Journal of Hospitality and Tourism Management

Volume 29, December 2016, Pages 60–68

Performance evaluation of the hotel industry in an emerging tourism destination: Case of

Oman

Amar Oukil, Nabil Channouf, Asma Al-Zaidi

Sultan Qaboos University

Abstract:

This study is concerned with evaluating the performance of the hotel industry in the Sultanate of

Oman through a two-stage data envelopment analysis (DEA) procedure. In the first stage, DEA-

bootstrap is used to estimate point and interval efficiency ratios of the hotels, identify the

benchmark hotels and suggest a potential ranking. In the second stage, a truncated regression

model based on the double bootstrapping procedure of Simar & Wilson (2007) is implemented to

identify potential sources of hotels’ operational inefficiency. In addition, an empirical approach

is introduced to quantify the attractiveness of tourism destinations through a weighting scheme.

The benchmarking analysis is carried out on a sample of 58 hotels, and revealed that (1) the

majority of hotels in Oman are technically inefficient; (2) most of the efficient hotels are located

in the capital, Muscat; (3) star rating and cultural attractions are the most important factors

influencing hotels’ efficiency. Practical implications of these findings are also discussed.

Keywords:

Hotel performance; Data Envelopment Analysis (DEA); Truncated

regression; Bootstrap; Hospitality management; Oman.
1. Introduction

The sultanate of Oman is located on the southern tip of the Arabian Peninsula with, on its

borders, the United Arab Emirates (U.A.E.), Saudi Arabia (S.A.), and Yemen. Oman covers an

area of 309,500 km², with rugged mountains and rocky deep-water fjords to the north, the

mountains and green hills of the Dhofar region to the south, and the Wahiba Sands in the center

(Choufany & Younes, 2005). Lying on the Tropic of Cancer, Oman is one of the world’s hot and

arid regions, though part of the south of the country has a tropical climate (Figure 1).

[ Figure 1]

Oman’s economy is oil based, with an oil activity accounting for 30% of Gross Domestic Product

(GDP) and representing 61% of total exports, estimated to $53bn in 2012 (QNB, 2013). Oman

has been successful at turning its oil wealth into broad-based economic growth, stirred by the

government’s strategy of diversifying the economy and reducing dependence on petroleum

resources. Although the latest among the Gulf countries to join the tourism "race", Oman is

emerging as one of the most attractive tourism destinations on the Arabian Peninsula with the

number of tourists increasing every year (Winckler, 2007). Moreover, tourism industry is

perceived among the key alternatives to petroleum based economy (Subramoniam, Al-Essai, Al-

Marshadi, & Al-Kindi, 2010) and set as one of the top targets of the long-term socio-economic

plan, namely, "Oman 2020" (Winckler, 2007). The industry's total contribution to GDP nears 7%

in 2012, with 77,500 jobs, equivalent to 7% of total employment (WTTC, 2013). The forecast

for 2023 is 117,000 jobs supported by the industry (WEF, 2013).

With a sector expanding so rapidly, measures are being taken by the Omani government to boost

tourism competitiveness, expand tourist base, facilitate travel activities, and endorse innovative

initiatives (Assaf & Barros, 2011). As the largest and arguably the most important actors of
tourism industry, hotels must compete globally to attract customers and achieve high profits

(Tarim, Dener, & Tarim, 2000). Viewed from this perspective, conducting a performance

evaluation of the hotel industry is a necessary step to developing a meaningful set of benchmarks

for best practices and successful hotel businesses (Min, Min, & Joo, 2009). Such a focused study

can help stakeholders to determine current competitive positions of different hotels in the Omani

market, in addition to supporting decisions pertaining to the improvement of operational

performance, downscaling specific operations, or deferring scheduled expansions (Assaf &

Barros, 2011). To the authors’ best knowledge, performance of the hotel industry has never been

researched in Oman’s context specifically. Therefore, the present paper adds to previous research

in this field.

This study uses a two-stage approach (Barros, Botti, Peypoch, & Solonandrasana, 2011; Shang,

Wang, & Hung, 2010). The approach starts with a Data Envelopment Analysis (DEA) evaluation

of the hotels’ technical efficiencies, followed by a statistical regression of the efficiency scores

over a set of contextual factors. The objective of the second stage is to identify the factors that

contribute more significantly to the efficiency of the hotels. A truncated regression model with a

double bootstrapping procedure (Simar & Wilson, 2007) is implemented to identify these factors,

but also to assess the consistency of the DEA efficiency scores.

In the light of the above, the contribution of the present study to the hospitality and tourism

literature is two-fold. First, the study investigates efficiency measures of the Omani hotel

industry, a topic that has not been addressed hitherto, in spite of its pertinence to such a growing

industry. Second, the study examines the contextual factors that impact the hotel industry in

Oman, with possible extension to other tourism destinations with similar characteristics.

Concurrently, an empirical approach is introduced to quantify the attractiveness features of

touristic destinations.
The remainder of the paper unfolds as follows. In the next section, a brief review of the literature

pertaining to the two-stage approach in the hotel industry is presented. Section 3 outlines the

methodology of the study. Section 4 is dedicated to the discussion of the results related to point

and interval estimation of the hotels’ efficiency scores. In Section 5, the potential relation

between the hotel contextual factors and efficiency levels is discussed. The paper concludes with

some recommendations and possible venues for future research.

2. Literature review

In recent years, the measurement of efficiency in the hotel industry has mostly been addressed

through frontier efficiency methods, namely, the stochastic frontier (Greene, 2008) and data

envelopment analysis (Cooper, Seiford, & Tone, 2002). The stochastic frontier analysis (SFA)

requires the output of the decision making units (DMUs) to be expressed as an explicit function of

a set of inputs, an inefficiency factor, and a random error whose distribution is assumed a priori

(Coelli, Rao, O'Donnell, & Battese, 2005). Some leading studies that use SFA in the hotel industry

include Anderson, Fish, Xia, & Michello (1999), Barros (2004, 2006), Chen (2007), and Hu, Chiu,

Shieh, & Huang (2010). Unlike SFA, data envelopment analysis (DEA) is a non-parametric approach

that does not impose functional forms on the data nor does it need to use probability distributions

(Barros, Botti, Peypoch, & Solonandrasana, 2011). Furthermore, DEA has the potential to

evaluate the efficiency of DMUs that employ multiple inputs (resources) to produce multiple

outputs (products and/or services).

According to Wöber (2007) “Although efficient frontier methods have been used extensively in

the past, it has been just recently that tourism researchers have discovered DEA for examining

efficiency in their industry”. Indeed, the share of tourism is estimated to only 1.34% of all DEA

application papers (Liu, Lu, Lu, & Lin, 2013). Hruschka (1986) and Banker & Morey (1986a) are
first to apply DEA to the hospitality industry, more specifically, to restaurants. Later, Bell &

Morey (1994, 1995) use DEA to determine best practices for corporate travel agencies. The

application of DEA to the hotel industry is pioneered by Morey & Dittman (1995).

