43Review Article UDC: 338.485:316.64(497.16)
doi: 10.5937/menhottur2101103M
How do residents assess the social impact of tourism?
Srdan Milosevic1*, Dusan Peric1, Iva Skrbic1
1 Educons University, Faculty of Sport and Tourism, Novi Sad, Serbia
Abstract: Tourism plays an important role in the economic and cultural development of society, producing many positive and negative effects. This study aimed to (a) assess the impact of tourism on modern life in Montenegro based on the attitudes of residents and (b) standardize instrument (questionnaire) for assessing these impacts. The sample formed by 521 adults from three different tourist regions (coastal, continental, mountainous). The questionnaire has high validity and reliability. Two components (two independent scales) have been extracted - the negative and positive social impact of tourism. It was found that the increase in positive impacts is followed by an increase in negative effects too. Type of the region, level of municipal development and engagement of respondents in tourism influence significantly the differences between attitudes of examinees. Those who live by tourism most respect its importance for local development.
Keywords: tourism, social impact, residents attitude, questionnaire, Montenegro JEL classification: Z32
Kako rezidenti procenjuju socijalni uticaj turizma?
Sazetak: Turizam igra vaznu ulogu u ekonomskom i kulturnom razvoju drustva, generisuci mnoge pozitivne i negativne efekte. Cilj ovog rada je bio da: (a) proceni uticaj turizma na savremeni zivot u Crnoj Gori na osnovu stavova rezidenata i (b) standardizuje instrument (upitnik) za procenu ovih uticaja. Uzorak je sacinjen od 521. punoletnog stanovnika Crne Gore iz tri razlicite turisticke regije (primorska, kontinentalna, planinska). Upitnik ima visoku validnost i pouzdanost. Izvucene su dve komponente (dve nezavisne skale) -negativni i pozitivni socijalni uticaj turizma. Utvrdeno je da porast pozitivnih uticaja prati i porast negativnih efekata. Tip regije, nivo opstinske razvijenosti i angazovanje ispitanika u turizmu znacajno uticu na razlike izmedu stavova ispitanika. Oni koji zive od turizma najvise vrednuju njegovu vaznost za lokalni razvoj.
Kljucne reci: turizam, socijalni uticaj, stav rezidenata, upitnik, Crna Gora JEL klasifikacija: Z32
srdjan.milosevic@tims.edu.rs
k^^Hjl^H This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http ://creativecommons .org/licenses/by/4.0/).
*
1. Introduction
The effects of tourism are predominantly linked with economic indicators (García et al., 2015; Liu & Wu, 2019) rather than with cultural and ecological changes, which results in social life of the local population being much richer in content (Coban & Yildiz, 2019). The measuring of effects of tourist development provokes a number of dilemmas. The economic development does not automatically lead to the satisfaction of the local population (Kim et al., 2013). For example, the income revenue grows due to the multiple increase in the volume of traffic, however the traffic jams and problems with parking increase simultaneously (Andereck & Vogt, 2000; Kuvan & Akan, 2005). New tourist facilities are built, but the amount of waste grows as well (Kuvan & Akan, 2005; Vargas-Sanchez et al., 2011). The number of tourists is increasing but safety is decreasing (Andereck et al., 2005; Haralambopoulos & Pizam, 1996). A different perception of tourism by the state institutions and by the residents widens this gap (Antonakakis et al., 2015). The ministry or the municipality measures the effects of tourism primarily by the economic results and presents them through numerical data (the number of tourists, the number of nights at the location, charged tax). Residents evaluate the effect oftourism very subjectively (García et al., 2015). For them, the most important factor is to not be disturbed by tourism, that their previous habits are not significantly impacted, that they do not come in conflict with tourists (Coban & Yildiz, 2019), that there are no issues with crowdedness and parking; simply, that their quality of life does not decrease (Boley & McGehee, 2014; Boley et al., 2014). The opinions of residents are of great importance for the objective evaluation of the impact of tourism industry (Joo et al., 2019; Lundberg, 2017), which is why they frequently serve as research target. Previous research studies have identified the positive and negative effects of tourism (Ap, 1992; García et al., 2015; Ko & Stewart, 2002; Lankford & Howard, 1994). Residents mostly recognize the economic benefits (primarily the increase in employment) as the main positive impact oftourism (García et al., 2015). Observations have been recorded regarding the positive impact on socio-cultural aspects of life, primarily on the improvement of services offered by the community (Andereck et al., 2005; Kim et al., 2013). The residents show interest in preserving their cultural heritage and lifestyle, which strengthens national pride and cultural identity (Andereck et al., 2005). Furthermore, the examinees also perceive the increase of traffic density and problems with parking as the most significant negative influences (García et al., 2015; Ko & Stewart, 2002). The residents queried in other studies note the increase in delinquency and vandalism as a negative consequence of tourism development (Andereck et al., 2005; Haralambopoulos & Pizam, 1996; Lankford & Howard, 1994). The analyses of ecological aspects oftourism also indicate both negative and positive effects. Residents gave a positive evaluation of the protection of natural resources, while environmental pollution and creating of a large quantity of waste were identified as negative influences ofmodern tourism (McGehee & Andereck, 2004; Teye et al., 2002). A significant section of tourist experiences was conditioned by supporting activities that do not exclusively depend on the direct participants in tourism industry (Ap, 1992; Lankford & Howard, 1994). This conditionality is seen in all those activities taking place in the everyday life ofresidents. The manner in which the local population treats the development of tourism is in direct correlation with the success of tourism at a certain destination (Andereck et al., 2005; García et al., 2015; Ko & Stewart, 2002). Considering that the integral product of a tourist destination is unimaginable without the local inhabitants (Cooper & Hall, 2008; Kripendorf, 1982; Laws, 1995; Leiper, 1989), all future efforts that have as their goal the evaluation of tourism impact, must seriously take into consideration the opinions of residents. The attitudes regarding the social impact of tourism are the starting point in overviewing the potential of tourism (Diedrich & García, 2009; Long et al., 1990; Vargas-Sanchez et al., 2011; Yoon et al., 1999), and can surely serve in defining developmental
strategies of countries that strive for competitive positions. Therefore, this study aimed to (a) assess the impact of tourism on modern life in Montenegro based on the attitudes of the residents and (b) standardize the instrument (questionnaire) for assessing these impacts. In defining the research subject, the assumption that tourism in the Republic of Montenegro would significantly contribute to the prosperity of the population is taken as the starting point. The research subject consists of the opinions of the local population regarding the evaluation oftourism impact on the entire social and economic development ofMontenegro. The main data sources used in the analysis are the opinions and attitudes of the residents rather than the official statistical data regarding the economic effects. Specific characteristics of residents and destination complexity required the construction of a suitable instrument for data collecting (which is the secondary aim of this study).
