An importance-performance analysis of destination competitiveness factors: case of Jablanica district in Serbia

Abstract The principal aim of this paper was to explore destination competitiveness factors of Jablanica District (Serbia), by examining the attitudes of tourists. For research purposes, 32 attributes of destination competitiveness were rated with two parallel five-point Likert type scales – one by which tourists rated the importance of the attribute and the other by which they expressed their satisfaction with the same. A total of 378 validly completed questionnaires were collected and served as a basis for data analysis. By using a factor analysis, the most important destination competitiveness factors were extracted. Based on this factor solution, Importance-Performance Analysis (I.P.A.) was performed. Using I.P.A., this article examines the efficiency of the resources allocation based on isolated factors, and the possibility of creating management strategies to improve the competitiveness of this destination. The results of this study indicate that the factors of food and environment are of primary importance for tourists when choosing this tourism destination. Research results will be of great importance for tourism managers in the destination, as well as for authorities of local governments in the district, to better understand strengths and weaknesses of identified competitiveness factors and utilise them as a starting point for tourism development.


Introduction
There are many factors that influence the choice of a destination, and its respective competitiveness. In the academic literature, but also in practice, it is widely accepted that the satisfaction with destination factors is the main prerequisite for tourists when choosing a destination (Sánchez-García & Currás-Pérez, 2011). Shonk, Greenwell, and Divers (2012) define destination satisfaction as 'a multi-dimensional assessment of its individual attributes characteristics, and their evaluation during and after returning from the destination' . Satisfaction is observed as a factor in the ability of a destination to meet functional, service, social and psychological tourist needs. Satisfaction research, at the level of destination

Literature review
Compared to 140 countries in the world, the competitiveness of Serbian tourism is not favourable. According to the World Economic Forum (2015), Serbia is at the bottom of the list of European countries, just in front of Bosnia and Herzegovina, and Moldova. Moreover, when we observe the internal tourism conditions in the country, we can conclude that the share of Jablanica district in the tourist traffic of the Republic of Serbia is extremely low. Thus, this district participates only with 0. 86% (2006) up to a maximum of 1.06% (2014) of the total number of tourist arrivals in Serbia (Serbian Statistical Office, 2015), which makes it uncompetitive on both domestic and international tourism markets.
In this paper, the authors intended to explore the importance and satisfaction of tourists with certain destination attributes of Jablanica district in order to better understand the gap and provide useful information about future priority areas for investment and improvement. It was necessary to analyse all the extracted factors of tourism competitiveness in the Jablanica District and obtain a broader perspective in relation to prior work, where the factor of food was identified as a factor of primary importance for tourists when choosing a Jablanica District tourism destination (Stamenković & Djeri, 2016).
The quadrant of each attribute suggests a different managerial strategy, as described in Figure 1. Attributes positioned in the Quadrant I pose the greatest weakness of destinations/organisation, and require urgent managerial attention in order to improve quality and performance of attributes -concentrate here. Attributes that are positioned in the Quadrant II suggest that managers are doing the right thing and that, in the future, should strive to preserve the quality of these attributes -keep up the good work. Attributes in Quadrant III are considered as attributes of low priority, and do not require additional financial resources or improvement of performance attributes -low priority. The attributes that fall into Quadrant IV are called 'possible overkill' , and managerial suggestions are aimed at allocating funds to the attributes that have greater importance for the consumer.

