analysis of variance (ANOVA) test on empathy statements, paraphrasing statements, and the speed of response
Hotels that respond to unfavourable reviews in a more personal and less generic way - by way of empathy or paraphrasing statements - can increase their ratings.
Perez-Aranda et al. Spain
(2018)
Reviews of 335 hotels collected from TripAdvisor, and questionnaires and follow-up telephone interviews
Correlation and regression analyses were conducted on the predictors and factors that directly affect positive eWOM reviews
Competence and commitment have a major impact on positive eWOM. This suggests that finding these predictors can improve the valence of eWOM and maximise hotel resources and efforts.
Mariani et al. (2019) Italy and Russia Nearly 500,000 online hotel reviews in Italy (i.e., Rome) and Russia (i.e., Moscow) Ordinary least squares (OLS) regressions were conducted to investigate the extent to which the use of consumers' native language influences their online review behaviour. The language skills of hotel service providers have a positive impact on online review ratings, while the experience of reviewers in writing reviews is critical in accurately reflecting the travel experience.
Sánchez-Franco & Alonso-Dos-Santos (2021) USA (New York) 30,338 online reviews (54% by females and 46% by males) Text mining was performed to analyse the distribution of vocabulary terms in online reviews The study showed that gender influences the price predictions of Airbnb services, with female guests being more motivated and emphasising location and local experiences and male guests focusing on hygiene and facilities.
Fastrich (2024) The UK and Germany 160 online English reviews and 160 German reviews An analysis was conducted on the first-person perspective (FPP) and cognitive linguistic theories of construal English speakers prefer 'person orientation', while German speakers prefer 'content orientation'. The FPP appeared more frequently in English reviews, thereby implying that the
cognitive processing of the hotel stay was more personalised.
gender influenced price predictions for Airbnb services, with female guests being more intrinsically motivated and emphasising location and local experiences and male guests prioritising hygiene and facilities. Fastrich (2024) found that English speakers preferred person orientation, whereas German speakers preferred content orientation. English reviews had a higher frequency of FPP, suggesting a more personalised cognitive processing of the hotel stay. Based on these studies, it can be concluded that the communication landscape of online hotel media is shifting away from thematic issues and towards more specific linguistic characteristics.
The present study seeks to build on these important findings by examining the linguistic characteristics of negative online reviews and response strategies. Existing research has yet to discover the linguistic dynamics in the interactions between guests and hotel managers, particularly the use of conjunctions to explore the logical relationships between clauses and sentences. In this study, systemic functional linguistics (SFL) can be used to analyse logical relationships and linguistic patterns in online complaints. SFL views language as a social semiotic system that generates meaning in social situations (Halliday & Matthiessen, 2013). SFL is a comprehensive framework for analysing how
by Yau Ni Wan
language choices and text structures reflect social relations and dynamics in the context of use (Bloor & Bloor, 2013; Schleppegrell & Oteíza, 2023). It allows for an in-depth examination of the linguistic resources used by dissatisfied guests to project their negative experiences, as well as the responses of luxury hotels to negative eWOM.
2.2. Conjunctions and logic relations in guest-hotel management interactions
Halliday and Hasan (1976) first used grammar-based approaches to study conjunctions as a grammatical resource for linking phrases, clauses, and sentences in texts. Later, Martin and Rose (2007) proposed the use of conjunctions as a discourse-semantic system for analysing conjunctive relations. Conjunctions
such as since and because indicate causal relationships and reasons for dissatisfaction, whereas contrast conjunctions such as but and yet indicate contrast and emphasise the difference between expectations and reality (Martin & Rose, 2007). These external conjunctions are important in shaping discourse structures and communicating the expectations and concessions expressed by hotel guests and management. They implicitly establish the order of events and contribute significantly to the overall linguistic structure and intensity of complaints and repairs, but they have also been overlooked in most previous studies, as shown in Table 1. Conjunctions can be classified as internal or external (Martin & Rose, 2007). The main difference between internal and external conjunctions is that internal conjunctions are more commonly used to connect ideas in an academic essay; for example, a writer might use internal conjunctions thus, also and further to build the arguments and support the central ideas (Martin & Rose, 2007). However, external conjunctions such as besides, instead of, even if and as soon as are more commonly used to connect different activities and events when narrating an experience (Martin & Rose, 2007). The present study focuses on external conjunctions to reveal the logical relationships between how hotel participants logically organise different events, reflect on their emotions and experiences, and construct meanings in the texts. External conjunctions have four types of logical relationships: addition, which most often adds activities together; comparison, which compares activities as similar or different; time sequences, which order activities chronologically; and consequences, which refer to how activities are causally related (Martin & Rose, 2007). These conjunction dimensions aid in exploring the relationships between processes when using addition, comparison, sequencing, and explanation. The external conjunctions in Table 2, as developed by Martin and Rose (2007, p. 153), are intended as a useful reference for identifying the role of conjunctions in textual analysis for the present study.
Continuatives are a small collection of connectors used to create logical relationships that represent addition (too, as well), comparison (only, just), and time (again, already) (Martin & Rose, 2007, p. 141). Continuatives manage expectations in terms of comparison and time. For example, comparative con-tinuatives indicate that an event is more or less than expected,
whereas time continuatives indicate that something will happen sooner, later, or take longer than expected (Martin & Rose, 2007). By analysing these conjunctions and continuatives in complaints and responses, we can systematically investigate the event sequence, underlying reasons, motivations, and feelings that drive negative word of mouth (WOM) about the hotel industry on social media platforms.
2.3. Emotional labour in the hospitality industry and the appraisal system (judgement resources)
The sociologist Hochschild (1979) coined the term emotional labour to describe the process of controlling one's emotions in response to external pressures in the workplace. Employees who interact with customers must regulate their emotions to project a positive image (Ashforth & Humphrey, 1993; Gong et al., 2020). This concept involves controlling and expressing appropriate emotions to maintain positive customer relationships. Previous linguistic research has explored the idea of emotional labour with respect to VIP members of frequent flyer programmes (Tuzovic, 2010) and customer loyalty (Lee et al., 2023). Tuzovic's (2010) study examined online customer behaviour, specifically negative WOM and response strategies regarding airline frequent flyer programmes. Their study found that loyalty programmes can lead to negative feelings, manipulation, retaliation, low ratings, and non-recommendation in complaints. The study provided a theoretical 'customer frustration-defection' framework for understanding dysfunctional online customer behaviour and loyalty programmes. The findings have implications for tracking the perceived quality of loyalty programmes and developing proactive response strategies. In addition, Lee et al. (2023) examined customer loyalty and service dimensions in Airbnb reviews. Using topic modelling and expectancy confirmation theory, they identified 'information' as a major theme and 'homeliness' as a distinct dimension associated with behavioural and attitudinal loyalty. The findings indicated that customer loyalty is enhanced by helping hosts and home-sharing platforms identify customer concerns. The present study extends these findings by exploring the concept of emotional labour in hotel service providers, which is used to guide communication between guests, such as frequent travellers or VIP members, and hotel staff regarding service delivery and complaint resolution. This concept facilitates research into the interpersonal strategies used by luxury hotels to deal with guest complaints. In addition to external conjunctions, the appraisal system developed by Martin and White (2005) in SFL allows for a systematic interpersonal analysis of the attitude, engagement, and gradation of emotions expressed in language, thereby revealing strategies associated with negative eWOM. The evaluative meanings extracted from customer complaints and hotel management responses are evaluated. In particular, the appraisal resources highlight the emotional labour that luxury hotel employees must perform. An in-depth examination of these resources can enhance our understanding of a hotel's responsiveness and ability to address customer concerns.
