JOURNAL OF COMPARATIVE POLITICS 73 THE LEVEL OF POLITICAL KNOWLEDGE IN SLOVENIA: WHO IS (NOT) POLITICALLY INFORMED? Meta NOVAK 1 ………………… ……… ………………… ……… ………………… ……… … Delli Carpini and Keeter showed that political knowledge is more or less general, meaning that informed citizens are usually informed in all aspects of political knowledge. Unfortunately, political knowledge is not evenly distributed. This is even more problematic when we keep in mind that the politically informed are also more politically active. The aim of this paper is to find out how informed Slovenians are as citizens of a young post-socialist democracy and who is informed. We used data from the European Election Study 2009 and performed a simple comparison of mean values and a linear regression model. Slovenians turned out to be moderately knowledgeable, with higher levels of EU political knowledge in comparison to national political knowledge. Slovenians were also among the most knowledgeable citizens in the EU member states. Differences in the level of knowledge are present between males and females, education groups and classes, based on the level of news attention and political interest. Key words: political knowledge, knowledge inequality, Slovenia, European Union, European Election Study. 1 INTRODUCTION Democracy as a form of government is based upon the presumption that citizens are active in political participation, informed and engaged in political activity (Delli Carpini and Keeter 1996). Furthermore, a supposition of democracy is ‘that decision making of the public is based on well-informed and sophisticated political reasoning (Scheufele et al 2002, 427)’. The ideal citizen in a democratic system is thus informed, doubtful, partial to public affairs, attentive towards positions of his or her nation and the quality of its leaders (Iyengar and Kinder 1987, 119) but also always concerned about issues that have an impact and take place in his community (Lippman 1961, 269). However, the research shows that the average citizen is poorly informed, with scarce knowledge of political institutions, public policies, socio-economic circumstances and political actors, while this level of information – or to be more precise, the lack of information – is stable over time (Delli Carpini 2000, 129; Lupia and McCubbins 2003, 17). Despite scarce knowledge of 1 Meta NOVAK is research assistant at the Faculty of Social Sciences, University of Ljubljana. JOURNAL OF COMPARATIVE POLITICS 74 governmental affairs and politics, the average citizen is regularly confronted with the need to form an opinion towards public issues. Ultimately, citizens have to decide to whom they give their vote to and how they will make decisions about programmes and candidates (Stokes 1962, 690). Notwithstanding the high level of political ignorance, Delli Carpini and Keeter (1996, 289) claim that voters do not need all the information that is out there to make reasonable decisions (see also Lupia and McCubbins 2003). Iyengar (1990, 182) also argues that we should abolish the model of an informed voter, which is completely unrealistic. After all, a low level of information does not mean complete ignorance, and at least it does not mean general ignorance (Delli Carpini and Keeter 1994, 19). With the acquisition of new information, voters can at least familiarize themselves about current political events (Popkin 1994). The question thus no longer is whether the public is politically informed, but rather who is informed and what he is informed about (Delli Carpini and Keeter 1994, 19). Therefore we are setting the following three research questions: 1. Who in Slovenia is politically informed? 2. Who is knowledgeable about national issues compared to EU issues? 3. How knowledgeable are Slovenians in comparison to citizens in other EU member states? While surveys of political behaviour, attitudes and opinions in Slovenia are regularly conducted, measures of political knowledge, although present, are not that uniform. This article, by use of data collected from the European Election Study 2009, examines sources of variation in political knowledge of Slovenian citizens, comparing also the level of knowledge in Slovenia to other EU-member states. Due to the recent elections for the European parliament (May 2014), the question of who possesses political knowledge about national and EU issues is very topical. Results can offer us an insight into which groups are information poor and calls the attention towards which groups information campaigns should be directed to in future. 2 WHAT IS POLITICAL KNOWLEDGE Political knowledge 2 is the best measure for political intellectuality, sophistication and awareness (Mondak 1999, 58). Delli Carpini and Keeter (1996, 10), as the authors who might have studied political knowledge the most, systematically defined it as a set of factual information on politics stored in long-term memory. They understand political knowledge as a competence, as a resource that can be upgraded. The political knowledge, that a voter needs in order to be a good citizen, can thus be acquired also by an individual with average competence (Delli Carpini and Keeter 1993, 1186). And factual knowledge is that fundamental knowledge necessary for building voters competence (Delli Carpini and Keeter 1994, 21). It includes knowledge of political systems and institutions, knowledge of the functioning of political systems and tasks of government, knowledge of political actors and their 2 In literature, we come across different terminological variations that denote similar concepts. From ‘political sophistication’ (Luskin 1990, 331–361), which is wider concept and also includes motivation and competence, similarly to ‘political awareness’ (Zaller 1990, 125–153) and ‘political expertise’ (Fiske et al 1990, 31–48), which also includes the organisation of concepts as well as factual knowledge, exposure to media and information, experiences, inclusion and interest towards ‘political information’ (Grönlund and Milner 2006, 389), which denotes only information without cognitive dimension and, last but not least, ‘political knowledge’ (see e.g. Delli Carpini and Keeter 1993, 1179–1206; Eveland and Hively 2009, 205– 224; Fraile 2013, 119–143), which includes factual knowledge and will be used in this paper as a synonym for all of the above-mentioned concepts. JOURNAL OF COMPARATIVE POLITICS 75 jurisdictions, knowledge of current economic and social