Survey instrument
The great number of studies regarding tourism destinations, have used either structured (scale format) or unstructured (open-ended) measurement techniques. Within the studies with structured techniques of measurement, the Likert scale is most commonly used (Baloglu, 1997;Chon, 1991;Milman & Pizam, 1995), while with an unstructured approach, open-ended survey questions or focus groups are usually employed. The combination of these two techniques is recommended by the authors Echtner and Ritchie (1993), in order to carry out a complex evaluation of all the attributes of destinations competitiveness. Crouch and Ritchie (1999) defined 36 basic attributes for measuring the competitiveness of tourist destinations, divided into five dimensions: tourist attractions (7 attributes); supporting factors and resources (6 attributes); destination management (9 attributes); determinants of the limitations and expansion (6 attributes); destination policy, planning and development (8 attributes). Dwyer and Kim (2003) developed an integrated model of destination competitiveness, made up of six indicators: inherited factors (2 attributes); created resources (4 attributes); contributory factors (5 attributes); destination management (5 attributes); situational analysis (6 attributes) and market performance indicators (7 attributes). Cooper, Fletcher, Gilbert, and Wanhill (1998) categorised destination components into four factors: attractions; amenities; access; ancillary services. Laws (1995) grouped tourism destination components into: primary (natural and cultural resources); and secondary (tourism infrastructure and superstructure) factors. The list of tourism destination attributes in this work was adopted from the work of the author Armenski (2014), which examined a development of tourism destination loyalty model, taken from Geng-Qing Chi (2005). In that paper, author Geng-Qing Chi (2005) developed the questionnaire from a comprehensive review of the previous literature on the elements of destination, content analysis of tourism literature, and promotional brochures, and summarised all destination attributes in 7 categories: accommodation and facilities for the provision of accommodation, food and facilities for providing food and drink, shopping possibilities, natural and cultural attractions, activities and events, environment, and accessibility. Within these categories, crucial attributes of destination were identified based on an extensive literature review of the subject, and through focus groups with tourism sector representatives (academic experts, representatives of tourism organisations, employees of travel agencies and hotels). Experts in this area were asked to discuss and suggest what attributes are essential in the creation of the measuring instrument. Based on the qualitative research, 32 destination attributes were defined (Armenski, 2014). Five-point Likert type scales have been used to measure the importance and performance levels. Scales are ranging from 1, least important to 5, most important, in the Importance part, and from 1, strongly disagree, to 5, strongly agree, in the Performance part. The questionnaire was prepared in two languages: Serbian and English, due to both Serbian and foreign respondents. When translating Serbian back to English, the original terminology was preserved.

Data collection
The study sample consists of tourists who have visited Jablanica district during the five research months (May to September) in 2013. According to the official reports of Serbian Statistical Office, there were a total of 23,416 tourists registered in Jablanica district in 2013. Official statistical data have been used to create the sampling frame from which the sampling units were taken.
To examine the adequacy of the sample size, the Raosoft calculator for the sample size estimation was used (http://www.raosoft.com/samplesize.html). The sample size was estimated using the same parameters (margin of error 5%, confidence level 95%, variance in population 50%) on a population size N = 23,416. The recommended sample size on the 95% confidence level is 378 respondents (Table 1).

Data analysis
In this paper, socio-demographic profiles of the tourists were described by using descriptive statistics. Exploratory factor analysis (E.F.A.) was applied on 32 destination attributes which are used to assess tourists' satisfaction level with individual elements of the destination. Cronbach's Alpha reliability coefficient was calculated to test the stability of variables retained in each factor. I.P.A. is then used in establishing tourist satisfaction with the performance of tourism products and services in the destination. Afterwards, an independent sample t-test was applied with the aim of comparing the answers of men and women, and to determine the statistical significance of their differences.

Characteristics of respondents
Based on data from Table 2 it can be seen that, 52.9% of respondents are female respondents and 47.1% are male respondents. Most respondents are between 19 and 49 years old (76.4%) and have higher education (44.7%). The largest percentage of the respondents earns less than 20,000 € (69.1%) per annum. As for the nationality structure, the majority of the respondents were locals with Serbian nationality (92.9%), but Romani people are

