Influence of the Country’s Information Development on Its Tourist Attractiveness Yuliia Lola1, Svetlana Prokopovich1, Olena Akhmedova1 1 Simon Kuznets Kharkiv National University of Economics, 9-A, Ave.Nauki, Kharkiv, 61166 Ukraine yuliia.lola@hneu.net, prokopovichsv@gmail.com, yelena.akhmedova@hneu.net Abstract. A number of studies have researched the effects of tourism on trans- portation system, hotel industry, economic efficiency and environment. This paper examines the influence of the information and communication tech- nologies development on the inbound tourism intensity. The correlation and regression analysis has been used to identify the relationship between the Travel and Tourism Competitiveness Index, the Information and Communication Technology Development Index and International tourism arrivals. The results demonstrate that there is a close link between the countries’ tourist attractive- ness and the level of their information and communication development. How- ever, it is not equal for different countries, which are grouped by the level of in- tensity of tourism arrivals, the level of the country’s attractiveness and its in- formation and communication technologies development. Besides, the coun- try’s information and communication technologies development has little effect on the inbound tourism intensity Keywords: tourist attractiveness of a destination, information and communica- tion development of a country, the travel and tourism competitiveness index, in- ternational tourism arrivals. 1 Introduction The contemporary unification of the world society into a single information and communication network and the transformation of information technologies into the generative force of socio-economic development contributes to the close interconnec- tion between countries, regions and societies of different nations. Most countries are actively using Internet space to shape the country’s image (including tourist image). Mainly because it is one of the most important tools for creating an appropriate image of the country as a tourist destination that can greatly increase the intensity of the inbound tourism flow. In this regard, the study of the impact of the country’s informa- tion and communication technologies development on the tourism and travel devel- opment is relevant and requires further theoretical and practical research. 2 Theoretical development and hypotheses formulation Tourism development affects the development of a range of other areas of economic activity. Inbound and outbound tourism has bidirectional causality with air transporta- tion (Syed Abdul Rehman Khan and other, 2017). Air transport and tourism are highly connected. Researches show that tourist-oriented airports may achieve higher efficiency levels than non-touristic ones (Xosé Luis Fernández, Pablo Coto-Millán, Benito Díaz-Medina, 2018). The level of tourism development is estimated by the global index. The Travel & Tourism Competitiveness Index has been the subject of some methodological criti- cism, such as the arbitrary weighting of variables. There is an alternative methodol- ogy for calculating this index based on two points of reference to propose a new stan- dardization. А synthetic index that measures the state of the pillar in the worst posi- tion, as well as other alternative indices, is calculated (Juan Ignacio Pulido-Fernández, Beatriz Rodríguez-Díaz, 2019). Depending on how the variables are included in the underlying technology specifi- cation, the same tourism index can be oriented towards the assessment of either the private or the public sector’s effectiveness (Walter Briec and other, 2018). Trade openness, climate change and intensity of market competition increase tour- ism efficiency in China. Tourism efficiency improvement in China was mainly driven by technological improvement (Sami Chaabouni, 2019). UNESCO’s World Heritage inscription is considered to positively influence tour- ism demand. However, relevant econometric research has yielded inconsistent results. A sub-group analysis identifies different factors in developing vs. developed countries and cultural vs. natural WHS types. (Yang LanXuе, Thomas E.Jones, 2019). The expansion of tourism translates