2. Case study region
Montenegro is a relatively small country, but it has favorable geographical position and rich natural resources for tourism development. It is located on the Balkan Peninsula and is connected with the Adriatic Sea. According to the last population census from 2011, Montenegro has 621 810 inhabitants living in 21 municipalities and 1 256 settlements. Tourism in Montenegro is developed in three geographically different areas that are tre ated in official documents (Ministry of Tourism and Environmental Protection of Montenegro, 2007; Ministry of Sustainable Development and Tourism of Montenegro, 2014) as three regions - coastal region (southern region), central (continental) region and mountain region (northern region) (Table 1).
Ta ble 1: Sample by defined criteria
Crite ria Group %of respondents Respondents by municipalities Group definition crite ria
Region Coastal 34% Bar, Budva, Herceg Novi, Kotor, Tivat, Ulcinj Ministry of Tourism and Environmental Protection, 2007; Ministry of Sustainable Development and Tourism, 2014.
Continental 39% Cetinje, Danilovgrad, Niksic, Podgorica
Mountain 27% Andrijevica, Berane, Bijelo Polje, Kolasin, Mojkovac, Plav, Pljevlja, Pluzine, Rozaj, Zabljak
Tourism level High 20% Bar, Budva, Herceg Novi, Ulcinj Statistical Office of Montenegro -MONSTAT, 2014.
Medium 47% Cetinje, Kolasin, Kotor, Podgorica, Tivat, Zabljak
Low 33% Andrejevica, Berane, Bijelo Polje, Danilovgrad, Mojkovac, Niksic, Plav, Pljevlja, Pluzine, Rozaje
Engagement Professionals 31% All municipalities Instrument (questionnaire)
No engagement 49%
Season 21%
Gender Male 45% All municipalities Instrument (questionnaire)
Female 55%
Source: Author's research
What is characteristic ofthese regions is their large diversity of attractions, which contributes to the development of numerous forms of tourism (Milosevic, 2017). According to official data, year after year, tourism in Montenegro is becoming increasingly important in overall economic development. The total contribution of Travel and Tourism to GDP was EUR988.2mn, 23.7% of GDP in 2017 and is forecast to rise by 8.9% in 2018, and to rise by 3.9% pa to EUR1,582.3mn, 27.9% of GDP in 2028. The direct contribution of Travel and Tourism to GDP was EUR459.1mn, 11.0% of total GDP in 2017 and is forecast to rise by 9% in 2018, and to rise by 4.2% pa, from 2018-2028, to EUR752.6mn, 13.3% of total GDP in 2028 (WTTC, 2018, p. 7). The importance of tourism is primarily seen in creating new workplaces. Increase in employment can be noted not only within the basic tourism sector (hotels and restaurants), but also in the supporting activities sector. In 2017 Travel and Tourism directly supported 14,500 jobs (7.6% of total employment). This is expected to rise by 4.5% in 2018 and rise by 1.1% to 17,000 jobs (8.1% of total employment) in 2028. In 2017, the total contribution of Travel & Tourism to employment, including jobs indirectly supported by the industry was 19.3% oftotal employment (36,500 jobs). This is expected to rise by 7.7% in 2019 to 39,000 jobs and rise by 1.3% to 45,000 jobs by the year 2028 (WTTC, 2018, p. 8). During 2019, 2,510 million tourists visited Montenegro, which shows 20.8% growth compared to the previous year (UNWTO, 2020).
3. Materials and methods 3.1. Sample
The study included 521 adults (232 male and 289 female) who live and work on the territory of Montenegro. The number of examinees was proportionate to the size of the region. Therefore, the largest number of examinees originated from the continental region (N2=203), without a doubt the largest region, followed by the coastal region, which is at the same time the most developed in terms oftourism (Nj=177), while the lowest number of examinees was from the mountain region (N3=141). All respondents were familiar with the research aim and participated in the survey voluntarily. According to the current statistical criteria used to determine sample size (Creative Research System, 1982), this number of examinees is on the level of significance of 0.05 (Confidence Level = 95%)), which enabled conclusions to be drawn with confidence interval (ConfidenceInterval) of 4.29, and where a number of adult residents ofthe Republic ofMontenegro were used as the basic set (Population) from which the sample was drawn. The acquired confidence interval (±4.29) can be considered acceptable for this type of demographic study. By inspecting the education structure of the examinees, results have shown that most ofthe examinees (49.1%) had high education, and then secondary education (24.6%). Regarding the employment status, the majority of the examinees had permanent employment, then part-time employment, while the lowest number was made up of students. The sample included almost all social stratums in terms ofthe most significant socio-demographic criteria: students, unemployed, and those retired. The representation of all stratums was proportionate to their total number represented in the entire Montenegrin society. Apart from the residential status, gender, age, and education level, data regarding employment were collected as relevant for this research, data regarding the significance of tourism to the everyday life of the examinees, as well as their ability to impact societal flows, primarily those connected to tourism. Among the examinees, most were not employed in tourism (48.6%), while those that regularly conduct tourism related affairs made up less than 1/3 of the sample (30.5%). To the question regarding the importance of tourism in ensuring material existence, almost half of the examinees responded as not depending on tourism (49.5%), while slightly more than a quarter (27.3%) stated that tourism is only additional activity used to improve their financial status. Only
5.6% of the examinees stated that they live off tourism, while a bit less than a fifth of the sample (17.7%) stated that they depend on tourism significantly.
3.2. Instrument design
The initial questionnaire was formed from the claims related to various aspects of tourism that were applied in previous papers researching similar problems. The first part of the instrument contains the relevant socio-demographic data of the respondents: gender, age, education level, employment status, engagement in tourism, knowledge of the municipal economy, importance oftourism for their financial status and the possibility of influencing decision-making in the place of residence (Andereck & Vogt, 2000; Boley & McGehee, 2014; Ko & Stewart, 2002; Kuvan & Akan, 2005; Lankford & Howard, 1994; Teye et al., 2002). The second part ofthe instrument was formed from 30 claims used by the examinees to state their opinion on a five-point Likert-type scale. After comparative content analysis of the most frequent questions, 30 claims/items were selected for this study that comprised the initial questionnaire (Table 2).