Training, Language and Culture rudn.tlcjournal.org
Volume 8 Issue 2, 2024, pp. 76-93 doi: 10.22363/2521-442X-2024-8-2-76-93
Table 2 External conjunctions (Martin & Rose, 2007, p. 153)
TYPES EXAMPLES
Addition Additive Adding Subtracting and, besides, both... and nor, neither... nor
Alternative or, either... or, if not., then
Comparison Similar like, as if
Opposite whereas, while
Different Replacing instead of, in place of, rather than
Expecting except that, other than, apart from
Time Successive Sometime Immediate after, since, now that, before once, as soon as, until
Simultaneous as, while, when
Cause Expectant because, so, therefore
Concessive although, even though, but, however
Means Expectant by, thus
Concessive even by, but
Condition Open Expectant Concessive even if, even then even if, even then
Closed unless
Purpose Desire Expectant Concessive so that, in order to, in case even so, without
Fear lest, for fear of
To summarise, this study builds on previous Systemic Functional Linguistics (SFL) research to explore the use of external conjunctions and appraisal resources (judgement) in negative eWOM in the luxury hotel industry. It aims to explore the linguistic dynamics that drive online customer complaints and management responses in the hotel industry. The first research question is concerned with identifying the external conjunctions used to frame online complaints and responses, and analysing their discourse patterns to discover the structure in which customers express dissatisfaction and how hotel management responds and repairs. In the second research question, we explore how these conjunctions, together with judgment resources, shape the emotional labour performed by hotel staff, and we also interpret the formation of customer identity in the context of negative eWOM. The research could advance our knowledge of linguistic strategies used in negative eWOM communication in the luxury hotel sector on social media.
3. MATERIAL AND METHODS
A linguistic corpus of complaints and responses from a social media travel forum for 38 five-star luxury hotels in Hong Kong from 2010 to 2023 was collected and analysed. Data was collected from various online travel forums, with reviews and responses selected based on specific criteria such as complexity of the complaints, sufficient length of review, date range (2010-2023) and relevance to the focus of the study. These reviews included guest complaints and hotel management responses to negative eWOM, which were then used to investigate the complex dynamics of customer relationship repair in luxury hotel reviews.
Guest complaints are rarely formal, and dissatisfied guests casually describe their experiences. Some of the complaints were written informally, with grammatical and spelling errors, as well as incomplete sentences and paragraphs. However, the responses from hotel management representatives were generally
by Yau Ni Wan
well-written and formal. The studied travel forum allows guests to rate their stay on a scale from five, most satisfied, to one, least satisfied. To investigate the most serious and complex complaints, this study only considered guest comments with the least satisfied scores in these hotel reviews. In this public forum, not every guest comment received a formal response
from hotel management. We found that only the most serious negative WOM attracted the attention of hotel management. Our dataset includes 76 complaints and responses, comprising 38 guest complaints (11647 words) and 38 hotel management responses (5697 words), for a total of 17344 words from 18 hotels, as shown in Table 3.
Table 3
Summary of the corpus of guest complaints and management responses
TEXT POST TOPIC GUEST COMPLAINTS (WORDS) MANAGEMENT RESPONSE (WORDS) HOTEL
1a, 1b Dreadful staff (and rooms) 348 (1a) 132 (1b) A
2a, 2b Very noisy and poor management 407 (2a) 90 (2b) B
3a, 3b Disappointed... Overrated? 376 (3a) 91 (3b) B
4a, 4b So disappointed! 145 (4a) 255 (4b) C
5a, 5b Theft and poor service 163 (5a) 367 (5b) D
6a, 6b Terrible experience 151 (6a) 212 (6b) D
7a, 7b Very disappointing 188 (7a) 139 (7b) E
8a, 8b Zero attention 105 (8a) 151 (8b) F
9a, 9b We got Clockenflapped 621 (9a) 323 (9b) F
10a, 10b Pathetic. No delivery of goods. 581 (10a) 65 (10b) F
11a, 11b Our third visit - STAFF RUNE EVERYTHING 222 (11a) 85 (11b) G
12a, 12b So disappointed and let down 151 (12a) 175 (12b) G
13a, 13b Yeah, right 60 (13a) 236 (13b) H
14a, 14b Overpriced Dinosaur 122 (14a) 202 (14b) H
15a, 15b Unbelievably expensive and terrible management 268 (15a) 127 (15b) H
16a, 16b Worst service in a 5-star hotel... 250 (16a) 180 (16b) H
17a, 17b Nightmare being turned away with unaccommodating, Unfriendly staff 525 (17a) 154 (17b) H
18a, 18b Staff need more training 211 (18a) 175 (18b) H
19a, 19b Very Disappointed 228 (19a) 188 (19b) H
20a, 20b Stay from hell 127 (20a) 139 (20b) I
21a, 21b Do not try the Classic Room! 424 (21a) 253 (21b) I
22a, 22b 5-star Hotel Price with 1 -star Hotel Service 291 (22a) 155 (22b) J
23a, 23b Oh Dear! 283 (23a) 113 (23b) K
24a, 24b Told our room not available when we are on the way to check in 224 (24a) 66 (24b) K
25a, 25b Fall from grace! 541 (25a) 51 (25b) L
Volume 8 Issue 2, 2024, pp. 76-93
doi: 10.22363/2521-442X-2024-8-2-76-93
TEXT POST TOPIC GUEST COMPLAINTS (WORDS) MANAGEMENT RESPONSE (WORDS) HOTEL
26a, 26b Stayed 3 nights, changed room twice!!! 397 (26a) 84 (26b) L
27a, 27b Disappointed...maybe a bad day? 474 (27a) 99 (27b) M
28a, 28b Stay away if you're sensitive to fragrances 203 (28a) 101 (28b) N
29a, 29b Extremely Dishonest hotel 226 (29a) 97 (29b) N
30a, 30b Not 5 star 508 (30a) 136 (30b) N
31a, 31b I wanted to love this place 899 (31a) 129 (31b) N
32a, 32b Never sit at lobby, SO MANY MOSQUITOS 171 (32a) 175 (32b) O
33a, 33b Bad staff (ill mannered) and discriminative 497 (33a) 61 (33b) O
34a, 34b Awful... Do not be fooled by 5* rating 341 (34a) 132 (34b) O
35a, 35b Unpleasant experience 217 (35a) 130 (35b) O
36a, 36b First impressions count 233 (36a) 129 (36b) P
37a, 37b The Worst Hotel in HK? 221 (37a) 107 (37b) Q
38a, 38b Worst Hotel Ever 248 (38a) 193 (38b) R
TOTAL 11647 words (I) 5697 words (II) 18
17344 words (I) + (II)
The data that were analysed in the following stages are shown in Figure 1 below.