conditions, and the main daily issues and positions of political leaders towards these issues (Delli Carpini and Keeter 1996). This knowledge can be divided in three categories of political knowledge: (1) “what government is”, which includes knowledge of the rules of the game; (2) “what government does”, which includes the content of politics and (3) “who government is?”, which includes knowledge of the actors, politicians and parties. The more that a voter knows about these categories (breadth) and the more in detail he or she knows these areas (depth), more capable he or she will be to take political action (Delli Carpini and Keeter 1996, 65). Similarly, Luskin understands the political knowledge of an individual as a union of the extent of political knowledge (how much information someone knows), how wide-ranging it is (e.g. whether this includes knowledge of both national and international politics) and how organised this knowledge is (Luskin 1990, 332). With this definition of political knowledge, Delli Carpini and Keeter (1996, 10–11) distinguish knowledge from other voter characteristics, such as electoral behaviour, values, opinion and belief, but also from judgments, verdicts, and decisions and from participation, education and the use of media; 3 at the same time, they define factual knowledge as the best indicator of political knowledge (Delli Carpini and Keeter 1993, 1180) and as a necessary condition of a working democracy (Delli Carpini and Keeter 1994, 21). Zaller, similarly to Delli Carpini and Keeter, argues that political knowledge is best to be measured with variables of political information, with simple factual questions about politics (Zaller 1989, 186). According to these authors, factual knowledge helps voters when they want to enforce their own interests, connect their interest with the general interests of society and when they want to express their interest in political participation (Delli Carpini and Keeter 1996, 1). Even though factual knowledge will be analysed in our article, we should point out that all scholars are not in favour of this definition of political knowledge. Scholars who define knowledge as subjective, contextually specific and consistent in terms of different opinions do not agree that knowledge could be measured as objective and universal, regardless of how the knowledge is perceived from an individual’s point of view (Hindman 2009, 792). Political knowledge can thus be differentiated based on depth, structure and understanding and general in contrast to specific knowledge. Jeffres and colleagues differentiate political knowledge between: (1) ‘confirmatory knowledge’, which is knowledge of political and social systems and includes knowledge of executive power, current political actors, and legislative content; (2) ‘domain-specific knowledge’, which is knowledge of position towards different issues (such us knowledge of current events that usually have high media coverage); (3) ‘relational knowledge’, where individuals compare their own positions with positions of political actors and (4) ‘piecewise knowledge’, where an individual does not possess complete knowledge but only some information (Jeffres et al 2011, 33). Similarly, Eveland and Hively (2009, 212) distinguish between ‘issue stance knowledge’—which is a concept similar to Jeffres’ relational knowledge—‘general political knowledge’ and ‘knowledge structure density’, which includes knowledge and an understanding of the connections and relationships between gained concepts and an understanding of the connections between individuals, political questions and institutions. Most political knowledge typologies are formed based on two dimensions, depth of knowledge and extent of knowledge. Most authors undertake Park’s distinction between ‘knowledge of acquaintance’ as knowledge that we gain without systematic learning (Park 1940, 669–686); Genova and Greenberg 3 We will dedicate the next chapter to how political knowledge can be intertwined with other characteristics of voter. JOURNAL OF COMPARATIVE POLITICS 76 (1979, 79–91) name this ‘factual knowledge’, and ‘knowledge about’ or ‘structural knowledge’ as a consequence of the analytical process, observation and judgment of information. Extent of knowledge is distinguished between ‘general knowledge’ and ‘domain specific knowledge’ (Zaller 2003). For an individual to gain new knowledge, he or she needs to be exposed to new information through some communication medium and accept this information, which means understanding and storage of information in memory (Zaller 1989, 186). Besides understanding of new information, an individual also has to be motivated to find new information (Luskin 1990, 331–361). Luskin (1990, 331–361) thus defines political knowledge as a factor of three dimensions: (1) motivation, or the readiness to accept new knowledge; (2) the information to which an individual is exposed to; and (3) the capability to organise and retain information. Luskin suggests that individuals who have been completely isolated from political information since birth cannot acquire political knowledge regardless of their capability and motivation. Also, individuals who are motivated and exposed to information cannot acquire political knowledge without the right amount of capability. As well, capability and exposure to information are not enough if an individual does not have the right motivation (Luskin 1990, 338). 3 WHY DO WE CARE ABOUT POLITICAL KNOWLEDGE? POLITICAL KNOWLEDGE, PARTICIPATION AND POWER Political knowledge is of collective benefit to the entire society. A more knowledgeable society is more democratic and tolerant; what is more, broader participation increases the legitimacy of the government and puts limits on the abilities of decision makers to manipulate public interests (Delli Carpini and Keeter 1996, 219–220). Political knowledge is not an independent characteristic of each citizen. As stressed by Delli Carpini and Keeter (1996, 218–267), there are five dimensions on which political knowledge contributes to good citizenship, which should be of more importance for young democracies like Slovenia. Political knowledge (1) promotes civic virtues; (2) promotes active participation; (3) helps construct stable, consistent opinions; (4) helps identify political interests and connect them to political attitudes and (5) helps link citizen’s attitudes with participation. Special interest in political knowledge