A study of expected and perceived service quality in Croatian and Slovenian hotel industry

The main purpose of present research is the comparison of guests’ expectations and perceptions of service quality in Croatian and Slovenian hotels. In order to measure hotel guests’ expectations and perceptions modified SERVQUAL model was developed. The questionnaire was designed in accordance with researches conducted by Parasuraman et al. (1988), Zeithaml et al. (1990), Snoj and Ogorelc (1998), Pizam and Ellis (1999) and Marković (2003). The questionnaire was divided in three parts: measurement of expected service quality, measurement of perceived service quality and demographic questions. Data were collected in hotel settings in Opatija Riviera (Croatia) and Slovenia’s coastal region. A total of 253 (Croatian sample) and 172 (Slovenian sample) completed and usable questionnaires were gathered. Data were analyzed using descriptive statistics, independent samples t-test, exploratory factor analysis and reliability analysis. The conducted statistical analysis revealed similarities in both surveys. The research results identified high level of guests’ expectations and perceptions regarding the hotel service quality, and importance of intangible aspects of hotel service. A study of expected and perceived service quality in Croatian and Slovenian hotel industry. European Journal of Tourism Research 6(1), 36-52.


Introduction
Providing high quality services and maintaining customers' satisfaction are important factors that increase competitiveness and lead to the business success (Anderson et al., 1997;Parasuraman, 2002). Thus, understanding hotel guests and heaving in mind the importance of hotel service attributes are important criterions for gaining competitive advantages in tourism and hospitality marketplace (Baloglu et al., 2003;Tsaur and Lin, 2004).
A literature review indicates that studies related to hotel service quality in South Central Europe are rather limited (Snoj and Mumel, 2002;Marković, 2003;Marković and Raspor, 2010). What is more, existing studies have not compared expected and perceived service quality in Croatian and Slovenian hotel industry in such an extent as it was done in present research. Therefore, the present research aims to contribute to the existing literature in this field.
The focus point of this research was guests' expectations and perceptions of service quality in Croatian and Slovenian hotel industry. The specific aims were to assess the level of guests' expectations and perceptions, to identify main factors of expected and perceived service quality and to determine potential differences between Croatian and Slovenian hotels.
The paper is organized in two parts. First, a brief conceptual background of main concepts of interest is provided. In the second part, results of empirical research are presented, followed by discussion and conclusion.

Literature review
In order to set the research framework, in following sections the basic concepts of present research are explained.