Figure 1. Stages of data analysis
After data collection, we removed all personal information, as this study focused solely on communication analysis and did not wish to reveal the identity of individual customers or hotel brands. Special codes such as Hotel A and Hotel B were used to replace the actual names, and spelling errors were corrected. The data was analysed using a systematic approach to identify external conjunctions and appraisal resources. Each text was carefully examined for the motivation and content. We then
used the list of external conjunctions (Table 2) to locate external conjunctions (e.g., and, besides, while) in the text. Later, the texts were manually reviewed again to find judgement resources for adjectives and related phrases such as rude, helpful, respectful. Finally, the raw frequency, relative frequency and log likelihood of these conjunction and judgement resources were calculated. After conducting these steps, the external conjunctions and appraisal resources, particularly judgement resources,
by Yau Ni Wan
found in the complaints and management responses were identified and classified to determine their attitudinal and evaluative meanings. These lexical items were counted manually and examined in detail for the raw frequency and relative frequency. The log likelihood test was used to determine the similarities and differences in the patterns of external conjunctions and appraisal resources between guests and hotel management. The log likelihood ratio was calculated using the online log likelihood wizard (LL Wizard, 2024) of Rayson and Garsides (2000). In the findings, items with a log likelihood greater than 3.84 are marked with an asterisk, which indicates significant use in the corpus at the p < 0.05 level (as per the significance level scale used on the above website). Only items with a significant difference greater than p < 0.05 were selected for further investigation in the data analysis. Two colleagues from the research team assisted in double checking the results of the log likelihood tests in the corpus.
External conjunction patterns in the texts were identified and analysed to gain a better understanding of the users' linguistic strategies for expressing and responding to their dissatisfaction with their hotel experiences.
Finally, the logical relationships between the conjunctions and the surrounding text were studied and systematised in terms of frequency, distribution, and contextual relevance using SFL's conjunction and the appraisal system.
4. RESULTS AND DISCUSSION
4.1. External conjunctions and continuatives in online hotel complaints and responses
This section examines the use of external conjunctions to indicate the different logical relations implied, focusing on those involving addition, comparison, time, and consequences. Table 4 shows the raw frequency, relative frequency, log likelihood (LL), and p-value of the external conjunctions and continuatives found in the guest and hotel management complaints and responses, respectively. Our preliminary findings suggest that there are significant differences in the frequency of use of conjunctions and continuatives between guests and hotel management. Dissatisfied guests tend to use external temporal conjunctions (e.g., then, before, after, until, finally, and when), temporal continuatives (e.g., still), consequential conjunctions (e.g., because, but, and without), and comparative continuatives (e.g., only and even) to frame negative eWOM. Luxury hotels, however, use addition conjunctions (e.g.,further), similar comparison conjunctions (e.g., like), expectant follow-on conjunctions (e.g., so that), and time continuatives (e.g., again) to acknowledge problems, assure service quality, and deliver positivity.
The results show that dissatisfied guests used comparative continuatives (e.g., just and even) when recounting events to reflect customer expectations, whereas these continuatives were not found in the management responses. Continuatives are mainly used within a clause rather than at the beginning (Martin & Rose, 2007). Even works to indicate that the customer's dissatisfaction is greater than expected. Even (LL=21.5; p <
0.0001) is used when complaining about irrational policies (1), unhygienic environments leading to discomfort (2), and unener-getic staff (3-4).
(1) She accused us of taking photos in the hotel room with the photography crew on my wedding day, while pointing out that even using an iPhone to take photos in the room is prohibited! (Text 22a).
(2) I had 12 mosquito bites and more each on my legs and arms and even under my clothes (Text 32a).
(3) People seemed to look tired and unenergetic, which made me feel even more depressed (Text 8a).
(4) The staff didn't even acknowledge the greeting with more than a dirty look that told me they weren't happy to see me back (Text 25a).
Just (LL=19.91; p < 0.0001) was another important comparative continuative that we found in the data, and it indicated that the service received was below customer expectations and unacceptable. Similar to the use of even, just was also used in the data to report poor staff performance (5-6). Based on the frequency and use of just and even, we found that hotel guests had high expectations of hotel staff performance in maintaining interpersonal relationships, as they were expected to be pleasant, happy, and friendly. This phenomenon is closely related to the concept of emotional labour in the service industry, which is further explored in RQ2.
(5) I just don't understand how she could look at me like that. I am really disappointed at how unfriendly and unaccommodating she was at reception (Text 17a)
(6) I was actually bumped into twice as staff walked past me and they didn't apologise or even acknowledge that they had just made physical contact with me. They just walked on (Text 25a).
Temporal conjunctions were frequently used to establish logical relations between events and activities in the complaints. The most significant examples were when (LL=19.41; p < 0.0001), then (LL=17.52; p < 0.0001), finally (LL=8.76, p < 0.01), before (LL=7.36, p < 0.01), until (LL=7.17; p < 0.01), still (LL=7.17; p < 0.01), and after (LL=5.25; p < 0.05).
When is a simultaneous temporal conjunction that denotes a hypotactic relationship, or it can also be understood as an unequal dependency relationship between a dependent clause and an independent (dominant) clause (Marin & Rose, 2007, pp. 120-121). The when--clause acts as the context in which the dependent clauses take place, for example, when the basics aren't done right in (7) and when you expected a 5-star hotel room from Hotel I in (8). The other clause is the independent clause: it is quite disappointing, and you got a 3-star room instead.
(7) It's quite disappointing with such a high reputation when the basics aren't done right (Text 3a)
(8) When you expected a 5-star hotel room from Hotel I, you got a 3-star room instead (Text 21a).
As per Table 4, the significant successive temporal conjunctions in the complaint texts were then, before, after, until, finally. Time conjunctions are used to represent temporal sequences or events. Narratives are consistently expected to follow a specific
Volume 8 Issue 2, 2024, pp. 76-93 doi: 10.22363/2521-442X-2024-8-2-76-93
Table 4
The raw frequency, relative frequency, loglikelihood (LL), and p-value of external conjunctions and continuatives
LOGICAL RELATION
EXAMPLE GUEST/ RAW FREQ. HOTEL
RELATIVE LOG
FREQ. (X LIKELIHOOD 1000 (LL)
P-VALUE
EXTERNAL CONJUNCTIONS
Addition
Additive Adding Further
Guest 0 0
Hotel 8 1.4
Guest 12 1
Hotel 16 2.8
Guest 22 1.9
Hotel 0 0
Guest 16 1.4
Hotel 1 0.2
Guest 31 2.7
Hotel 6 1.1
Guest 9 0.8
Hotel 0 0
Guest 11 0.9
Hotel 0 0
Guest 44 3.8
Hotel 3 0.5
Guest 0 0
Hotel 2 0.4
Guest 16 1.4
Hotel 1 0.2
Guest 68 5.8
Hotel 9 1.6
Guest 2 0.2
Hotel 5 0.9
Guest 5 0.4
Hotel 0 0
17.81**** p < 0.0001
Comparison Similar
Like
6.94*
p < 0.01
Time
Then
Successive Sometime
Before
After
Immediate Until
Terminating Finally
When
Simultaneous
Meanwhile
17.52**** p < 0.0001
7.36** p < 0.01
5.25*
8.76*
4.45*
p <0.05
7.17** p < 0.01
p < 0.01
19.41**** p < 0.0001
p <0.05
Expectant Because
Cause
Concessive But
Consequence
Expectant So that
Purpose
Concessive Without
7.36*
4.35*
3.98*
p < 0.01
18.65**** p < 0.0001
p <0.05
p <0.05
Note: *p < .05 p < .01 *** p < .001 **** p < .0001
by Yau Ni Wan
LOGICAL RELATION
EXAMPLE GUEST/ RAW FREQ. RELATIVE LOG P-VALUE
HOTEL FREQ. (X 1000 LIKELIHOOD
WORDS) (LL)
CONTINUATIVES
Comparison Less than Just Guest 25 2.1 19.91**** p < 0.0001
Hotel 8 1.4
More than Even Guest 27 2.3 21.5**** p < 0.0001
Hotel 0 0
Persistent Still Guest 9 0.8 7 17 ** p < 0.01
Time Hotel 0 0
Repetitive Again Guest 3 0.3 36 99**** p < 0.0001
Hotel 24 4.2
order of events (Marin & Rose, 2007). In their complaint texts, the dissatisfied guests used temporal conjunctions such as then to organise how the events occurred in sequence, i.e., from the beginning to the last step. Then supports the formation of successive logical relationships between clauses and sentences. In (9), the guest was given an unexpected HKD $200 bill, and the manager threatened to call the police if the guest refused to sign.