into an environmental deterioration of the des- tination (risk dimension) and, furthermore, it substantiates that there are specific vari- ables connected to environmental sustainability (regulatory dimension) that contribute to greater tourism growth, so that the relationship between tourism and environmental sustainability is bidirectional (Juan Ignacio Pulido-Fernández, Pablo Juan Cárdenas- García, Juan Antonio Espinosa-Pulido 2019). The studies suggest that the effect of growth rate of total foreign tourist arrivals on hotel equity return is asymmetric and state-dependent, conditional on the distributions of hotel equity return. The study further identified that GTA has a significant influ- ence only on equity returns of hotels with a small size (Ming-Hsiang Chen, 2016). The shift of our view on information technology in tourism research from a primar- ily a marketing-driven tool to a knowledge creation tool due to new technological conditions such as the smartphone, drone, wearables, new connectivity and big data is recognized. Some possible future research problems and challenges regarding our existing views of the relationship between information technology and tourism are studied (Zheng Xiang, 2018). Not only ICTs empower consumers to identify, customise and purchase tourism products but they also support the globalisation of the industry by providing effective tools for suppliers to develop, manage, and distribute their offerings worldwide (Bu- halis, 1998). Buhalis (1998) stated that potential tourists have become moreindependent and so- phisticated on using a wide range of tools to arrange for their trips (such as Expedia, Googleand Kayak, visitbritain.com), web 2.0portals, wayn and tripadvisor, kelkoo). Information Search is a significant part of the purchase decision process and was revolutionised as a result of the Internet. ICTs not only reduce uncertainty and per- ceived risks but also enhance the quality of trips (Fodness &Murray, 1997). The quality of the website, Digital Marketing, Social Networking, Multimedia, Mobile Technologies and Intelligent Environments are the main keys factors of ICT in Tourism (Elisabete Paulo Morais & other, 2016). A Virtual Travel Community (VTC) makes it easier for people to obtain informa- tion, maintain connections, develop relationships, and eventually make travel-related decisions (Stepchenkova, Mills & Jiang, 2007). Increasingly the impacts of ICTs are becoming clearer, as networking, dynamic in- terfaces with consumers and partners and the ability to re-develop the tourism product proactively and reactively are critical for the competitiveness of tourism organizations (Buhalis, D., & Law R., 2008). The analysis of the mentioned resources has allowed hypothesizing the following: Hypothesis 1. Information development of the society contributes to the improve- ment of the country’s tourist attractiveness. Hypothesis 2. The development of information and communication technologies in the countries across the globe positively influences the inbound tourism intensity. 3 Methods The methods of multivariate statistical analysis, such as Descriptive Statistics, the multiple regression, the cluster analysis were used to study the influence of informa- tion and communication technologies on tourism. These statistical methods were im- plemented with the StatSoft's software package Statistica. This package is well bal- anced with the “power / convenience ratio”, has a wide range of functional data analysis algorithms and has wide graphical capabilities for data visualization. To carry out the research, the global indices and variables of tourism development were selected: The Travel and Tourism Competitiveness Index (TTCI), which reflects the level of the country’s attractiveness for both tourists and also investors and representatives of the tourism business. This index includes the characteristics of the following frame- work: Enabling Environment, Travel and Tourism Policy and Enabling Conditions, Tourism and Transport Infrastructure, Natural and Cultural Resources [12]; The Information and Communication Technologies Development Index (ICT) re- flects the level of networked infrastructure and access to ICTs, the level of use of ICTs in the society and more efficient and effective ICT use [11]. International tourism arrivals (ITA) is one of the main indicators that reflects the effectiveness of all the measures adopted for the development of tourism in the coun- try [3]. The objects of research are 80 countries of the world. The variables are the data for 2016. The countries without sufficient data were excluded from the database. 