Table 2: Results of Scale reliability analysis for initial questionnaire of 30 items
No Statements / Variables Cronbach's Alpha if Item Deleted
1. In my municipality, development of tourism is insufficiently encouraged 0.840
2. Politicians do not work sufficiently on tourism promoting of my municipality 0.843
3. My municipality can become an attractive tourism destination 0.835
4. Tourism development contributes to gaining reputation of my municipality 0.831
5. Tourism development increases traffic problems, pollution and noise 0.846
6. Investment in tourism development is the only safe investment of my municipality 0.832
7. Tourism development offers numerous possibilities for resident employment 0.829
8. Tourism development is an important diplomatic activity 0.829
9. Tourism will have a major economic role at my municipality in the future 0.831
10. Government incentives for tourism development are insufficient 0.854
11. Tourism development increases the crime rate in my municipality 0.843
12. Tourists negatively impact the lifestyle in my community 0.841
13. Tourism development will secure more parks and recreational spaces 0.834
14. Only a small number ofresidents at this municipality have the benefits of tourism 0.841
15. Tourism development in my municipality will attract investors and spending 0.831
16. The living standard will significantly increase by developing tourism 0.827
17. Tourism strengthens the image about my town in the country and the world 0.828
18. Tourism development ensures high standard of roads and public facilities 0.830
19. Tourism development incentivizes the restauration ofhistorical places 0.830
20. The importance of tourism is underestimated in our country 0.853
21. Tourism significantly increases the tax revenues of the municipality 0.836
22. Tourism can be the cause of changes in the traditional culture of the municipality 0.851
23. Tourism development should be a priority in my municipality and country 0.829
24. Tourism development contributes to good international relations 0.830
25. My municipality and state would collapse without tourism 0.838
26. The benefits of tourism outweigh the negative impacts 0.835
27. I support the building of new tourist facilities that will attract more tourists 0.834
28. My municipality has become overcrowded with tourists 0.853
29. Tourism is the best ambassador of my city and country 0.831
30. Tourism development is a chance to exit anonymity 0.830
Cronbach's Alpha 0.860
Source: Author's research
The key criteria for the selection of the aforementioned claims from previous studies were related to the: (1) evaluation of tourism impact based on opinions of the residents; (2) representation of the claims that had dominant sociological and economic approach; (3) statistically proven instrument reliability; (4) publishing of studies in referential journals; (5) sample and climate specificities taking into consideration the level of tourism development of Montenegro and the historical burden that it carries as one of the youngest countries in Europe. During the selection, priority was placed on the universal claims (regarding employment, income, destination preservation, safety) excluding certain claims that were not adequate to the context of Montenegro (e.g. statement about mass tourism, as well as claims related to highly developed destinations). Special attention was given to the claims used to evaluate the benefits oftourism for building a positive destination image. Those claims (No. 4, 8, 17, 24, 29, 30) were directly taken from previous studies that had as their aim to research destination attractiveness based on tourism development (Boley & McGehee, 2014; Boley et al., 2014; Kuvan & Akan, 2005; Lee, 2016; Stylidis et al., 2014). The remaining claims included in the sample were adapted to the diversity of the geographical area of Montenegro, which reflects the experience of the residents in the field of tourism engagement, as well as the level oftourism development of the destination. Therefore, the instrument contains claims that can be used to evaluate the primary economic dimension of tourism impact (No, 6, 7, 9, 15, 21), and which have been taken from previous studies (Andereck & Vogt, 2000; Ko & Stewart, 2002; Lankford & Howard, 1994). The economic dimension is the main reason for positive attitudes of the residents. Most of claims represented in the instrument (No, 11, 12, 13, 14, 18, 19, 22, 25, 27, 28) evaluate the socio-cultural dimension of tourism impact (Andereck & Vogt, 2000; Kuvan & Akan, 2005; Lankford & Howard, 1994; Teye et al., 2002). Tourism has an effect on local sociocultural characteristics at different moments in the lives of residents, threatening their cultural identity and social reality (García et al., 2015; Stylidis et al., 2014). The influences that were noted on the environment, such as problems in traffic, then pollution, noise, etc., were evaluated by using claims from previous proven studies (Ko & Stewart, 2002; Lankford & Howard, 1994). Also, the attitudes ofresidents towards tourism represented in the paper (No. 1, 2, 3, 10, 16, 20, 23) have been assessed in several previous studies (Andereck & Vogt, 2000; Haralambopoulos & Pizam, 1996; Lankford & Howard, 1994; Stylidis et al., 2014). The attitudes ofresidents towards tourism include general support of residents for further development oftourism, financing tourism development, increase in tourist volume, etc. (Ko & Stewart, 2002; Latkova & Vogt, 2012; McGehee & Andereck, 2004).