is present above all due to the correlation of political knowledge, political participation and political behaviour (Mondak 1999, 72–73; Grönlund and Milner 2006, 386; Eveland and Hively 2009, 205). Research shows that political knowledge itself is highly correlated with political motivation, political interest, political activity and political participation (Jennings 1996, 228–252), though it is not particularly clear what has an effect on what; it appears that they have an inter-correlation affect. Horstmann showed that motivation is higher when there is an increased level of political knowledge; at the same time, motivation contributes to a higher level of political knowledge (Horstmann 1991, 83). Unequal interest in political issues also has an effect on unequal exposure towards political content in media (Zaller 2003, 1). Those who show less interest in political issues will search for content that addresses current political activity to a lesser degree. However, higher levels of interest and activity should contribute to higher levels of political information (Jennings 1996, 228–252). JOURNAL OF COMPARATIVE POLITICS 77 Political knowledge also has an effect on political participation and not just political interest. Citizens will more likely participate in political activity if they know when and where they take place. They will also protest, take part in demonstrations and vote for a particular candidate with more confidence when they have knowledge and opinions about the political background (Cho and McLeod 2007, 224). Consequently, more informed citizens are more effective in politics, more interested in political issues, and more likely to follow politics and discuss it and to form consistent, meaningful and stable opinions that do not change with every new piece of relevant information. Even with misleading information, it is easier for them to make political decision, and they are more likely to act in line with their interests and vote for candidates that support their position. They are less likely to vote based on a candidate’s personal characteristics and are more likely to support democratic norms and values and show a higher level of political participation in various ways by taking part in elections, participating in the activities of their political party and attending community meetings (Zaller 1989, 222; Luskin 1990, 332–333; Delli Carpini and Keeter 1996, 218–226; Delli Carpini 2000, 142–143; Grönlund and Milner 2006, 393). However, political knowledge is not evenly distributed. Research shows that more knowledgeable citizens are more educated, with a higher socio-economic position, and they are typically male and older (Delli Carpini and Keeter 19996, 135–177; Jennings 1996, 228–252). And ‘the unequal distribution of knowledge is as troubling as the low average levels of political literacy’ (Fraile 2013, 119), even more so if we keep in mind that unequal distribution of knowledge is connected with socioeconomic position (Gaziano 2012, 17), translates into unequal possession of power and maintains the current social structure. After all, information and control over information are the foundations of social power (Tichenor et al 1980, 22). Access and gained knowledge enable individuals and groups to restore, maintain and increase their positions of social power (Gaziano 1988, 351–352), which furthermore obstructs class mobility and changes of social structure (Hwang and Jeong 2009, 523). Therefore, a relevant question is: “Who is politically informed?” 4 METHODS AND DATA 4.1 Data When assessing the level of political knowledge in Slovenia, especially from a comparative perspective, we face a scarce availability of comparative data. This is why we are using the European Election Study 2009, Voter Study (EES) (Egmond et al 2013), which provides us with a knowledge test that consists of seven questions about the functioning of EU institutions and national political actors with varying levels of difficulty. Despite the fact that some questions measure national issues, they are designed in a way to imply the same level of difficulty for interviewers, which allows us to also compare cross-country levels of political knowledge (Fraile 2013, 119–143). The data were collected between 5 June 2009 and 9 July 2009, while the European Elections were taking place between 4 June and 7 June 2009. The survey took place in all the 27 countries that were member of the European Union at that time, with a total sample of 27069 units, of which 1000 Slovenians were interviewed. Data was collected by CATI phone interviews. JOURNAL OF COMPARATIVE POLITICS 78 4.2 Dependent variable: Political knowledge The dependent variable in our analyses is the level of political knowledge. We operationalised political knowledge as accurate information about politics. To measure political knowledge, we used seven knowledge test questions and formed an additive index of political knowledge. We used a similar approach to the one that Marta Fraile (2013, 119–143) used in her article. Several variables that measure political knowledge were computed. Wrong or missing answers 4 were coded as value = 0, representing absence of knowledge. Correct answers were coded as value = 1, representing the presence of knowledge. Afterwards, we added values for all variables into one index ranging from value 0 = absence of knowledge to value 7 = high knowledge. As most people are generalist in what they know about politics (Delli Carpini and Keeter 1996), adding all variables of political knowledge into one dependent variable is acceptable. The seven questions included in our index of political knowledge are the following true/false statements: 1. Switzerland is a member of the EU. 2. The European Union has 25 member states. 3. Every country in the EU elects the same number of representatives to the European parliament. 4. Every six months, a different member state becomes president of the Council of the European Union. 5. The [Specific Minister] is [Correct name]. 5 6. Individuals must be 25 or older to stand as candidates in [COUNTRY] elections. 