Hotel guests' expectations and perceptions
A review of literature suggests that customer expectations have been consistently acknowledged as the basis on which service quality and customer satisfaction judgments are formed (Parasuraman et al., 1988;Oliver, 1997). Thus, to improve the service quality in hotel industry, it is important to understand hotel guests' expectations.
Hotel service providers should be aware of the fact that, their customers may use multiple types of expectations when evaluating hotel service. There are two most often referred to types of expectations, namely predictive expectations and normative expectations (Coye, 2004;Tam, 2005). Predictive expectations are generally defined as customer's beliefs about the level of service that a service provider would be likely to offer (Coye, 2004;Tam, 2005). These expectations represent customer's predictions about what is likely to happen and are frequently used as a standard of reference against which satisfaction judgments are made (Churchill and Surprenant, 1982). Normative expectations are generally conceptualized as customer's ideal level of service, i.e. what customers believe a service provider should offer (Coye, 2004;Tam, 2005). This type of expectations has been used as the standard against which customers evaluate service quality (Zeithaml et al., 1993). While seemingly distinct, these two perspectives of expectations are often used interchangeably.
Furthermore, Teas (1993a, b) believed that expectations can have six, somewhat different interpretations: service attribute importance (expectations represent levels of importance), forecasted performance (expectations represent predictions of performance customers would expect), ideal performance (expectations represent optimal performance), deserved performance (expectations represent the level of performance that "should be" delivered), equitable performance (expectations represent the level of performance in comparison with the costs) and minimum tolerable performance (expectations represent "must be" performance).
Customers form their expectations based on the provider's promotional activities, previous experience, word-of-mouth, first impression (Zeithaml and Bitner, 2003). Culture (e.g. customer's cultural values) is also one of the factors that determine customer's expectations in the service sector (Kueh and Voon, 2007). In addition, Fernandez- Barcala et al. (2009) argue that hotel guest's expectations are influenced by price, hotel category, promised services and the location. Hence, the expectations can be created by the factors that either can or can not be controlled by the service provider.
In accordance with Parasuraman et al. (1988) work, in the present research the term "expectations" is defined as customers' desires or wants. It is used to describe what hotel guests believe about the capability of the hotel service provider. Specifically, expectations represent what guests feel a hotel should offer. As such, expectations of hotel guests in this study are conceptualized as normative expectations.
On the other hand, according to Zeithaml et al. (1990), perceived service quality is the extent to which a firm successfully serves the purpose of customers. Rowley (1998) argued that perceived service quality is an attitude related to, but not the same, as satisfaction.
Customers determine the perceived or cognitive value of service based on their experience with delivered service. Ghobadian et al. (1994) stated that customers' expectations, service delivery process and service outcome have an impact on perceived service quality. Yoo and Park (2007) found out that employees, as an integral part of a service process, are a critical element in enhancing perceived service quality. Furthermore, Edvardsson (2005) pointed out that service quality perceptions are formed during the production, delivery and consumption process. The author concluded that customers' favourable and unfavourable experience, as well as their positive and negative emotions may have an important impact on perceived service quality. Similarly, O'Neill and Palmer (2003) have reported that customers' perceptions of service quality may, to a large extent, be influenced by the degree of their prior experience with a particular service.
In tourism context, perceived service quality has been viewed mostly as the quality of the opportunities available at a destination and is considered to be related to a tourist's quality of experience (Crompton and Love, 1995).
In present research perceived service quality is defined as customer's subjective experience. It represents the assessment of actually delivered service. Specifically, hotel guests' perceptions represent guests' attitude and evaluation of hotel service performance.

Measurement of expected and perceived hotel service quality
A major contribution to service quality conceptualization and measurement have made Parasuraman et al. (1985;1988) who proposed that service quality is a fivedimensional concept and results from comparison of customers' expectations and perceptions of service performance. Based on these arguments they developed a model for measuring service quality, called SERVQUAL model. During the years, the model has become the most widely adapted and tested conceptualization of service quality (Ladhari 2008;2009).
The SERVQUAL model uses 22 statements to measure both expectations and perceptions. The statements cover five service quality dimensions, namely reliability, responsiveness, tangibles, assurance and empathy. The level of agreement with the given statement is assessed on the Likert-type scale.
The applicability of SERVQUAL methodology in the hotel industry has been demonstrated by several authors. Fick and Ritchie (1991) measured hotel service quality using original SERVQUAL model. Knutson et al. (1990;1992) developed LODGSERV, a model utilized to measure expected service quality in the lodging industry. The model was based on five original SREVQUAL dimensions and contained 26 items on various aspects of a hotel experience. Getty and Thompson (1994) introduced another specific model for hotel settings, called LODGQUAL. Heung and Wong (1997) used LODGSERV model to measure guests' expectations in Hong Kong hotels. Mei et al. (1999) (Jin et al., 2008), China (Law and Yip, 2010), Poland (Goranczewski andPuciato, 2011) andGermany (Eckert et al., 2011) In addition, several authors measured perceived service quality regardless on SERVQUAL methodology. Choi and Chu (2001) measured travellers' perceptions of hotel attributes in Hong Kong's hotels, Juwaheer (2004) in hotels of Mauritius, Poon and Lock-Teng Low (2005) in Malaysian hotels. In addition, importance-performance analysis was performed to assess customer perceptions of hotel service quality in New Zeland (Mohsin, 2007) and in India (Mohsin and Lockyer, 2010).

