Examining key hotel attributes for guest sleep and overall satisfaction

Abstract

Introduction:

Despite the fact that hotels provide a venue for sleeping, there is surprisingly little research that has explored sleep among hotel guests. The aim of this study is to identify the relationship between hotel attributes (e.g., light in the guestroom, bed linens), guest sleep, and overall guest satisfaction.

Methods:

Cross-sectional survey data were obtained from frequent business and leisure travelers (N = 609). Guest sleep satisfaction and overall guest satisfaction were measured on 5-point Likert scales. Participants were asked to report the extent to which hotel attributes (e.g., “Room too light or too dark”) related to their sleep on a scale from 1 (not at all) to 5 (very much). We used ordinal logistic regression to predict guest sleep and hotel satisfaction using hotel attributes as predictors while controlling for age, sex, and relationship status.

Results:

Hotel guest sleep did not differ between business and leisure travelers. Hotel guest sleep was inversely associated with “uncomfortable bed linen,” “uncomfortable pillows,” and “sound from the air conditioning unit or heater.” Regression revealed that sleep satisfaction was a strong predictor of overall hotel satisfaction.

Conclusions:

Our study suggests that guest sleep is be a critical component of the guest satisfaction equation. Importantly, our study illuminated the hotel attributes that offer the biggest contribution to hotel guest sleep and the specific steps for improving guest sleep.

Introduction

It is recommended that adults sleep at least 7 hours per night (Hirshkowitz et al., 2015; Watson et al., 2015), or almost a third of the day. However, poor sleep, when experienced by a guest during a hotel stay, has the potential to cloud an otherwise positive experience, for insufficient or poor quality sleep is associated with reduced problem solving ability (Deak and Stickgold, 2010), impaired mood (Calkins et al., 2013), reduced alertness (Thomas et al., 2000), and lower general well-being (Colten and Altevogt, 2006). Realizing this, some hotels have made guest sleep a top priority. For example, the Premier Travel Inn’s “Good Night Guarantee” offers a full refund to guest who are unsatisfied with their sleep during their stay (Olsen and Zhao, 2008). Also, Westin Hotels invested $30 million in their “Heavenly Bed” campaign to upgrade mattresses and bed linens in their guestrooms (Beldona et al., 2018). Despite the importance of sleep for guests, the time spent sleeping during a typical hotel stay compared to other activities (e.g., dining or spa) and the fact that several hotels have enacted guest sleep-related programs and policies, the academic literature on hotel sleep, as well as the techniques to improve sleep among guests, is relatively scant.

Literature review

Guest satisfaction may be understood through the customer-centric service quality gap model, which articulates that guest quality is a function of the difference between two factors: guest expectations (i.e., beliefs about future service) and guest perceptions (e.g., subjective assessments of services experienced) (Parasuraman et al., 1985). According to this model, customers actively compare their perceptions to their experience, thus a smaller gap between perception and experience will result in higher service quality and additional favorable service outcomes, including repeat visits, positive word of mouth, and customer referrals (Berezan et al., 2013; Cadotte and Turgeon, 1988; Cooil et al., 2007; O’Neill and Mattila, 2010; Zeithaml et al., 1996). According to Zeithaml and Bitner (2003), guest service perceptions are focused on three key elements: physical environment, interactions and outcomes. Accordingly, guests will evaluate hotel services based on their perception of the guest bedroom, interactions with staff relating to sleep, and the quality of their sleep while on property.

Sleep might be viewed as basic function of a hotel experience, but perhaps not one where additional value may be derived: as long as the basic needs are met, additional investment may be perceived as achieving diminishing returns (Roberts and Shea, 2017). Indeed, it has been argued that creating a clean and safe sleeping environment is the most fundamental goal of a hotel (Roberts and Shea, 2017). However, we propose that hotels have an opportunity to design programs and make small changes to their guestrooms in order to clearly communicate to guests that their sleep is a top priority. In doing so, and in line with the above findings (Berezina et al., 2016; Xiang et al., 2015), guestrooms that feature thoughtful changes for improving the sleep experience or programs specifically designed for better sleep (e.g., pillow menus) may constitute a unique selling point that will aid in guest satisfaction.