(9) We went down to the restaurant. But the food was not very attractive, we then went to check out and were given a bill of $200 for lunch. It was bad enough that we had to pay $60 for 2 cokes by the pool. I told him I would not sign the bill, but he said he had my credit card details. I was then told that if I did not sign immediately, then he would call the police (Text 15a).
To analyse this further, the first then is an expectant temporal conjunction that clearly links all the activities that have occurred in rapid succession. Then can also be used to implicitly indicate accumulated anger after detailing several events (Martin & Rose, 2007). In (10), the last two instances of then indicate that the situation has changed to a new negative and counter-expectant situation of calling the police. These conjunctions were used not only to organise activities but also to skilfully manage expectations within the context of events (Martin & Rose, 2007, p. 119). It is important to note that then is not typically counter-expectant but that it was in this example.
Successive temporal conjunctions such as finally, before, until, still, and after are also significant. Finally signals the end of the event reported, as in (10), as well as when receiving a lower upgrade price for the bill in response to the complaint. This included a sense of expectation of the outcome (Martin & Rose, 2007). In (11) and (12), the before and after conjunctions allowed the writer to move the sequence of events backwards and forwards. Until represents an immediate moment in the sequence; for example, in (13), the guest was ignored by the managers until she threatened to notify the media and book another
hotel. Tuzovic (2010) distinguished three types of frustration intensity in airline WOM: low venting, intense protest, and strong protest. Low venting leads to passive avoidance (e.g., I rate this airline 1 out of 5 in the online ranking system), whereas strong protest leads to active avoidance, e.g., not using the remaining credit or miles. Strong protest extends to influencing others to leave this airline. The present study found that booking another hotel was an escalation of previous WOM, as similarly found by Tuzovic (2010), with angry guests showing the highest level of frustration, such as threatening press conferences and lawsuits in (13). Lastly, still is a persistent continuat-ive that indicates that the activities will continue for a predetermined amount of time, as evidenced in (14), where the guest was still waiting for a response from the general manager.
(10) We threatened to move to a better hotel. The reservation stafffinally offered us a discounted upgrade price (Text 21a).
(11) The bed sheets were also covered in dirty make-up. I wonder if the room was properly cleaned before I arrived (Text 6a).
(12) After waiting for a few hours in the room they put me in, the reception did not contact me and went home (Text 17a).
(13) I could not get through to any manager at all when the incident happened until I threatened to call a press conference, book another hotel and sue the hotel (Text 24a).
(14) At the time of writing, I'm still waiting for the GM to contact me (Text 4a).
But and because are categorised as consequent conjunctions of cause (Martin & Rose, 2007). But is a concessive conjunction that contradicted the expectation in (16). It was used to control the reader's expectations (Martin & Rose, p. 57). However, this conjunction realised an unexpected consequence.
(16) The hotel had a better than good rating and we expected a lot, but it just did not deliver (Text 38a).
(17) The only reason it gets a 1-point rating on this review is because I can give no lower in the ranking system (Text 33a).
Volume 8 Issue 2, 2024, pp. 76-93 doi: 10.22363/2521-442X-2024-8-2-76-93
It must be noted that because is an expectant conjunction, meaning that one event obliges another to occur as a cause and effect (Martin & Rose, 2007, p. 128). For example, as the lowest score in the ranking system is 1, the expectant effect is that Hotel O will receive a score of 1 in the review due to the guest's dissatisfaction.
As shown in Table 4, luxury hotels tended to use addition conjunctions (further), similar comparison conjunctions (like), expectant consequence conjunctions (so that), time conjunctions (meanwhile), and time continuatives (again) to acknowledge problems, assure service quality, and deliver positivity. Addition conjunctions, such as further in (18), were used to add phrases one after the other to improve the quality of the service. The hotel management used comparative conjunctions to indicate similarity, such as like. This comparative conjunction emphasises the valuable review that was provided by the guest in (19). Lastly, conditional conjunctions like so that reveal the structure and organisation of the response, as well as causal relationships, thus allowing management to correct mistakes, restore guest trust, and increase the chances that the guest will return to the hotel, as shown in (20).
(18) Please be assured that your concerns have been brought to the attention of the relevant departmental managers, who will be looking for ways to further improve our guest experience (Text 30b).
(19) Please be assured that comments like yours are very important to us and will be reviewed with the relevant departments to ensure that similar situations don't happen again (Text 6b).
(20) Could you please let us know the period of your stay and registered name so that we can investigate the matter? (Text 27b).
Meanwhile is a time conjunction that indicates the simultaneous state of noticing feedback and apologising, as evidenced in (21). Again is a time continuative that repeats the meaning of the communication; this was evidenced in (22), where the hotel management apologised twice.
(21) We have noted your feedback regarding the location and design of the rooms and will take this into account when planning future improvements to the hotel. Meanwhile, we apologise for not providing you with the high level of service to which we aspire (Text 14b).
(22) Again, I'm sorry that we didn't find a way, either during or after your stay, to find out more about the deeper source of your unhappiness (Text 38b).
In the above, we show how guests and hotel management use different conjunctions to connect their messages in the discourse. Table 5 extends this analysis by illustrating the flow of conjunctions in the complaints and responses. Text samples were taken to examine the logical organisation of and to demonstrate the logical relations in the discourse. Texts 36a and b deal with credit card/debit card problems and unhelpful staff.
Table 5
The conjunctions and continuatives used between the discussed stages in the complaint and response texts
STAGES TYPES OF EXTRACT FROM THE SAMPLE TEXT
CONJUNCTIONS AND CONTINUATIVES
TEXT 36A
Orientation The receptionist, Karen, was so rude that I was left speechless.
Incident 1 We presented our local HK debit card as a hold card throughout our stay,
Comparison (contrast) but it was rejected as it didn't meet their criteria.
Incident 2 Time (successive) Then we gave them our UK debit cards, the same thing again.
Incident 3 Time (successive) Then we dug out my credit card (thank God!).
Time (successive) It finally worked.
Incident 4 Time (successive) After some initial muttering amongst the staff, presumably about us, we were able to get on with checking in!