7. There are [150 percent of real number] members of the [COUNTRY Parliament]. General political knowledge in Slovenia will be examined and compared in terms of EU political knowledge and national political knowledge. An index of national political knowledge was formed in a similar way as the index of political knowledge, this time by adding up the correct answers to the three questions about national political actors (question 5, 6 and 7). The index for national knowledge ranged from value 0 to value 3. In a similar way, we added up the correct answers to four questions that measure knowledge about EU institutions (questions 1, 2, 3 and 4). The index for EU knowledge ranged from value 0 to value 4. In order to be able to compare both indexes as well as the index of general political knowledge, we standardised the indexes to range from value 0 to value 1 with the following formula: K X − Xm in Xm ax − Xm in 4 When analysing political knowledge, we can come across: (1) correct answers, (2) incorrect or semi-correct answers and (3) absence of answers (Don’t know answers) (Mondak 1999, 57– 82). While the absence of answers implies also the absence of knowledge and receiving information, wrong answers imply there has been some information present, but not received or wrongly understood (Mondak 1999, 57–82). Mondak thus argues that “don't know answers” should be treated differently than wrong answers. We decided to treat both values as absence of knowledge, since research also shows that wrong answers can also be an attempt to guess and are not always evidence of partial knowledge (Luskin and Bullock 2011, 547–557). Furthermore, some other surveys also treated “don't know answers” as incorrect answers (see e.g. Jennings 1996, 228–252; Fraile 2013, 119–143). 5 For the case of Slovenia, the statement was: “The Slovenian Minister for education and sport is dr. Igor Lukšič.” JOURNAL OF COMPARATIVE POLITICS 79 4.3 Independent variables: Country, level of education, age group, gender, political interest, self exposure to news media and self assessment of class Previous research has shown that political knowledge is not the same in every country, and context plays a vital role (Fraile 2013, 119–143). In this respect, we will compare levels of political knowledge among EU members after we take a closer look at who is informed in Slovenia by analysing political knowledge in connection to level of education (3-point scale), class (dichotomy), age (4-point scale) and gender (dichotomy) and also political interest (4-point scale) and news attention (dichotomy). Previous research has shown that there is an existing knowledge gap between more and less educated citizens and among socioeconomic groups (Delli Carpini and Keeter 1996; Holbrook 2006, 343– 352; Fraile 2011; Fraile 2013, 119–143) that is even wider in periods of infusion of media information (Tichenor et al 1970, 159–170). Some researches stress the importance of individual characteristics, such us media attention (Iyengar and Kinder 1987), political interest and motivation (Ettema and Kline 1977, 179–202) when explaining the level of political knowledge. More knowledgeable citizens also tend to be older (Delli Carpini and Keeter 1996). Research also shows a gender gap in political knowledge (Fraile 2014). 5 ANALYSIS The analyses will be done in two steps. In the first step, we will analyse political knowledge in Slovenia. We will be interested in who is knowledgeable. To examine the impact of each independent (demographical and structural) variable on political knowledge, we performed a simple t-test analysis for dummy independent variables: gender, news attention and socioeconomic position, and we conducted an analysis of variance with a post-hoc Bonferroni test for interval independent variables: level of education, age group and political interest in relation to political knowledge. We will conclude the analyses of general political knowledge in Slovenia with a simple linear regression model, where our independent variables will be included in the regression model as predictors of political knowledge. Last but not least, we will also determine if the same groups are the most knowledgeable about national issues and on factual questions on European Union. Paired sample t-test will be performed to check if the mean value of EU political knowledge is different than the mean value of national political knowledge, and a bivariate correlation test will examine if the values of both indexes are correlated. In the second step, we will examine how knowledgeable Slovenians are compared to citizens of other EU member states. To observe the level of political knowledge in Slovenia and other member states, we compared the mean values of the index of general political knowledge, national political knowledge and EU political knowledge. In this step, predesigned sample weights per country will be used. 5.1 Who is informed in Slovenia? Before we look into who is informed in Slovenia, we present the frequency distribution of political knowledge for Slovenia in Figure 1. Each asterisk represents one percent of respondents. Along with the distribution of political knowledge in Slovenia in Figure 1, three hypothetical models of political knowledge are also represented as introduced by Delli Carpini and Keeter (1996, 152). Their model of managerial democracy presumes a high level of knowledge for the few and a low level of knowledge for the majority. The exact JOURNAL OF COMPARATIVE POLITICS 80 opposite is the model of strong democracy, where the majority has a high level of knowledge and a minority has a low level of knowledge. The pragmatic model, on the other hand, presumes that the majority has a middle range of knowledge, a few are uninformed and a few are highly informed. The distribution of political knowledge in Slovenia is closest to the last described model – the pragmatic democracy model. The majority has a middle range of knowledge, while those with high knowledge levels (6 or 7 questions answered correctly) are more (26,3%) than those who are completely uninformed (0 or 1 question answered correctly), who represent only 9,2% of respondents. As Delli Carpini and Keeter state: This distribution /…/ would be the product of a political culture in which the acquisition of political information was a civic norm, political information was reasonably accessible through the schools and the mass media, and most citizens had enough motivation and cognitive skill to gather and retain at least a moderate amount of factual knowledge (Delli Carpini and Keeter 1996, 153). FIGURE 1: HYPOTHETICAL AND REAL DISTRIBUTION OF POLITICAL KNOWLEDGE Source: Delli Carpini and Keeter (1996, 152). Though the majority of voters in Slovenia turned out to be moderately informed, a difference in the amount of knowledge that