3.3. Procedures
The questionnaire that collected the data was completed in two ways: online version or classically, using the paper-pencil model. The survey was anonymous. Only fully completed questionnaires were taken for the final analysis. The survey included participants who: (1) are employed in tourism (professionals) (2) are not primarily engaged in tourism (seasonal), (3) do not work in tourism and have no experience in the tourism business (no engagement). The questionnaires were sent to the first group via e-mail. Online distribution was done by announcing the sending of a questionnaire by telephone or e-mail. The questionnaire was sent to employees of all registered public, private and civil organizations that carry out tourism-related activities in the territories ofthe municipalities covered by the survey. The Central Register ofEconomic Entities ofMontenegro and direct contacts with local tourist organizations were used to search for e-mail addresses. Participants in the other two groups filled out the questionnaire by hand. Nine interviewers have distributed the questionnaires to participants. They contacted the participants and gave them sufficient time to answer all survey questions (no more than two weeks). Interviewers contacted residents on the street or in restaurants. Participants doing seasonal jobs in tourism (private accommodation, animators, beach bar, and season workers) were surveyed at workplaces. The survey was voluntary and completed by anyone who wished to do so. The only condition for the participants was to have a residence address at the municipality where the study was conducted. The questionnaire was available in the Serbian (Montenegrin) language. Participants completed the questionnaire electronically or in hard copy and expressed their opinion of each claim by selecting the proper position on a five-point Likert type scale. Position 1 marked the lowest, and position 5 the highest level of agreement. The scale reliability analysis confirmed that the questionnaire has good internal concordance in view of the fact that the Cronbach's Alpha was higher than the theoretically recommended value of 0,7 (DeVellis, 2003). In the first phase, participants from the coastal region (subsample Nj) completed the initial questionnaire of 30 items. Explorative factorial analysis confirmed good metric characteristics for only 20 items. As the remaining 10 statements (items 6, 9, 10, 14, 20, 21, 22, 25, 26) did not have good metrics, they were excluded from the questionnaire. The largest number of variables were left out due to commonality lower than 0.3, and a smaller number was left out due to simultaneous saturation ofboth extracted factors (items 6, 9, 27). Three variables (items 10, 14, 20) were left out because they did not have significant correlation (r) with any factor (r<0.3). Items 6, 20 and 25 contain an extreme and insufficiently clear assertion, while items 9, 22 and 27 are related to the assessment of uncertain effects oftourism in the future. The four other excluded variables (items 10, 14, 21, 26) require precise expert data that residents usually do not have. Such formulations have created uncertainty in the examinees regarding the selection of adequate responses and probably caused poor metric characteristics of excluded items. The reduced questionnaire of 20 items was tested by confirmative factorial analysis conducted on two new sub-samples -the residents from continental region (sub-sample N2) and the residents from mountain region (sub-sample N3). All three separate factor analyses (one explorative and two confirmative) gave very similar matrices with 2 factors (Table 3).
Table 3: Oblimin rotation factor loadings of the explorative PCA (Coastal region) and two _confirmative PCA (Continental and Mountain region)_
Loading on (Pattern Matrix)
Coastal region Continental region Mountain region
Variable Factor 1 Factor 2 Factor 1 Factor 2 Factor 1 Factor 2
S17 0.801* -0.090 0.854* -0.094 0.726* -0.029
S18 0.749* -0.080 0.806* -0.052 0.650* 0.032
S16 0.739* -0.127 0.748* 0.095 0.679* -0.181
S19 0.721* -0.041 0.723* -0.134 0.658* -0.048
S23 0.713* -0.072 0.638* -0.005 0.688* -0.156
S8 0.711* 0.009 0.732* 0.076 0.699* -0.029
S29 0.702* 0.035 0.692* -0.047 0.559* 0.197
S7 0.696* -0.046 0.719* 0.088 0.620* -0.126
S30 0.686* 0.031 0.716* -0.041 0.700* 0.032
S24 0.675* 0.088 0.782* -0.142 0.599* 0.141
S4 0.637* 0.076 0.673* -0.046 0.586* 0.093
S15 0.633* 0.043 0.765* -0.074 0.577* -0.148
S13 0.564* -0.061 0.639* 0.041 0.422* 0.033
S3 0.486* 0.126 0.464* 0.230 0.488* 0.062
S1 0.188 0.746* 0.274 0.460* 0.292 0.378*
S2 0.075 0.643* 0.256 0.451* 0.294 0.568*
S28 0.085 0.604* 0.049 0.642* 0.047 0.518*
S11 -0.189 0.572* -0.104 0.616* -0.262 0.451*
S12 -0.109 0.546* -0.298 0.682* -0.098 0.702*
S5 -0.062 0.546* -0.169 0.447* -0.045 0.583*
Note: * Significant coefficients of correlations between the variables and the factors; S -
s tatement
Source: Author's research
The final 20-item instrument validation was conducted on a unique sample (Ntotal=N1+N2+N3=521). Norms (averages) for both factors were calculated for total sample and also for specific sub-samples (according to gender, the level of tourism development in the municipality and engagement of the residents in tourism).
3.4. Statistical analysis
The data gathered were processed using descriptive and comparative statistical procedures. Validity of multi-items questionnaire was assessed by Factor analysis (model of Principal Components Analysis - PCA), with Direct Oblimin method of rotation and Kaiser Normalization. Descriptives (Mean and Std. Deviation) were calculated according to scalar values used by the examinees to express their opinion regarding individual claims from the questionnaire. For testing the significance of differences between arithmetic means gained on specific subsamples, One-Way ANOVA was used (for testing the differences between geographic regions, between groups with various degree of tourism development and residents with various tourism engagement) and T-test for independent samples (when comparing average scalar values ofmale and female). All conclusions were realized on 0.05 level of significance (p<0.05). Portable IBM SPSS v.21 application (License Stats Prem: 761b17dcfd1bf20da576 by Hearne software) was used for complete statistical analysis.
4. Results
4.1. Factorial validity of the questionnaire
For the purpose of explaining the latent structure of the purified 20-item questionnaire, factorial analysis of the main components (PCA) was conducted. The explanation of the main components was preceded by the evaluation of data suitability for the factorial analysis. By inspecting correlational matrix gained from the initial sample (N1=177), many coefficients of 0.3 value and higher have been recorded. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) was 0.902, which significantly exceeded the 0.6 value recommended by Kaiser (1970; 1974). The Bartlett's test of sphericity (Bartlett, 1954) also indicated statistical significance of the gained factorial model (Chi-Square=1457.833; Sig.<0.001). These data have indicated that the initial correlational matrix has good factorability. Main components analysis gained after oblimin rotation has revealed the presence of five components with Eiegenvalues over 1.
