THE ROLE OF AGRITOURISM’S IMPACT ON THE LOCAL COMMUNITY IN A TRANSITIONAL SOCIETY: A REPORT FROM SERBIA*1

Abstract This study assessed local residents’ attitudes in Serbia toward the impact of agritourism in their surroundings, using a Tourism Impact Attitude Scale (TIAS). Till now, analysis of the impact of tourism on the attitudes of residents in rural areas of Serbia and other Balkan transitional countries is insuffi ciently researched. The analyzed items of the TIAS were grouped into four factors: personal and community benefi ts (grouped eight items); negative impacts (seven items); concern for the local tourism development (fi ve items); and general opinion about tourism development (three items). The factors explain 47.47% of the variance. Furthermore, the results showed that residents consider the possibility to have more money to spend as the most important impact of tourism development. It is followed by the support of local authorities to promote tourism development. The third relevant issue for the residents is related with encouragement of tourism in the local community. These are the key propositions to start an initiative for the local communities to actively participate in agritourism development. The results provide residents, tourism organizers and local authorities with important community perceptions pertaining to the agritourism’s impact.

larger' (р. 513). The change of the social, economic and political system in Serbia at the beginning of the new millennium also marks the start of a new phase in the development of agritourism. According to Todorović and Bjeljac (2009), this form should be one of the main carriers of tourism off er and a factor of integral development of rural areas in Serbia and the rest of the Balkans.
The main subject of this research is the examination of how agritourist activities aff ect rural surroundings in a typical rural, transitional society, from communism and socialism to the capitalist era. How do individual and/or community benefi ts correlate with support for tourism development? How do concerns for tourism development correlate with the general opinion about tourism? And how do impacts of tourism development correlate with the community support for tourism? Therefore, the general aspects of agritourism's impact, as well as its eff ect on the local residents, will be explained. The study has three objectives. The fi rst objective, drawing on the study of Lankford and Howard (1994), is to test the Tourism Impact Att itude Scale (TIAS), in order to examine the att itude of residents in the observed area. The second one is to examine if there are any diff erences in att itudes among residents, divided into groups according to the results of the exploratory factor analysis and to compare with previous similar studies. The third objective is to demonstrate relations between the obtained factors by using Pearson r statistical correlation.

Impacts of agritourism on local community
Globally speaking, agritourism is a very important and increasing segment of travel industry. Nickerson, Black and McCool (2001) stated that the rural area is the basic resource for the development of agritourism and that it relies on the city residents' need for peace and outdoor space for recreation. Agritourism is used more conventionally for notions that are related to products and services, which are directly related to the agrarian environment, agricultural products and types of farm-stay. Such activities involve staying in such environment, educational visits, recreational activities or selling agricultural and homemade products (Sznajder, Przezborska and Scrimgeour, 2009). As many other branches of tourism, agritourism has certain impacts on local surroundings. It brings numerous consequences on the environmental, economic, social and (even) psychological changes ( Table 1).
The eff ect of (agri)tourism activities and events on att itudes and behavior of the locals was explained in several research papers. Some of the studies showed that residents think tourism helps the local economy (Ritchie, 1988;Choi and Sirakaya, 2005), aff ects the general increase of the life standard of a community (Var and Kim, 1989;Choi and Sirakaya, 2005), but also stimulates the entrance of foreign currency in a host country (Ahmed and Krohn, 1992). Furthermore, the studies have shown that tourism directly stimulates the opening of new jobs (Milman and Pizam, 1988;Var and Kim, 1989; Ahmed and Krohn, 1992;Tosun, 2002) and increases the income of numerous segments of local economy (Sett ina and Richmond, 1978;Tosun, 2002). However, Var and Kim (1989) noted that tourism also leads to the employment of locals out of their (poor) communities. Nevertheless, numerous studies (Liu and Var, 1986;Weaver and Lawton, 2001;Andereck et al., 2005;Blešić et al., 2014b;Vujko and Gajić, 2014;Vujko and Plavša, 2014) emphasized the prevailing positive eff ects of tourism on life standard, employment and income. On the other hand, those activities are often the main ones 'to blame' for numerous social issues (the rate of poverty, crime, overconsumption of alcohol, prostitution, hazardous games, etc.), which could result in devastation of traditional culture, customs and the beliefs of the community (Ahmed and Krohn, 1992; Nunkoo and Ramkissoon, 2011) and even to the process of acculturation (Weaver and Lawton, 2001). Nevertheless, tourism considerably contributes to the building of numerous facilities for cultural, artistic and sport events (Liu and Var, 1986;Yu, Chancellor and Cole, 2011). In addition, the signifi cance of mutual infl uence between the local community and visitors frequently leads to progressive changes and development for both sides (Aref, Gill and Aref, 2010).
When it comes to the environmental issues, many authors stated numerous negative cases of the tourism's impact on the environmental pollution, such as devasta-tion of natural resources, noise, and damaging cultural heritage (Var and Kim, 1989;Ahmed and Krohn, 1992;Andereck and Vogt, 2000;Brankov, Jovičić and Milijašević, 2015;Srdanović and Pavić, 2015). Besides the mentioned impacts, the tourism economy results in great amounts of solid refuse, since hospitality facilities, traffi c companies and visitors themselves leave behind loads of rubbish, which is very often not easily degradable. The problem is far more obvious in developing countries, where the absence of sustainable refuse management and adequate recycling measures is extremely apparent (Andereck and Vogt, 2000). Even though in the recent years, tourism shows a 'protective' att itude towards the environment, the understanding of this issue is still quite limited. All the previous studies comprehensively explained that (agri)tourism is a relevant factor in positive, as well as negative changes in the local area and that it might aff ect the residents' behavior, att itudes and even lifestyle.