Incident 5 Time (successive) When I finally managed to get my room key,
Continuative (comparison) Karen didn't even bother to tell me my room number,
Addition (alternative) show me how to use the lift or even ask if I needed any help with our luggage.
Interpretation Consequence (cause) As the title suggests, first impressions count
Addition and this was certainly the case during our stay at Hotel G.
Coda Old, tired and lacking in charm or consideration for its guests.
by Yau Ni Wan
STAGES TYPES OF CONJUNCTIONS AND EXTRACT FROM THE SAMPLE TEXT
TEXT 36B
Orientation I am concerned about your experience
Addition and would like to apologize most sincerely for not meeting your expectations.
Responses to Incidents 1 -3 We had experienced challenges with debit card issue,
Consequence (cause) as previous cases had shown that debit cards could be used for pre-payment,
Comparison (contrast) but unfortunately were not suitable as a holding card due to the lack of a credit function.
Consequence (means) Thus, to avoid the overcharge and refund scenario, we usually recommend that guests establish credit by credit card.
Follow-up action Please be assured that we will review the situation
Addition and discuss your honest feedback with our department managers to make improvements.
Coda Addition (alternative) If you would like to discuss your concerns further, please contact me at comments@hotelPcom.
A guest used the comparison (contrast) conjunction but to indicate that he had a problem with his credit card/debit card. Later, he used several temporal conjunctions, such as then, after, and when, to connect each phase with the previous events discussed. The guest ended his narrative with the conjunction finally, e.g., It finally worked, and I finally got my room key. Additionally, from the log likelihood frequency analysis, we found that the guest complaints often used a series of temporal conjunctions that also emphasised the recounting of past events and activities. We also noticed that many of the examples in the complaint texts were related to the concept of emotional labour, which was found to seriously shape the guests' expectations of the staff's performance or actions, e.g., the staff were so rude, did not tell me the room number, or did not ask if I needed help with my luggage. Text 36a was entitled 'First impressions count' in the forum, and we found that service quality issues relating to the helpfulness of reception staff were consistently the most common complaint. This is consistent with the previous findings obtained by Levy et al. (2013) and Danaher and Mattson (1994), who found that negative check-in experiences caused by unhelpful front desk staff highlighted the importance of first impressions in service evaluations.
Compared to the customer, the hotel manager rarely responded to complaints with a list of temporal conjunctions, relying instead on consequence and addition conjunctions to establish a common ground for repairing the conflict and thus potentially achieving a positive experience in future engagements. In Text 36b, the hotel manager addressed customer expectations and offered a solution to the debit card problem. Levy et al. (2013) also suggested that a more effective response strategy for the hotel could have been explaining the cause of the complaint and then apologising for not meeting the customer's expectations. We also found that many of the complaints were regarding the
perception of poor staff performance. This finding is consistent with that of a previous study by Levy et al. (2013), in which one of the most common complaints was about staff service errors, disrespectful behaviour, and a reluctance to help. Customers typically expect hotel staff to have a more positive attitude, which is closely related to the concept of emotional labour in the service industry; this topic is discussed further in RQ2.
4.2. Customer identity and emotional labour in eWOM on social media
As the reviews were pulled from a public online forum, personal information such as real names, identities, and specific details about the hotel stay were often not included in the complaints. However, we found that some customers did identify themselves. According to the data, certain customers identified themselves as loyal hotel fans or frequent travellers, or they mentioned their VIP membership. Some examples of frequent travellers and VIP members are shown in (23-25):
(23) I wondered why the hotel would try to give me a low-class room when they knew I had stayed there before, from the records they had in my membership with them? (Text 21 a)
(24) As a frequent visitor to Hong Kong, I thought I could find another hotel to stay at besides Hotel G and Hotel H, but today's experience has disabused me of that notion (Text 35a).
(25) He wanted to charge me for a snack, which is supposed to be free if you pay with a credit card. Then I insisted on not paying and informed him that I had Hotel O chain privileges! (Text 11a).
These guests may have chosen to reveal their identity in their complaints to demonstrate a high level of industry knowledge or expertise, thereby increasing the credibility of their negative reviews. VIP guests have been shown to have higher expectations of hotel service quality. For example, Mariani et al. (2019) found that business travellers generally rated hotels
Volume S Issue 2, 2024, pp. 76-93 doi: 10.22363/2521-442X-2024-S-2-76-93
lower than individuals travelling with family or friends. As a result, the dissatisfied guests expressed disappointment and the potential violation of their expectations by highlighting that the hotel did not provide the expected level of service. In addition,
frequent travellers and VIP members are existing customers, whereas the hotel fans in Text 12a appeared to be new customers; these guests were a couple staying at the hotel for their anniversary, as shown in Table 6.
Table 6
The conjunctions and continuatives used between the discussed stages in the complaints of hotel fans and responses of the hotel
TEXT 12A
STAGES TYPES OF CONJUNCTIONS EXTRACT FROM THE SAMPLE TEXT
AND CONTINUATIVES
Orientation Time (successive) After hearing so much about this beautiful hotel,
it has always been our dream to have a stay here for a special occasion.
I decided to save some money and celebrate our anniversary with my husband.
Incident 1 I mentioned this in my booking request,
Comparison (contrast) but they either didn't know or didn't care.
Incident 2 On arrival, no congratulations on our anniversary
Addition and we were treated like any other traveller in a lower class hotel.
Incident 3 No welcome drink, no card, not even a little treat for our anniversary.
Interpretation My husband still enjoyed the stay very much,
Comparison (contrast) but I felt differently.
I had high expectations for such an expensive place,
Comparison (contrast) but I was let down.
Coda That's not how you treat your fans.
TEXT 12B
Orientation I am very concerned to learn of your disappointment and unpleasant experience during your recent
stay at our hotel.
Responses to Incidents 1-3 Please accept our sincere apologies that your special celebration was not as memorable as you had expected.
I understand the frustration you feel,
Comparison (contrast) but please be assured that this is not an intentional act.
Follow-up action Consequence (cause) As the satisfaction of our guests is of paramount importance to us,
we would like to investigate this matter further.
I will send you a private message
Addition and would be grateful if you could give me more details.
We will review the situation to make further improvements.
Coda Thank you again for bringing this to my attention
Addition and I look forward to hearing from you.
by Yau Ni Wan
Upon further investigation of these data, we discovered that, when guests refer to themselves as fans, they may want to establish a personal and cognitive connection with the hotel. This linguistic representation in the text was included to establish a stronger relationship with the hotel, where the expectation was for more personalised and special treatment. As fans, they wish to be treated positively or desire for the quality of service to be improved to emphasise their importance. We found that the logical relations represented by the external conjunctions in both Texts 12a and 12b were highly consistent with the conjunction patterns established in RQ1. However, in terms of the explicit-ness or implicitness of the issue, this fan complaint differed from those of frequent traveller or VIP complainants, who experienced explicitly offensive events; these fan complaints were more associated with individual emotions, especially the negative implicit emotion of disappointment. In addition, the extract My husband still enjoyed his stay, but I felt differently shows that gender differences also seem to influence guest satisfaction. Sánchez-Franco & Alonso-Dos-Santos (2021) investigated the impact of the gender of guests on accommodation services, and they found that female reviewers prioritised a pleasant experience and atmosphere during their stay, whereas male guests emphasised hygiene and facilities. This may explain why couples react differently to the same event. Hotel managers were also found to use different linguistic strategies in their corresponding responses, focusing on showing extra empathy and care so as to repair the negative reception from fans, i.e., I am very concerned to learn of your disappointment, I understand the frustration and not an intentional act. According to Min et al. (2015), managers should respond with empathy by carefully reviewing the complaints and rephrasing them in their responses. Although the issue of fans who are new customers is rarely discussed in the literature (Table 1), it is also a critical factor that influences customer expectations, as expectation gaps can lead to significant guest dissatisfaction. Our findings regarding hotel fans brought new perspectives to the analysis of negative eWOM for hotel services. To summarise, guests mentioning their special customer identities can be understood as an attempt to receive personal attention and preferential treatment from hotel staff, where the aim is to increase the likelihood that their complaint will be addressed and resolved (Cetin & Walls, 2016). Conversely, because of previous positive experiences or their status in a hotel membership programme, they may expect the hotel to feel a greater sense of responsibility in addressing their concerns.