Figure 1: Scree plot for results obtained in initial subsample (Ni=177)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Source: Author's research
The Scree plot (Figure 1) indicated that the breaking point is located after the third component. Based on the Cattell (1966) criterion, it was decided to retain only two components that were above the scree point. This decision was supported by the results of a parallel two-component analysis, the characteristic values ofwhich exceed the corresponding values of the statistic threshold (Watkins, 2000) gained using equally large random numbers matrix (20 variables x 177 subjects). This two-component solution explained the total of 44.957% of the variance, where the first component contributed with 33.404% and the second with 11.553%. All the communalities were over 0.3, which meets the recommended statistical criterion (Pallant, 2013; Thurstone, 1947) significant for a variable to be retained in the system. Following the oblimin rotation, each of the 20 variables had substantial factorial significance for only one of the two main components. The same PCA procedure was repeated in two additional confirmative analyses, one of which was realized on a sub -sample ofthe continental (N=203), and the other on a sub-sample of the mountain region (N=141). KMO and Bartlett's test confirmed high factorability of correlational matrices in both cases (in group Continental region, the values were: KM0=0.874, Chi-Square=2011,514, Sig.<0.001; in Group Mountain region, the values were: KM0=0.781, Chi-Square=1019.363, Sig.<0.001). In the Continental region group, two-component
solution explained the total of 47.292% of the variance (the contribution of the first component was 37.287%, and the second 10.005%,), while in the Mountain region 38.498% of the variance was explained (contribution of the first component was 28.816%, and the second 9.682%). In both confirmative analyses, all commonalities were higher than 0.3, which confirmed a significant contribution of all 20 items to explaining the total variability. Considering the great similarity of correlation matrices acquired through the explorative and two confirmative factorial analyzes, merging of the three groups of examinees (N1, N2, N3) was conducted in the final phase of instrument validation, and the PCA procedure was realized on the complete sample (Ntotal=521). The data definitely confirmed high factorability of the two-component solution (KMO=0.89; Chi-Square=4276.297; Sig.<0.001) and determined the hierarchical value of the variables significant for explaining the extracted factors. Once again, the same variables saturated the first and second factor. The communalities of all 20 variables were statistically significant. Two components of the final matrix explained the 45.95% of the total variance together, where the first component contributed with 33.174% and the second with 12.776%. The factor of positive impact retained the first hierarchical position. This indicates that the majority ofthe residents clearly recognizes the positivity coming from the tourism industry, and that that positive features better explain the variability between the opinions and the attitudes of the examinees. The most significant for the final validation of the questionnaire were the data regarding the structure of extracted components (Table 3). In both Pattern matrices gained from the two sub-samples clearly show that the same 14 variables that saturated the first factor have been abstracted (statements: 17, 18, 16, 19, 23, 8, 29, 7, 30, 24, 4, 15, 13, 3). At the same time, the six remaining variables (statements: 1, 2, 28, 11, 12, 5) statistically influenced only the formation ofthe second factor. The hierarchal relation ofthe factor was the same in all three matrices. Minor differences found were related to the order of certain variables within the same factor. All three factorial analyses (explorative and two confirmative) resulted in very low inter-factorial correlation coefficients (coastal region: r=-0.066; continental region: r=0.12; mountain region: r=0.047). This indicates that the factors gained are relatively independent, meaning that applied questionnaire contains two scales that can be independently used in similar research studies. By analyzing the content ofthe 14 statements that saturated the first factor, it can be noted that they predominantly refer to the positive effects of tourism (promotion of municipality and the country in the world, economic progress, building ofroad). The first factor was labeled as Positive Social Impact of Tourism (PSIT). The remaining 6 statements indicate negative phenomena that accompany the development of tourism (crowds and jeopardizing the environment due to an increased number oftourists, rise in the crime rate, showing personal weaknesses). The second factor was labeled as Negative Social Impact of Tourism (NSIT).
4.2. Normative scale data
It was noticed from the results that the positive social impact of tourism in all groups was significantly higher than negative (Table 4).
Table 4: Positive (PSIT) and negative (NSIT) social impact of tourism - Scale Means and _Standard deviation for different groups (sub-samples)_
PSIT Scale NSIT Scale
N Mean Std. Dev. Mean Std. Dev.
Region
Coastal 177 4.224 0.654 2.932 0.769
Continental 203 3.967 0.741 2.231 0.573
Mountain 141 4.030 0.651 1.967 0.594
ANOVA F=6.979 Sig. = 0.001 F=97.062 Sig.<0.001
Tourism level
High 106 4.329 0.580 2.833 0.732
Medium 243 4.024 0.751 2.430 0.766
Low 172 3.980 0.644 2.083 0.625
ANOVA F=9.662 Sig.=0.000 F=36.537 Sig.<0.00l
Engagement
Professionals 159 4.232 0.677 2.610 0.806
No engagement 253 3.942 0.696 2.249 0.693
Seasonal 109 4.136 0.671 2.433 0.780
ANOVA F=9.323 Sig.=,000 F=ll.544 Sig.<0.00l
Gender
Male 232 4.108 0.669 2.346 0.796
Female 289 4.042 0.717 2.439 0.733
T-test t=l.07l Sig.=0.285 t=-l.379 Sig. = 0.l68
Total 521 4.071 0.696 2.398 0.762
Source: Author's research
Both on the level ofthe total sample, and within each specific subsample, the average scalar value for PSIT was over 4, while for NSIT was always below 3. This, without a doubt, is a proofthat the residents recognize tourism as a significant factor of general development. Most examinees gave significant advantage to the positive effects of tourism. However, what is important is that the higher marks for PSIT were regularly accompanied by higher marks for NSIT. This indicates that the residents are aware that by developing tourism, the chances for the negative social phenomena thatjeopardize the quality of life in the community to be manifested also increase. The average scalar values gained for both factors (PSIT and NSIT) showed a statistically significant difference in relation to the region, degree of development and engagement of the examinees in tourism. Significant differences between the scalar averages of men and women were lacking in both factors (Table 4). Positive impacts of tourism were recognized in the highest degree by the examinees professionally engaged in tourism, followed by those from the most touristically developed municipalities, as well as those living in the coastal region. This finding is logical considering the fact that tourism in Montenegro is the most developed precisely in the municipalities of the coastal region. The data indicating the lowest scalar averages recorded in the municipalities where tourism is not developed (level low) are in accordance with this. Post Hoc Tests (Tukey HSD) indicated that the differences between the specific groups were more pronounced regarding the negative than the positive social impacts of tourism. All absolute differences between the values for the NSIT in various regions and municipalities with different level of tourism development were statistically significant. Post Hoc Tests conducted according to the criteria of examinee engagement as an only source of variation revealed significantly lower marks given by the examinees that are not professionally tied to tourism, both for PSIT and for NSIT. While overviewing PSIT, the main source ofvariability for various regions were the significantly higher marks given by the examinees from the coastal region accompanied by a lack of significant differences between the marks ofthe continental and the mountain region. When PSIT was analyzed in relation to the level of tourism development, the main source of variability were the significantly higher marks given by the examinees from the most developed municipalities. The marks for PSIT of the examinees from the municipalities with low or medium level of development did not show statistically significant difference. In summary, these data clearly indicate that tourism is most valued by those for whom it is a primary source of income. At the same time, they are the most aware of the dangers accompanied by the development of tourism.