4 Results In order to study the influence of the country’s information and communication de- velopment on tourism development, the following algorithm of the research has been proposed: Stage 1. Selection of the initial variables. Stage 2. Research of the basic statistical characteristics of the selected variables. Stage 3. Verification of the first hypothesis on the basis of the correlation- regression analysis methods. Stage 4. Verification of the second hypothesis on the basis of the correlation- regression and cluster analysis methods for the whole array of initial data and within the scope of separate groups of countries, which are similar according to the level of tourism activity. For implementation of the first stage of the algorithm, the following variables were selected: Travel and Tourism Competitiveness Index (TTCI), Information and Com- munication Technologies Development Index (ICT) and International Tourism Arri- vals (ITA). The descriptive statistics was used to process, systematize and provide quantitative description of the empirical data by means of the main statistical indicators. The im- plementation of the second stage of the study presupposed the calculation of the fol- lowing characteristics: Mean, Median, Mode, Frequency of Mode, Minimum, Maxi- mum, Variance, Standard Deviation, Coefficient of Variation, Skewness, Kurtosis, as well as histograming. The results of calculation are presented in Table.1. Table 1. Descriptive Statistics Descriptive Statistics Variable Variance of Mode Valid N Median Kurtos. Skewn. Mode Mean Coef. Freq. Dev. Max Var. Min Std. TTCI 80 4,13 4,125 3,910 3 3,09 5,43 0,36 0,60 14,542 0,2079 -0,7614 ICT 80 6,52 6,875 – – 3,03 8,98 2,66 1,63 24,946 -0,4806 -0,7991 ITA 80 12684,2 5460,0 – – 121,0 82600,0 316498734 17790,4 140,256 2,3697 5,6854 The results of the histograming of distribution for each of the studied variables are presented in Fig. 1 – 3. Histogram: TTCI K-S d=,06011, p> .20; Lilliefors p> .20 Shapiro-Wilk W=,97569, p=,13091 30 25 20 No. of obs. 15 10 5 0 2,5 3,0 3,5 4,0 4,5 5,0 5,5 X <= Category Boundary Fig. 1. TTCI variable distribution histogram Histogram: ICT K-S d=,10402, p> .20; Lilliefors p<,05 Shapiro-Wilk W=,94517, p=,00184 22 20 18 16 14 No. of obs. 12 10 8 6 4 2 0 2 3 4 5 6 7 8 9 X <= Category Boundary Fig. 2. ICT variable distribution histogram Histogram: Arrivals K-S d=,24004, p<,01 ; Lilliefors p<,01 Shapiro-Wilk W=,67618, p=,00000 60 50 40 No. of obs. 30 20 10 0 -10000 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 X <= Category Boundary Fig. 3. ITA variable Distribution Histogram According to the results of the analysis of the obtained statistical characteristics (see Table 1) and distribution histograms (see Figures 1-3), the following conclusions can be made: - the TTCI variable has a distribution close to normal. This is evidenced by the prox- imity of the mean, mode and median, as well as small values of the skewness and kurtosis. This variable has the least value of the coefficient of variation (14,52); - the ICT variable also has a distribution rather close to normal. Its average value is close to the median. But unlike the TTCI variable, it has a slightly larger range (from 3.03 to 8.98) and a larger coefficient of variation (24.95). It should be noted that al- most half of the world’s countries (38 out of 80) fall into the last two intervals with values of 7 to 8 and from 8 to 9. This means that a significant part of the countries in the considered group has a high level of development of information and communica- tion technologies (ICT); - the ITA variable is significantly different from the previous two. First of all, it has a completely different unit of measurement and dimension, therefore, during further research with the simultaneous use of the TTCI and ICT