Over 63% of related publications cover destinations in the Asian Pacific region (Keh, Chu, & Xu,

2006; Assaf, 2012), with around 50% dealing with the hotel industry only in Taiwan (e.g., Assaf,

Barros, & Josiassen, 2010; Chin, Wu & Hsieh, 2013; Huang, Ho, & Chiu, 2014). Research on

the performance of hotels in the Middle East using DEA is very scarce. The few existing

publications consider cases in Turkey (e.g., Tarim, Dener, & Tarim, 2000; Önüt & Soner, 2006;

Tumer, 2010), Iran (Shirouyehzad, Hosseinzadeh Lotfi, Shahin, Aryanezhad, & Dabestani, 2012)

and Israël (Hadad, Friedman, & Israeli, 2005). Apart from the study in Assaf & Barros (2011)

which involves hotel chains from S.A., the U.A.E. and Oman, there is no known research

dedicated specifically to performance analysis of the hotel industry in Oman. Therefore, the

present work enriches the literature in this field through a systematic analysis of hotels’

performance in Oman with a view to identify benchmarks for best practices and support

stakeholders’ operational decisions.

Our methodological approach covers two-stages. The first stage uses DEA to estimate the

hotels’ efficiency scores. In the second stage, an econometric analysis is conducted to discern

possible correlation between the DEA efficiency scores and the contextual factors. The latter are

often exogenous factors that are neither inputs nor outputs, but can still influence the operating

process (Jeong, Park, & Simar, 2010). The objective is to identify the factors that might influence

efficiency significantly.

The application of the two-stage approach in the hotel industry is quite recent. Early studies

have investigated the effect of hotel contextual factors on efficiency using ordinary least squares
(OLS) estimation (e.g., Sun & Lu, 2005). However, the OLS estimation has been considered

unsuitable for explaining the efficiency scores since the latter variables are bounded. Instead,

Tobit regression models have been used in subsequent research. In Hu, Shieh, Huang, & Chiu

(2009), DEA is adopted to evaluate the operational performance of international tourist hotels

(ITHs) in Taiwan through cost, allocative, technical, and scale efficiency ratios. In the second

stage, each of these ratios is regressed on a set of environmental variables using Tobit model

(Tobin, 1958). A similar approach is also used in Chen, Hu, & Liao (2010), Honma & Hu (2012)

and Huang, Mesak, Hsu, & Qu (2012). Simar & Wilson (2007) argue that the efficiency

estimates are serially correlated, which renders the standard inference approaches used in the

conventional two-stage DEA procedure statistically invalid. Therefore, the truncated regression

model is used to deal with the bias problems in the second stage of the DEA approach. Under the

assumption that the distribution of efficiency is truncated normally with a mean of zero, Barros &

Dieke (2008) examine the determinants of efficiency of African hotels. More recently, the

truncated regression is applied with a bootstrapping procedure in Barros, Botti, Peypoch &

Solonandrasana (2011), Chen, Hu, & Liao (2010), Tundis, Corsino, & Zaninotto (2012), Fang

(2013), Hu, Yeh, & Tsai (2014) and Hathroubi, Peypoch, & Robinot (2014). More extensive

reviews and references can be found in, e.g., Manasakis, Apostolakis, & Datseris (2013) and

Fang (2013).

In this paper, a truncated regression model with a double bootstrapping procedure (Simar &

Wilson, 2007) is used (1) to estimate the bias and produce confidence intervals for the efficiency

scores of the hotels in Oman, and (2) to discriminate the contextual factors that have substantial

effect on these scores.

3. Methodological framework

The DEA models that are most frequently applied in the hotel industry are CCR (Charnes,

Cooper, & Rhodes, 1978), which assumes constant returns to scale (CRS), and BCC (Banker,

Charnes, & Cooper, 1984), which allows variable returns to scale (VRS). VRS implies

disproportionate variation in outputs when inputs are increased. Under either CRS or VRS

assumption, the managerial purposes of efficiency analysis, in a competitive context, are the

measurement of relative efficiency ratios as an essential step to setting industry’s benchmarks

besides achieving more profit. Therefore, the output-oriented versions of CCR and BCC models

are more suitable. Other models are used in the literature, depending on the contexts and

managerial objectives.

3.1. Efficiency estimation

Assume a set of K hotels, each hotel k defined with N inputs x and M outputs y. With

reference to the underlying production technology, hotel (xk, yk) is fully defined with the observed

values xik and yjk , with i=1,.., N and j=1,.., M. To estimate the efficiency score h of hotel (xh, yh)

and set production targets for inefficient hotels, the output-oriented formulation of CCR model

can be represented as follows.

max θh (1)
Subje ctto :
K
(CCR) x
k 1
k ik  xih i  1,...,N (2)
K

y
k 1
k jk  θh y jh j  1,...,M (3)

k  0 k  1,...,K (4)
The efficiency h of hotel (xh, yh) represents the maximal radial increase of outputs that is

required to reach the efficiency frontier for a specified level of inputs. The vector λ measures the

weights of peers in producing the projection of hotel (xh, yh) on the efficiency frontier. Constraints

(2) and (3) state that reference points are linear combinations of the input and output values of

efficient peers for hotel (xh, yh).

BCC model can be obtained from (CCR) by adding the convexity constraint that guarantees that

 k  1.
K
only weighted averages of efficient hotels enter the reference set, i.e. k 1

The adequate choice of inputs and outputs for a DEA based benchmarking problem lies

often on the dicta “less is better” and “more is better”, respectively (Cook, Tone, & Zhu, 2014).

Thus, with respect to the specific context of our study, we identified 4 outputs and 4 inputs.

The output variables are Annual revenue (Chiang, Tsai, & Wang, 2004; Barros & Mascarenhas,

2005; Neves & Lourenco, 2009; Pulina, Detotto, & Paba, 2010), Number of guests (Barros,

2005b), Number of nights (Barros, 2005b; Barros & Mascarenhas, 2005; Sigala, Jones,

Lockwood, & Airey, 2005) and Occupancy rate (Chiu, Huang & Ting, 2012; Ting & Huang,

2012; Yang & Lu, 2006). Annual revenue includes incomes from the rental of the hotel rooms,

food and beverages served to customers, phone call bills, as well as laundry services. Number of

guests counts hotel’s guests, regardless of the duration of their stay. Number of nights provides a

cumulative value of full nights spent in the hotel. Occupancy rate refers to the proportion of hotel

capacity effectively used over a specific time period (e.g. one year), i.e. number of rooms rented

out over the total number of rooms available. Occupancy rate has been used recently and it is

managerially useful (Perrigot, Cliquet, & Piot-Lepetit, 2009). The input variables are Number of

beds (Manasakis, Apostolakis, & Datseris, 2013), Number of rooms (Anderson, Fok, & Scott,
2000; Assaf et al., 2010; Barros, 2005b; Chen, Hu, & Liao, 2010), Number of employees (Chiang,

Tsai, & Wang, 2004; Barros & Mascarenhas, 2005; Hwang & Chang, 2003), and Salary of

employees (Assaf & Agbola, 2011; Morey & Dittman, 1995; Reynolds, 2003).