5. Discussion
Taking the number ofthe examinees included in this study (N=521) as the starting point, the analyzed sample can be characterized as very representative. It is approximately 0.001% of the total electoral body of the Republic of Montenegro. If parallels were made with the largest European countries, this percentage would be equivalent to a sample of 45000 persons in Great Britain or France, and to as high as 60000 examinees in Germany. The opinions of the local population regarding tourism (to be more precise: regarding the perception oftourism impact) have been a subject ofresearch for over 30 years (Andereck & Vogt, 2000). Early studies that focused on the opinions of the residents were directed towards measuring the impact oftourism (Jafari, 1986). These studies would usually include a series of statements that referred to several types of tourism impact (Liu & Var, 1986). Most research studies discovered one or more positive impacts or one or more negative impacts oftourism (Andereck & Vogt, 2000). Many more studies have resulted in positive opinions regarding tourism as the population believed that tourism benefits the local community (Andereck & Vogt, 2000). In most studies (Boley & McGehee, 2014; Kim, et al., 2013; Lankford & Howard, 1994; McCool & Martin, 1994) the residents did not indicate being overly concerned with the negative impacts of tourism, with the exception of a study conducted by Johnson et al. (1994), where the results indicated that the population does not have a positive opinion regarding tourism and believes that tourism has a negative impact on their local community. The results ofprevious studies regarding negative social impact of tourism are mostly in accordance with the results of this paper. The population in Montenegro also did not show great concern for the negative impacts of tourism. In most previous studies, two-factor models were constructed by applying factorial analysis. One of these studies, which was used to define the questionnaire in this paper, was conducted by Lankford and Howard (1994). They are the authors of TIAS scale (Tourism Impact Attitude Scale) that was widely used in the following years. The aforementioned authors grouped all items into two factors that were named as follows: (1) care for the local tourism development (18 items) and (2) benefits for the individuals and the local community (9 items). Wang et al. (2006) also had a two-factor structure that has proven to be highly applicable for measuring the opinions of the local population regarding the impact of tourism. These two factors explained 51% of the variance in the opinions towards the development of tourism, which is very close to the amount of variability explained in this paper. Woosnam (2012) also extracted two factors, first of which he named supporting tourism development, and the second - contribution to the community. Of course, there are similar research studies where more than two factors have been identified. However, what they all have in common is that the authors evaluated the factors through the positive and negative influence on the local community. So, Andereck and Vogt (2000) conducted a research study in local communities in the USA which resulted in a three-factor model: community development, negative impact, and the quality of life. The study of Latkova and Vogt (2012) also resulted in three factors, marking them as: personal benefit from tourism (2 items), positive impact (12 items), and negative impact (8 items). Boley et al. (2014) also had 7 factors in a study conducted in Virginia (USA). Their factors were labeled as follows: psychological empowerment, social empowerment, political empowerment, personal economic benefits oftourism, support from tourism, positive impact, and negative impact. The opinions of the local population towards the social impact oftourism in Montenegro were analyzed in relation to 4 criteria: sex of the examinees, the development level oftourism in the municipality, region of the municipality, level of examinee engagement in the field oftourism (Table 1). Only a significant influence of sex was lacking, while the remaining three predictors had an impact on significant differences with both extracted factors (Table 4). Such findings indicate without a doubt that the examinees from touristically developed municipalities gave higher marks on average
compared with the examinees from the municipalities where tourism is less developed. What is significant is that the examinees from touristically developed municipalities have shown a lower level of concern for the negative social impact oftourism. These findings confirm the conclusions ofprevious studies where the authors (Butler, 1980; Johnson et al., 1994; Yoon et al., 1999) stated that the opinions ofthe residents depend on the condition (development phase) ofa tourism destination from which the examinees originate. However, studies can be found in which the level of tourism development did not have a significant impact on forming the opinions of domicile population regarding tourism (Andriotis & Vaughan, 2003). In a commentary ofthese results García et al. (2015) note that they were obtained on Cypress, which is a mature tourist destination, and where most of the population lives off tourism, which is why most of them provide strong support for the dev elopment of tourism without serious criticism. For this reason, these results cannot be used as a rule for all destinations. In favor of these arguments are the results of most newer and older studies that state that the local population, regardless of the development level of their residing destination, always recognizes both positive and negative social impacts of tourism. At the same time, in most ofthe queried individuals there is a noticeable growth in the perception of positive and negative social impacts accompanied by the growth of tourism development level (Diedrich & García, 2009; Long et al., 1990). The findings of our study fully support the aforementioned observations ofprevious research studies. It is clear that the development of tourism in the local community is accompanied by a change in the inhabitants' perception regarding its positive and negative impacts. Positive impact of tourism is most noted by those that live in touristically developed municipalities, and as far as Montenegro is concerned those are the destinations from the coastal (touristically most developed) region. The inhabitants of the touristically most developed municipalities recognize the benefits of tourism industry the most, which is why, in time, they become increasingly more tolerant to the negative impacts of tourism. Of course, this phenomenon explains why statistically significant lower marks for the second factor (NSIT) were recorded from the inhabitants of the leading tourist municipalities. The results of our study have shown that the engagement of the examinees in the field of tourism was statistically significant for the height of the average grades given by the examinees from various groups to the negative and positive impacts oftourism. Examinees who do not conduct business related to tourism, marked the first factor (PSIT) with lower grades on average than those who are occasionally or professionally engaged in tourism. Practically speaking, the examinees who value tourism less show a lower awareness of its positive impacts on the local community, while on the contrary, those who deal with tourism for a living evaluate it with higher marks. Even though the average grades for the second factor (NSIT) did not show a statistically significant difference, it can be noted that the scalar averages of the examinees who are professionally tied to tourism are slightly higher than the average grades of the examinees who are not engaged in tourism. This indicates that those who use tourism for a living are aware that tourism does not only have positive but also negative impacts on the society. This observation fully corresponds with the results of previous research studies (Andereck et al., 2005; Bujosa & Rosselló, 2007; Davis et al., 1988; García et al., 2015; Lankford & Howard, 1994) where it was determined that the examinees engaged in tourism related occupations have a better perception ofits impact than those who are not directly engaged, as well as that they have more positive opinions regarding its development.