Several studies have measured sleep among hotel guests (Choi and Chu, 2001; Gunderson et al., 1996). For instance, in an examination of hotel reviews on a popular travel website (TripAdvisor), Hargreaves (2015) found that most guests (70%) rated their sleep in the hotel as “very good” or “excellent”, suggesting that most hotels are delivering upon a good night’s rest for their guest. However, Chen and colleagues found that business travelers reported overall worse sleep at hotels as compared to their home environment (Chen et al., 2018).

Research has also examined elements of the hotel that relate to guest sleep. For instance, in their study with business and leisure travelers, Hon and Fung (2019) found a strong relationship between comfortable bed amenities (e.g. quality mattress and pillow) and both guest satisfaction and intention to return to the hotel. Pallesen et al. (2016) developed a questionnaire asking participants to report the extent to which, on average, 15 sleep-damaging hotel attributes (e.g., high temperature, uncomfortable bed) affected their sleep during a typical hotel stay. In this survey-based study, Pallesen et al. (2016) reported that unsupportive pillows and high temperature in the guestroom were determinants of poor sleep among business travelers. Another study analyzed hotel reviews posted by travelers to TripAdvisor, finding that comments about sleep addressed either the hotel’s geographic location, sleep-related facilities, or guestroom components such as pillow and mattress (Mao et al., 2018).

Our study addresses two important gaps in the literature. First, there is a paucity of literature on hotel guest sleep satisfaction and overall guest satisfaction. While research has measured guest ratings of the sleep environment and guest satisfaction, the sleeping environment is subsumed in these analyses under perceptions of housekeeping services (Hargreaves, 2015). Second, while Pallesen et al. (2016) examined guest perceptions of 15 hotel attributes and their relationship with guest sleep quality, we investigate guest perceptions in reaction to a more comprehensive list of attributes (26 in total) and examined the relationship between guest sleep and overall gust satisfaction. Building on the service gap model, we examine the relationship between guest perceptions of sleep-related attributes and overall guest satisfaction (Zeithaml and Bitner, 2003). Our study also draws upon work demonstrating the importance of guest satisfaction for favorable outcomes such as repeat visits and positive word of mouth (Berezan et al., 2013; Cooil et al., 2007; Zeithaml et al., 1996). Using data from business and leisure travelers in the U.S., this study explored the following research questions:

Research Question 1: What is the relationship between specific hotel attributes and hotel guest sleep?

Research Question 2: Which hotel attributes predict unique aspects of hotel guest sleep satisfaction, separate from all of the others?

Research Question 3: What is the relationship between guest sleep satisfaction and overall guest satisfaction?

Research Question 4: What is the relationship between the sleep-related hotel attributes and guest satisfaction?

Research Question 5: Could a composite score reflecting sleep-related hotel attributes reliably predict hotel guest satisfaction?

Methods

Participants

To explore the association between guest perceptions, hotel guest sleep, and guest satisfaction, we collected cross-sectional surveys using geographically representative sampling. Surveys were administered by Qualtrics, LLC (Provo, Utah), a survey management company (Miller et al., 2020). Qualtrics maintains an online respondent pool of participants. Participants for this panel were recruited using a combination of social media, television and radio advertisements as well as offline mailers. Qualtrics achieved geographic representation using quota sampling. Quotas were set based on the United States Census data for gender, race, household income, and geographic region. All participants were above age 18. Eligible individuals reported traveling and staying in a hotel in the past 30 days. During enrollment for the present study, quota sampling was also based upon participant-reported primary purpose of travel as either business or leisure, so that a balanced sample comprised of equal parts business and leisure guests was recruited. The recruitment procedure was successful, so that the sample was indeed comprised of approximately one-half business travelers and one-half leisure travelers.

Measures

Hotel guest sleep satisfaction

Participants were asked to provide a rating of their satisfaction with their sleep during their most recent hotel stay. Specifically, participants were asked “Overall, how satisfied were you with the quality of sleep?” on a scale from 1 (not at all satisfied) to 5 (extremely satisfied).

Hotel attributes.

The 15-item sleep damaging hotel attribute scale developed by Pallesen et al. (2016) was employed in the current study as a starting point. Like in the work by Pallesen et al. (2016), our team was comprised of hotel researchers and sleep experts but added 11 additional hotel attributes to the scale developed by Pallesen et al. for a total of 26 hotel attributes. These additional attributes were identified through team brainstorming in several stages. First, we developed a long list of potential sleep issues guests may have. We then cross-referenced this list with Pallesen, and made final adjustments to avoid overlap. We then administered the 26-item scale to assess sleep damaging hotel attributes in this study. The full list of hotel attributes included the following factors:

• Disruptive noise (from the air-conditioning unit or heater, the hallway, another room, or outside the hotel).