different groups posses still remained (Table 1). The level of political knowledge is not different between age groups but similar to other research results (Fraile 2014) men have a higher level of political knowledge. While on average, women answered three and a half questions correctly (out of seven), men answered more than four and a half questions correctly (out of seven). The difference is also present among different socioeconomic classes, where the voters who identified themselves as middle, upper middle or upper class answered more questions correctly than those who identified themselves as working or lower class. Similarly, the level of political knowledge is also different among groups with different educational levels. Post-hoc Bonferroni tests showed that the mean values of political knowledge are different between all three educational groups. Voters with an elementary school education or less have a lower level of political knowledge than voters graduating from a vocational school, who answered fewer questions correctly than voters with a high school degree or more. Those who show a higher level of political interest are also more knowledgeable. A post-hoc Bonferroni test revealed a statistically significant difference in the level of political knowledge among all groups with a different political interest, except between the group that is very interested and somewhat interested and between the group that is somewhat interested and a little interested. There the differences are too small to talk about real differences in the level of political knowledge. Last but not least, the group that follows the news every day answered more questions correctly than the group who follows the news less often. JOURNAL OF COMPARATIVE POLITICS 81 TABLE 1: LEVEL OF POLITICAL KNOWLEDGE AMONG DIFFERENT GROUPS OF VOTERS IN SLOVENIA Source: The authors’ own calculations based on the European Election Study 2009, Voter Study (EES) data. To find out how much the above factors predict the level of political knowledge, we also conducted a simple linear regression model and entered the following independent variables in the model: political interest, news attention, level of education and gender. We left out age group (because the previous analyses showed us that the level of political knowledge does not differ between age groups) and class (because the question did not offer real diversification). The variables political interest (4-point scales), news attention (7-point scale) and education (8-point scale) were all treated as interval variables, while gender was treated as a dummy variable (0 = female, 1 = male). Our model explained 28,5% of the variance in political knowledge. 6 Although all the predictors in our model explain the level of political knowledge, there is still room to improve the explanatory power of the model. Evidently, some other factors should be taken in consideration. Among the predictors that we entered in the model, the strongest one was level of education, followed by gender, political interest and news attention (the least powerful factor but still significant). A voter who is male, follows the news more often, has a higher level of political knowledge and education should also have a higher level of political knowledge. TABLE 2: PREDICTORS OF POLITICAL KNOWLEDGE 7 IN SLOVENIA – LINEAR REGRESSION WITH REGRESSION COEFFICIENTS (B) AND STANDARDISED REGRESSION COEFFICIENTS (BETA) Source: The authors’ own calculations based on the European Election Study 2009, Voter Study (EES) data. 6 Multicollinearity was not a problem. 7 Standardised values of political knowledge from 0 to 1. JOURNAL OF COMPARATIVE POLITICS 82 5.2 Political knowledge of national system compared to EU system Delli Carpini and Keeter (1996) showed that groups that tend to have political knowledge from one field also tend to have political knowledge in other fields (e.g. individuals who are knowledgeable about national political system also have political knowledge about current international affairs), proving that voters are generalists in terms of what they know about politics. We examined this for Slovenia by comparing the level of national political knowledge with the level of EU political knowledge among different groups. We can reasonably suspect that the knowledge about EU issues would be scarcer. The reason for this would be Slovenia’s relative short period of membership in the EU. 8 Furthermore, the elections for the European parliament are still seen as second- level elections (Kropivnik 2010, 28–29), with less than 29% of voters turning out in 2009 and 2004 and with 24,09% voters turnout in 2014. 9 The survey results showed quite the opposite; the whole sample of Slovenian voters is more knowledgeable about EU issues, with a mean value 0,65 of EU political knowledge index, while the mean value of national political knowledge index was 0,51. 10 The higher level of EU political knowledge compared to national political knowledge could be attributed to the European elections that were taking place at the time of the survey. The informational effect of the election campaign might have contributed to the higher EU political knowledge of respondents (Holbrook 2006, 343–352). In order to check whether the level of EU political knowledge is related to the level of national political knowledge, we calculated the Pearson correlation coefficient value. The coefficient (Pearson correlation = 0,487, p<0,05) value indicates a relatively strong correlation between levels of national and EU-level political knowledge. Higher levels of national knowledge correlate with higher levels of EU political knowledge. A comparison of the level of EU and national political knowledge in relation to different independent variables showed us some differences. For general political knowledge as well as for EU and national political knowledge, there are no statistical differences between age groups in what people know. Though the ones who are more politically interested are more knowledgeable about EU and national affairs, the differences are bigger when we observe national political knowledge. The post-hoc Bonferroni test revealed that all the groups with different levels of political interest significantly differ among themselves with the exception of the difference between the group that is somewhat interested and the group that is a little interested. The difference between these two groups is too small. On the other hand, the difference in the level of EU political knowledge is statistically significant only between the group that is not at all interested and other groups based on their interest levels. Political interest apparently plays a more vital role when it comes to gaining knowledge about national issues than EU issues. There could be various reasons for this. Due to the election campaign for the European parliament, the environment might have been information rich, which made it easier for respondents to gain EU- related knowledge (Fraile 2013, 119–143). It is also possible that the effort to inform Slovenian citizens about the EU – as part of the whole association process to the EU – may have been very successful in educating Slovenians. It 8 On 1 May 2014, Slovenia with nine others countries celebrated its 10-year anniversary of membership. 