6. Conclusion
Based on the instruments used in previous research studies dealing with social impacts of tourism, a questionnaire suitable for the assessment of the opinions of the local population regarding tourism was constructed. The sample includes the residents of all municipalities of
the Republic of Montenegro, both those touristically developed and non-developed. The authors made sure that those persons professionally engaged in tourism were included amongst the examinees, as well as those who are not involved in tourism business. The starting point was the presupposition that the integral product of a tourism destination is unimaginable without the local residents and that the social impact of tourism cannot be overviewed without a serious analysis of residents' opinions. Their contribution to the reputation ofthe country is priceless, precisely because the residents establish direct and indirect contacts with tourists, and thus, represent themselves and the culture of their community. By analyzing the data gathered using the questionnaire constructed during this study, two factors were defined that can be used to explain the social impact of tourism -positive and negative. The results have indicated that the level of tourism development, as well as the degree of examinee engagement in tourism (active or passive) significantly impacts the creation of the residents' opinions. The residents of the municipalities where tourism is highly developed, as well as those that are actively involved in tourism industry are better at recognizing the positive impact. The economic effects were recognized as the most significant ones. In parallel with recognizing the positive impacts, the examinees clearly show awareness of the potential negative impacts accompanying tourism development, amongst which nuisances revolving around traffic and the fear of increased crime rate occupy the most prominent position. As the specific opinions of the residents constantly change under the influence of tourism development, it would be advisable to repeat this research after a certain time period in the same (already observed) municipalities. Repetition of this research would be of particular importance in the municipalities where significant changes in tourism activities are taking place. Theoretically, this research supports various studies that had a similar goal in focus. The results obtained provide useful in formation on impact as sessment by population. Montenegro is a "new" destination in the Mediterranean whose focus in the coming years will be on tourism development. The information on the reactions of residents to tourism development is the main practical contribution ofthe study. Residents in the early stages of development are very sensitive to the positive impact oftourism and have shown awareness of some of the negative impacts, even when do not live at a tourist destination. These facts are important for planning development policies and guidelines for choosing the best strategies and implementing them. Finally, although studies such as that presented here can provide useful information to tourism development organizations, additional insights can be gained using qualitative methods such as interviews with community residents and business people.
Conflict of interest
The authors declare no conflict of interest. References
1. Andereck, K. L., & Vogt, C. A. (2000). The relationship between residents' attitudes toward tourism and tourism development options. Journal of Travel Research, 39, 2736. https://doi.org/10.1177/004728750003900104
2. Andereck, K., Valentine, K., Knopf, R., & Vogt, C. (2005). Residents' perceptions of community tourism impacts. Annals of Tourism Research, 32(4), 1056-1076. https://doi.org/10.1016Zj.annals.2005.03.001
3. Andriotis, K., & Vaughan, R. (2003). Urban residents' attitudes toward tourism development: The case of Crete. Journal of Travel Research, 42, 172-185. https ://doi.org/10.1177/0047287503257488
4. Antonakakis, N., Dragouni, M., & Filis, G. (2015). How strong is the linkage between tour ism and economic growth in Europe? Economic Modelling, 44, 142-155. https://doi.org/10.1016/j.econmod.2014.10.018
5. Ap, J. (1992). Residents' perceptions on tourism impacts.Annals ofTourismResearch, 19(4), 665-690. https://doi.org/10.1016/0160-7383(92)90060-3
6. Bartlett, M. S. (1954). A note on the multiplying factors for various x2 approximations. Journal of the Royal Statistical Society, 16(2), 296-298.
7. Boley, B. B., & McGehee, N. (2014). Measuring empowerment: Developing and validating the resident empowerment through tourism scale (RETS). Tourism Management, 45, 85-94. https://doi.org/10.1016Zj.tourman.2014.04.003
8. Boley, B. B., McGehee, N., Perdue, R. R, & Long, P. (2014). Empowerment and resident attitudes toward tourism: Strengthening the theoretical foundation through a Weberian lens. Annals of Tourism Research, 49, 33-50. https://doi.org/10.1016/j.annals.2014.08.005
9. Rosselló, J., & Bujosa, A. (2007). Modelling environmental attitudes toward tourism. Tourism Management, 28(3), 688-695. https://doi.org/10.1016/j.tourman.2006.04.004
10. Butler, R. (1980). The concept of a tourist area life cycle of evolution: Implications for management of resources. Canadian Geographer, 19(1), 5-12. https ://doi.org/10.1111/j. 1541-0064.1980.tb00970.
11. Cattell, R., B. (1966). The scree testfor the number offactors. Multivariate Behavioral Research, 1 (2), 245-276. https://doi.org/10.1207/s 15327906mbr0102_10
12. Coban, G., & Yildiz, O. S. (2019). Developing a destination management model: Case of Cappadocia. Tourism Management Perspectives, 30, 117-128. https://doi.org/10.1016/j.tmp.2019.02.012
13. Cooper, C., & Hall, C. M. (2008). Contemporary tourism, an international approach. Oxford: BH Elsevier.
14. Davis, D., Allen, J., & Cosenza, R (1988). Segmenting local residents by their attitudes, interests, and opinions toward tourism Journal of Travel Research, 28(2), 2-8. https ://doi.org/10.1177/004728758802700201
15. DeVellis, R. F. (2003). Scale development: Theory and applications (2nd ed.). Thousand Oaks, California: Sage.
16. Diedrich, A., & García, E. (2009). Local perceptions of tourism as indicators of destination decline. Tourism Management, 30, 512-521. http s://doi.org/10.1016/j .tourman.2008.10.009
17. García, F., Balbuena, A., & Macías, R. (2015). Resident's attitudes towards the impacts of tourism. Tourism Management Perspectives, 13, 33-40. https://doi.org/10.1016/j.tmp.2014.11.002
18. Haralambopoulos, N., & Pizam, A. (1996). Perceived impacts of tourism: The case of Samos. Annals of Tourism Research, 23, 503-526. https://doi.org/10.1016/0160-7383(95)00075-5
19. Jafari, J. (1986). A systemic view of sociocultural dimensions of tourism. Literature review: The president's commission on American outdoors. Washington DC: Government Printing Office, 33-50.