variables, the calculations will be made on the basis of the standardized data. Secondly, the distribution of this variable is quite distant from normal. This is evidenced by the large difference be- tween the mean and the median (12 684.2 and 5 460.0 thousand persons respectively), as well as the statistical criteria of the Kolmohorov-Smirnov (K-S test), Shapiro-Wilk test and Lillifors test. For the distribution of this variable the right-side bias is charac- teristic (the skewness equals 2.37) and significant elevation (the kurtosis is equal to 5.69). In 2016, this variable was significant (from 121 to 82,600 people), more than 50 countries had the value of this variable up to 10,000 thousand people, in France this value was more than 80,000, and in Spain and United States – it ranged from 70,000 to 80,000 thousand people. The verification of the first hypothesis that the information development of the so- ciety contributes to the improvement of the country’s tourist attractiveness was car- ried out during the implementation of the third stage of the study. A pair correlation coefficient between the TTCI and ICT variables was calculated according to the data from all 80 countries. In 2016 it was equal to 0.711 that indicates a fairly close direct linear relationship between these variables. The graphic representation of this connec- tion is given in Fig. 4. Fig. 4. Dispersion field (correlation field) between factors The dispersion field proves a linear relationship between TTCI and ICT, therefore, we can accept the hypothesis 1 that the greater the information development of the country is, the better the tourism and travel sector is developed in this country. This allows putting forward the second hypothesis that the development of information and communication technologies in the countries of the world positively influences the intensity of inbound tourism. Thus, the implementation of the fourth stage involves verification of the second hypothesis that involves the following sequence of steps: Step 1. Determination of the pair correlation coefficients between the resulting ITA variable and factor variables of TTCI and ICT. Step 2. Construction of the multiple regression based on the standardized data of the type: where a1 та a2 are unknown parameters that are evaluated by the least squares method. Step 3. Distribution of countries into homogeneous groups according to the ITA, TTCI and ICT variables on the bassis of the cluster analysis methods. Step 4. Construction of the multiple regression (1) for each of the clusters. Step 5. Making conclusions as to the hypothesis acceptance or rejection. In the result of implementation of the first step the following values of the pair cor- relation coefficients have been received: ; . These values mean that there is a direct linear link of the moderate level between ITA and TTCI, but between ITA and ICT there is a direct but weak link. The obtained values are not sufficient for accepting or rejecting the hypothesis 2. Therefore, during the second step, the following equation of multiple regression was constructed: This regression equation is statistically significant in terms of Fisher’s criterion (F = 40.32), and separate parameters according to Student’s criterion ( , ). The coefficients of the multiple correlation (R = 0.713), the determina- tion ( )) and the corrected determination coefficient ( ) indi- cate a sufficient quality of the model. There is no autocorrelation of the errors in this model (the statistics of Darbine-Watson are approximately equal to 2, and the cyclic coefficient of autocorrelation is close to 0). Thus, this model can be used for analysis and forecasting. We have analysed the problem under study according to this model. As can be seen from the obtained regression equation, compared with the pair correlation coeffi- cients, there is a significant increase of the influence of TTCI on ITA (from 0.6574 to 0.9293), the simultaneous change of direction and the increase of the influence of ICT on ITA (from + 0.2734 to -0.3915). To answer the question whether these changes are only due to the multicollinearity that are present in the model, or in fact there is an inverse relationship between ICT and