In addition to customer identities, the concept of emotional labour also attracted our attention, as it appeared frequently in the data and was found to influence customer expectations and satisfaction. Emotional labour is the effort required to manage and express specific emotions during work (Grandey, 2000). Hotel employees are expected to manage and control their emotions to maintain professional and friendly behaviour, resolve potential conflicts, and provide a satisfying customer experience (Kim, 2008; Kim et al., 2012; Shani et al., 2014; Wang, 2020). This concept is particularly important in the service industry,
where expressions of a pleasant attitude, empathy, and attentive service can reinforce a customer's positive impression. RQ2 was mainly evidenced via a log likelihood analysis of the appraisal items, which were used to examine how different parties used judgement resources to reflect their behaviour. Of the various subcategories in the judgement system, the tokens related to normality, capacity, and propriety were tested. Table 7 shows that it was only the use of judgement resources (capacity) that differed significantly between hotel management and guests.
Cannot, nice, and helpful were the terms more commonly used by guests, which can be seen in (26-28), whereas best and able to were more frequently used by hotel management.
(26) The hotel manager said he cannot ask his bellboys to take so many bags up to our rooms (Text 33a).
(27) Even though I arrive late at night, I expect some kind of nice welcome (Text 8a).
(28) Banquet staff were helpful when asked but stood around and did little most of the time (Text 25a).
Example (27) was regarding check-in issues, and we believe that check-in issues are critical. This is consistent with the previous findings obtained by Lee et al. (2023) and Zhang (2019), who advised that hosts or service providers should provide a smooth check-in process for guests, even if some arrive late, to develop positive interpersonal interactions. Welcoming guests by a host is an example of important professional behaviour that can be associated with customer loyalty (Lee et al., 2023). In addition, our previous findings from RQ1 also show that the studied hotels used a variety of emotional labour techniques to address negative eWOM, such as empathy, apologies, and reassurance. In (29) and (30), positive judgement (capacity) items such as best and able to were used in the responses that were provided to achieve a favourable customer relationship result.
(29) At Hotel N Hong Kong, we strive for excellent service delivery and are always looking for ways to ensure that every guest experiences the best we have to offer (Text 31b).
(30) We will be able to handle your complaints and room assignments in a most satisfactory manner (Text 1b).
The data also showed that many hotel responses included follow-up actions such as emails and phone calls, but it was also found that concise monetary or non-monetary compensation was rarely discussed. As Levy et al. (2013) point out, these compensations are rarely mentioned online due to privacy concerns. Responding to negative eWOM is seen as a challenge for hotel managers. They often have to strike a delicate balance between managing customer expectations of 'appropriate behaviour' from hotel staff, investigating the real problem, and maintaining positive interpersonal relationships while managing their own emotions and reactions both online and offline.
Based on our findings about external conjunctions, we show luxury hotel managers how to linguistically deconstruct and interpret guest complaints. When responding to these guests, they should aim to establish a clear logical connection in their responses to negative eWOM. For example, in management response texts, they need to acknowledge and paraphrase the
Volume S Issue 2, 2024, pp. 76-93
doi: 10.22363/2521-442X-2024-S-2-76-93
Table 7
Appraisal analysis of judgement resources (capacity)
APPRAISAL SYSTEM GUEST/ HOTEL RAW FREQ. RELATIVE FREQ. (X1000 WORDS) LOG LIKELIHOOD (LL) P-VALUE
Best Guest 4 0.3 22.4'*" p < 0.0001
Hotel 18 3.2
Able to Guest 3 0.3 7.31" p <0.001
Hotel 8 1.4
Judgement (capacity) Cannot Guest 7 0.6 5.57" p < 0.05
Hotel 0 0
Nice Guest 7 0.6 5.57" p < 0.05
Hotel 0 0
Helpful Guest 6 0.5 4.78" p < 0.05
Hotel 0 0
Note: "p < .05; "p < .01; *** p < .001; and "" p < .0001
problem, provide a clear reason or explanation for failing to meet the customer's expectations, offer a sincere apology, show understanding, and demonstrate commitment and eagerness for future improvement that can help repair customer relationships and mitigate the impact of negative reviews to ensure service quality. In addition, our findings on emotional labour and customer identity suggest that luxury hotel managers should pay attention to the importance of the check-in process, which leads to many complaints, remind reception staff to give the guest a warm welcome and a helping hand, even at midnight check-in, and all staff should have a positive attitude with pleasant facial expressions to build respect and create good impressions. Finally, hotel managers can be more sensitive to the complaints of VIP members or loyal fans, who may have a better knowledge of the hotel industry or a stronger emotional connection, and therefore a much higher expectation of service quality than other customers. More personalised, empathetic and sincere language strategies are needed to deal with these complaints.
5. CONCLUSION
The present study was conducted to respond to Tuzovic's (2010) and Schuckert et al.'s (2015) call to investigate online community consumer review sites, as more hotels are incorporating social media into their marketing research efforts. It is also in line with service marketing research, which has highlighted the importance of language in service interactions (Holmqvist et al., 2017; Mariani et al., 2019). By examining the linguistic framing of negative eWOM, as well as how luxury hotel management manages customer relationships on social media platforms,
this study extends previous research (Fastrich, 2024; Mariani et al., 2019; Sánchez-Franco & Alonso-Dos-Santos, 2021) on the role of lexical-grammatical features in online reviews. This study uses the SFL and discourse semantic approach, with a focus on logical relations and emotional labour, to examine the external conjunctions and continuatives used in online hotel complaints. RQ1 was assessed using log likelihood and qualitative analyses to identify patterns in external conjunctions and to examine the logical relationships expressed by these conjunctions within the discourse structure. The results show that there are signiflcant differences in the strategies used by guests and hotel management in complaints and responses. Guests tend to use additional temporal conjunctions to list the multiple events that they experienced, while management responses often contain logical relations such as comparison (similarity) and consequence (purpose) to indicate the follow-up actions that are being taken to address customer concerns. This analysis provides insights into the dynamics of complaints and responses, as well as suggestions on how hotel management can reduce negative reviews, that is, by acknowledging and rephrasing a customer's complaint, apologising, offering solutions, and reassuring the customer. RQ2 was explored by investigating how a customer's speciflc identity, for example, as a frequent traveller, VIP member, or fan, may affect their satisfaction. The results of this analysis can help hotel management to improve customer experiences and to attract loyal customers. In summary, this study contributes to our understanding of effective business communication in the service industry, which can be used to improve customer relationships and maintain digital reputations in social media. The flndings of
by Yau Ni Wan
this study have certain practical implications for luxury hotels and their management. Firstly, this study highlights the importance of acknowledging and addressing customer complaints on social media, as this can help to repair customer relationships and prevent negative electronic word of mouth. Secondly, this study supports the notion that hotel management should be mindful of the language that they use in their responses, as this can influence customer satisfaction. For example, to reduce negative reviews, luxury hotel managers should carefully analyse and interpret guest complaints. When responding to these complaints on social media, they should acknowledge the problem, provide an explanation, and show understanding and commitment. The hotel can focus on the check-in process, greeting visitors warmly and maintaining a cheerful attitude. They can pay more attention to the dissatisfaction of loyal fans or VIP members, who may have higher standards and require more individualised and empathetic language techniques.