20. Johnson, D. J., Snepenger, J. D., & Akis, S. (1994). Residents ' perceptions of tourism development. Annals ofTourism Research, 21, 629-642. https://doi.org/10.1016/0160-7383(94)90124-4
21. Joo, D., Cho, H., & Woosnam, K. M. (2019). Exploring tourists' perceptions of tourism impacts. Tourism Management Perspectives, 31, 231-235. https://doi.org/10.1016/j.tmp.2019.05.008
22. Kaiser, H. (1970). A second generation Little Jiffy. Psychometrika, 35, 401-415.
23. Kaiser, H. (1974). An index of factorial simplicity. Psychometrika, 39, 31-36.
24. Kim, K., Uysal, M., Joseph, M. S. (2013). How does tourism in a community impact the quality of life of community residents? Tourism Management, 36, 527-540. https://doi.org/10.1016/j.tourman.2012.09.005
25. Ko, D. W., & Stewart, W. P. (2002). A structural equation model ofresidents ' attitudes for tourism development. Tourism Management, 23, 521-530. https://doi.org/10.1016/S0261-5177(02)00006-7
26. Kripendorf, J. (1982). Towards new tourism policies: The importance of environmental and socio-cultural factors. Tourism Management, 3, 135-148. https://doi.org/10.1016/0261-5177(82)90063-2
27. Kuvan, Y., & Akan, P. (2005). Residents ' attitudes toward general and forest-related impacts of tourism: The case of Belek, Antalya. Tourism Management, 26, 691-706. http s://doi.org/ 10.1016/j .tourman.2004.02.019
28. Lankford, S., & Howard, D. (1994). Developing a tourism impact attitude scale. Annals of Tourism Research, 21, 121-139. https://doi.org/10.1016/0160-7383(94)90008-6
29. Látková, P., & Vogt, C. A. (2012). Residents ' attitudes toward existing and future tourism development in rural communities. Journal ofTravel Research, 51(1), 50-67. https ://doi.org/10.1177/0047287510394193
30. Laws, E. (1995). Tourism destination management. London: Routledge.
31. Lee, K. H. (2016). The conceptualization ofcountry attractiveness: A review ofreseaich. International Review of Administrative Sciences, 82(4), 807-826 https ://doi.org/10.1177/0020852314566002
32. Leiper, N. (1989). Main destination ratios: Analyses oftourist flows. Annals of Tourism Research, 16(4), 530-541. https://doi.org/10.1016/0160-7383(89)90007-8
33. Liu, J., & Var, T. (1986). Resident attitudes toward tourism impacts in Hawaii. Annals ofTourism Research, 13, 193-214. https://doi.org/10.1016/0160-7383(86)90037-X
34. Liu, A., & Wu, D. C. (2019). Tourism productivity and economic growth. Annals of Tourism Research, 76, 253-265. https://doi.org/10.1016Zj.annals.2019.04.005
35. Long, T., Perdue, R., & Allen, L. (1990). Rural resident tourism perceptions and attitudes by community level of tourism. Journal ofTravel Research, 28(3), 3-9. https ://doi.org/10.1177/004728759002800301
36. Lundberg, E. (2017). The importance of tourism impacts for different local resident groups: A case study of a Swedish seaside destination. Journal ofDestination Marketing & Management, 6(1), 46-55. https://doi.org/10.1016/jjdmm.2016.02.002
37. McCool, S. F., & Martin, S. (1994). Community attachment and attitudes toward tourism development. Journal of Travel Research, 32(3), 29-34. https ://doi.org/10.1177/004728759403200305
38. McGehee, N., & Andereck, K. (2004). Factors predicting rural residents ' support of tourism. Journal of Travel Research, 43(2), 131-140. https ://doi.org/10.1177/0047287504268234
39. Milosevic, S. (2017). Analysis of objective indicators of tourism development in Montenegro. Tims. Acta - Journal ofSport Sciences, Tourism and Wellness, 11(1), 3143. https://doi.org/10.5937/timsact11-12325
40. Ministarstvo odrzivog razvoja i turizma Crne Gore [Ministry of Sustainable Development and Tourism of Montenegro] (2014). Informacije o kvalitativnim i kvantitativnim efektima Ijetnje turisticke sezone [Information on the qualitative and quantitative effects ofthe summer tourist season]. Podgorica.
41. Ministarstvo turizma i zastite zivotne sredine Crne Gore [Ministry of Tourism and Environmental Protection of Montenegro] (2007). Nacionalna strategija odrzivog razvoja [National sustainable development strategy]. Podgorica.
42. Pallant, J. (2013). SPSS survival manual. New York: McGrow-Hill Education.
43. Stylidis, D., Biran, A., Sit, J., & Szivas, E. (2014). Residents' support for tourism development: The role of residents' place image and perceived tourism impacts. Tourism Management, 45, 260-274. https://doi.org/10.10167j.tourman.2014.05.006
44. Teye, V., Sonmez, F. S., & Sirakaya, E. (2002). Residents' attitudes toward tourism development. Annals ofTourismResearch, 29, 668-688. https://doi.org/10.1016/S0160-7383(01)00074-3
45. Thurstone, L. L. (1947). Multiple factor analysis. Chicago: University ofChicago press.
46. Vargas-Sanchez, A., Porras-Bueno, N., & Plaza-Mejia, M. D. (2011). Explaining residents' attitudes to tourism is a universal model possible? Annals of Tourism Research, 38, 460-480. https://doi.org/10.1016Zj.annals.2010.10.004
47. Wang, Y., Pfister, R. E., & Morais, D. (2006). Residents' attitudes toward tourism development: A case study of Washington, NC. In R Burns & K. Robinson (Eds.), Northeastern Recreation Research Symposium - Collection of Papers (pp. 411-419). NY: Bolton Landing.
48. Watkins, M. W. (2000). Monte Carlo PCA for parallel analysis [computer software]. State College, PA: Ed and Psych Associates.
49. Woosnam, K. M. (2012). Using emotional solidarity to explain residents' attitudes about tourism and tourism development. Journal of Travel Research, 51(3), 315-327. https ://doi.org/10.1177/0047287511410351
50. World Travel and Tourism Council (WTTC) (2018). Travel and tourism economic impact 2018Montenegro. London: World Travel and Tourism Council.
51. World Tourism Organization (UNWTO). (2020). UNWTO tourism highlights, 2020 edition. Madrid: UNWTO.
52. Yoon, Y., Gursoy, D., & Chen, J. (1999). An investigation of the relationship between tourism impacts and host communities' characteristics. International Journal ofTourism and Hospitality Research, 10(1), 29-44. https://doi.org/10.1080/13032917.1999.9686970
Received: 29 March 2021; Sent for revision: 29 April 2021; Accepted: 8 May 2021