The data used for this study have been collected from the Ministry of Tourism through direct

access to the database of hotels available at the department of Statistics & Geographic

Information. All required information was obtained for 58 hotels, spread over seven regions of

Oman (Muscat, Dhofar, Al-Buraymi, A'Dakhiliyah, A'Sharqiyah, Al-Batinah, and Musandam). A

statistical summary of the corresponding inputs and outputs is given in Table 1.

[ Table 1]
Note that Number of rooms and Number of beds are strongly correlated, with a correlation

coefficient rbr=0.9768. The same holds for the input variables Number of employees and Salary

of employees, with res=0.9087. Therefore, Number of beds and Salary of employees are the only

input variables we consider for the analysis.

3.2. Identification of efficiency drivers

Commonly, the efficiency estimation is carried out without considering contextual factors that

may influence the outcome of the hospitality operations. The second stage analysis is conducted

to assess the cross-sectional association of these factors with the DEA efficiency scores. In an

output orientated DEA model, these scores’ estimators are biased upward for this data

configuration and bounded on the left at 1 (1h) that is, h–1 is the proportional increase in

outputs that could be achieved by hotel h with input quantities held constant. If zh denotes the
vector of contextual variables and  the associated coefficients in a regression model, we have

h=  zh+ 1    1-  zh , where  is the error term,  N(0,).

Therefore, a truncated regression of the inefficiency scores h–1 against the contextual variables

zh can be used to identify the factors that may influence more significantly the efficiency

estimates.

In the hotel industry, examples of contextual factors include hotel size (Assaf, Barros, &

Josiassen, 2010), location (Barros, 2005a; Bernini & Guizzardi, 2010; Tundis et al., 2012), and

type of ownership (Barros & Dieke, 2008), which are found to be strong determinants of hotel

efficiency in many case studies. Other variables could also be pertinent, like star rating (Assaf &

Cvelbar, 2010), used essentially to reflect quality of service, even though it is far from being a

wholly satisfactory proxy for such an operational factor (Oliveira, Pedro, & Marques, 2013).

Based on previous studies, we consider four contextual variables: Type of ownership, Hotel size,

Star rating and Attractions. The variable Attractions is introduced to investigate the influence of

hotel’s location on its efficiency. Bernini & Guizzardi (2010) suggest that location is positively

correlated with technical efficiency, especially for sun and beach destinations, as well as cities

with renowned cultural importance. Thus, resources that may contribute to the attractiveness of a

hotel’s location need to be conserved (Gomezelj & Mihalič, 2008). The latter being nominal, it

cannot be used in a regression model without a prior quantification. For that reason, the number

of attractions is used as a quantitative substitute.

Based on the classification of the Ministry of tourism, there are three categories of attractions:

Nature, Culture, and Activities. The items that fall under each category are as follows:

 Nature: Reserves, valleys, strait of Hormuz, mountains, caves, deserts, beaches, islands,

water springs, lagoons, rocks park, canyon, Muscat geo-site.

 Culture: Aflaj system, traditional villages, souqs, world heritage, museums, forts, castles,

archeological and religious sites, crafts, frankincense, cities.

 Activities: Scuba diving, boating, climbing, Via Ferrata, trekking, camping, caving, golf,

kite-surfing, kite-boarding, shopping, watching (whales, birds, turtles, dolphins), racing

(camels, horses), off-road, Muscat geo-heritage.

In order to gauge the individual effect of each category and draw more focused decisions, we

consider them as separate variables.

Accordingly, the variable location is represented with three variables, whose values are

calculated as follows. First, we identify all potential attraction sites and activities related to each

destination. Next, we cluster these items based on the above classification scheme. Finally, we

count the number of items for each category. Each number translates the weight of each location

with respect to each attraction category. The values obtained are presented in Table 2.

[ Table 2]
For instance, the value of variable Nature is 16 for Muscat, that is, there are potentially 16

touristic sites in Muscat corresponding to, at least, one of the items listed under category Nature.

Similar reasoning applies to the other variables. Muscat is, apparently, the most attractive with

respect to cultural sites, while Dhofar is leading with its natural sites. The majority of regions

offer some sort of activities, except Al-Batinah.

Regarding the other contextual variables, Type of ownership is a dichotomous variable taking a

value 1 if the hotel is part of a chain of hotels, a value 0 otherwise. For hotel size, we use values 0,

1 or 2 depending on whether the hotel is small, medium or large, respectively, that is, the number

of rooms is less than 100, between 100 and 300, or more than 300. Star rating refers to the

number of stars assigned to a hotel for the previous year’s exercise, a number varying between 1
and 5. Ray & Phillips (2005) and Assaf & Agbola (2011) suggest that the number of stars and

efficiency are positively correlated, that is, the more stars, the better the performance.

The summary statistics for the contextual variables are given in Table 3.

[ Table 3]

3.3.Double bootstrapping procedure

According to Simar & Wilson (2007), conventional inference methods used in the two-stage

DEA procedure are based on efficiency estimates that are serially correlated. As a result, related

statistical inference might not be reliable. To enable consistent inference on the efficiency scores,

Simar and Wilson (2007) develop a double bootstrap algorithm.

The bootstrapping concept is based on the idea that simulating the sampling distribution of

interest is possible by mimicking the data-generating process (DGP). Under the assumption that

the original data sample is generated by the DGP, the DEA efficiency scores are re-estimated

with the ‘simulated’ data. Through multiple replications of this process, a Monte Carlo

approximation of the sampling distribution is derived from the empirical distribution of the

bootstrap values.

The double-bootstrapping procedure (Simar and Wilson, 2007) works as follows

Step 1: Compute the efficiency score h for each hotel (xh, yh) by solving model (CCR).

Step 2: Use truncated maximum likelihood estimation to regress h against a set of contextual

variables zh and provide an estimate β̂ of the coefficient vector β and an estimate σ̂ ε of σ ε , the

standard-deviation of the residual errors  .

Step 3: Repeat the next sub-steps B1 times for each hotel h (h=1, …, K) to produce a set of B1

bootstrap estimates ˆhb for b=1, …, B1.

3.1. Generate the residual error  h from the normal distribution N(0, σ̂ ε 2 ) with left-

truncation at (1  β̂zh ) .

3.2. Calculate h*  β̂zh  εh .

3.3. Construct a pseudo data set ( x*h , y *h ) where xh*  xh and yh*  yh h / h* .

3.4. Run model (CCR) with the pseudo data set ( x*h , y *h ) to compute an estimate ˆh* of the

“real” efficiency score.