Factor analysis
In order to estimate the latent dimensions of tourist satisfaction with attributes of destination competitiveness in the tourism destination of Jablanica district, we applied E.F.A. Through factor analysis we carried out the reduction of data to a smaller number of dimensions which explained the majority of variance in the structure of satisfaction. Principal components analysis (P.C.A.) was conducted on 32 attributes of destination, which are used to evaluate the tourist level of satisfaction by individual elements of destination. Factorability of the matrix was indicated by results of the Kaiser-Meyer-Olkin indicator, in the value of 0.881, which exceeded the recommended value of 0.60 (Kaiser, 1974). Bartlett's test of sphericity showed us a statistically significant value (p = 0.000), so the validity of applying E.F.A. was confirmed (Bartlett, 1954). In the selection of factors and defining the dimensions, several criteria were used: eigenvalue, percentage of variance, Cronbach's alpha coefficient, extracted communalities from individual motives, Scree plot, and factor loadings with their structure. At the beginning of the factor analysis, only those factors whose eigenvalue was greater than 1 were extracted. To obtain the appropriate factor solution, we took into consideration only those factor structures that explained more than 60% of the total variance. Visual identification of the optimal number of extracted factors and the existence of fracture point behind the seventh component was determined with the Scree plot. Seven factors with characteristic values over 1, were discovered by P.C.A. with the explained variance of 65.79%. Communality values are calculated in the range from 0.407 to 0.794. Therefore, this kind of seven factor solution shows that the variability of the original items is explained sufficiently. For further research, it was decided to retain all seven factors, based on Catell criteria (Catell, 1966). The structure of the isolated dimensions is shown in Table 3. Varimax rotation was used, in which there is no correlation between the extracted dimensions. The objective of the rotation is that each variable is represented, if possible, with the lowest number of factors, and with better spatial distribution. Based on the items that constitute the seven-factor solution, factors were appointed to: F1 -Accommodation, F2 -Dining, F3 -Environment, F4 -Accessibility, F5 -Activities and events, F6 -Shopping and F7 -Attractions.
The application of Cronbach's Alpha Reliability Coefficient (α) showed the reliability of the measuring instrument. It represents the most common instrument used for measuring the internal consent of the scale, and the level of relatedness of items of which the scale is made (Pallant, 2007). This coefficient in the ideal case should be higher than 0.7 (DeVellis, 2003). For proposed seven factor solution, Cronbach's alpha coefficient is: α = 0.817, suggesting adequate internal consistency of selected factors. Cronbach's alpha coefficient for the whole scale of 32 questions amounts to α = 0.904.

I.P.A. matrix
Determination of tourist satisfaction with performances of tourism products and services in tourism destination of Jablanica District, was achieved using I.P.A. For research purposes, 32 attributes were identified, grouped into seven factors, which were rated with two parallel measuring five-point scales.
With the first scale respondents rated attributes according to their importance with scores from 1 -completely unimportant, to 5 -extremely important. With the measuring scale, respondents rated level of satisfaction with individual attributes that have been identified in a tourist destination, with scores from 1 -very dissatisfied to 5 -very satisfied. Based on the results of E.F.A., the 32 tourism destination attributes were divided into seven factors: accommodation, dining, environment, accessibility, activities and events, shopping, and attractions. The resultant overall mean values of importance and performance when evaluating dimensions of the tourism destination are calculated on the whole sample and shown in Table 4.
These values were then used as coordinates to create the I.P.A. matrix of importance-performance. The overall average values for the importance of evaluated tourism destination components have been entered on the vertical (y) axis, while the overall performance average of the evaluated tourism destination component values, were inserted on the horizontal (x) axis. The overall average values of all components of importance (4.05) and performance (3.80), were chosen as the intersection point of the (x) and (y) axes in the coordinating system of I.P.A. matrix, from which we could define four quadrants. The components are then analysed according to the position in the I.P.A. matrix of the importance and performance, i.e., the corresponding quadrant on which they are located. Components in the upper left quadrant M (1.1) were rated as very important, but with a below average level of satisfaction. Components in the upper right quadrant M (1.2) were rated as very important and had an above average level of satisfaction. Components in the lower left quadrant M (2.1) are considered to be less important and had a below average level of satisfaction. Finally, the components in the lower right quadrant M (2.2) were estimated above average on the scale of satisfaction, but also estimated below average on the scale of importance.
The I.P.A. matrix shows that factors Dining (F2) and Environment (F3) are positioned in the quadrant `keep up the good work`; Factor Attractions (F7) is positioned in the quadrant `possible overkill`; while factors Accommodation (F1), Accessibility (F4), Activities/Events (F5) and Shopping (F6) are located in the quadrant of `lower priority` (Figure 2). When choosing a tourism destination Jablanica district, tourists believe that the factors Dining and Environment are, for them, the factors of greatest importance, which confirms the research hypothesis H1. The city of Leskovac, as the centre of Jablanica District, has profiled itself as a "city of barbecue", which explains the significance of Dining factor in their choice of tourism destination. In addition to food, Jablanica district has a very friendly and hospitable local people, positive southern atmosphere and municipalities with a high level of safety and security, which are recognised by tourists within the framework of Environmental factor, when choosing a tourist destination. In order to identify the level of importance and performance with individual attributes within destination factors, an individual overview of the scores by destinations components is shown in Table 5.
Within the Accommodation factor (F1) tourists gave the greatest importance to attributes of quality and cleanliness, the price of accommodation and accommodation services. In Dining factor (F2) tourists gave most importance to the food quality, diversity of cuisine and service in restaurants. The attribute of safety and security is the most important for tourists within the Environment factor (F3). Availability of transport/taxi service and level of accessibility are of primary importance in factor Accessibility (F4). The variety of special events/festivals, variety of outdoor recreation and reasonable price for activities are of greatest importance in the factor Activities/Events (F5). In the factor Shopping (F6), the attribute of reasonable price of merchandise has the greatest importance for respondents., In the factor Attractions (F7) attribute of the greatest importance is variety of natural attractions. Within the factors of greatest importance for tourists (F2 and F3), when choosing a tourism destination Jablanica district, high-performance attributes have been identified, but in lower importance categories, so we can conclude that the research hypothesis H2 was also confirmed. In the factor Dining, the attribute of appropriate food and beverage is positioned in the field M (2.2) `possible overkill`, while in the factor Environment, the attributes of friendliness of local people and peaceful and restful atmosphere are also positioned in this quadrant. A strategic decision for tourism managers in the District in this case would be allocation of financial resources towards attributes of those factors which have greatest importance for tourists.
Lastly, it should be pointed out that managers and other relevant stakeholders should keep an eye on all attributes that are not of great importance to the respondents and are low priority in terms of development, and continue to conduct research in order to react if necessary.