ITA, partial correlation coefficients have been calculated and their statistical significance checked. The results of calculations are given in Table 2. Table 2. Results of correlation coefficients calculations Variables currently in the Equation; DV: ITA Partial Semipart b* in Tolerance R-square t(77) p-value Variable Cor. Cor. TTCI 0,936420 0,684528 0,658439 0,494413 0,505587 8,23980 0,000000 ICT -0,392394 -0,366155 -0,275910 0,494413 0,505587 -3,45277 0,000905 As can be seen from Table 3, the value of the partial coefficients is: ; and they are statistically significant according to the Student’s criterion. During the next, third step, using the cluster analysis methods, we obtain homoge- neous groups of countries. The grouping of countries is based on the hierarchical method of full communication, which allows clearly divide the countries into two, three, or four clusters. The division into 2 clusters is not informative. If we divide countries into 4 clusters, then the last cluster will consist of only three countries (28, 69 and 77). Therefore, it is rational to divide countries into three clusters, which cor- responds to the logical distribution of countries with high, medium and low intensity of foreign tourists’ arrivals. Based on the iterative method of clustering k-means, the following cluster results have been obtained. The first cluster includes 30 countries with the low inbound tour- ism activity (Cluster contains 30 cases). These countries are listed in Table 3. Table 3. Members of Cluster Number 1 and Distances from Respective Cluster Center Саse Coun- Case Dis- Case Dis- Distance Country Country No. try No. tance No. tance Alba- Nicara- C_1 0,2130 C_29 Georgia 0,370 C_56 nia gua 0,569 Arme- Guate- 0,533 Para- C_3 0,3328 C_32 C_58 nia mala 6 guay 0,448 Azer- Hondu- 0,562 C_6 0,5073 C_33 C_59 Peru baijan ras 3 0,479 Philip- C_9 Bhutan 0,4266 C_36 India 0,961 C_60 pines 0,119 Bosnia Iran, and Roma- C_10 0,4561 C_37 Islamic 0,287 C_63 Herze- nia Rep, govina 0,626 Cam- C_13 0,5963 C_41 Jamaica 0,166 C_65 Serbia bodia 0,649 Co- Kyrgyz Sri C_17 0,3328 C_44 0,455 C_70 lombia Republic Lanka 0,421 Do- mini- C_23 can 0,1642 C_51 Moldova 0,723 C_75 Ukraine Repub- lic 0,428 Vene- Ecua- Mongo- C_24 0,3597 C_52 0,244 C_79 zuela, dor lia RB 0,288 El Monte- C_25 Salva- 0,4453 C_53 0,585 C_80 Vietnam negro dor 0,348 Besides, these countries have a very low level of tourism potential (TTCI), infor- mation, and communication technologies development (ICTs). Under the current conditions of certain instability, Ukraine is referred to this group. The second cluster includes 38 countries with an average level of inbound tourism activity (Cluster contains 38 cases) and is presented in Table 4. Table 4. Members of Cluster Number 2 and Distances from Respective Cluster Center Саse Dis- Саse Dis- Саse Dis- Country Country Country No. tance No. tance No. tance Argen- 0,42326 C_2 0,29058 C_5 Nether- 0,47531 C_2 Estonia tina 0 6 9 4 lands 0 Austra- 0,75429 C_2 0,22647 C_5 New Zea- 0,44557 C_4 Finland lia 2 7 6 5 land 0 Barba- 0,51231 C_3 0,55392 C_5 0,42966 C_7 Greece Norway dos 5 1 3 7 1 0,20128 C_3 0,42158 C_6 0,44142 C_8 Belgium Hungary Poland 8 4 9 1 8 0,54964 C_3 0,55850 C_6 0,41838 C_11 Brazil Iceland Portugal 9 5 0 2 3 0,33395 C_3 0,22727 C_6 Russian 0,57500 C_12 Bulgaria Ireland 7 8 4 4 Federation 4 0,69699 C_3 0,53658 C_6 0,52116 C_14 Canada Israel Singapore 3 9 2 6 2 0,47380 C_4 Korea, 0,55880 C_6 Slovak 0,48333 C_15 Chile 4 3 Rep, 7 7 Republic 7 Costa 0,47433 C_4 0,44958 C_6 0,26416 C_18 Latvia Slovenia Rica 4 5 2 8 8 0,21570 C_4 0,49697 C_7 0,34754 C_19 Croatia Lithuania Sweden 6 6 7 1 2 0,37245 C_4 Luxem- 0,42012 C_7 Switzer- 0,69462 C_20 Cyprus 7 7 bourg 3 2 land 1 Czech 0,22669 C_4 0,73836 C_21 Repub- Malaysia 7 8 2 C_7 0,75310 lic Uruguay 8 3 Den- 0,40213 C_4 0,26146 C_22 Malta mark 6 9 8 Countries in the cluster 2 are characterized by the highest level of tourism potential (TTCI) and the average level of ICT development. The third cluster includes 12 countries with the highest level of inbound tourism activity (Cluster contains 12 cases). This claster is presented in Table 5. Table 5. Members of Cluster Number 3and Distances from Respective Cluster Center Саse No. Country