Study area, research procedure and sample
The analysis of this paper should point to the signifi cance of the impact of agritourism development on the life of local population in the selected village sett lements. The selection of the village sett lements has been done according to the recommendations of a National Project 'Wealth of Diversity', supported by the Provincial Government of Vojvodina in Northern Serbia. According to their established criteria 1 , the project 'Wealth of Diversity' has evaluated and made a list of the 17 most representative villages in Vojvodina Province with respect to agritourism. In this regard, the research involved only those villages which already have agritourist activities and amenities.
The main problem issues of the paper refer to the general aspects of agritourism's impact, as well as its eff ect on the residents. In this respect, the analysis of data was conveyed several times during 2014 (from September to December) and 2015 (from March to June), by using the Tourism Impact Att itude Scale (TIAS). All the interested respondents in the observed villages participated in the survey. The only condition was that their domicile address was in the researched villages. The poll was anonymous, i.e. the names of the examinees were not relevant for the selected data. The examination of the target groups was done with the technique 'face to face'. Their socio-demographic characteristics are shown in detail in Figure 1.
Of the 300 distributed questionnaires in total, 228 were fi lled correctly and used in the statistical procedure. This number represents 76% of the response rate. According to Babbie (1986), the response rate on the level of ≥70% is considered to be a good indicator of the measurement scale acceptance. The sample in this research (N ≥51) is adequate for meaningful statistical assessments (Bagozzi, 1981

Research instrument
In order to propose measurement solutions for the tourism impact eff ects, Lankford and Howard (1994) wrote a paper titled 'Developing a Tourism Impact Att itude Scale'. Contrary to the previous similar scales (Pizam, 1978;Liu and Var, 1986;Milman and Pizam, 1988;Ap, 1992), the authors' intention was to overcome the suggested omissions by using a multivariate Likert Scale (Likert, 1967). The authors have presented a unique model for measuring tourism's impact on the att itude of local population (most often in rural tourism), called Tourism Impact Att itude Scale -TIAS. The TIAS consists of original variables, respectively 27 dependent variables (Lankford and Howard, 1994, p. 130) and 15 independent variables (Lankford and Howard, 1994, p. 132). After the main study, the TIAS has been used in the researches conveyed mostly in economically developed countries, such as Canada (Rollins, 1997) (Schneider, Lankford and Oguchi, 1997) and Taiwan (Lankford, Chen and Chen, 1994). For this reason, it will be highly challenging to test TIAS in the conditions of Serbia (Vojvodina Province), as an example of typical transitional country, similar with other transitional Balkan states. Bearing in mind that the observed country has specifi c economic, geographical and sociological aspects, it should give a great number of possibilities of the later application on the level of the whole region. Till now, closest to the observed region (Northern Serbia) was only the TIAS testing in Western Serbia (Blešić et al., 2014a) and in Slovakia (Sabolova, 2013). In this respect, it is highly necessary to discover the coherence between tourism's impacts and locals, and work to improve them. A 5-points Likert scale (from 1 'strongly disagree' to 5 'strongly agree') was used for the measuring of the elements.