Research questions, objectives and hypotheses
The research was conducted in two phases. First, a qualitative review of previous research findings regarding expected and perceived hotel service quality was performed. These results are provided in conceptual background section. In the second phase, the quantitative approach was taken to answer following research questions: 1. What is the difference between Croatian and Slovenian hotels regarding guests' expectations? 2. What is the difference between Croatian and Slovenian hotels regarding guests' perceptions? 3. What are the differences in expected service quality factors between Croatian and Slovenian hotels? 4. What are the differences in perceived service quality factors between Croatian and Slovenian hotels?
Specifically, the research aimed to (a) empirically investigate hotel guests' expectations and perceptions, (b) identify expected and perceived service quality factors, and (c) determine potential differences between Croatian and Slovenian hotels.
In order to meet study's objectives and to answer research questions, following hypotheses were proposed: H1: There is no statistically significant difference between Croatian and Slovenian hotels regarding guests' expectations.
H2: There is no statistically significant difference between Croatian and Slovenian hotels regarding guests' perceptions. H3: The factor structure of expected service quality is similar between Croatian and Slovenian hotels. H4: The factor structure of perceived service quality is similar between Croatian and Slovenian hotels.

Research instrument
The empirical research was conducted using primary data. The instrument for this study was on-site and self-administered questionnaire.
As a foundation for questionnaire development, the SERVQUAL model was used. By reviewing previous studies in context of examining hotel service quality, the questionnaire used in this study is adapted to the specific features of hotel service and is based on researches conducted by Parasuraman et al. (1988), Zeithaml et al. (1990, Snoj and Ogorelc (1998), Pizam and Ellis (1999), Marković (2003).
The model modification included following procedures. The original items were slightly modified to suit the hospitality setting. For example, instead of "XYZ Company has modern-looking equipment", the statement was modified to "Hotel has modern-looking equipment". The original item "Guests feel safe in their transactions with employees" was replaced by the item "Guests feel safe and secure in their stay". The reason for this change is the confusing meaning of the word "transactions" and the fact that safety and security are regarded as an important factor in a hotel stay. Moreover, in order to measure attributes specific to hotel environment, following items were added: "parking area" (Pizam and Ellis, 1999), "appropriate location", "available and clear information", "variety of facilities" (Snoj and Ogorelc, 1998), "clean and tidy hotel", "feeling safe and secure", "ease of finding a way around the hotel" and "typical service quality for hotel category" (Marković, 2003). All the statements in the questionnaire were positively worded. Finally, the modification resulted in the deletion of one original SERVQUAL item and the inclusion of eight new items, leaving a total of 29 hotel attributes. These attributes represented seven dimensions: five original SERVQUAL dimensions (tangibles, reliability, responsiveness, assurance, empathy) and two new dimensions, named as accessibility and output quality.
The questionnaire consisted of three parts. First two parts examined respondents' expectations and perceptions of hotel service quality. These were measured on the basis of 29 hotel attributes. A 7-point Likert-type scale was adopted to assess respondents' ratings (1 = "strongly disagree", 7 = "strongly agree"). The third part consisted of demographic questions. This included country of residence, age, gender, purpose of visit, duration of staying at a hotel, and level of education.
The research instrument was prepared in the Croatian language and was additionally translated into the Slovene, English, Italian and German language to capture both domestic and international hotel guests.

Sampling procedure
The research was carried out in Croatia and Slovenia, resulting with two samples. One sample was taken from the hotel settings in the Opatija Riviera (Croatia), while the second sample was taken from the hotel settings in Slovenia's coastal region. To make the research more representative, the samples included hotels of different sizes and categories. Before the data collection started, hotel managers were contacted for permission to take part in the study. Thus, the questionnaires were administered only in those settings whose managers agreed to participate.
The reception desk employees helped to distribute and collect the survey sheets from the participating customers. Participation was voluntary. Thus, the data were collected using a convenience sampling approach. Questionnaires were distributed to guests who were willing to participate in the research at the reception desk. Specifically, 960 questionnaires were distributed in the Croatian survey and 770 questionnaires in the Slovenian survey. Of 265 returned questionnaires in the Croatian survey, 12 were not included in the analysis because of incompleteness. In the Slovenian survey 177 questionnaires were collected and 5 of them were incomplete. Therefore, data collection resulted with 253 usable questionnaires in Croatian sample (26.35% response rate) and with 172 usable questionnaires in Slovenian sample (22.34% response rate).