• Too much or too little light.

• Lack of blackout curtains.

• Aspects of the hotel room itself (small size, poor location, poor view).

• Room odor.

• Aspects of the sleeping surface (uncomfortable bed linens, pillows, or mattress, not enough pillows).

• Aspects of cleanliness (bedspread, room in general).

• Inability to control temperature, poor quality of room (amenities, towels, furnishings, electronics, WIFI).

• No soaking tub.

• No secure locks or alarm.

• No fresh air.

• No electric outlets/phone charging station close to bed.

Participants were asked, “Please indicate the extent the following hotel attributes influenced your quality of sleep” on a scale from 1 (“did not bother me at all”), 2 (“slightly bothersome”), 3 (“somewhat bothersome”), 4 (“moderately bothersome”), and 5 (“extremely bothersome”). We re-coded the responses to dichotomous variables to distinguish between an attribute that was not at all bothersome (a response of 1) and an attribute that was slightly, somewhat, moderately, or extremely bothersome (a response of 2, 3, 4, or 5).

Overall guest satisfaction.

Hotel satisfaction was examined by asking participants three questions pertaining to their stay (Back, 2005). Specifically, participants were asked on a scale from 1 (strongly disagree) to 5 (strongly agree) with the statements “I am happy about my decision to stay at this hotel,” “I believe I did the right thing when I stayed at this hotel,” and “overall, I am satisfied with the decision to stay at this hotel.” The three questions had strong internal consistency (Cronbach’s alpha = 0.78) and therefore averaged to create a single scale assessing guest satisfaction.

Demographic covariates.

Potential confounders were controlled for in the analyses, which included demographics characteristics (sex, age, and marital status).

Analysis

All significance tests were evaluated at the .05 level. As an initial approach, univariate statistical tests (i.e., analysis of variance for continuous variables and chi-square for categorical variables) compared responses by sleep satisfaction as reported by guests. As a manipulation check, sleep satisfaction was compared by leisure and business travelers, but did not vary significantly, in relation to these groups (p>.05) and therefore these two groups were combined in our analyses. Due to the fact that the hotel attributes were written as negative aspects (e.g., “room too light or too dark”), for ease of interpretation, sleep satisfaction and hotel satisfaction items were reverse coded, such that higher numbers indicated greater degree of complaint. Ordinal logistic regression analyses were performed to examine the relationship between each hotel attribute as predictor and poor sleep satisfaction as an ordinal outcome (Research Question 1). This analysis was then repeated using a step-wise approach whereby attributes were entered into the ordinal logistic regression model one at a time, which disallows multicollinear variables from entering into the model selection procedure, and therefore illuminating whether any attributes contributed unique predictive value in terms of sleep satisfaction (Research Question 2). The step-wise approach was conducted so as to reduce multicollinearity.

Then, ordinal logistic regression was used to examine both the relationship between overall guest satisfaction as an outcome and guest sleep satisfaction as a predictor (Research Question 3) and the relationship between each hotel attribute identified to be important for sleep satisfaction and overall satisfaction (Research Question 4). Finally, all hotel attributes that individually predicted hotel guest sleep were combined to compute a composite score, which was then used to evaluate whether the composite score predicted overall guest sleep satisfaction and guest satisfaction (Research Question 5). All models were controlled for covariates, which included age, sex, and relationship status. We report the McFadden Pseudo R² for all combined ordinal regression models. All analyses were conducted in Stata, Version 16 (College Station, TX, USA).

Results

Table 1 displays summary statistics describing demographic characteristics and hotel attributes. Within our sample, 38% were between the ages of 25 and 35 years. Our sample was comprised of more males (56.7%) than females (43.3%). The majority of the sample was married (53.9%). Hotel guest sleep satisfaction was reported to be “very good” or “excellent” by approximately 50% of the sample, and “neither good nor bad” by approximately 32% of respondents.

Table 1.

Descriptive statistics of the sample of travelers (n=609).