Step 4: Calculate the bias-corrected estimator ˆh for each hotel h (h=1,…, K) using the bootstrap

estimator of the bias b̂h (Simar and Wilson, 1998, 2000) where ˆh  h  b̂h and

 1 B1 
bˆh   ˆhb*    h .
 B1 b1 

Step 5: Use truncated maximum likelihood estimation to regress ˆh on the contextual variables zh

and provide an estimate β̂* for β and an estimate σ̂ * for σ ε .

Step 6: Repeat the next sub-steps B2 times to yield a set of B2 pairs of bootstrap estimates

(β̂*b* , σ̂ *b* ) with b=1, …, B2.

6.1. Generate  h from the normal distribution N(0, σ̂ *2 ) with left-truncation at (1  β̂* zh ) for

each hotel h (h=1,…, K).

6.2. Calculate ˆh** for each hotel h (h=1,…, K) so that ˆh**  β̂* zh  εh .
 6.3. Use truncated maximum likelihood estimation to regress ˆh** on the contextual

variables zh and provide an estimate β̂** for β and an estimate σ̂ ** for σ ε

Step 7: Construct the estimated 1  α  % confidence interval of the j-th element β j of the vector

β , that is, [Lowerα, j , Upperα, j ]  [β̂*j  âα , β̂*j   b̂α ] with Prob( b̂α  β̂*j*  β̂*j  âα )  1  α

4. Point and interval efficiency evaluation

The efficiency evaluation is performed using a code implemented under R version 3.0.1. In both

parts of the double bootstrapping procedure, the computations are conducted over 3,000 bootstrap

iterations, i.e., B1=B2=3,000. All the results required less than 1 hour of computer time, running

on a desktop PC (HP double processor*3.40 GHz, 8 GB RAM).

Table 4 provides, for each hotel, the initial efficiency estimate h obtained from model (1)-(4), the

bias-corrected efficiency value ˆh , and the lower and the upper bounds of the efficiency score’s

confidence interval computed at 95% significance level.

[ Table 4]
The values of h show that only 8 hotels out of 58 are technically efficient under CRS and almost

thrice (22 hotels) under VRS. Over the seven regions involved in the study, almost all the

efficient hotels are located in Muscat. Such a high concentration is primarily justified, knowing

that Muscat is the capital and hosts most of the important touristic sites, besides more than half

the number of hotels. For the inefficient hotels, the associated reference sets (potential

benchmarks) are given in Table 5 for VRS assumption.

[ Table 5]
The average results indicate that there is a considerable potential for efficiency improvement in

terms of output increase while keeping the level of input constant. Indeed, the figures in Table 5

show that the inefficient hotels are required to expand their outputs by less than 1.8%, on

average, except for the occupancy rate which needs to be increased by more than 25%. Hence,

more focus could be put on the outputs Annual revenue, Number of guests, and Number of nights

since the corresponding distances to the efficiency frontier are smaller.

[ Table 5]
Meanwhile, the interval estimates of pure technical efficiency scores, constructed with

bootstrapping, reveal that, at 95% confidence, the average interval width is 0.393 for an average

variance of 0.022. These intervals are relatively wider for the efficient hotels, with an average

width of 0.414 and an average variance of 0.018, a minimum width of 0.098 and a maximum of

0.878. In the case of hotel performance research, this is an important finding if the purpose of the

frontier estimation is to identify best and worst performing hotels. The narrower the widths of

the confidence intervals the better one’s position to statistically identify specific clusters of hotels

in terms of relative efficiency.

[ Table 6]
For instance, the benchmark hotels under VRS technology can be ranked based on the widths of

the associated confidence intervals, as in Table 6, where H18 (Safeer Continental Hotel) and H57

(Al-Shumukh Guest House) can be presented as, respectively, the first and the last ranked hotels

in the group of the best performing hotels.

5. Identifying performance drivers

The results of the truncated regression analysis with the double bootstrap are displayed in Table 7

for both CRS and VRS technologies. The statistical significance of our results is assessed using

95% confidence intervals.

[ Table 7]

Since inefficiency is the regressand of the truncated model, parameters β̂** whose values are

negative denote a potential for improvement and, as a result, the corresponding factors are source

of efficiency.

With respect to individual significance levels, the results show that variables Type of ownership,

Hotel size, Nature and Activities are statistically insignificant. With the exception of Activities,

the coefficients of the latter variables are positive, suggesting a negative impact on hotel

efficiency. In the meantime, the Activities variable affects positively technical efficiency in both

CRS and VRs specifications albeit statistically insignificant.

Although the negative effect of the variable Nature is found to be statistically significant only

under CRS assumption, such a result is unexpected as it conflicts with the trend of domestic

tourism, known for being strongly influenced by natural factors. Yet, this result can be justified as

the sample of hotels considered for our study includes very few hotels from nature-based touristic

destinations, like Salalah.

On the other hand, the negative impact of hotel ownership on efficiency does not conform to the

findings of related studies (e.g., Barros & Dieke, 2008). In practice, chain ownership is expected

to boost efficiency through better management abilities, more accessibility to novel technologies,

and higher capital at lower cost. On an aggregated level, only 32.76% of the hotels belong to a

chain but account for 74.57% of the annual revenue, 56.03% of the number of guests, 62.79% of
the number of nights, and 41.70% of the occupancy rate. Yet, these figures are not sufficient to

assess the impact of hotel ownership on efficiency.

Regarding size, the hypothetical relationship is a positive relationship between hotel size and

profit opportunity and, hence, hotel efficiency. However, the existing literature does not present

converging results on the matter. While Barros & Dieke (2008) show that the larger the more

efficient applies for African hotels, other researchers (e.g., Chen, 2007; Hwang & Chang, 2003)

find that efficiency is not affected by size in the case of Taiwanese hotels, which also contradicts

recent findings of Assaf et al. (2010). In our case, only 31.03% of the hotels fall under the

category medium or large (more than 100 rooms) but represent all together 82.78%, 64.59%,

68.62% and 35.80% of the annual revenue, the number of guests, the number of nights and the

occupancy rate, respectively. Again, these proportions being restricted to an aggregated cluster of

hotels, they cannot be used to support decisions on an individual hotel level.