Data analysis
The collected data were analyzed using statistical package SPSS. Data analysis included descriptive statistics, independent samples t-test, exploratory factor analysis and reliability analysis.
Descriptive statistics was used to examine demographic profiles of the respondents and to evaluate service quality expectations and perceptions of hotel guests. The independent samples t-test was performed to determine the significance of differences between perceived and expected scores of service quality in studies conducted in Croatia and Slovenia. At this stage, first two hypotheses were tested. Exploratory factor analysis was employed to derive factors from hotel service attributes for expectation and perception scale. This method was used to test third and fourth hypotheses. To test the reliability of the scales and to assess the inner consistency of each extracted factor, the reliability analysis was conducted.
This study adopted principal component analysis with varimax rotation as the method for identifying service quality factors in the hotel service. In order to adequately apply this technique, several conditions should be respected. First, Kaiser-Meyer-Olkin's measure (KMO) should be greater than 0.7, and is inadequate if it is less than 0.5 (Stewart, 1981). Further, Bartlett's sphericity test should be significant (i.e. a significance value should be less than 0.05) (Leech et al., 2005). Finally, items with eigenvalues equal or greater than 1, factor loadings above 0.4, and factors, which contain at least three items, were retained (Hair et al., 2006).
To test the reliability, Cronbach's alpha coefficients were calculated. Coefficients higher than 0.6 were considered acceptable, indicating reasonable internal consistency and reliability (Hair et al., 2006).

Respondents' profile
Descriptive statistical analysis was run on respondents' demographic variables. The results are shown in Table 1.
As shown in Table 1, in both, Croatian and Slovenian sample male respondents (51.8% and 52.9%, respectively) outnumbered female hotel guests. In terms of age distribution, in both samples the majority of the respondents were between 36 and 45 years old. The differences can be seen in age group between 16 and 35 years, indicating that Slovenian sample had more young respondents than Croatian sample. On the other hand, in the Croatian sample, respondents between 56 and 65 years outnumbered those from Slovenian sample. Furthermore, in Croatian sample, a majority of the respondents (more than 83%) were foreign visitors, while in Slovenian sample about 54% of the respondents were domestic visitors. In terms of education level, nearly 61% of the respondents in Croatian sample reported they had higher or university education. In the Slovenian sample, almost 70% of the respondents reported the same level of education. In both surveys the majority of the respondents indicated that the main purpose of their hotel stay was vacation and about half of them stayed at the hotel between four and seven days.