Variable	Category/units	Percent
Age	Under 25 years of age	6.4%
	25 to 30 years of age	19.4%
	31 to 35 years of age	18.7%
	36 to 40 years of age	14.1%
	41 to 45 years of age	7.2%
	46 to 50 years of age	6.7%
	51 to 55 years of age	6.6%
	56 to 60 years of age	7.1%
	61 to 65 years of age	4.8%
	over 65 years of age	9.0%
Gender	Male	56.7%
	Female	43.4%
Relationship Status	Unmarried	24.8%
	Married	71.1%
	Divorced/Widowed/Separated	4.1%
Hotel Attributes	Sound of the air-conditioning unit or heater	54.2%
	Hotel staff or other noise in the hallway	54.8%
	Noise from another room	54.8%
	Room too light or too dark	46.6%
	Room too small	39.1%
	Smell of the room (e.g., musty)	45.3%
	Bed linen uncomfortable	46.3%
	Pillows are uncomfortable	58.8%
	Poor location of the room within the hotel	44.8%
	Mattress too hard or soft	53.4%
	Noise from outside the hotel	51.6%
	Not enough pillows	46.1%
	Bedspread old or dirty	41.7%
	Room not maintained/clean	41.5%
	Unable to control temperature	49.5%
	Poor quality of bathroom amenities	43.8%
	Poor quality of was towels	42.9%
	poor maintained furnishings	42.2%
	No blackout curtains	43.8%
	No soaking tub	42.0%
	No secure locks or alarm	40.9%
	Not available or not working electronics (e.g. TV)	41.2%
	No stunning views	45.3%
	No fresh air	48.3%
	No WIFI	43.5%
	No electric outlets/ charging station close to bed	55.3%

Table 2 reports the relationship between poor guest sleep satisfaction and hotel attributes after adjusting for demographic covariates (Research Question 1). Only two attributes were not related to guest sleep satisfaction, including “room too small” (p>.05) and “no soaking tub” (p>.05). All other hotel attributes were individually significant predictors of poor guest sleep satisfaction. Increased likelihood of worse sleep satisfaction was associated with all other attributes, ranging lack of WIFI (OR = 1.37, 95% CI: 1.00–1.87) to uncomfortable pillows (OR = 2.47, 95% CI: 1.79–3.39).

Table 2.

Ordinal logistic covariate-adjusted regressions examining the relationships between hotel attributes and guest sleep satisfaction (n=609).

	Poor sleep satisfaction (1= Extremely; 5=Not at all)
		95% CI
Hotel Attribute	OR	Lower	Upper	p-value
1. Sound of the air-conditioning unit or heater	2.03	(1.50	2.74)	<.0001
2. Hotel staff or other noise in the hallway	2.01	(1.47	2.74)	<.0001
3. Noise from another room	2.07	(1.52	2.81)	<.0001
4. Room too light or too dark	1.75	(1.28	2.39)	<.0001
5. Room too small	1.36	(0.98	1.87)	0.063
6. Smell of the room (e.g., musty)	1.76	(1.30	2.40)	<.0001
7. Bed linen uncomfortable	2.47	(1.79	3.39)	<.0001
8. Pillows are uncomfortable	2.92	(2.13	4.00)	<.0001
9. Poor location of the room within the hotel	2.00	(1.46	2.73)	<.0001
10. Mattress too hard or soft	2.47	(1.80	3.38)	<.0001
11. Noise from outside the hotel	2.37	(1.74	3.23)	<.0001
12. Not enough pillows	1.68	(1.23	2.30)	0.001
13. Bedspread old or dirty	1.57	(1.14	2.15)	0.005
14. Room not maintained/clean	1.51	(1.10	2.08)	0.010
15. Unable to control temperature	1.55	(1.14	2.10)	0.005
16. Poor quality of bathroom amenities	1.52	(1.11	2.09)	0.009
17. Poor quality of was towels	1.40	(1.02	1.90)	0.036
18. Poor maintained furnishings	1.52	(1.11	2.09)	0.009
19. No blackout curtains	1.71	(1.25	2.35)	0.001
20. No soaking tub	1.36	(0.99	1.87)	0.059
21. No secure locks or alarm	1.41	(1.03	1.93)	0.032
22. Not available or not working electronics (e.g. TV)	1.47	(1.07	2.03)	0.016
23. No stunning views	1.48	(1.09	2.02)	0.012
24. No fresh air	1.49	(1.10	2.02)	0.011
25. No WIFI	1.37	(1.00	1.87)	0.048
26. No electric outlets/ charging station close to bed	1.61	(1.18	2.21)	0.003

Models adjust for age, sex, and relationship status. Bolded ORs significant at the .05 level.