The variables Star rating and Culture are statistically significant and, consequently, are proven to

be important sources of efficiency for hotels. The factor Star rating has more effect on efficiency

than Culture as revealed from the corresponding parameters β̂ * * . Indeed, the contributions

(β̂*3* , β̂*5* ) of these factors under CRS and VRS specifications are, respectively, (-0.394,-0.081)

and (-0.285,-0.065). The positive impact of Star rating on technical efficiency is in line with the

findings of Ray & Phillips (2005) as well as Assaf & Agbola (2011). Moreover, this result is

practically consistent with the market parity between quality of service and room price (revenue),

assuming that star rating reflects effectively the expected service quality.
6. Conclusions, implications and future research

The present study provided the first performance analysis of the hospitality industry in Oman

based on a DEA-double bootstrap procedure. The point estimation of technical efficiency

revealed that 13.8% and 37.9% of Oman’s hotels are efficient under CRS and VRS assumptions,

respectively. The interval estimation produced ranges of efficiency scores that are relatively

narrow, which allowed us to attempt a ranking of the efficient hotels and distinguish Safeer

Continental Hotel and Al-Shumukh Guest House as the best and the worst efficient hotels. The

slack analysis showed that faster efficiency improvement might be achieved through expanding

the outputs Annual revenue, Number of guests, and Number of nights. About 72.7% of the

efficient hotels located in Muscat. Such a high concentration is probably due to the attractiveness

of Muscat, being the capital of the country and the key business place. These facts, in spite of

being positive indicators for potential investors, may also reflect strong centralization of current

tourism operations. Indeed, the statistics of the Ministry of Tourism reveal that there are 54

hotels in the capital Muscat alone, with a lodging capacity of 4,602 rooms (approximately 62% of

the total capacity). Dhofar follows with 15 hotels and 914 rooms. The other regions, all together,

host only 38 hotels, that is, about 25% of the total capacity. Henceforth, more measures ought to

be taken to promote tourism in the other regions of the country, for example, through discount

packages from airlines and hotels, targeting foreign and domestic tourists.

Potential factors of inefficiency have been investigated via truncated regression analysis using the double

bootstrapping procedure suggested by Simar & Wilson (2007). The results showed that the factors Type

of ownership (independent or chain dependent), Hotel size, Nature and Activities have no impact on the

efficiency of a hotel, whilst Star rating and Culture appear as the most influential factors. The positive

effect of Culture may also translate the dominant profile of hotels’ customers. This can inform the
marketing operations on the customers’ populations to be targeted in order to promote other attractions,

like nature and activities. These findings could also benefit the Omani government in the process of

strategy improvement. Future strategies might focus on setting clear policies for the

rehabilitation of the existing cultural heritage which consists of 748 major archaeological sites in

addition to more than 2,660 archaeological and historic buildings and landmarks over the country.

Furthermore, incentive schemes might be developed to encourage construction of new hotels in

areas close to cultural sites.

Future research may be enriched with more input and output variables, together with a horizon

extension covering more than one year, so that the dynamics of the efficiency measures can be

captured. In addition, one could consider incorporating all of the variables (discretionary and

non-discretionary) into the same model using extended DEA models (e.g. Banker & Morey,

1986a,b).

References:

Anderson, R.I., Fish, M., Xia, Y., & Michello, F. (1999). Measuring efficiency in the hotel
industry: a stochastic frontier approach, International Journal of Hospitality
Management, 18(1), 45–57.
Anderson, R.I., Fok, R., & Scott, J. (2000). Hotel industry efficiency: An advanced linear
programming examination. American Business Review, 18(1), 40–48.
Assaf, A.G. (2012). Benchmarking the Asia Pacific tourism industry: A Bayesian
combination of DEA and stochastic frontier. Tourism Management, 33, 1122-27.
Assaf, A.G., & Agbola, F. (2011). Modelling the performance of Australian hotels: a DEA
double bootstrap approach. Tourism Economics, 17(1), 73-89.
Assaf, A.G., & Barros, C. (2011). Performance analysis of the Gulf hotel industry: A
Malmquist index with bias correction. International Journal of Hospitality
Management, 30(4), 819–26.
Assaf, A.G., Barros, C.P., & Josiassen, A. (2010). Hotel efficiency: A bootstrapped
metafrontier approach. International Journal of Hospitality Management, 29(3), 468-
75.
Assaf A.G., & Cvelbar L.K. (2010). The performance of the Slovenian hotel industry:
evaluation post-privatisation. International Journal of Tourism Research, 12(5), 462–
71.
Banker, R.D., Charnes, A., & Cooper, W.W. (1984). Some models for estimating technical
and scale inefficiencies in data envelopment analysis. Management Science, 30(9),
1078-92.
Banker, R.D., & Morey, R.C. (1986a). Efficiency analysis for exogenously fixed inputs and
outputs. Operations Research, 34(4), 513-21.
Banker, R.D., & Morey, R.C. (1986b). The use of categorical variables in data envelopment
analysis. Management Science, 32(12), 1613-27.
Barros, C. P. (2004). A stochastic cost frontier in the Portuguese hotel industry. Tourism
Economics, 10(2), 177–92.
Barros, C.P. (2005a). Evaluating the efficiency of a small hotel chain with a Malmquist
productivity index. The International Journal of Tourism Research, 7(3), 173-84.
Barros, C.P. (2005b). Measuring efficiency in the hotel sector. Annals of Tourism Research,
32(2), 456-77.
Barros, C.P. (2006). Analyzing the rate of technical change in the Portuguese hotel industry.
Tourism Economics, 12(3), 325–346.
Barros, C.P., Botti, L., Peypoch, N., & Solonandrasana, B. (2011). Managerial efficiency and
hospitality industry: the Portuguese case. Applied Economics, 43(22), 2895-905.
Barros, C.P., & Dieke, P.U.C. (2008). Technical efficiency of African hotels. International
Journal of Hospitality Management, 27(3), 438-47.
Barros, C.P., & Mascarenhas, M.J. (2005). Technical and allocative efficiency in a chain of
small hotels. International Journal of Hospitality Management, 24(3), 415–36.
Bell, R.A., & Morey, R.C. (1994). The search for appropriate benchmarking partners– A
macro approach and application to corporate travel management. Omega, 2(5), 477-
90.
Bell, R.A., & Morey, R.C. (1995). Increasing the efficiency of corporate travel management
through macro benchmarking. Journal of Travel Research, 33(3), 11-20.
Bernini, C., & Guizzardi, A. (2010). Internal and locational factors affecting hotel industry
efficiency: evidence from Italian business corporations. Tourism Economics, 16(4),
883-913.
Charnes, A., Cooper, W.W., & Rhodes, E. (1978). Measuring the efficiency of decision
making units. European Journal of Operational Research, 2(6), 429-44.
Chen, C.F. (2007). Applying the stochastic frontier approach to measure hotel managerial
efficiency in Taiwan. Tourism Management, 28(3), 696–702.
Chen, C.T., Hu, J.L., & Liao, J.J. (2010). Tourists’ nationalities and the cost efficiency of
international tourist hotels in Taiwan. African Journal of Business Management,
4(16), 3440–46.
Chiang, W., Tsai, M., & Wang, L.S. (2004). A DEA evaluation of Taipei hotels. Annals of
Tourism Research, 31(3), 712–75.
Chin, J.B., Wu, M.C., & Hsieh, L.F. (2013). Strategic planning of optimal resource allocation
in response to global financial crisis – A study of international tourist hotels. Applied
Economics, 45(23), 3316–28.
Chiu, Y., Huang, C., & Ting, C. (2012). A non-radial measure of different systems for
Taiwanese tourist hotels’ efficiency assessment. Central European Journal of
Operations Research, 20(1), 45–63.
Choufany, H. M., & Younes, E. (2005). Oman today’s spotlight, tomorrow’s destination. HVS
International London, available at: http://www.hvs.com/Content/1687.pdf
Coelli, T., Rao, D.S.P. O'Donnell, C.J., & Battese, G.E. (2005). An Introduction to efficiency
and productivity analysis, Springer, New York.
Cook, W.D., Tone, K., & Zhu, J. (2014). Data envelopment analysis: Prior to choosing a
model. Omega, 44(1), 1-4.
Cooper, W.W., Seiford, L.M., & Tone, H. (2002). Data envelopment analysis: A
comprehensive text with models, applications, references, and DEA-Solver software,
Kluwer Academic Publisher, N.Y.
Fang C.Y. (2013). Did the financial crisis and policy deregulation for Chinese tourists affect
the efficiency and output slacks of Taiwanese hotels? In Matias, A., Nijkamp, P., &
Sarmento, M. (Eds), Quantitative methods in tourism economics, Springer-Verlag
Berlin Heidelberg
Gomezelj, O., & Mihalič, T. (2008). Destination competitiveness - applying different models,
the case of Slovenia. Tourism Management, 29(2), 294-307.
Greene, W. (2008). The econometric approach to efficiency analysis. In Fried, H.O., Lovell,
C.A.K., & Schmidt, S.S. (Eds), The measurement of productive efficiency and
productivity growth, Oxford University Press.
Hadad, Y., Friedman, L., & Israeli, A.A. (2005). Evaluating hotel advertisements efficiency
using data envelopment analysis. Journal of Business Economics and Management,
6(3), 145-53.
Hathroubi, S., Peypoch, N., & Robinot, E. (2014). Technical efficiency and environmental
management: The Tunisian case. Journal of Hospitality and Tourism Management,
21(1), 27-33.
Honma, S., & Hu, J.L. (2012). Analyzing Japanese hotel efficiency. Tourism & Hospitality
Research, 12(3), 155–67.
Hruschka, H. (1986). Ansätze der efflzienzmessung von betrieben, Journal für
Betriebswirtschaft : management review quarterly, 36(2), 76-85.
Hu, J.L., Chiu, C.N., Shieh, H.S., & Huang, C.H. (2010). A stochastic cost efficiency analysis
of international tourist hotels in Taiwan. International Journal of Hospitality
Management, 29(1), 99–107.
Hu, J.L., Shieh, H.S., Huang, C.H., & Chiu, C.N. (2009). Cost efficiency of international
tourist hotels in Taiwan: A data envelopment analysis application. Asia Pacific
Journal of Tourism Research, 14(4), 371-84.
Hu, J.L., Yeh, F.Y., & Tsai, Y.C.(2014). Regional traffic and medical services and cost
efficiency of international tourist hotels in Taiwan. International Business Research,
7(10), 36-46.
Huang, C., Ho, F.N., & Chiu, Y. (2014). Measurement of tourist hotels' productive efficiency,
occupancy, and catering service effectiveness using a modified two-stage DEA model
in Taiwan. Omega, 48, 49–59.
Huang, Y., Mesak, H.I., Hsu, M.K., & Qu, H. (2012). Dynamic efficiency assessment of the
Chinese hotel industry. Journal of Business Research, 65(1), 59–67.
Hwang, S.N., & Chang, T.Y. (2003). Using data envelopment analysis to measure hotel
managerial efficiency change in Taiwan. Tourism Management, 24(4), 357–69.
Jeong, S.O., Park, B.U., & Simar, L. (2010). Nonparametric conditional efficiency measures:
asymptotic properties. Annals of Operational Research, 173, 105–122.
Keh, H.T., Chu, S., & Xu, J. (2006). Efficiency, effectiveness and productivity of marketing
in services. European Journal of Operational Research, 170(1), 265–76.
Liu, J.S., Lu, L.Y.Y., Lu, W.M., & Lin, B.J.Y. (2013). A Survey of DEA applications.
Omega, 41(5), 893–902.
Manasakis, C., Apostolakis, A., & Datseris, G. (2013). Using data envelopment analysis to
measure hotel efficiency in Crete. International Journal of Contemporary Hospitality
Management, 25(4), 510-35.
Min, H., Min, H., & Joo, S.J. (2009). A data envelopment analysis on assessing the
competitiveness of Korean hotels. The Service Industries Journal, 29(3), 367–85.
Morey, R., & Dittman, D. (1995). Evaluating a hotel GM's performance: A case study in
benchmarking. Cornell Hotel and Restaurant Administration Quarterly, 36(5), 30-35.
Neves, J., Lourenco, S. (2009). Using data envelopment analysis to select strategies that
improve the performance of hotel companies. International Journal of Contemporary
Hospitality Management, 21(6), 698–712.
Oliveira, R., Pedro, M.I., & Marques, R.C. (2013). Efficiency and its determinants in
Portuguese hotels in the Algarve. Tourism Management, 36, 641-49.
Önüt, S., & Soner, S. (2006). Energy efficiency assessment for the Antalya Region hotels in
Turkey. Energy and Buildings, 38(8), 964–71.
Perrigot, R., Cliquet, G., & Piot-Lepetit, I. (2009). Plural form chain and efficiency: Insights
from the French hotel chains and the DEA methodology. European Management
Journal, 27(4), 268-80.
Pulina, M., Detotto, C., & Paba, A. (2010). An investigation into the relationship between
size and efficiency of the Italian hospitality sector: a window DEA approach.
European Journal of Operational Research, 204(3), 613-20.
QNB (2013). Oman Economic Insight 2013, Qatar National Bank, available at:
http://www.qnb.com.qa/cs/Satellite?blobcol=urldata&blobheader=application%2Fpdf
&blobkey=id&blobtable=MungoBlobs&blobwhere=1355479823300&ssbinary=true
Ray, P., & Phillips, P. (2005). Performance comparisons of hotels in China. International
Journal of Hospitality Management, 24(1), 57-73.
Reynolds, D. (2003). Hospitality-productivity assessment using data envelopment analysis.
Cornell Hotel and Restaurant Administration Quarterly, 44(2), 130-37.
Shang, J., Wang, F., & Hung, W. (2010). A stochastic DEA study of hotel efficiency. Applied
Economics, 42(19), 2505-18.