Distance Саse No. Country Distance C_5 Austria 0,701654 C_50 Mexico 0,941073 C_16 China 0,727240 C_69 Spain 1,028845 C_28 France 1,230744 C_73 Thailand 0,926716 C_30 Germany 0,654982 C_74 Turkey 0,916710 C_40 Italy 0,171680 C_76 United Kingdom 0,678250 C_42 Japan 0,932930 C_77 United States 0,968470 Countries of the latter cluster are characterized by the highest level of tourism po- tential development (TTCI) and rather high level of ICT. It should be noted that the received country grouping by the level of tourism activ- ity is sustainable, since the hierarchical method of complete dependence and the k- medium method yielded identical results, except for the country number 78 (Uruguay) which, according to the first method, was reffered to the low-income countries devel- opment, and, according to the second method, it was reffered to countries with an average level of development. As the final result we accept the one that gives the k- medium method, since this method minimizes intragroup variance and maximizes the intergroup, thus providing higher-quality clusterization. The average means of the variables, according to which the clusterization was car- ried out, are presented in Fig. 6 Fig. 6. Average means of TTCI, ICT and ITA variables by clusters Analysis of the means given in Fig. 6, allows to draw the following conclusions. First, the cluster number 1 is formed by the countries with the lowest values of the TTCI, ICT and ITA variables. The second cluster consists of the countries with an average level of TTCI and ITA, but with the highest level of ICT. The third cluster includes the countries with the highest levels of TTCI and ITA with high (but not the highest) level of ICT. In addition, the countries of the second and third clusters are characterised by the inverse dependence between the ITA and ICT variables. This is also confirmed by the calculation of the pair correlation coefficients between the vari- ables for each cluster separately. The results of calculations are presented in the Table 6. Table 6. Matrices of the pair correlation coefficients for each cluster Claster 1 Claster 2 Variable TTCI ICT ITA Variable TTCI ICT ITA TTCI 1 -0,0763 0,4156 TTCI 1 0,4883 0,3444 - ICT -0,0763 1 -0,1822 ICT 0,4883 1 0,1534 ITA 0,4156 -0,1822 1 ITA 0,3444 -0,1534 1 Claster 3 Variable TTCI ICT ITA TTCI 1 0,8328 0,3931 ICT 0,8328 1 0,1444 ITA 0,3931 0,1444 1 Let us analyze the means of the obtained coefficients in more detail. Thus, cluster 1, in comparison with other clusters, is characterised by the strongest direct dependence between TTCI and ITA (0.4156) and the inverse dependence be- tween ICT and ITA (-0.1822). Besides, the countries within this cluster have almost no dependence between TTCI and ICT (-0,0763), that is, for countries of this group we reject the hypothesis 1. The absence of multicollinearity between the TTCI and IST factors enables to con- struct a two-factor regression model (1): This regression equation is statistically significant in general according to Fisher’s criterion (F = 3.395, Significance F = 0.048). According to Student’s criterion, only the influence of the TTCI factor ( ) is statistically signifi- cant, whereas the influence of ICT is not statistically significant ( ). The coefficients of the multiple correlation (R = 0.442), the determination ( ) and the corrected determination coefficient ( ) indicate insufficient qual- ity of the model. Thus, on the basis of the above stated, for the countries with the low level of tourism activity the hypothesis 2 is rejected. Cluster 2 is characterised by a weak dependence between TTCI and ICT (0.4883), as well as between TTCI and ITA (0.3444). There is a weak inverse dependence be- tween the ICT and ITA values (-0.1534), that is, for countries of this group we accept the hypothesis 1. The lack of multicollinearity between the TTCI and ICT factors within the cluster 1 allows to construct a two-factor regression model (1): This regression equation is statistically significant in terms of Fisher’s criterion (F = 6.143, Significance F = 0.005). According to Student’s criterion, the influence of both factors is statistically significant: і ТТСІ ( ), і ІСТ ( ). The coefficients of the multiple