Table 3 displays results from the combined ordinal logistic regression examining the predictors of poor guest sleep satisfaction that contributed unique variance to sleep satisfaction (Research Question 2). In this model, the variables that contributed unique variance to poor sleep satisfaction were sound of the air conditioning or heating system (OR = 1.57, 95% CI: 1.06–2.34), uncomfortable bed linens (OR = 2.63, 95% CI: 1.46–4.75), uncomfortable pillows (OR = 2.49, 95% CI: 1.53–4.06) and outside noise (OR = 2.23, 95% CI: 1.41–3.54) – these were all associated with increased likelihood of poor sleep satisfaction (McFadden’s Pseudo R² = 0.07).

Table 3.

Ordinal covariate-adjusted logistic regression to identify the hotel attributes that are unique predictors of guest steep satisfaction (N=609).

	Poor Sleep satisfaction (1= Extremely; 5=Not at all)
		95% CI
Hotel Attribute	OR	Lower	Upper	p-value
Sound of the air-conditioning unit or heater	1.57	(1.06	2.34)	0.025
Hotel staff or other noise in the hallway	0.88	(0.54	1.42)	0.597
Noise from another room	1.36	(0.87	2.12)	0.172
Room too light or too dark	1.19	(0.74	1.92)	0.478
Room too small	0.44	(0.25	0.80)	<.001
Smell of the room (e.g., musty)	1.02	(0.61	1.72)	0.938
Bed linen uncomfortable	2.63	(1.46	4.75)	0.001
Pillows are uncomfortable	2.49	(1.53	4.06)	<.001
Poor location of the room within the hotel	1.25	(0.70	2.22)	0.444
Mattress too hard or soft	1.23	(0.74	2.06)	0.428
Noise from outside the hotel	2.23	(1.41	3.54)	0.001
Not enough pillows	1.02	(0.60	1.71)	0.950
Bedspread old or dirty	1.00	(0.51	1.94)	0.992
Room not maintained/clean	0.52	(0.24	1.11)	0.090
Unable to control temperature	0.97	(0.59	1.62)	0.921
Poor quality of bathroom amenities	1.27	(0.65	2.48)	0.482
Poor quality of was towels	0.57	(0.29	1.11)	0.098
Poor maintained furnishings	0.95	(0.46	1.97)	0.888
No blackout curtains	1.42	(0.78	2.58)	0.250
No soaking tub	0.68	(0.36	1.29)	0.234
No secure locks or alarm	0.91	(0.48	1.73)	0.767
Not available or not working electronics (e.g. TV)	0.91	(0.47	1.79)	0.793
No stunning views	0.86	(0.52	1.45)	0.579
No fresh air	0.78	(0.45	1.33)	0.362
No WIFI	0.61	(0.34	1.10)	0.102
No electric outlets/ charging station close to bed	1.34	(0.71	2.53)	0.368

Model adjusted for age, sex, and relationship status. Bolded ORs significant at the .05 level.

To address Research Question 3, ordinal logistic regression examining sleep satisfaction and hotel satisfaction showed a strong association with overall hotel satisfaction after adjusting for demographic covariates (OR = 3.48, 95% CI: 2.86 to 4.23; McFadden’s pseudo R²=0.15). Figure 1 displays the positive overall relationship between hotel satisfaction and sleep satisfaction.

Figure 1.

Sleep satisfaction and guest satisfaction as reported by business and leisure guests (n = 609).

Table 4 displays the relationships between individual hotel attributes and guest satisfaction (Research Question 4). Results show that all hotel attributes were significantly related to hotel satisfaction (range: uncomfortable pillows: OR = 3.29, 95% CI: 2.28–4.73; uncomfortable bed linens: OR = 5.06, 95% CI: 3.45–7.43).

Table 4.

Ordinal covariate-adjusted logistic regressions to identify the relationship between hotel attributes and guest satisfaction (N=609).