Descriptive and bivariate analyses
The results of descriptive and bivariate analyses are presented next. Table 2 reports the results for the respondents' expectations and perceptions of hotel service quality, as well as significance of difference in the mean scores between Croatian and Slovenian hotels.
As noted in Table 2, the mean scores of guests' expectations in Croatian hotels ranged from 5.92 to 6.69. The lowest expectation items were "offering variety of facilities" and "modernlooking equipment". On the other hand, the highest expectations considered "clean and tidy hotel", followed by "feeling safe and secure", "courteous hotel staff" and "neat hotel staff". The overall mean score for service quality  Note: a Expectations and perceptions means range from 1 to 7; b study conducted in Croatia (N=253); c study conducted in Slovenia (N=172); values in parentheses are standard deviations; * p < 0.01; ** p < 0.05. expectation items was 6.36. This score indicates high expectations of hotel guests regarding the service quality in Croatian hotels.
Similarly, in study conducted in Slovenian hotels, the mean scores of hotel guests' expectations ranged from 5.58 to 6.64. The lowest expectations were related to the attributes "hotel staff provides personal attention" and "modern-looking equipment", while the highest expectations were related to the items "clean and tidy hotel", "neat hotel staff" and "interest in solving guests' problems". The overall mean score for expectations attributes in Slovenian sample was 6.17, indicating high service quality expectations.
Furthermore, the analysis of hotel guests' perceptions indicated following results. The mean scores of guests' perceptions in the Croatian survey ranged from 4.77 to 6.34. The lowest perception item was "offering a variety of facilities", which indicates that hotels did not provide enough suitable facilities that could enhance hotel quality. On the other hand, hotel guests' highest perceptions were regarding the "ease of finding a way around the hotel". Furthermore, guests highly assessed the following hotel attributes: "feeling safe and secure", "willingness for helping guests" and "courteous hotel staff". These indicate that a hotels' staff has one of the crucial roles in performing high service quality. The overall mean score for service quality perceptions items in Croatian sample was 5.92. This score indicates rather high perceptions of hotel guests regarding service quality.
In the Slovenian survey mean scores for guests' perceptions ranged from 5.71 to 6.35. The lowest perceptions were related to the attribute "parking area", suggesting that hotel guests' were not pleased with parking possibilities. On the other hand, the highest perceptions were related with items "feeling safe and secure", "willingness for helping guests", "courteous hotel staff ", and "typical service quality for hotel category". The overall mean score for perceptions attributes in Slovenian sample was 6.09, indicating high service quality perceptions.
Moreover, Table 2 presents dimensions' mean scores, as well. The most important expectations dimension in both surveys appears to be assurance. In the Croatian survey this is followed by dimensions responsiveness, reliability, accessibility, tangibles, empathy, and output quality. Similarly, in the Slovenian survey dimension assurance is followed by reliability, responsiveness, accessibility, output quality, empathy and tangibles.
When analyzing importance of dimensions for perceptions scale, the result indicate that in the Croatian survey the most important perception dimension was accessibility, followed by assurance, responsiveness, reliability, empathy, tangibles, and output quality. On the other hand, the highest mean score in the Slovenian survey was given to dimension output quality, followed by assurance, responsiveness, accessibility, reliability, tangibles, and empathy.
The analysis of difference in service quality expectations between Croatian and Slovenian hotels indicated higher expectations score in the Croatian sample than in the Slovenian sample. Only for one item ("available and clear information in a hotel") the scores were higher in the Slovenian sample.
The results of independent samples t-test for expectations scale show that in 15 out of 29 hotel attributes significant differences were found between Croatian and Slovenian surveys. For these hotel attributes guests staying in Croatian hotels had significantly higher service quality expectations than those staying in Slovenian hotels.
Furthermore, the analysis of difference in service quality perceptions between Croatian and Slovenian hotels resulted with higher perceptions scores in the Slovenian sample than in the Croatian sample for the majority of hotel attributes. Only attributes "appropriate location", "performing services right the first time", "hotel staff has time to answer guests' questions", "hotel staff provides personal attention", and "ease of finding a way around the hotel" were assessed better for Croatian hotels.
However, the results of independent samples ttest for perceptions scale indicate that in 11 out of 29 hotel attributes, significant differences were found between Croatian and Slovenian hotels. These attributes were "modern-looking equipment", "visually appealing physical facilities", "visually appealing materials (pamphlets, web-sites)", "clean and tidy hotel", "appropriate location", "parking area", "errorfree service", "hotel staff has knowledge to answer questions", "ease of finding a way around the hotel", "offering variety of facilities", and "typical service quality for hotel category". Majority of these attributes were assessed significantly better for Slovenian hotels. Only for attributes "appropriate location" and "ease of finding a way around the hotel" guests staying in Croatian hotels had significantly higher service quality perceptions than those staying in Slovenian hotels.
However, in the majority of hotel attributes there are no significant differences in perception scores between Croatian and Slovenian surveys, which implies that guests in Croatian and Slovenian hotels have similar perceptions of service quality for majority of hotel attributes.

Factor and reliability analyses
The objective of factor analysis was to identify the main factors of expected and perceived hotel service quality, using the principal component method with varimax rotation. The analysis was performed on both, the Croatian and the Slovenian sample. The results are presented next. Table 3, KMO values for both surveys are high, indicating sufficient items for each extracted factor. The Bartlett's Test is significant (p < 0.05) meaning that there are strong correlations between the items in each factor. Hence, it is justified to conduct exploratory factor analysis for expectations scale.