Data analysis
In order to explore agritourism's impact on the locals' att itudes in selected villages, factor analysis, descriptive statistics and statistical correlations were applied. The data was processed with the statistical program SPSS 18.0. Exploratory factor analysis is used for the analysis of the gathered data about interconnections of the sets of variables. It is not meant for testing research hypotheses, but for the reduction of the amount of input data (Pallant, 2011) and their bett er interpretation. The reliability of the measurement instrument was checked by using Cronbach's Alpha Reliability Coeffi cient. For the needs of the further results analysis, the descriptive statistical measurements have also been applied (Mean, Standard deviation, Median and Mode or Dominant value). Finally, the connection among the selected factors will be expressed by Pearson's r statistical correlation, since it deals with the interval variables.

Research hypotheses
A research goal of the paper states: Interconnection of items grouped into factors, which explains the level of agritourism impact on the att itudes of local population, shows positive correlation values. Explanation: The goal is formed according to the supposition that TIAS variables, grouped into factors, positively inter-correlate. In the results of the paper, the hypothesis will be proved or refuted by using Pearson's r statistical correlation and includes six sub-hypotheses ( Figure 2): 1. H1a: Benefi ts from tourism development for individuals and local community and negative impacts of tourism development show positive correlation value; 2. H1b: Benefi ts from tourism development for individuals and local community and concern for local tourism development show positive correlation value;

Factor analysis fi ndings
Exploratory factor analysis is used for the overview of the gathered data about interconnections of the sets of variables. According to the previous fi ndings ( (Kaiser, 1974). In addition, Bartlett 's test of sphericity has achieved the needed statistical signifi cance (p=0.000), which confi rms the justifi cation of the application of the exploratory factor analysis. The main components analysis has discovered the presence of four components with characteristic values above one (1), which is explained by 17.17% (F1), 11.58% (F2), 9.70% (F3) and 9.01% (F4) of the variance ( Table 2). After the forming of factors, the rotation was done by using the method of Varimax rotation.
The reliability of the measurement instrument was checked by using Cronbach's Alpha Reliability Coeffi cient. This instrument is among the most commonly used for inner closeness of the items composing the scale (Pallant, 2011). In an ideal case, Cronbach's coeffi cient should be above 0.70 (DeVellis, 2003), but the values of this instrument are very sensitive to the number of items on the scale. As Pallant (2011) stated, short scales (fewer than 10 items) usually have quite small Cronbach's coeffi cient (below 0.50), so in that case it is more appropriate to calculate the mean inter-item correlation. In this case, the recommended values are from 0.20 to 0.40, as optimal scope of inter-item correlation (Briggs and Cheek, 1986). Even though the reliability coeffi cients that are below 0.70 are generally considered unacceptable, sometimes the coeffi cients above 0.60 are accepted. According to Lehman et al. (2005), the ideal value of internal consistency value is in the interval from 0.80 to 0.90. The coeffi cient value for the fi rst, third and fourth factor exceeds the recommended value of 0.70 (F1=0.88, F3=0.71, F4=0.71), while the value of the second factor is close to the recommended value (F2=0.69). According to previous authors, who used similar grouping of the items (put in the brackets beneath), the factors are titled in the way presented in the Table 2. Cronbach's coeffi cient for the whole scale of 23 items is F1-F4=0.86, which is above the value of 0.70. After the conveyed factor analysis, the pure factor structure has been obtained, with high coeffi cients. Four items have been excluded from the model, due to their low values of factor loading coeffi cients (below 0.40). Thus, a model with 23 items grouped into four factors, which explain 47.47% of the variance has been obtained.

F1
F1a Better roads and sidewalks due to tourism. .729 F1b Public services (health services, water supply, fi re protection) improved due to tourism.
.772 F1c Have more money to spend.
.850* F1d Tourism has increased my standard of living.
.725 F1f Tourism provides highly desirable jobs.
.540 F1g Shopping opportunities are better.
.723 F1h Tourism will play major economic role. .558

F2
F2a Should not attract more visitors.
.549 F2c Noise level not appropriate.
.605 F2d More litter from tourism.

F3
F3a Encourage tourism in community.
.649 F3c Like to see tourism be main industry.
.471 F3d Will provide more jobs in community.

F4
F4a Encourage more intensive development.