	Low Guest Satisfaction (1=Extremely; 5=Not at all)
		95% CI
Hotel Attribute	OR	Lower	Upper	p-value
Sound of the air-conditioning unit or heater	3.06	(2.17	4.31)	<.001
Hotel staff or other noise in the hallway	3.30	(2.31	4.71)	<.001
Noise from another room	3.33	(2.34	4.75)	<.001
Room too light or too dark	3.85	(2.68	5.53)	<.001
Room too small	4.70	(3.24	6.84)	<.001
Smell of the room (e.g., musty)	4.12	(2.87	5.91)	<.001
Bed linen uncomfortable	5.06	(3.45	7.43)	<.001
Pillows are uncomfortable	3.29	(2.28	4.73)	<.001
Poor location of the room within the hotel	5.20	(3.57	7.57)	<.001
Mattress too hard or soft	3.87	(2.68	5.57)	<.001
Noise from outside the hotel	3.74	(2.61	5.36)	<.001
Not enough pillows	3.50	(2.44	5.01)	<.001
Bedspread old or dirty	4.33	(2.99	6.26)	<.001
Room not maintained/clean	4.37	(3.03	6.33)	<.001
Unable to control temperature	3.41	(2.39	4.85)	<.001
Poor quality of bathroom amenities	5.04	(3.46	7.34)	<.001
Poor quality of was towels	4.22	(2.93	6.06)	<.001
Poor maintained furnishings	4.65	(3.21	6.75)	<.001
No blackout curtains	4.23	(2.93	6.11)	<.001
No soaking tub	5.06	(3.47	7.38)	<.001
No secure locks or alarm	4.60	(3.17	6.67)	<.001
Not available or not working electronics (e.g. TV)	4.54	(3.13	6.59)	<.001
No stunning views	4.60	(3.19	6.63)	<.001
No fresh air	3.59	(2.52	5.11)	<.001
No WIFI	5.27	(3.62	7.66)	<.001
No electric outlets/ charging station close to bed	5.09	(3.49	7.42)	<.001

Model adjusted for age, sex, relationship status, and sleep satisfaction. Bolded ORs significant at the .05 level.

The composite score was a sum of twenty-two possible attributes found to be most strongly related to hotel guest sleep. Scores ranged from 0–22. The mean composite score was 10.41, SD = 8.86. Scores were divided into 5 categories: none (0), low (1–5), moderate (6–10), high (11–15), and very high (16–22). The degree to which overall score and score category predicted sleep satisfaction and overall satisfaction are reported in Table 5. The relationship between composite score and both sleep satisfaction and overall guest satisfaction were robust (OR = 1.04 for sleep satisfaction and OR = 1.12 for guest satisfaction, all p < 0.05). Figure 2 displays the composite scores by overall hotel rating, binned in whole numbers 1–5. Regarding the cutoff score, compared to a score of 0, those with low, moderate, high and very high scores were 3.8, 10.3, 14.7, and 5.9 times as likely to report worse sleep satisfaction, respectively (all p < 0.05). Also, those with low, moderate, high, and very high scores were 2.6, 10.9, 15.3, and 18.2 times as likely to report a lower hotel rating, respectively, compared to those with a score of 0 (all p < 0.05). All of these relationships remained significant even after controlling for overall sleep satisfaction, except that “low” scores were no longer associated with worse guest satisfaction.

Table 5.

Ordinal logistic regression examining relationship between hotel attribute composite score and guest sleep satisfaction and overall hotel satisfaction.

	High Satisfaction (1=Not at all; 5=Extremely)
		95% CI
Guest Sleep or Overall Hotel Satisfaction	OR	Lower Upper	p
Guest Sleep Satisfaction
Overall Score	1.04	(1.02, 1.06)	<.001
Low (1–5)	3.78	(2.33, 6.15)	<.001
Moderate (6–10)	10.26	(5.66, 18.59)	<.001
High (11–15)	14.74	(7.04, 30.86)	<.001
Very High (16 and higher)	5.89	(3.58, 9.69)	<.001
	McFadden’s Pseudo R2=0.05
Overall Guest Satisfaction
Overall Score	1.12	(1.10, 1.14)	<.001
Low (1–5)	2.59	(1.54, 4.36)	<.001
Moderate (6–10)	10.8	(5.68, 20.72)	<.001
High (11–15)	15.33	(7.01, 33.52)	<.001
Very High (16 and higher)	18.22	(10.33, 32.15)	<.001
	McFadden’s Pseudo R2=0.12
Overall Guest Satisfaction (Adjusted for Sleep Satisfaction)
Overall Score	1.12	(1.09, 1.14)	<.001
Low (1–5)	1.47	(0.83, 2.63)	0.189
Moderate (6–10)	4.50	(2.22, 9.14)	<.001
High (11–15)	5.89	(2.56, 13.58)	<.001
Very High (16 and higher)	11.32	(6.14, 20.88)	<.001
	McFadden’s Pseudo R2=0.22

Bolded ORs significant at the .05 level.