t-Test analysis
The t-test of independent samples was applied with the aim of comparing the attitudes of two groups of respondents, males and females, and to determine the statistical significance of their differences. This type of statistical analysis is used to determine whether there is a statistically significant difference in the average score measurements of some characteristics with two groups (Turjačanin & Čekrlija, 2006). A result of Levene's test for equality of variances determines the exact t value that we will be using in our analysis. The statistical significance of differences between the two groups of respondents is determined by observing the results of the t-test of equality of variances and by examining the significance column (Sig. (2-tailed)). When significance level is less than 0.05, we can conclude that there is a statistically significant difference between the mean values of the dependent variable in each of the two groups (Pallant, 2007). When determining statistical significance of the values obtained by the t-test, probability of risk of 5% and 1% was taken. For large samples (N ≥ 200) at a significance level of 5% or less (p ≤ 0.05), t ≥ 1.98 (-sign is not important), and with level of significance of 1% (p = 0.01), t ≥ 2.63 (-sign is not important), the t-test was performed at the significance level of p ≤ 0.01. In case of this type of analysis, the t-test, shows whether there exists a statistically significant difference between independent variables (gender) and dependent variables (Destination competitiveness grouped into seven factors).
The results of the t-test for independent samples showed that there is a statistically significant difference in the attitudes of male and female tourists when it comes to importance of competitiveness factors, which confirmed the research hypothesis H3. In six factors (F1, F2, F4, F5, F6, F7), it was shown that the attitudes of males and females differ when it comes to the importance of those factors of destination competitiveness. Female respondents evaluated the destination attributes with higher marks than male respondents. In one factor (F3 -Environment), there are no statistically significant differences in male-female attitudes of the respondents ( Table 6).
The results obtained in this study, using t-test, can significantly help tourism managers in the Jablanica district in the identification of market segments. Unlike the results of importance of destination attributes, there are no statistically significant differences in the attitudes of male and female respondents in the results of their performances for all seven factors. This can be interpreted by the fact that in the analysed municipalities of Jablanica District, tourists of both genders have a relatively similar position on performance factors of tourism destination.
Identification of a group of customers who have certain common features when choosing a particular product plays an important role in the market positioning (Blešić et al., 2014). Analysing the perception of competitiveness in terms of different segments of destinations and tourism can help managers in the District to create marketing strategies which will meet the specific needs of each segment. Age, gender, education, occupation, marital status, as well as their wishes and needs, together with their level of satisfaction with destination specific attributes can affect their decision when choosing tourism destination.