9 National Electoral Commission. 10 We performed pair-sample t-tests that examine the difference in mean values for two different variables on the whole sample. The difference between mean values of national political knowledge and EU political knowledge was significant (t = 13,995, p = 0,000). JOURNAL OF COMPARATIVE POLITICS 83 might also be that knowledge about the EU is not associated to a large extent to political interests, while it might be associated with other kinds of interest. For general knowledge and for EU political knowledge and national knowledge, male and voters from the middle, upper middle and upper class answered more question correctly than women and voters from the working and lower middle class. Higher EU and national political knowledge is also present for the groups that finished high school than for voters with vocational school or voters with elementary school or less. Post-hoc Bonferroni tests revealed that there are statistically significant differences between all pairs of educational groups for EU and national knowledge. While voters that follow the news every day tend to have a higher level of national knowledge, there is no difference between voters who follow the news every day and those who don’t in terms of EU knowledge. Apparently, knowledge about national issues is introduced in the media, while other channels may be used for gaining EU knowledge. Again, this could also be a consequence of the European election campaign. During the campaign, information and messages are clear, concentrated and simplified so that any voter can understand them. Furthermore, during an election campaign, information is more accessible, while voters need to invest less effort to get new information (Franklin 1991, 1193–1214). TABLE 3: LEVEL OF NATIONAL AND EU POLITICAL KNOWLEDGE AMONG DIFFERENT GROUPS OF VOTERS IN SLOVENIA Source: The authors’ own calculations based on the European Election Study 2009, Voter Study (EES) data. JOURNAL OF COMPARATIVE POLITICS 84 5.3 Political knowledge in Slovenia compared to other EU member states Now that we have an idea who posses information in Slovenia, we should also look into how knowledgeable Slovenians are in comparison to other citizens from EU member states. The figure bellow represents mean values of the index of political knowledge. Countries are distributed in increasing order of political knowledge. ANOVA test shows that differences in political knowledge of EU member states are statistically significant (F- statistic = 108.933 and p<0,001). Post-hoc Bonferroni test revealed that the level of political knowledge in Slovenia is different from all EU members (p<0,05) except for those from Estonia, France, Austria and Finland, where the differences in political knowledge are too small to talk about real differences. FIGURE 2: LEVEL OF POLITICAL KNOWLEDGE IN EU MEMBER STATES: Source: The authors’ own calculations based on the European Election Study 2009, Voter Study (EES) data. As we can see from Figure 2, the differences in the level of political knowledge between the country with the highest level, Denmark (Denmark joined EU in 1973), and the country with the lowest level, Romania (Romania joined EU in 2007), is substantial. On average, the voters from Denmark replied correctly to almost two and a half questions more in comparison to voters from Romania. We can say that the level of political knowledge among the EU member states is far from equal. Slovenia has a mean value of political knowledge of 0,59, which place it among the five most knowledgeable EU member states. Along with Denmark, Luxembourg, Sweden, Austria, Finland and France, Slovenia is one of seven countries that answered correctly on more than 4 questions out of seven on average and is also the only country in this group that was not one of the old EU members, EU15. On the other side of the continuum, we have the citizens of Ukraine, Poland, Lithuania, Spain and Romania, who answered less than 3 questions correctly on average. Except for Spain, the rest of the countries from this group joined the EU in 2004 or 2007 and are treated as new member states. JOURNAL OF COMPARATIVE POLITICS 85 Besides the year of accession to the EU, other contextual factors could also contribute to the different levels of political knowledge, such us the level of support for the EU and characteristics of the election campaign. Since our index of political knowledge also includes questions about national political actors we should consider also socioeconomic, political and communicational contexts (Fraile 2013, 119–143). Since national level factors can contribute to different levels of political knowledge, we also decided to compare Slovenia to other EU member states based on national political knowledge and EU political knowledge separately. Figure 3 represents the results (countries in Figure 3 are distributed in the same order as in Figure 2). The ANOVA test shows that while the differences in national and EU political knowledge of member states are statistically significant (F- statistic for national political knowledge = 107.520; p<0,001 and F- statistic for EU political knowledge = 108.161; p<0,001), a post-hoc Bonferroni test revealed that the level of EU political knowledge in Slovenia is different from almost all other EU members 11 (p<0,05), but the level of national political knowledge in Slovenia is different only from half of the EU member states 12 (p<0,05). The level of national political knowledge in Slovenia is just below the EU average and similar to 12 other member states, but Slovenia ranks near the top for the level of EU political knowledge along with Austria and Luxemburg. FIGURE 3: LEVEL OF NATIONAL AND EU POLITICAL KNOWLEDGE IN EU MEMBER STATES: Source: The authors’ own calculations based on the European Election Study 2009, Voter Study (EES) data. 11 The level of EU political knowledge in Slovenia is similar to the level of EU political knowledge in France, Austria and Luxemburg. 