References
Appel, G., Grewal, L., Hadi, R., & Stephen, A. T. (2020). The future of social media in marketing. Journal of the Academy of Marketing science, 48(1), 79-95. https://dx.doi.org/10.1007/s11747-019-00695-1
Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labor in service roles: The influence of identity. Academy of Management Review, 18(1), 88-115. https://dx.doi.org/10.5465/amr.1993. 3997508
Baker, M. J., & Hashimoto, B. (2024). Expression of customer (dis)satis-faction in online restaurant reviews: The relationship between adversative connective constructions and star ratings. International Journal of Business Communication, 61 (1), 148-180. Bloor, T., & Bloor, M. (2013). The functional analysis of English: A Halli-
dayan approach. Routledge. Cetin, G., & Walls, A. (2016). Understanding the customer experiences from the perspective of guests and hotel managers: Empirical findings from luxury hotels in Istanbul, Turkey. Journal of Hospitality Marketing & Management, 25(4), 395-424. https:// doi.org/10.1080/19368623.2015.1034395 Colmekcioglu, N., Marvi, R., Foroudi, P., & Okumus, F. (2022). Generation, susceptibility, and response regarding negativity: An in-depth analysis on negative online reviews. Journal of Business Research, 153, 235-250. https://dx.doi.org/10.1016/j.jbusres.20 22.08.033
Danaher, P J., & Mattsson, J. (1994). Customer satisfaction during the service delivery process. European Journal of Marketing, 28(5), 5-16. https://doi.org/10.1108/03090569410062005 Dávid, L. D., El Archi, Y., Zhu, K., Varga, I., Benbba, B., & Csizmadia, N. (2024). Inbound and outbound tourism in China: A perspective article. International Journal of Engineering and Management Sciences, 9(1), 91-107. https://dx.doi.org/10.21791/IJEM S.2024.006
Deng, X., Qiu, W., Zhang, L., & He, J. (2023). Within-group multimarket contact, service diversity and consumer rating: Evidence from the hotel industry. Journal of Hospitality Marketing & Management, 32(1), 122-149. https://dx.doi.org/10.1080/19368623. 2023.2134248
The present study has some limitations, such as a relatively small sample size, which means that the results of this study cannot be used to draw broad generalizations to the hotel industry. As such, future expansions of this topic of research should involve a larger sample size.
Furthermore, as we only looked at the lowest rating scores, we left out some of the representative complaints and responses. Therefore, future research could explore the effectiveness of language strategies across different rating scores, contexts, cultures, and regions, thereby providing valuable data and observations for hoteliers in terms of managing their global online reputation.
This study also indicates that it is worth investigating the dissatisfied reactions of guests to manager responses for the purpose of linguistic effectiveness.
Lastly, additional research could be conducted on investigating social media use in other service industries.
Dwivedi, Y. K., Ismagilova, E., Hughes, D. L., Carlson, J., Filieri, R., Jacob-son, J., ... & Wang, Y. (2021). Setting the future of digital and social media marketing research: Perspectives and research propositions. International Journal of Information Management, 59, 102168. https://doi.org/10.1016/j.ijinfomgt.2020.102168 Fastrich, B. (2024). (Im)personalization in German and English negative online reviews: Contrasts, comparisons, and cognitive implications. International Journal of Business Communication, 61(1), 39-69. https://doi.org/10.1177/23294884231200249 Grandey, A. A. (2000). Emotional regulation in the workplace: A new way to conceptualize emotional labor. Journal of Occupational Health Psychology, 5(1), 95-110. https://psycnet.apa.org/doi/ 10.1037/1076-8998.5.1.95 Gong, T., Park, J., & Hyun, H. (2020). Customer response toward employees' emotional labor in service industry settings. Journal of Retailing and Consumer Services, 52, 101899. https://dx.doi.org/ 10.1016/j.jretconser.2019.101899 Grishechko, E. G. (2021). Communicative strategy of politeness in modern English: Multi-level means of implementation. RUDN University Press. EDN: EHERPQ Gupta, V (2019). The influencing role of social media in the consumer's hotel decision-making process. Worldwide Hospitality and Tourism Themes, 11(4), 378-391. https://dx.doi.org/10.1108/ WHATT-04-2019-0019 Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. Longman. Halliday, M. A. K., & Matthiessen, C. M. (2013). Halliday's introduction to functional grammar. Routledge. https://dx.doi.org/10.4324/9 780203431269
Heskett, J. L., Jones, T. O., Loveman, G. W., Sasser, W. E., & Schlesinger, L. A. (1994). Putting the service-profit chain to work. Harvard Business Review, 72(2), 164-174. Hochschild, A. R. (1979). Emotion work, feeling rules, and social structure. American Journal of Sociology, 85(3), 551-575. Holmqvist, J., Van Vaerenbergh, Y., & Grônroos, C. (2017). Language use in services: Recent advances and directions for future research. Journal of Business Research, 72, 114-118. https:// doi.org/10.1016/j.jbusres.2016.10.005
Training, Language and Culture Volume 8 Issue 2, 2024, pp. 76-93
Kietzmann, J., & Canhoto, A. (2013). Bittersweet! Understanding and managing electronic word of mouth. Journal of Public Affairs, 13(2), 146-159. https://doi.org/10.1002/pa.1470 Kim, H. J. (2008). Hotel service providers' emotional labor: The antecedents and effects on burnout. International Journal of Hospitality Management, 27(2), 151-161. https://dx.doi.org/10.1016/ j.ijhm.2007.07.019 Kim, T., Jung-Eun Yoo, J., Lee, G., & Kim, J. (2012). Emotional intelligence and emotional labor acting strategies among frontline hotel employees. International Journal of Contemporary Hospitality Management, 24(7), 1029-1046. https://dx.doi.org/10.110 8/09596111211258900 Kucukusta, D., Pang, L., & Chui, S. (2013). Inbound travelers' selection criteria for hotel spas in Hong Kong. Journal of Travel & Tourism Marketing, 30(6), 557-576. https://dx.doi.org/10.1080/105 48408.2013.810995 Kulakhmetova, G., Aktymbayeva, A., Assipova, Z., Baoleer, B., & Koshkimbayeva, U. (2022). Current problems in the tourism and hotel industry taking the world's tourist cities as an example. GeoJournal of Tourism and Geosites, 43(3), 841-849. https://doi.org/10.30892/gtg.43301 -895 Lai, X., Wang, F., & Wang, X. (2021). Asymmetric relationship between customer sentiment and online hotel ratings: The moderating effects of review characteristics. International Journal of Contemporary Hospitality Management, 33(6), 2137-2156. https:// doi.org/10.1108/IJCHM-07-2020-0708 Law, R., Leung, R., Lo, A., Leung, D., & Fong, L. H. N. (2015). Distribution channel in hospitality and tourism: Revisiting disinterme-diation from the perspectives of hotels and travel agencies. International Journal of Contemporary Hospitality Management, 27(3), 431-452. https://dx.doi.org/10.1108/IJCHM-11-2 013-0498
Lee, C. K. H., Tse, Y. K., Zhang, M., & Wang, Y. (2023). What have hosts overlooked for improving stay experience in accommodation-sharing? Empirical evidence from Airbnb customer reviews. International Journal of Contemporary Hospitality Management, 35(2), 765-784. https://doi.org/10.1108/IJCHM-12-2021 -1544 Li, Z. C., Tao, W., & Wu, L. (2023). The price of good friendships: Examining the roles of relationship norms and perceived controllability in service failure encounters. International Journal of Business Communication, 60(1), 105-129. https://dx.doi.org/ 10.1177/2329488420907119 Levy, S. E., Duan, W., & Boo, S. (2013). An analysis of one-star online reviews and responses in the Washington, DC lodging market. Cornell Hospitality Quarterly, 54(1), 49-63. https://dx.doi.org/ 10.1177/1938965512464513 LL Wizard. (2024). Log-likelihood and effect size calculator. Lancaster University UCREL Nanural Language Pprocessing Group. https://ucrel.lancs.ac.uk/llwizard.html Martin, J. R., & Rose, D. (2007). Working with discourse: Meaning beyond the clause. Bloomsbury Academic Publishing / Continuum Publishing.