Conclusion
Importance Performance Analysis is an excellent guide for the allocation of limited financial resources of companies and directing their actions toward the development of the attributes that have the highest value for the consumer. Bruyere et al. (2002) find that the accuracy and reliability of the results, their interpretation, and undertaking of concrete strategic actions of the companies increases if the consumers are, on some basis, segmented into more homogeneous groups. If there is no clear segmentation of demand, it is highly probable that the limited resources of enterprises may not be able to allocate the most important attributes based on which the competitive advantage in the market can be achieved.
In this paper, by applying E.F.A., 32 attributes of tourism destinations competitiveness were classified into seven factors: accommodation, dining, environment, accessibility, activities/events, shopping, and attractions. Research results confirmed that in the tourist destination of Jablanica district, there were no indicated competitiveness factors which are of great importance but low satisfaction for tourists (quadrant 1.1 I.P.A. grid -Concentrate here). This has a great importance for tourism managers when creating a destination strategic plan of the forthcoming period, because tourist discontent with service quality, environment, interconnection, or with tourism infrastructure is a limiting factor for further growth and destination development. Tourist dissatisfaction could also result a negative verbal propaganda and create an obstacle in forming the base of loyal tourists. Decreased number of tourist reduces the income of tourism stakeholders in the destination, which results in less money for investment in tourism infrastructure.
On the other hand, tourists are very satisfied with restaurant services such as: Meal plan, Service in restaurants, Food quality and Reasonable prices. Also, attributes related to Friendliness of local people, Safety and security, Peaceful and restful atmosphere, and Cleanliness received high marks from the respondents. At the same time, these factors of competitiveness have great importance for the tourists staying in the area. These are precisely the values of Jablanica district in which it is necessary to continue the ongoing investments to maintain the level of achieved quality, because they represent a pillar of tourism development of the whole tourist destination (quadrant 1.2. I.P.A. grid -Keep up the good work). Research results also showed that tourism managers in Jablanica district need to reallocate funds currently invested in: Variety of natural attractions, Variety of cultural options and Variety of historic/cultural sites, because these attributes do not have significance for the tourist in destination (quadrant 2.2. I.P.A. grid -Possible overkill). According to research results, attributes that received a low score on a scale of both importance and performance re: Quality and cleanliness of accommodation, Level of accessibility, Variety of Activities and events, Quality of merchandise and Friendliness of service. Considering that the stated dimensions of tourist destinations quality are not significant for the respondents within the sample, investing in their development should not be of primary importance (quadrant 2.1. I.P.A. grid -Lower priority).
This study has certain limitations. We can indicate a structure of research sample as one of the most important limitation factors. Namely, the sample structure is dominated by domestic tourists (91.8%) with revenues of less than 5,000 euros per year (69%), and the research results are gained predominantly by tourists who have visited the Jablanica district during the period of five research months (May to September) in 2013. Also, there are limitations that arise from relying on research from a single year (2013), whereas, in order to form an accurate insight of the competitiveness factors in observed tourist destination, a multi-year longitudinal research with repeated measures should be implemented. Results of this longitudinal research could have important theoretical and practical impacts on the creation of strategic management directions in the tourism sector of Jablanica district.
Our recommendations for future research of competitiveness of tourism destination of Jablanica district, using I.P.A. refer to the extension of the research sample, which could also include managers and employees in tourism companies in this district. This survey supplement would create a complete image of situation of tourism sector in all six municipalities of Jablanica District, which would contribute to more comprehensive profiling of tourism competitiveness attributes of this destination.

Disclosure statement
No potential conflict of interest was reported by the authors.