As shown in
The results of factor and reliability analyses of hotel guests' expectations in Croatian and Slovenian surveys are presented in Table 4.
In the Croatian survey exploratory factor analysis extracted four factors that explained 65.8% of the total variance in the data. Most of the factor loadings were greater than 0.6, meaning that correlation of the items with the factors on which they were loaded is reasonably high. After examining the item descriptions, the four factors that represent main dimensions of expected service quality in Croatian hotels were interpreted as follows.
Factor 1 -"reliability" contains eleven items and indicates employees' politeness and willingness for helping, as well as the safe and confident service performed at promised time. Factor 2 -"empathy" is loaded with eight items and refers to providing personal attention to customers and employees' readiness to answer questions and provide error-free service. Factor 3 -"tangibles" includes five items reflecting appropriate hotel location, appealing appearance of facilities, equipment and employees, as well as provision of parking area. Factor 4 -"accessibility" contains five items that indicate ease of communication (e.g. available information, appropriate promotional materials) and finding the way around the hotel.
The results of the reliability analysis showed that Cronbach's alpha coefficients of the extracted factors ranged from 0.819 to 0.936. That is well above the minimum value of 0.60, which is considered acceptable as an indication of scale reliability (Hair et al, 2006). Thus, these values suggest good internal consistency of the factors. Finally, Cronbach's alpha value for the overall expectation scale in the Croatian survey is 0.961 and indicates its high reliability.
On the other hand, seven factors were derived from the factor analysis in the Slovenian survey. The extracted factors explained 66.1% of the total variance in the data. Most of the factor loadings were greater than 0.6. However, two factors (F5 and F7) contained only two items each, and could not be considered as factors. Thus, final solution retained five factors that represent main dimensions of expected service quality in Slovenian hotels. These factors were interpreted as follows: Factor 1 -"assurance and responsiveness" contains eleven items and refers to employees' readiness for helping. Factor 2 -"reliability" includes five items reflecting consistent, prompt, and error-free service. Factor 3 -"empathy" gathered three items and indicates understanding guests' interests and meeting their needs. Factor 4 -"individual attention" is loaded with three items that reflect providing personalized treatment and convenience to hotel guests.
Factor 6 -"tangibles" contains three items reflecting appealing appearance of facilities and communication materials, as well as the ease of finding the way around the hotel.
The results of the reliability analysis showed that Cronbach's alpha coefficients of the extracted factors ranged from 0.541 to 0.910. These values suggest good internal consistency of all factors, except one. Namely, Factor 6 had Cronbach's alpha coefficient lower than minimum acceptable value of 0.60, indicating low reliability of the factor. However, Cronbach's alpha value for the overall expectations scale in the Slovenian survey is 0.947 and indicates its high reliability.