F4c
Council right in promoting tourism (importance of local authorities' support in tourism development and affi rmation). From the results in Table 3, it can be concluded that there are signifi cant diff erences in residents' att itudes considering the impact of agritourism on their local communities. The results in the F1 show that respondents consider the possibility to have more money to spend, thanks to tourism development, as the most important item. In addition, this item shows the highest factor loadings among all the others in factor analysis. F1 grouped equal items, which are titled the same as it was done in some previous papers (Lankford and Howard, 1994;Lankford, Chen and Chen, 1994;Schneider, Lankford and Oguchi, 1997). The largest diff erence of 0.164 compared to those in the paper of Lankford and Howard (1994) was noticed in the case of F1g. It can be explained by the fact that, in the observed villages, the shopping opportunities have not yet risen under the impact of tourist activities, as it is present in the analyzed USA villages, where agritourism is much more widespread. On the other hand, the research results from Taiwan (Lankford, Chen and Chen, 1994) have also grouped seven items into this factor, and it can be noticed that six items are identical and with similar factor loadings. The greatest diff erence is noticed with item F1f (0.28), which can be explained by the fact that the jobs which are provided by (agri)tourism in Taiwan are more att ractive than the jobs which are present in the observed villages of Vojvodina. It leads to the conclusion that in rural areas of Taiwan, agritourism is more developed and that the locals consider the jobs provided via tourism development much more att ractive, which is present on a much lower level in the analyzed villages in a transitional country, such as Serbia. The only left-out factor in the case of this research is F3d, which has been set in the factor Concern for local tourism development in this paper. This statement is explainable by the fact that locals, having the opinion that tourism will aff ect a large number of the employees in their sett lement, at the same time have concerns about the general well-being and the development of their community, which can be considered as a justifying result.
The fi ndings in Table 3 indicate that the most important item in the F2 is the fact that crime could be increased due to tourism development, which is usually the main locals' concern in many other presented tourism study areas. Even though Rollins (1997) got fi ve items within this factor, all the items are identical with the items grouped in this factor and with similar factor loading, except F2d. This diff erence can be explained by the fact that the increase in the amount of litt er is still not a relevant problem in the observed villages, due to the absence of mass participation in agritourism. Harrill and Pott s (2003) also have got fi ve items grouped in this factor. However, item F3e, which is in this paper within the factor Concern for local tourism development (the same as in the research by Lankford and Howard, 1994), in the case of these authors, it was comprised in Negative impacts of tourism. Such a fact is not worrying, since in Rollins's research, this item is in the factor General opinion about tourism development. In the case of this study, the objection to building tourist facilities, which will att ract a large number of visitors to the rural areas, can certainly be considered a concern for tourism development by the locals in the observed Serbian villages.
The most important issue to the respondents in the F3 is similar with the F2, which also have connection with the tourism encouragement in the local community. The greatest diff erence in the loadings to some previous results (Wang, Pfi ster and Morais, 2006; Wang and Pfi ster, 2008; Woosnam, 2012) is noticed in the variable F3c, which can be explained by the fact that a large share of the number of examinees in Serbia think that tourism should become the main industry (together with agriculture). Wang, Pfi ster and Morais (2006), and Wang and Pfi ster (2008) have also got a large number of items grouped within this factor (12 in total). Although, the grouped items have similar factor loadings, the largest diff erence is noticed in F3a. This can also be explained with the conclusion that the residents in the observed rural areas may not have completely understood the meaning of the statement that in their sett lement tourism development should be encouraged.
Results emphasized that support of council right to promote tourism, which refl ects the importance of local authorities in tourism development and affi rmation, is the most important one (F4). All the obtained items in this paper completely match Rollins's fi ndings (1997), as well as factor loadings where the diff erences are small. The largest diff erence is noticed in F4a, which leads to the conclusion that the examinees in both study areas (the Island of Vancouver in Canada and Vojvodina Province in Serbia), in a relatively similar amount, think that their communities should encourage more intensive development, with the aim of a more successful tourist affi rmation of the rural areas.

Descriptive statistical analysis fi ndings
Beside factor analysis, for the needs of the results testing, some descriptive statistical measurements have also been used.  (Table 4).
Based on the results presented in Table 4, the values of arithmetic mean on the level of total values within the defi ned factors range from 3.21 (F1), as a lowest value, to 4.62 (F2), as a highest value. The lowest arithmetic mean is closest to score 3, while the highest arithmetic mean is closest to score 5. Based on that, it can be concluded that the values in the defi ned factors are relatively high. This means that the local population perceived negative impacts and concern for agritourism development and that they are aware of the potential challenges and risks that it may pose to them. In the overall factor set of questions, the biggest marks were given to the following: 'Against new tourism development', 'Encourage tourism in community' and 'Community should become destination'. The lowest mark was given to the question 'Like to see tourism be main industry', which points out that tourism is still developing in these areas and that locals give advantage to tourism to be an economically important industry in future (in addition to the traditional agriculture). Together with these, it is also proved by the mode (the most frequent score), which is 5 in all the cases on the level F1-F4. For the individual questions, in 75% (18)   cases, while among the individual questions in 17 out of 23 questions it has a value of 5, and in the remaining six questions it has the value 4. Since the three values are closely the same in questions, it can be said that it is a symmetrical frequency distribution. Standard deviation, which shows the mean value of the deviation of individual scores from the arithmetic mean, exceeds the value 1 in seven questions, while in the remaining 16 questions it ranges in the interval from 0.57 to 0.99.