12 The level of national political knowledge in Slovenia is different from the level of national political knowledge in Belgium, the Czech Republic, Denmark, Estonia, Greece, Spain, Latvia, Luxemburg, Poland, Finland, Sweden, the United Kingdom and Romania. JOURNAL OF COMPARATIVE POLITICS 86 EU political knowledge is higher in comparison to national political knowledge not just in Slovenia, as previously noted, but noticeably also in the Czech Republic, Austria, France, Germany, Luxemburg, Spain and Cyprus. Here there is no difference between old, new and founding member states. Support for the EU could play a role when explaining why EU political knowledge is higher in comparison to national political knowledge. 13 With the exceptions of Austria and the Czech Republic, the rest of these countries are also among the top six countries that support EU membership. There also seems to be a connection between the level of EU political knowledge and new and old member states. Besides Cyprus, the Czech Republic and Slovenia, the rest of the new member states (accession in 2004 or 2007) have below average levels of EU political knowledge but not necessarily also below average levels of national political knowledge: Estonia, Latvia, Hungary, Bulgaria and Slovakia show above average levels of national political knowledge. Although the same socio-demographic variables explain the level of national and EU political knowledge in Slovenia, the member state countries represent a different context when explaining national versus EU political knowledge. While citizens can be generalist in political knowledge (Delli Carpini and Keeter 1996), differences between EU and national political knowledge are clearly present at the national level. Some member states are disposed to have similar levels of EU and national political knowledge, but others have better knowledge in one domain. 6 CONCLUSION The aim of this article was to identify information-poor groups in Slovenia. While the level of political knowledge turned out to be higher in Slovenia in comparison to new EU member states and also to many of older member states, especially when we look at EU political knowledge, the majority of Slovenian voters tend to have moderate political knowledge. Some groups, nevertheless, are better politically informed than others. As the more informed citizens turned out to be those who are very interested in politics, follow the news every day and have a better socioeconomic position. While education turned out to be the most powerful predictor of political knowledge, a difference in political knowledge is also present among both genders, where males turn out to be more informed in comparison with women. We can conclude that the most vulnerable groups are women, the less educated and those with less interest. Although we have not checked for the impact of the interaction of these three variables on political knowledge, a reasonable doubt exists that the least knowledgeable citizen groups are where the effect of inequality in political knowledge is intersectional. Information campaigns during periods of election should thus be directed towards these groups above all. The same socio-demographic groups that are well informed about general political issues also have high levels of EU and national political knowledge. There is also a relatively strong positive correlation between EU knowledge and national political knowledge. Surprisingly, the level of EU knowledge is higher in comparison to national political knowledge and less dependent on news attention and political interest. This could be the result of the European parliament election campaign that took place around the time of the survey. 13 The respondents were asked the following: Generally speaking, do you think that [COUNTRY’S] membership of the European Union is a good thing, a bad thing, or neither good nor bad? JOURNAL OF COMPARATIVE POLITICS 87 Thus, there is undoubtedly potential for further research on the effects of election campaigns on the levels of political knowledge. Besides the individual factors that explain political knowledge, researchers should in future dedicate some attention to country level contextual explanatory factors, such as how the level of nation and EU political knowledge differs between EU member states. While old member states seem to be disposed to higher levels of EU political knowledge, it appears that a positive attitude towards EU membership is connected with higher levels of EU political knowledge in comparison to the level of national political knowledge. Even though this article offers us an interesting insight into the level of political knowledge in Slovenia, the analysis however was limited due to the available data. Above all, we have to point out the positions of critics in terms of the measures of political knowledge and the knowledge gap hypothesis. Even though the survey knowledge questions are comparable between countries, they are at the same time factual questions that present an advantage for middle class respondents (Bonfdelli 2002, 65–84) and do not offer an insight into deep knowledge or so-called structural knowledge (Genova and Greenberg 1979, 79-91). In future research on the political knowledge of Slovenians, this has to be taken into account. Last but not least, let us point out that in last few years, the European Union has invested in the EU knowledge of their citizens, especially throughout the European Year of Citizens in 2013. These activities continued in 2014 through the election campaign for the European parliament. We can reasonably suspect that the level of political knowledge about the EU has improved in the last year and will continue to improve and may also diminish the knowledge gaps between the knowledge-rich and knowledge-poor groups, at least in terms of EU political knowledge. REFERENCES Bonfadelli, Heinz. 2002. “The Internet and Knowledge Gap: A Theoretical and Empirical Investigation.” European Journal of Communication 17 (1): 65–84. Cho, Jaeho and Douglas M. McLeod. 