When
analyzing factor structure for expectations scale in both studies, it can be noted that some factors ("reliability", "empathy", and "tangibles") are common for both sample groups, while others overlap. Therefore, findings reveal certain similarities.
Next, the results of factor and reliability analyses of hotel guests' perceptions in Croatian and Slovenian survey are provided. Table 5 indicates that KMO values for both surveys are high, indicating sufficient items for each extracted factor. The Bartlett's Test is significant (p < 0.05) meaning that there are strong correlations between the items in each factor. Hence, it is justified to conduct exploratory factor analysis for perceptions scale. Table 6 shows that five factors were extracted in the Croatian survey, which accounted for 65.1% of total variance in the data. Most of the factor loadings were greater than 0.6, indicating that correlation of the items with the factors on which they were loaded is reasonably high.
One factor (F5) contained only two items and could not be considered as factor. Thus, the final solution retained four factors that represent main dimensions of perceived service quality in Croatian hotels. The four remaining factors are labelled as follows: Factor 1 -"reliability" contains nine items and indicates solving guests' problems and performing error-free service at promised time. Factor 2 -"empathy and competence of staff" is loaded with seven items that refer to staff knowledge and ability to provide individual attention. Factor 3 -"accessibility" gathered eight items reflecting appropriate location of the hotel and ease of communication and finding the way around the hotel. Factor 4 -"tangibles" includes three items referring to appearance of the facilities, equipment and communication materials.
The results of the reliability analysis showed that Cronbach's alpha coefficients of the extracted factors ranged from 0.785 to 0.917. That is well above the minimum value of 0.60, which is considered acceptable as an indication of scale reliability (Hair et al., 2006). Thus, these values suggest good internal consistency of the factors. Finally, Cronbach's alpha value for the overall perceptions scale is 0.953 and indicates its high reliability.
Regarding the Slovenian survey, exploratory factor analysis derived five factors that explained 63.2% of total variance in the data. These factors represent main dimensions of perceived service quality in Slovenian hotels and are labelled as follows: Factor 1 -"reliability and responsiveness" contains eight items indicating service timeliness, promptness and accuracy. Factor 2 -"tangibles" is loaded with six items that refer to the appearance of the facilities, equipment, communication materials and output quality. Factor 3 -"empathy and accessibility" gathered seven items that indicate providing personal attention, information availability, and finding the way around the hotel. Factor 4 -"assurance" includes five items referring to employees' courtesy, knowledge, instilling confidence, as well as the safety of hotel guests.
Factor 5 -"surrounding" contains three items reflecting hotel location and provision of parking area.
The results of the reliability analysis showed that Cronbach's alpha coefficients of the extracted factors ranged from 0.617 to 0.907. These values suggest good internal consistency of the factors. Finally, Cronbach's alpha value for the overall perceptions scale in the Slovenian survey is 0.947 and indicates its high reliability.
Comparing the factor structure for perceptions scale in both studies, it can be seen that one factor ("tangibles") is common for both samples, while others overlap. This reveals similarities to a certain extent.