Statistical correlation fi ndings
The connection among the selected factors (F1-F4) will be expressed by Pearson's r correlation, since it deals with the continuous (interval) variables. As Cohen (1988) stated, the gained values in Pearson's r correlation can present: low correlation (r=0.10-0.29); middle correlation (r=0.30-0.49); and high correlation (r=0.50-1.00).
According to the results shown in Table 5, it can be noticed that all the obtained correlation between the observed factors are positive and that nowhere high correlations have been obtained. Thus, it is important to comment the obtained middle correlations, i.e. those which exceed the value of 0.30. From the correlation patt ern, it is concluded that the positive correlation of the middle strength, between F3 and F1, equals 0.41. It means that concern for local tourism development shows a positive correlation with the benefi ts that the individuals and the local community have from tourism development (proved sub-hypothesis H1b). This can be interpreted that locals, who are concerned and take care of their surroundings, also have the individual benefi ts from tourist activities in their area, and that benefi ts are potential for their surroundings as well. On the other hand, the results obtained by Woosnam (2012) showed that concern for local tourism development highly positively correlates (r=0.52) with the benefi ts that the local community has from tourism development. This can be explained by the diff erence in the number of the examinees that the author used in the research (N=446). In addition, there are diff erences in geographical and cultural regions where the researches were conveyed (Texas, USA), as well as in the number of the obtained factors (two factors). A bit of a lower correlation is noticed in the correlation of F3 and F4, which equals 0.39. Such a value shows that the concern for local tourism development correlates positive values with the general opinion about tourism development. This leads to the conclusion that the bett er the general opinion and att itude of individuals and of the community about tourism development in a destination is, the higher is the level of concern for it (proved sub-hypothesis H1f).
The correlation between F3 and F2 equals 0.30. It points to the fact that concern for local tourism development positively correlates with negative impacts of tourism development. Namely, the higher the concern for the surroundings in a tourist destination is, the more obvious are the negative impacts and vice versa (proved sub-hypothesis H1d). This statement can be explained by the fact that locals realize more negative impacts provided by tourism in their sett ings when they are directly involved in tourism business than when they are not involved. In this respect, they will be more concerned when they have their own interests in tourism development and pay more att ention to any potential threat or weakness, which can be provoked by tourists and tourism development.
Besides the presented results, low correlations (r=0.10-0.29) also have positive values. According to that, the other sub-hypotheses are also accepted -H1a, H1c and H1e. It points to the conclusion that individual and community benefi ts from tourism development in the Serbian selected villages positively correlate with the general opinion about tourism development and with the negative impact of tourism development. According to that, it can be stated that the higher the negative impact of tourism is, the general opinion about tourism development, which produces the feeling of benefi ts from tourist activity in the local area, rises. This phenomenon can be explained by many examples from the destinations worldwide (Bali, Costa Rica, Crete, Sicily, Turkey). Apparently, the residents perceived tourism impact (even negative) as a new way of economic activity, earning chance, enterprising possibility or even an event ('something happens') in their (passive) rural surroundings. Mostly, agriculture is the main and only industry in these rural areas, so tourism gives to the locals some new economic, social and cultural frameworks, which are usually very challenging for them.