Shirouyehzad, H., Hosseinzadeh Lotfi, F., Shahin, A., Aryanezhad, M.B., & Dabestani, R.
(2012). A DEA approach for comparative analysis of service quality dimensions with
a case study in hotel industry. International Journal of Services & Operations
Management, 12(3), 289–308.
Shyu, J., & Hung, S.C. (2012). The true managerial efficiency of international tourist hotels
in Taiwan: Three-stage data envelopment analysis. The Service Industries Journal,
32(12), 1991-2004.
Sigala, M., Jones, P., Lockwood, A., & Airey, D. (2005). Productivity in hotels: A stepwise
data envelopment analysis of hotels’ rooms division processes. The Service Industries
Journal, 25(1), 61–81.
Simar, L., & Wilson, P. (1998). Sensitivity analysis of efficiency scores: How to bootstrap in
nonparametric frontier models. Management Science, 44(1), 49-61.
Simar, L., & Wilson, P.W. (2007). Estimation and inference in two-stage, semi-parametric
models of production processes. Journal of Econometrics, 136(1), 31–64.
 Subramoniam, S., Al-Essai, S.A.N., Al-Marshadi, A.A.M., & Al-Kindi, A.M.A. (2010).
SWOT analysis on Oman tourism: a case study. Journal of Economics Development,
Management, IT, Finance and Marketing, 2(2), 1-22.
Sun, S., Lu, W.M. (2005). Evaluating the performance of the Taiwanese hotel industry using
a weight slacks-based measure. Asia Pacific Journal of Operational Research, 22,
487–512
Tarim, Ş., Dener, H.I., & Tarim, Ş.A. (2000). Efficiency Measurement in the hotel industry:
Output Factor Constrained DEA application. Anatolia: An International Journal of
Tourism & Hospitality Research, 11(2), 111-23.
Ting, C.T., & Huang, C.W. (2012). Measuring the effectiveness of mutual learning for
Taiwan's tourist hotels with the DEA approach. Cornell Hospitality Quarterly, 53(1),
65-74.
Tobin, J. (1958). Estimation of Relationships for Limited Dependent Variables.
Econometrica, 26(1), 24–36.
Tumer, N. (2010). Measuring hotel performance using data envelopment analysis. Anatolia:
An International Journal of Tourism & Hospitality Research, 21(2), 271-87.
Tundis, E., Corsino, M., & Zaninotto, E. (2012). Revealing the influence of managerial
practices and entrepreneurs’ characteristics on hotel efficiency. DISA Working Papers,
Trento University, Italy.
WEF (2013). The travel and tourism competitiveness report 2013. World Economic Forum.
Available at
http://www3.weforum.org/docs/WEF_TT_Competitiveness_Report_2013.pdf
Winckler, O. (2007). The Birth of Oman’s Tourism Industry. TOURISM - An International
Interdisciplinary Journal, 55(2), 221-34.
Wöber, K.W. (2007). Data Envelopment Analysis. Journal of Travel & Tourism Marketing,
21(4), 91-108.
WTTC (2013). Travel & tourism economic impact 2013, Oman. World Travel & Tourism
Council, available at:
http://www.wttc.org/site_media/uploads/downloads/world2013_1.pdf
Yang, C., & Lu, W.M. (2006). Performance benchmarking for Taiwan’s international tourist
hotels. INFOR, 44(3), 229–45.

Figure 1. Map of Oman

 Table 1. Summary statistics of Input and Output variables
Variables Unit Mean SD Min. Max.
Output
Annual revenue $ /year 6,911,989 13,731,079 3,004 78,795,452
Number of guests 17,864 20,273 597 96,877
Number of nights 23,882 28,091 669 147,084
Occupancy rate % 55 24 2 87
Inputs
Number of beds 135 138 23 937
Salary of employees $ /year 1,040,976 2,197,535 14,344 11,704,068

Table 2. Quantification of attraction categories

Region Nature Culture Activities
Muscat 16 16 13
Dhofar 19 8 13
Al Buraymi 5 5 13
A'Dakhiliyah 11 12 12
A'Sharqiyah 2 6 3
Al Batinah 1 3 0
Musandam 4 1 9