Figure 2.

Composite score of hotel attributes most important for sleep by overall hotel satisfaction (n = 609).

Discussion

The results from this study illuminated the specific hotel attributes that were most strongly related to guest sleep satisfaction and overall hotel satisfaction. Specifically, results showed that guests who reported a noisy air conditioner in their guest room were 1.57 times more likely to report a poor sleep experience. Also, guests who reported uncomfortable bed linens were 2.63 times more likely to also have poor sleep. Similarly, guests who reported uncomfortable pillows were 2.49 times more likely to report poor sleep, and those who reported noise from outside the hotel were 2.23 times more likely to report poor sleep. The results showed that hotel attributes were strongly associated with overall guest satisfaction.

Regarding hotel attributes that related to guest sleep satisfaction and overall satisfaction, our results highlighted that uncomfortable pillows or bed linens were strong predictors of poor sleep satisfaction. Results also showed that sound from an in-room air conditioning unit was also a significant predictor of poor sleep satisfaction. Previous research from Hon and Fung (2019) found that a comfortable guest room sleeping environment was associated with guest satisfaction and return intentions. Also, Pallesen et al. (2016) identified comfortable pillows as a strong predictor of guest sleep satisfaction. Our findings contribute to this research by illuminating the sleep-related hotel attributes that are associated with higher sleep satisfaction and overall satisfaction among guests.

Our findings, gleaned from quota sampling among travelers reporting a recent hotel stay, show that approximately one third of hotel guests rated their sleep as ‘neither good nor bad’ and just over one third reported their sleep was ‘very good,’ while 18% reported their sleep as ‘excellent.’ Previous research that examined select guest reviews of sleep posted to a travel website found that 70% of rated their sleep as very good or excellent (Hargreaves, 2015). Contrary to Hargreaves, our findings show a substantially higher proportion of guests rate sleep during their last hotel stay as ‘neither good nor bad.’ Hopefully, this finding can offer motivation for future research and practice to design targeted and tailored efforts to improve the guest sleep experience. Indeed, improvements to the guest sleep experience may be easy to implement and offer a high contribution to the overall guest experience. Our study also compared sleep between business and leisure guests. The results of this study showed no difference between business and leisure traveler sleep. Previous research has examined sleep among business travelers (Chen et al., 2018; Gunderson et al., 1996), but had not compared the sleep reported by business travelers to that of leisure travelers. It is somewhat surprising that we did not detect a difference between business and leisure travelers. One explanation for this may be that guest perceptions of hotel attributes may not differ widely between work and leisure travelers, yet future research is needed to explore these domains in greater detail. Another explanation could be that this finding was an artifact of the quota sampling technique.

According to the service gap model, guest satisfaction can be achieved by a focus on closing the gap between guest expectations and perceptions (Parasuraman et al., 1985; Zeithaml and Bitner, 2003). Given the strong relationship that we observed between guest perceptions of hotel attributes, guest sleep and overall satisfaction, our findings illuminate a number of aspects of a hotel environment (e.g., bed linens and pillows) that are important for guest sleep and in turn, guest satisfaction. Previous studies also suggest that hotel guests consider the sleeping environments at a property in forming personal perceptions about sleep quality (Hon and Fung, 2019; Mao et al., 2018). Another view of guest satisfaction offered by Xiang et al. (2015) articulates that the uniqueness of a hotel as imperative for satisfaction. Through this lens, and given our findings on the relationship between sleep and satisfaction, there could be an argument for hotels to make guest sleep a point of uniqueness and differentiation.

While our results illuminate the hotel attributes that are related to guest sleep and guest satisfaction, it is important to note that sleep is never fully within the control of a hotelier. Certainly, a guest with a preexisting sleep disorder may struggle to sleep despite being in an ideal environment. Alternatively, a hotel may be in close proximity to routes taken by emergency vehicles (e.g., fire engines), which may represent a significant external barrier to guests sleeping in streetfacing rooms. Therefore, a hotel must realize that there may be challenges to guest sleep which are outside their control.