2007. “Structural Antecedents to Knowledge and Participation: Extending the Knowledge Gap Concept to Participation.” Journal of Communication 57 (2): 205–228. Delli Carpini, Michael X. 2000. “In Search of the Informed Citizen: What Americans Know About Politics and Why it Matters.” The Communication Review 4 (1): 129 ‒164. Delli Carpini, Michael X. and Scott Keeter. 1993. “Measuring Political Knowledge: Putting First Things First.” American Journal of Political Science 37 (4): 1179 ‒1206. Delli Carpini, Michael X. and Scott Keeter. 1994. “The Public's Knowledge of Politics.” In Public Opinion, the Press, and Public Policy, ed. J. David Kennamer, 19–40. Westport, CT: Praeger. Delli Carpini, Michael X. and Scott Keeter. 1996. What Americans Know about Politics and Why It Matters. New Haven: Yale University Press. Egmond, Marcel van, Brug, Wouter van der, Hobolt, Sara, Franklin, Mark and Eliyahu V. Sapir. 2013. European Parliament Election Study 2009, Voter Study. GESIS Data Archive, Cologne. ZA5055 Data file Version 1.1.0, doi:10.4232/1.11760. Ettema, James S. and F. G. Kline. 1977. “Deficits, Differences and Ceilings: Contingent Conditions for Understanding the Knowledge Gap.” Communication Research 4 (2): 179–202. Eveland, William P. Jr. and Myiah Hutchens Hively. 2009. “Political Discussion Frequency, Network Size, and 'Heterogeneity' of Discussion as Predictors of Political Knowledge and Participation.” Journal of Communication 59 (2): 205–224. JOURNAL OF COMPARATIVE POLITICS 88 Fiske, Susan T., Richard R. Lau and Richard A. Smith.1990. “On the Varieties and Utilities of Political Expertise.” Social Cognition 8 (1): 31–48. Fraile, Marta. 2011. “Widening or Reducing the Knowledge Gap? Testing the Media Effects on Political Knowledge in Spain (2004–2006).” International Journal of Press/Politics 16 (2): 163–184. Fraile, Marta. 2013. “Do information-rich contexts reduce knowledge inequalities? The contextual determinants of political knowledge in Europe.” Acta Politica 48 (2): 119 ‒143. Fraile, Marta. 2014. “Do Women Know Less About Politics Than Men? The Gender Gap in Political Knowledge in Europe.” Social Politics 21 (2): 261–289. Franklin, Charles H. 1991. “Eschewing Obfuscation? Campaigns and the Perception of U.S. Senate Incumbents.” The American Political Science Review 85 (4): 1193–1214. Gaziano, Cecilie. 1988. “Community Knowledge Gaps.” Critical Studies in Mass Communication 5 (4): 351–357. Gaziano, Cecilie. 2012. “Antecedents of Knowledge Gaps: Parenting Knowledge and Early Childhood Cognitive Development-Review and Call for Research.” The Open Communication Journal 6 (1): 17–28. Genova, B. K. L. and Bradley S. Greenberg. 1979. “Interests in News and the Knowledge Gap.” Public Opinion Quarterly 43 (1): 79–91. Grönlund, Kimmo and Henry Milner. 2006. “The Determinants of Political Knowledge in Comparative Perspective.” Scandinavian Political Studies 29 (4): 386 ‒406. Hindman, Douglas Blanks. 2009. “Mass Media Flow and Differential Distribution of Politically Disputed Beliefs: The Belief Gap Hypothesis.” Journalism and Mass Communication Quarterly 86 (4): 790–808. Holbrook, M. Thomas. 2006. “Cognitive Style and Political Learning in the 2000 U.S. Presidential Campaign.” Political Research Quarterly 59 (3): 343–352. Horstmann, Reinhold. 1991. “Knowledge Gaps Revisited: Secondary Analyses from Germany.” European Journal of Communication 6 (1): 77–93. Hwang, Yoori and Se-Hoon Jeong. 2009. “Revisiting the Knowledge Gap Hypothesis: A Meta- Analysis of Thirty-five Years of Research.” Journalism and Mass Communication Quarterly 86 (3): 513–532. Iyengar, Shanto and Donald R. Kinder. 1987. News That Matters. Television and American Opinion. Chicago, London: The University Chicago Press. Iyengar, Shanto. 1990. “Shortcuts to Political Knowledge: The Role of Selective Attention and Accessibility.” In Information and Democratic Processes, eds. John A. Ferejohn and James H. Kuklinski, 160–185. Chicago: University of Illinois Press. Jeffres, Leo W., Atkin, David and Hanlong Fu. 2011. “Knowledge and the Knowledge Gap: Time to Reconceptualize the 'Content'.” The Open Communication Journal 5: 30–37. Jennings, Kent M. 1996. “Political Knowledge Over Time and Across Generations.” Public Opinion Quarterly 60 (2): 228–252. Kropivnik, Samo. 2010. “Slovenski volivci na volitvah v Evropski parlament 2009.” In Politične vsebine in volilna kampanja, ed. Simona Kustec Lipicer, 27–49. Ljubljana: Fakulteta za družbene vede. Lippman, Walter. 1961. Public Opinion. New York: The Macmillan Company. Lupia, Arthur and Mathew D. McCubbins. 2003. The Democratic Dilemma. Can Citizens Learn What They Need To Know? Cambridge: Cambridge University Press. Luskin, C. Robert. 1990. “Explaining Political Sophistication.” Political Behaviour 12 (4): 331 ‒361. Luskin, C. Robert and John G. Bullock. 2011. “Don't Know” Means “Don't Know”: DK Responses and the Public's Level of Political Knowledge.” The Journal of Politics 73 (2): 547‒557. Mondak, Jeffrey J. 1999. “Reconsidering the Measurement of Political Knowledge.” Political Analysis 8 (1): 57–82. National Electoral Commission. Available at http://www.dvk-rs.si/. Park, Robert E. 1940. “News as a Form of Knowledge: A Chapter in the Sociology of Knowledge.” American Journal of Sociology 45 (5): 669 ‒686. Popkin, Samuel L. 1994. The Reasoning Voter: Communication and Persuasion in Presidential Campaigns. Chicago: University of Chicago Press. JOURNAL OF COMPARATIVE POLITICS 89 Scheufele, A. Dietram, Shanahan, James and Sei-Hill Kim. 2002. “Who Cares about Local Politics? Media Influences on Local Political Involvement, Issue Awareness, and Attitude Strength.” Journalism & Mass Communication Quarterly 79 (2): 427–444. Stokes, Donald E. 1962. “Party Loyalty and the Likelihood of Deviating Elections.” The Journal of Politics 24 (4): 689–702. Tichenor, Phillip J., Donohue, George A. and Clarice N. Olien. 1970. “Mass Media Flow and Differential Growth in Knowledge.” Public Opinion Quarterly 34 (2): 159–170. Tichenor, Phillip J., Donohue, George A. and Clarice N. Olien. 1980. Community Conflict and the Press. Beverly Hills: Sage Publications. Zaller, John. 1989. “Bringing Converse Back In. Modelling Information Flow in Political Campaigns.” Political Analysis 1 (1): 181 ‒234. Zaller, John. 1990. “Political Awareness, Elite Opinion Leadership, and the Mass Survey Response.” Social Cognition 8 (1): 125–153. Zaller, John. 2003. The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press.