correlation (R = 0.504), the determination ( ) and the corrected determination coefficient ( ) indicate insufficient quality of the model. Thus, on the basis of the above stated, the hypothesis 2 for the countries with the average level of tourism ac- tivity is rejected. Cluster 3 is characterised by a weak dependence between TTCI and ITA (0.33931), as well as by a very weak dependence between ICT and ITA (0.1444). There is a strong direct dependence between TTCI and ICT (0.8328), that is, for the countries of this group we accept the hypothesis 1. The presence of strong multicollinearity between the factors of TTCI and ICT does not allow to construct a two-factor regression model (1). Thus, on the basis of the aforementioned, the hypothesis 2 for the countries with the high level of tourism activity is rejected. Let us analyze the means of the obtained coefficients in more detail. Thus, cluster 1, in comparison with other clusters, is characterised by the strongest direct dependence between TTCI and ITA (0.4156) and the inverse dependence be- tween ICT and ITA (-0.1822). Besides, the countries within this cluster have almost no dependence between TTCI and ICT (-0,0763), that is, for countries of this group we reject the hypothesis 1. The absence of multicollinearity between the TTCI and IST factors enables to con- struct a two-factor regression model (1): This regression equation is statistically significant in general according to Fisher’s criterion (F = 3.395, Significance F = 0.048). According to Student’s criterion, only the influence of the TTCI factor ( ) is statistically signifi- cant, whereas the influence of ICT is not statistically significant ( ). The coefficients of the multiple correlation (R = 0.442), the determination ( ) and the corrected determination coefficient ( ) indicate insufficient qual- ity of the model. Thus, on the basis of the above stated, for the countries with the low level of tourism activity the hypothesis 2 is rejected. Cluster 2 is characterised by a weak dependence between TTCI and ICT (0.4883), as well as between TTCI and ITA (0.3444). There is a weak inverse dependence be- tween the ICT and ITA values (-0.1534), that is, for countries of this group we accept the hypothesis 1. The lack of multicollinearity between the TTCI and ICT factors within the cluster 1 allows to construct a two-factor regression model (1): This regression equation is statistically significant in terms of Fisher’s criterion (F = 6.143, Significance F = 0.005). According to Student’s criterion, the influence of both factors is statistically significant: і ТТСІ ( ), і ІСТ ( ). The coefficients of the multiple correlation (R = 0.504), the determination ( ) and the corrected determination coefficient ( ) indicate insufficient quality of the model. Thus, on the basis of the above stated, the hypothesis 2 for the countries with the average level of tourism ac- tivity is rejected. Cluster 3 is characterised by a weak dependence between TTCI and ITA (0.33931), as well as by a very weak dependence between ICT and ITA (0.1444). There is a strong direct dependence between TTCI and ICT (0.8328), that is, for the countries of this group we accept the hypothesis 1. The presence of strong multicollinearity between the factors of TTCI and ICT does not allow to construct a two-factor regression model (1). Thus, on the basis of the aforementioned, the hypothesis 2 for the countries with the high level of tourism activity is rejected. 5 Conclusion Thus, 1) the tourist attractiveness of the country increases if there are developed informa- tion and communication technologies, because the quality of tourist information re- sources (including Internet resources), formation of the comfortable information envi- ronment, the mass use of platforms for travel services searching and comparing of their prices, the development of e-commerce in tourism in general contributes to the improvement of the tourism infrastructure; 2) there is no dependence of this type between tourism attractiveness and the de- velopment of information and communication technologies in the first group of coun- tries, which is characterized by low TTCI, ICT and low intensity of tourist arrivals. There is a