According to Hon and Fung (2019), the sleep environment and its relationship to sleep quality are limited and additional research is needed. No studies to our knowledge have examined a comprehensive list of hotel attributes and their relationship to guest sleep or examine the relationship between guest sleep an overall satisfaction. We advance the literature published in this area by examining guest sleep, determinants of guest sleep satisfaction, and the relationship between guest sleep satisfaction and overall hotel satisfaction among a sample of frequent travelers. Our research advances our understanding of hotel guest sleep and its determinants and in so doing offer utility to research and practice in hospitality and tourism.

Limitations and future research

Like any study, ours has a few limitations. The survey was administered by a panel of survey participants identified through a professional survey company. Unfortunately, the survey company did not disclose the response rate. Another limitation is that the data were cross-sectional. Also, our results could have been vulnerable to common method variance, or systematic variance attributable to measurement method rather than the constructs measured (Podsakoff et al., 2003). Further, the scale for all sleep-damaging hotel attributes collected responses from “not at all bothersome,” to “slightly bothersome,” “somewhat bothersome,” “moderately bothersome” and “extremely bothersome.” Yet, we dichotomized the hotel attribute response scales in order to aid in the interpretation of the findings in this study. Retaining the full scale may offer future researchers more statistical power. Further, the measure assessing bedroom light is double barreled (i.e., “too light or too dark”), which is a limitation. Finally, the sleep satisfaction question asked participants to reflect upon their last hotel and report their sleep quality during that stay. It could be that some participants may suffer from recall bias, and the responses about their sleep quality during their last hotel stay might be biased by their current sleep quality.

Our results identify several avenues for future research. Future research may consider experimentally examining different hotel guestroom environments and their effect upon either self-reported or objectively captured guest sleep satisfaction. While several of the attributes that we found to improve guest sleep may require capital improvements, such as quieter air heating or cooling systems, practical considerations for hotels interested in improving guest sleep may include providing guests with ear plugs upon request to reduce undesirable sound in the guest room. Future research may also examine additional guest satisfaction measures associated with dimensions of sleep satisfaction, such as the relationship between hotel guest sleep and likelihood of recommending a hotel to a friend.

Conclusion

Despite the importance of sleep, there is a surprisingly thin body of evidence that has examined sleep among guests in hotel environments, and the specific opportunities for improving the guest sleep experience. We examined hotel attributes that were related to guest sleep satisfaction and overall guest satisfaction. Our findings offer a unique contribution to the literature by identifying a strong relationship between hotel guest sleep and overall guest satisfaction, as well as the attributes that are the strongest predictors of high hotel guest sleep satisfaction.

Biographies

Author Biographies

Rebecca Robbins is a Postdoctoral Research Fellow at the Brigham and Women’s Hospital and Harvard Medical School. Her research is focused on novel interventions for improving sleep outcomes. She is the coauthor of “Sleep for Success.”

Michael Grandner is the Director of the Sleep and Health Research Program at the University of Arizona, Director of the Behavioral Sleep Medicine Clinic at the Banner-University Medical Center, and an Associate Professor in the Department of Psychiatry at the UA College of Medicine, with joint appointments in the Departments of Medicine, Psychology (UA College of Science), Nutritional Sciences (College of Agriculture and Life Sciences), and Clinical Translational Science. In addition, he is a faculty member of the Neuroscience and Physiological Sciences Graduate Interdisciplinary Programs. Grandner is the President of the Society for Behavioral Sleep Medicine.

Adam Knowlden, CHES, MS, MBA, Ph.D. is an Associate Professor of Health Science at the University of Alabama. Dr. Knowlden’s research focuses on the pathogenesis and prevention of adiposity and sleep-associated cardiometabolic disease risk factors. Adam Knowlden has been funded to support his research agenda by the National Heart, Lung, and Blood Institute. He previously served as Associate Editor for the peer-reviewed journal, Health Education & Behavior. Dr. Knowlden earned his MS and PhD degrees from the University of Cincinnati. He received his MBA from Franklin University while employed as a Legislative Analyst for the Columbus City Council (Ohio).

Kimberly Severt is an Associate Professor and the Director of the Hospitality Management program at The University of Alabama. Her research also includes incentive travel, investigating the inefficiencies of entering event specifications into hotel technology systems, hotel sleep issues, and destination familiarity and image.