Discussion and conclusion
The results of descriptive analysis suggested that hotel guests have high overall service quality expectations in both, the Croatian and the Slovenian survey (6.36 and 6.17, respectively). In the Croatian sample, the highest expectations considered "clean and tidy hotel", "feeling safe and secure", "courteous hotel staff", and "neat hotel staff". Similarly, in the Slovenian sample, the highest expectations were related to the "clean and tidy hotel", "neat hotel staff", and "interest in solving guests' problems". Thus, hotel guests expect clean facilities, safety, as well as polite and neat employees that are willing to help. It can be seen that hotel employees have an important role in meeting guests' expectations and providing service of a high quality. Thus, employees should have appropriate skills to provide expected service.
In addition, the comparison of respondents' expectations scores in the Croatian and the Slovenian survey indicated higher expectations in the Croatian sample. The overall mean score for service quality expectations items in the Croatian and the Slovenian study were 6.36 and 6.17, respectively. The independent samples t-test revealed that these differences were statistically significant in 15 out of 29 hotel attributes. Thus, hypothesis H1 can partially be accepted, implying that for about one half of the examined attributes guests staying in Croatian hotels had significantly higher expectations than those staying in Slovenian hotels.
On the other hand, the findings of descriptive analysis for hotel guests' perceptions indicated rather highly perceived service quality (overall mean scores for the Croatian and the Slovenian survey were 5.92 and 6.09, respectively). The highest perceptions in both surveys were related to items "feeling safe and secure", "willingness for helping guests", and "courteous hotel staff". These results imply that Croatian and Slovenian hotels provide safety and security, as well as polite employees that are willing to help their guests.
The comparison of respondents' perceptions scores resulted with higher perceptions in the Slovenian sample for majority hotel attributes. The overall mean score for service quality perception items in the Croatian study was 5.92, and in the Slovenian study 6.09. The results of independent samples t-test revealed that these differences were statistically significant in 11 out of 29 hotel attributes and majority of these attributes were assessed significantly better for Slovenian hotels. However, since for the majority of hotel attributes there are no significant differences in perceptions scores between Croatian and Slovenian samples, these findings confirm hypothesis H2.
Moreover, the findings of exploratory factor analysis for expectations scale identified slightly different factor structures between the two samples. Four factors were generated for the Croatian survey and five factors for the Slovenian survey. Three factors of service quality expectations are common for both samples, namely "reliability", "empathy", and "tangibles". In addition to these factors, in the Croatian survey factor structure is completed with factor "accessibility", and in the Slovenian survey with factors "assurance and responsiveness" and "individual attention". Thus, hypothesis H3 is also accepted.
These findings are somewhat consistent with similar studies conducted in hotel industry. The most important expectations factor in the Croatian study was "reliability". This is in accordance with previous researches (Fick and Ritchie, 1991;Knutson et al., 1992;Heung and Wong, 1997;Marković, 2003). On the other hand, in the Slovenian study, the most important expectations factor was "assurance and responsiveness", followed by "reliability". This is in accordance with study conducted by Douglas and Connor (2003), were factor "assurance" was the most important expectations factor, followed by "reliability".
According to these results, hotel guests' expectations can be explained with two main factors, "reliability" and "assurance". In the hotel industry this implies that guests usually expect reliable and error-free service, as well as polite and competent hotel staff. What is more, these results suggest that hotel guests' expectations are mostly determined by the intangible aspects of hotel service experience. This finding could be very useful for hospitality practitioners, who might believe that guests' expectations and needs can only be achieved through tangible elements, such as visual attractiveness of the facility.
The exploratory factor analysis for perceptions scale extracted following four factors in the Croatian study: "reliability", "empathy and competence of staff", "accessibility", and "tangibles". In the Slovenian survey five factors were derived, labelled as "reliability and responsiveness", "tangibles", "empathy and accessibility", "assurance", and "surrounding". It is evident that extracted factors in both studies overlap, confirming the hypothesis H4.
However, previous studies conducted in the hotel sector identified different outcomes with regard to the number and interpretation of factors guests use to assess perceived hotel service quality. Akan (1995) reported sevenfactor structure, labelled as "courtesy and competence of the personnel", "communications and transactions", "tangibles", "knowing and understanding the customer", "accuracy and speed of service", "solutions to problems" and "accuracy of hotel reservations". Mei et al. (1999) identified "employees", "tangibles" and "reliability" as key factors of service quality in hospitality industry. Moreover, Choi and Chu (2001) reported following seven factors: "staff service quality", "room qualities", "general amenities", "business services", "value", "security" and "IDD facilities". Marković (2003) identified three-factor solution, interpreted as "empathy and assurance of hotel staff", "reliability" and "physical quality". This implies that the number and definition of the extracted factors depend on the measurement context.
However, these results reveal similarities to certain extent. Namely, most common factors of perceived service quality in hotel industry appear to be "reliability", "employees", and "tangibles". This means that hotel guests usually perceive service quality as reliable and error-free, with courteous, professional and neat hotel employees and visually appealing physical facilities.
Finally, results of reliability analysis indicate that Cronbach alpha values for overall expectations and perceptions scales in the Croatian and Slovenian surveys are above 0.90, suggesting highly reliable measurement instrument employed in this study. Thus, the instrument is suitable for use by hotel managers in gaining easily interpretable and reliable data on hotel guests' expectations and perceptions of service quality.
There are several limitations that need to be acknowledged. The data were collected in small geographic areas. Although both areas are important tourist destinations in Croatia and Slovenia, they can not represent all tourist destinations in both countries. Also, the expectations and perceptions measurement was limited to 29 hotel attributes. Even though these attributes were included in other studies and their validity was tested, there could be other relevant hotel attributes that are likely to influence hotel guests' opinions about overall hotel experience. Particularly, this is when specific settings in hospitality industry are considered (e.g. hostels, boutique hotels, campsites, private accommodation).
The results of present research may broaden the knowledge of expected and perceived hotel service quality and are suitable for broader international comparison. In particular, the research itself is considered useful for Croatian and Slovenian academics and practitioners, since comparisons of guests' expectations and perceptions in the hospitality industry between these two countries are very rare.
Future research may expand the current study. The researchers and practitioners might be interested in testing the difference in service expectations and perceptions regarding the demographic characteristics of hotel guests. This could help to identify the role of demographic characteristics in guests' expectations and perceptions differences. Furthermore, future research could also assess hotel staffs' perceptions of service performance and compare them with guests' perceptions in order to identify the differences. Additionally, as guests' expectations and hotel performance may vary over time, similar research should be undertaken periodically.