significant dependence in the 2nd group of countries. There is a strong direct dependence in the third group, characterised by the highest level of tourist arri- vals and the highest level of tourist attractiveness. That is, the more the country is attractive for tourism, the stronger is the interdependence between the indicators of TTCI and ICT; 3) there is a weak inverse dependence between the countries’ information and communication technologies development and the intensity of tourist arrivals. More- over, this connection is not observed in any of the 3 distinct groups of countries. That is, it can be argued that the development of information and communication technolo- gies almost does not affect the intensity of inbound tourism. References 1. Buhalis, D., Law R.: Progress in information technology and tourism management: 20 yearson and 10 years after the Internet – The state of eTourism research. Tourism Man- agement 29, 609-623 (2008). 2. Buhalis, D.: Strategic use of information technologies in the tourism industry. Tourism Management 19 (5), 409–421 (1998). 3. Fodness, D., & Murray, B.: Tourist information search. Annals of Tourism Research 24 (3), 503-523 (1997). 4. Juan Ignacio Pulido-Fernandez, Beatriz Rodriguez-Diaz: Reinterpreting the World Eco- nomic Forum's global tourism competitiveness index. Tourism Management Perspectives, vol. 20, 131-140 (2016). 5. Juan Ignacio Pulido-Fernández, Pablo Juan Cárdenas-García, Juan Antonio Espinosa- Pulido: Does environmental sustainability contribute to tourism growth? An analysis at the country level. Journal of Cleaner Production. Vol. 213, 309-319 (2019). 6. Elisabete Paulo Morais, Carlos Rompante Cunh, João Paulo Sous: Information and com- munication technologies in tourism: Challenges and trends. Conference: Proceedings of the 27th International Business Information Management Association Conference - Inno- vation Management and Education Excellence Vision 2020: From Regional Development Sustainability to Global Economic Growth, IBIMA. Italy. (2016) Homepage, https://www.researchgate.net/publication/316830871_ Information_ and_communication_technologies_ in_tourism_Challenges_and_trends. 7. Ming-Hsiang Chen: A quantile regression analysis of tourism market growth effect on the hotel industry. International Journal of Hospitality Management. Volume 52, 117-120 (2016). 8. Syed Abdul Rehman Khan Dong Qianli, Wei SongBo, Khalid Zaman, Yu Zhang: Travel and tourism competitiveness index: The impact of air transportation, railways transporta- tion, travel and transport services on international inbound and outbound tourism. Journal of Air Transport Management, vol. 58, 125-134 (2017). 9. Sami Chaabouni: China's regional tourism efficiency: A two-stage double bootstrap data envelopment analysis. Journal of Destination Marketing & Management. Vol. 11, 183-191 (2019). 10. Stepchenkova, S., Mills, J. E., & Jiang, H.: Virtual travelcommunities: Self-reported ex- periences and satisfaction. In M. Sigala, L. Mich, & J. Murphy (Eds.). Information and communication technologies in tourism 2007. New York: Springer Wien. 163-174 (2007). 11. The ICT Development Index 2017 Homepage, https://www.itu.int/net4/ITU- D/idi/2017/index.html#idi2017rank-tab. 12. The Travel & Tourism Competitiveness Report 2017 Homepage, https://www.weforum.org/reports/the-travel-tourism-competitiveness-report-2017. 13. Walter Briec, Kristiaan Kerstens, Diego Prior, Ignace Van de Woestyne: Testing General and Special Fдre-Primont Indices: A Proposal for Public and Private Sector Synthetic In- dices of European Regional Expenditures and Tourism. European Journal of Operational Research. European Journal of Operational Research, Volume 271, Issue 2, 756-768 (2018). 14. Xosé Luis Fernández, Pablo Coto-Millán, Benito Díaz-Medina: The impact of tourism on airport efficiency: The Spanish case. Utilities Policy, Volume 55, 52-58 (2018). 15. Yang LanXuе, Thomas E.Jones: Tourism-enhancing effect of World Heritage Sites: Pana- cea or placebo? A meta-analysis. Annals of Tourism Research. Vol. 75, 29-41 (2019). 16. Zheng Xiang: From digitization to the age of acceleration: On information technology and tourism. Tourism Management Perspectives. Volume 25, 147-150 (2018).