Conclusion
The disadvantaged position of agritourism in Serbia during the 1990s, the absolute and relative small role and almost negligible signifi cance on a national scale are the results of the many limiting factors (political, economic, social, etc.). Some of them may be mentioned: surrounding areas that have high tourist values, more complete and bett er quality network of tourist infrastructure and a system of mediating factors with well-planned performance at the international market level, the lack of tourism development programs as systematic tourism policies, the eff ects of policy reforms and foreign currency as substantial reduction in border and other formalities of tourists' stay and so on.
Good position of agritourism of Serbia in the international market will actually be the fi nal result of transformation of interconnected quantitative and qualitative tourism resources in the regional tourism off er, as well as the overall tourism product. Achieving a good position in the tourism market will enable identifi cation of the objectives of tourism development that will contribute to the quantitative change of contemporary tourist off er. So far, there was not found a reliable general model of agritourism development, which would, in the shortest time, help a developing country (such as Serbia) to have a more stable position. It should be noted that the agritourism, where it had conditions for the formation of tourism supply, demand and support from the leadership of political and economic entities, within 10 years, had affected the development of many passive and undeveloped rural parts of the country.
For these reasons, this study had the aim to give a contribution through an analysis and overview of agritourism aspects, such as its impact on residents' att itudes. The research succeeds in this goal of implementing the Tourism Impact Att itude Scale (TIAS). The factor analysis, descriptive statistics and statistical correlation are used for the analysis of the gathered data about interconnections of the sets of variables and factors, comparing them with the fi ndings of similar researches. At the same time, it has been the focus of the paper to analyze the methodology and the results obtained earlier compared with the fi ndings of this study. The results show that respondents consider the possibility to have more money to spend, thanks to tourism development, as the most important one. The second important issue is the support of council right to promote tourism, which refl ects the importance of local authorities in tourism development and affi rmation. It is followed by the tourism encouragement in their local community. Less favored is the fact that crime rate could be increased due to tourism development.
Results also point to the conclusion that the higher the negative eff ect of tourism is, the more positively it aff ects the awareness of the general, as well as individual tourism benefi ts in the observed local areas in Serbia. Together with these, the study came to the conclusion that the locals who take care of their local surroundings may also benefi t from tourist activities in their area, and these benefi ts are potential for their environment as well. The obtained fi ndings confi rm previous results, which also proved that care about the local tourist development is in positive correlation with the benefi ts of tourism for the local community. On the other hand, results indicated that the higher the negative eff ect of tourism in Serbian villages is, the higher the opinion on tourism development of the locals, and that the higher the care about the area, the easier it is to recognize the potential negative eff ects and vice versa. With all these matt ers, our research emphasizes the fi nding that the higher the general opinion and att itude of an individual and of the community on tourism development in their local surroundings are, the higher is the care about the community. These statements in fact describe the general aspects of agritourism's impact, as well as its eff ect on the locals in rural areas of Serbia in the middle of the second decade of the 21 st century.

Introduction
Corruption is considered to be one of the most concerning problems for modern societies. In the last several years, even countries that were previously immune to this phenomenon, like Finland, encountered several problems that were considered unethical (European Commission, 2014b). As Klitgaard stated: 'Not so many years ago, corruption fl ourished in countries we now call 'developed'and indeed there is evidence that some forms of corrupt behavior are today on the rise in the most economically and politically 'advanced' nations. Corruption scandals in recent years have led to changes at the highest level of the government in Holland, West Germany, Great Britain, Israel, Japan and the United States' (Klitgaard, 1998, p. 10).
The matt er is a lot more serious in new European Union member states, especially in Romania and Bulgaria or in the possible future member states or partners of EU. All of the above mentioned countries are listed with high levels of corruption in Transparency International Corruption Perception Index (the FYR of Macedonia -rank 64/score 45 Specialists and scholars in this fi eld are rather pessimistic concerning the possibility to curb corruption worldwide. An analysis of the Corruption Perception Index for the last ten years indicates that the vast majority of the countries had decreased their CPI score and only a few improved it. In 2014, 69% of the surveyed countries scored below 50 out of 100. An interesting fi gure shows that among the G20 countries, presumably the most developed in the world, 58% had scores below 50 (Transparency International Corruption Perception Index, 2014). The European Union Report on Corruption states that the Eurobarometer survey conducted in 2013 showed a high level of concern among European citizens related to corruption (European Commission, 2014a). Thus, '... three quarters of respondents (76%) think that corruption is widespread in their own country' (Transparency International Corruption Perception Index, 2014) and 'a quarter of Europeans (26%), compared with 29% showed by the 2011 Euro Barometer, consider that they are personally aff ected by corruption in their daily lives'. However, there are important diff erences among countries. In some of them (Denmark, Sweden, Finland and Luxembourg), the perception is positive, while in others (Germany, the Netherlands, Belgium, Estonia and France) the general perception is negative but with very few individual experiences (Transparency International Corruption Perception Index, 2014). Thus, the report is revealing two facts: the increasing concern of the European citizens related to widespread corruption and the signifi cant diff erences that exist between various groups of states. This second aspect