Martin, J. R., & White, P. R. (2005). The language of evaluation: Appraisal in English. Palgrave Macmillan. https://dx.doi.org/10.1057/97 80230511910
Mariani, M. M., Borghi, M., & Kazakov, S. (2019). The role of language in the online evaluation of hospitality service encounters: An empirical study. International Journal of Hospitality Management, 78, 50-58. https://doi.org/10.1016/j.ijhm.2018.11.012
rudn.tlcjournal.org doi: 10.22363/2521-442X-2024-8-2-76-93
Min, H., Lim, Y., & Magnini, V P (2015). Factors affecting customer satisfaction in responses to negative online hotel reviews: The impact of empathy, paraphrasing, and speed. Cornell Hospitality Quarterly, 56(2), 223-231. https://dx.doi.org/10.1177/193896 5514560014
Olson, E. D., & Ro, H. (2020). Company response to negative online reviews: The effects of procedural justice, interactional justice, and social presence. Cornell Hospitality Quarterly, 61(3), 312-331. https://doi.org/10.1177/1938965519892902 Perez-Aranda, J., Guerreiro, M. M., & da Costa Mendes, J. (2018). Predictors of positive reviews on hotels: Hoteliers' perception. Online Information Review, 42 (2), 146-160. https://dx.doi.org/10.1 108/OIR-08-2016-0234 Rayson, P., & Garside, R. (2000, October 7). Comparing corpora using frequency profiling. In Proceedings of the Workshop on Comparing Corpora (pp. 1-6). Association for Computing Machinery. https://doi.org/10.3115/1117729.1117730 Sánchez-Franco, M. J., & Alonso-Dos-Santos, M. (2021). Exploring gender-based influences on key features of Airbnb accommodations. Economic research: Ekonomska istrazivanja, 34(1), 2484-2505. https://doi.org/10.1080A 331677X.2020.1831943 Schleppegrell, M. J., & Oteíza, T. (2023). Systemic functional linguistics: Exploring meaning in language. In M. Handford & J. P. Gee (Eds.), The Routledge handbook of discourse analysis (pp. 156-169). Routledge. Schuckert, M., Liu, X., & Law, R. (2015). A segmentation of online reviews by language groups: How English and non-English speakers rate hotels differently. International Journal of Hospitality Management, 48, 143-149. https://dx.doi.org/10.1016/JJJ HM.2014.12.007
Shani, A., Uriely, N., Reichel, A., & Ginsburg, L. (2014). Emotional labor in the hospitality industry: The influence of contextual factors. International Journal of Hospitality Management, 37, 150-158. https://doi.org/ 10.1016/j.ijhm.2013.11.009 Tuzovic, S. (2010). Frequent (flier) frustration and the dark side of word-of-web: Exploring online dysfunctional behavior in online feedback forums. Journal of Services Marketing, 24(6), 446-457. https://doi.org/10.1108/08876041011072564 Van Herck, R., Decock, S., De Clerck, B., & Hudders, L. (2023). The impact of employee empathy on brand trust in organizational complaint response emails: A closer look at linguistic realization. International Journal of Business Communication, 60(4), 1220-1266. https://dx.doi.org/10.1177/232948842110 32316
Varkaris, E., & Neuhofer, B. (2017). The influence of social media on the consumers' hotel decision journey. Journal of Hospitality and Tourism Technology, 8(1), 101-118. https://dx.doi.org/10.1 108/JHTT-09-2016-0058 Verma, S., & Yadav, N. (2021). Past, present, and future of electronic word of mouth (EWOM). Journal of Interactive Marketing, 53, 111-128. https://doi.org/10.1016/j.intmar.2020.07.001 Wan, Y. N., & Forey, G. (2024). Hospitality discourse on social media: Evaluating online complaints and service recovery for luxury hotels. Journal of Language Teaching and Research, 15(2), 364-373. https://doi.org/10.17507/jltr.1502.05 Wang, C. J. (2020). Managing emotional labor for service quality: A cross-level analysis among hotel employees. International Journal of Hospitality Management, 88, 102396. https://doi.org/ 10.1016/j.ijhm.2019.102396
by Yau Ni Wan
Wong, K. K., & Kwan, C. (2001). An analysis of the competitive strategies of hotels and travel agents in Hong Kong and Singapore. International Journal of Contemporary Hospitality Management, 13(6), 293-303. https://dx.doi.org/10.1108/0959611 0110400490
Xu, J., & Wu, Y. (2020). Countering reactance in crisis communication: Incorporating positive emotions via social media. International Journal of Business Communication, 57(3), 352-369. https:// doi.org/10.1177/2329488417702475
Yuan, Y. H., Tsao, S. H., Chyou, J. T., & Tsai, S. B. (2020). An empirical study on effects of electronic word-of-mouth and Internet risk avoidance on purchase intention: From the perspective of big data. Soft Computing, 24, 5713-5728. https://doi.org/10.1007/ s00500-019-04300-z Zhang, J. (2019). What's yours is mine: exploring customer voice on Airbnb using text-mining approaches. Journal of Consumer Marketing, 36(5), 655-665. https://dx.doi.org/10.1108/JCM-0 2-2018-2581
Yau Ni Wan
ORCID 0000-0002-7585-1745 H ynwan@hksyu.edu Hong Kong Shue Yan University, Hong Kong