Table 3. Summary statistics of hotel contextual factors

Variables Unit Mean SD Min. Max.
Type of ownership categorical 0.33 0.47 0 1
Hotel size categorical 1.33 0.51 1 3
Star rating categorical 2.81 1.41 1 5
Attractions:
 Nature 12.36 6.37 1.00 19.00
 Culture 11.55 5.33 1.00 16.00
 Activities 10.53 4.81 0.00 13.00
 Table 4. Point and Interval efficiency scores using DEA bootstrap
Table 5. CRS VRS
Hotel Lower Upper Lower Upper
h ˆh bound bound
h ˆh bound bound
H1 1.465 1.731 1.502 2.032 1.225 1.343 1.232 1.578
H2 1.844 2.343 1.892 2.819 1.098 1.189 1.103 1.363
H3 1.000 1.210 1.034 1.370 1.000 1.181 1.006 1.473
H4 1.000 1.305 1.030 1.552 1.000 1.185 1.006 1.515
H5 1.809 2.203 1.862 2.581 1.000 1.182 1.007 1.506
H6 1.396 1.630 1.431 1.823 1.099 1.181 1.105 1.336
H7 1.362 1.603 1.398 1.809 1.000 1.119 1.006 1.247
H8 1.049 1.233 1.077 1.377 1.000 1.156 1.006 1.355
H9 1.376 1.590 1.422 1.773 1.000 1.105 1.006 1.224
H10 2.059 2.356 2.114 2.635 1.434 1.531 1.444 1.623
H11 1.218 1.402 1.249 1.555 1.000 1.114 1.006 1.243
H12 1.656 1.874 1.697 2.087 1.060 1.133 1.067 1.226
H13 2.548 2.903 2.632 3.199 1.413 1.473 1.423 1.533
H14 1.000 1.327 1.027 1.543 1.000 1.196 1.006 1.609
H15 1.000 1.194 1.025 1.322 1.000 1.130 1.006 1.266
H16 1.000 1.362 1.030 1.563 1.000 1.196 1.006 1.568
H17 1.755 2.098 1.805 2.370 1.097 1.159 1.105 1.233
H18 2.359 2.796 2.434 3.145 1.000 1.056 1.005 1.103
H19 2.523 2.884 2.590 3.198 1.120 1.170 1.127 1.220
H20 2.079 2.440 2.129 2.807 1.618 1.764 1.627 1.950
H21 2.311 2.679 2.380 2.937 1.416 1.504 1.424 1.574
H22 2.655 3.194 2.740 3.674 1.252 1.306 1.260 1.374
H23 1.585 1.920 1.628 2.136 1.264 1.354 1.272 1.468
H24 2.127 2.588 2.179 2.929 1.000 1.094 1.008 1.163
H25 2.096 2.546 2.160 3.016 1.121 1.190 1.125 1.292
H26 1.016 1.168 1.042 1.312 1.000 1.105 1.004 1.251
H27 5.165 6.335 5.303 7.244 1.919 2.121 1.931 2.502
H28 1.000 1.294 1.034 1.531 1.000 1.199 1.006 1.687
H29 1.150 1.397 1.178 1.612 1.019 1.109 1.025 1.246
H30 1.111 1.327 1.128 1.542 1.000 1.114 1.007 1.294
H31 1.000 1.229 1.029 1.452 1.000 1.188 1.006 1.583
H32 1.004 1.136 1.022 1.267 1.000 1.158 1.006 1.352
H33 1.916 2.193 1.967 2.420 1.148 1.236 1.155 1.362
H34 4.602 5.483 4.732 6.131 1.637 1.725 1.647 1.804
H35 1.465 1.710 1.495 1.972 1.214 1.318 1.221 1.515
H36 9.459 11.088 9.613 12.742 8.542 9.224 8.610 10.051
H37 2.130 2.468 2.149 2.893 1.000 1.206 1.006 1.882
H38 2.253 2.668 2.317 3.035 1.203 1.330 1.210 1.619
H39 1.387 1.628 1.429 1.787 1.033 1.093 1.040 1.148
H40 1.000 1.288 1.027 1.470 1.000 1.177 1.006 1.477
H41 1.052 1.254 1.068 1.480 1.000 1.179 1.007 1.439
H42 1.931 2.209 1.976 2.461 1.142 1.238 1.148 1.384
H43 1.581 1.826 1.617 2.011 1.221 1.302 1.230 1.378
H44 2.000 2.387 2.040 2.786 1.545 1.683 1.555 1.953
H45 2.217 2.796 2.256 3.340 1.906 2.093 1.915 2.559
H46 9.351 11.264 9.499 13.278 8.940 9.815 8.992 11.555
H47 2.022 2.328 2.075 2.587 1.547 1.670 1.557 1.827
H48 3.357 4.073 3.460 4.703 2.165 2.359 2.176 2.621
H49 2.749 3.258 2.812 3.789 2.613 2.857 2.625 3.285
H50 3.062 3.660 3.131 4.031 1.747 1.841 1.757 1.963
H51 1.365 1.602 1.403 1.775 1.280 1.401 1.289 1.547
H52 1.217 1.416 1.253 1.567 1.000 1.067 1.006 1.128
H53 1.657 1.894 1.709 2.080 1.033 1.093 1.040 1.140
H54 1.649 1.864 1.661 2.192 1.632 1.762 1.640 1.985
H55 4.611 5.346 4.686 6.209 4.598 4.998 4.620 5.653
H56 1.185 1.398 1.204 1.638 1.024 1.134 1.029 1.386
H57 6.503 7.815 6.615 9.221 1.000 1.205 1.006 1.884
H58 1.294 1.457 1.335 1.615 1.151 1.232 1.158 1.333
 Reference sets of the inefficient hotels (VRS)
Hotel Reference set
H1 H3-H4
H2 H4-H5-H8
H6 H3-H8-H15
H10 H9-H11-H16-H24-H28
H12 H8-H15-H32
H13 H4-H9-H15-H18
H17 H3-H24-H28-H31-H32
H19 H15-H18-H31-H52
H20 H16-H24-H28-H40
H21 H8-H9-H11-H24-H28-H32
H22 H15-H18-H31
H23 H3-H9-H16-H24-H28-H31
H25 H24-H28-H31
H27 H16-H28-H37-H57
H29 H24-H28-H40
H33 H8-H9-H15
H34 H15-H18-H24
H35 H26-H30-H40
H36 H25-H28-H30
H38 H16-H37-H41
H39 H15-H24-H28-H31
H42 H3-H24-H32-H40
H43 H9-H16-H24-H28-H32-H40
H44 H16-H28-H40
H45 H24-H28-H40
H46 H28-H37-H41
H47 H3-H16-H24-H32-H40
H48 H3-H16-H24-H40
H49 H16-H28-H40
H50 H8-H9-H15
H51 H3-H8-H15-H32
H53 H4-H9-H15-H18-H24
H54 H28-H41
H55 H-H28-H30
H56 H28-H37-H41
H58 H3-H15-H24-H31-H32

Table 6. Average required expansion of the outputs per inefficient hotel (VRS)
Number of Required
Output Surplus
hotels expansion (%)
Annual revenue ($) 13 140625.66 1.79
Number of guests 27 4228.54 1.63
Number of nights 21 4967.26 1.55
Occupancy rate (%) 8 29.03 25.77
 Table 7. Interval width based ranking of the benchmark hotels
Interval Efficiency
Rank Hotel Location
width variance
1 H18 Safeer Continental Muscat 0.098 0.001
2 H52 Sohar Beach Al-Batinah 0.122 0.001
3 H24 Safeer Hotel Suites Muscat 0.156 0.002
4 H9 Radisson SAS Muscat 0.218 0.023
5 H11 Majan Continental Muscat 0.237 0.004
6 H7 Holiday Muscat 0.241 0.005
7 H26 Star hotel apartments Muscat 0.247 0.004
8 H15 Ruwi Muscat 0.260 0.006
9 H30 Corniche Muscat 0.287 0.005
10 H32 Hilton Salalah 0.346 0.012
11 H8 Golden Tulip Seeb Muscat 0.348 0.011
12 H41 Al Jawhara Al Buraymi 0.432 0.018
13 H3 Grand Hyatt Muscat 0.466 0.023
14 H40 Al Massa Al Buraymi 0.472 0.023
15 H5 Barr Al Jissah Resort Muscat 0.500 0.023
16 H4 The Chedi Muscat 0.509 0.025
17 H16 Coral Muscat 0.562 0.029
18 H31 Waffa Hotel Apartment Muscat 0.577 0.029
19 H14 Ramee Dream Resort Muscat 0.603 0.030
20 H28 Al-Hail Muscat 0.681 0.034
21 H37 Al Nasr Salalah 0.876 0.044
22 H57 Al-Shumukh guest house Al-Batinah 0.878 0.043

Table 8. Results of the Truncated double bootstrapped regression analysis

CRS VRS
Factor Lower Upper Lower Upper
β̂** SE
bound bound β̂** SE
bound bound
Intercept 5.672 1.190 1.840 6.547 4.313 0.938 0.140 3.871
Type of
0.077 0.867 -1.571 1.847 -0.131 0.651 -1.990 0.578
ownership
Hotel size 0.312 0.939 -0.986 2.680 0.291 0.720 -0.357 2.474
Star rating -0.394* 0.377 -1.951 -0.456 -0.285* 0.284 -1.541 -0.460
Nature 0.102* 0.117 0.075 0.537 0.050 0.097 -0.026 0.359
Culture -0.081* 0.101 -0.438 -0.035 -0.065* 0.082 -0.383 -0.058
Activities -0.089 0.139 -0.538 0.009 -0.013 0.117 -0.274 0.186