Journal of Political Marketing, 11:8–26, 2012 Copyright # Taylor & Francis Group, LLC ISSN: 1537-7857 print=1537-7865 online DOI: 10.1080/15377857.2012.642705 Political Marketing: Structural Models of Advertising Influence and Voter Behavior ANDRZEJ FALKOWSKI and WOJCIECH CWALINA Warsaw School of Social Sciences and Humanities, Warsaw, Poland Research on the influence of political advertising, the role of media, and the roles of affect and the cognitive organization of political reality in controlling voter behavior often overlooks a fundamental regularity in human behavior. Humans perceive political events in causal relationships. The image of a politician is related to people’s feelings toward him or her, and the two influence each other during the political advertising experience. The resulting mutual causal relationship determines a voter’s choice behavior, the perception of the function of the media in political campaigns, and the affect toward a politician’s voting record. Such a causal analysis allows a test of two approaches by which voters organize political reality: the realist and constructivist. It also can be used for more detailed cross-cultural comparisons. This study analyzes these issues based on the sequential model of the influence of advertising on voting behaviors and the structural model of voter’s choice behavior in Poland and the United States. KEYWORDS constructivist approach, Polish voting behavior, realist approach, sequential model of advertising influence on voting behavior, structural model of voting behavior, U.S. voting behavior SEQUENTIAL PROCESSES IN THE PERCEPTION OF POLITICAL ADVERTISEMENTS Since the moment the first political broadcast was shown in 1952 by Dwight D. Eisenhower running for the office of the U.S. president, televised electoral The authors thank Christine Williams for detailed, constructive comments that substantially improved the final version of the manuscript. Address correspondence to Andrzej Falkowski, Department of Marketing Psychology, Warsaw School of Social Sciences and Humanities, ul. Chodakowska 19=31, 03-815 Warsaw, Poland. E-mail: andrzej.falkowski@swps.edu.pl or wojciech.cwalina@swps.edu.pl 8 Structural Models of Advertising Influence and Voter Behavior 9 communication has, with time, become the dominating element of all political promotion strategies (Diamond and Bates 1992). Until the 1960s, hard-sell advertising played the main role; it was based on repeating the persuasion broadcast several times so as to ‘‘cram’’ it into potential voters’ minds. However, after that period an important change occurred in the way of appealing to the viewers. The transition to soft-sell advertising (i.e., advertising whose main aim is to affect the viewer through an emotional broadcast) is closely connected with the media specialist Tony Schwartz (1973). Arguing that people already had strong attitudes about political issues, Schwartz contended that the job of advertising is to surface the feelings voters already have and give them a direction by associating them with values and meaningful images. The influence of such an approach is still present in political advertising, with consultants now, as Kern (1989) observes, especially keen on creating negative broadcasts, attacking their political opponents. Influencing voters’ emotional relation to a candidate or political party is connected with creating images of particular participants in the electoral competition (Abelson et al. 1982; Cwalina and Falkowski 2000). An image is a picture of a special kind created for a particular aim that, by evoking certain associations, gives the object additional values (e.g., socialpsychological, ethical, or relating to personality) and in this way contributes to an emotional perception of him or her. The task of the image is to secure a more emotional reception of the advertised object without disfiguring its essence (Cwalina, Falkowski, and Kaid 2000, 2005). Studies conducted so far in many countries confirm the hypothesis that political advertising affects a change in voters’ pictures of the candidate and leads to a reconfiguration of the arrangement of features of which that picture is composed (Kaid and Holtz-Bacha 1995). A majority of analyses conducted so far have focused exclusively on the changes in the candidate’s image resulting from the advertisements (Kaid and Chanslor 1995; Kaid and Holtz-Bacha 1995). However, they do not try to reveal the relations between the created image, one’s emotional attitude toward the candidate, and the voting intention. These aspects of political advertising’s effect, however, provide little understanding for voters’ overall behavior if they are analyzed separately. One can only say that viewing the spots leads to a decrease or an enhancement of the candidate’s evaluation without a possibility of analyzing the consequences of this evaluation for the electoral decision. Thus, in order to find out whether the spots in fact change voters’ image of the candidate in such a way that she or he is more or less certain whether to vote for a particular candidate, or even is going to switch candidate preferences (i.e., change his or her decision), it is necessary to find a causal link among the following four components: (1) cognitive-affective elements (candidate image), (2) general feelings toward the candidate, (3) intention for whom to vote, and (4) decision for whom to vote. 10 A. Falkowski and W. Cwalina Falkowski and Cwalina (1999; Cwalina et al. 2000, 2005) presented a causal model of the influence of persuasive messages which, in reference to applying political advertising, is presented in Figure 1. The first analyses conducted in order to verify the model were conducted during the 1995 presidential elections in Poland (Falkowski and Cwalina 1999; Cwalina and Falkowski 2000). They were based on regression analyses in which the image of presidential candidates operationalized by semantic differential scales influenced emotional attitude measured by a feelings thermometer. In this way one could observe to which features of a presidential candidate one reacted in an emotional way and to which he or she was not sensitive. Because the affect is significantly correlated with voter intention, then, by manipulating through advertising those features of a candidate which are sensitive to emotions, one may, indirectly, influence a voter’s behavior. Conducting the analyses before and after watching political spots allowed us to demonstrate how the power of the appeal of a candidate’s image to voters’ feelings is changing and how the configuration of those features significantly influences the affect toward a candidate. In this way one could learn to what extent political advertising reinforces or weakens the power of a candidate’s influence on emotions and, on a more detailed level, which features of the image have been reinforced or weakened as a result of advertising. The concept following the sequential model was also tested during general and presidential elections in such countries as France, Germany, FIGURE 1 Sequential model of the influence of advertising on voting behaviors (Falkowski and Cwalina 1999, p. 228). Structural Models of Advertising Influence and Voter Behavior 11 and the United States (Cwalina and Falkowski 2003; Cwalina et al. 2000, 2005). The results obtained there confirm causal relationships between the image and the affect toward politicians in shaping preferences toward them. Generally, one can state that the proposed research methodology is a particular method of controlling voter behaviors based on advertising. The political consultant may ‘‘warm up’’ or ‘‘cool down’’ the feeling toward the candidate by manipulating the selected and above-mentioned characteristics of the candidates’ images in a promotional-advertising strategy. Therefore, determining the causal relationship between the image and the feelings toward a politician allows development of a marketing tool that one can use to determine the efficiency of political advertising. The sequential model of the influence of advertising on voting behaviors emphasizing causal relationships among the elements distinguished in it is part of a broader cognitive context, where two approaches toward human experiencing of the surrounding physical and social reality clash: the constructivist and the realist one. What differentiates them is the structure of causal relationships connecting the affect with cognition. CONSTRUCTIVIST AND REALIST APPROACHES IN PERCEIVING THE POLITICAL ADS In psychology the constructivist approach was very well presented in Neisser’s classic work (1967) and in the collection edited by Bruner (1973), where the authors analyzed in detail the psychological mechanisms of creating the representations of reality in the human mind. According to this approach, man ‘‘goes beyond’’ the perceived stimuli, constructing his or her own world, as a cognitive representation or image of the reality that surrounds him or her (see also Fodor 1991; Pinker 1999; Scott 1995). Thus, it is possible to control this process of perception and interpretation by manipulating individuals’ knowledge. In practice, this has been studied for many years by marketing specialists who create various images of politicians or consumer goods. Nowadays, the process of such creation has become so significant that it is even said that modern democracy has found itself in ‘‘the age of manufactured images,’’ as Newman (1999a) put it in the title of one of his books. Although the constructivist approach is the most dominant research paradigm for human cognition, some researchers argue that it is not correct (e.g., Gibson 1979; Reed 1996). Instead, they assume the so-called cognitive realism approach according to which humans directly perceive the information in their environment. Therefore, the other approach, termed realist, points to the relevance of affects in the formation of the image itself. Thus, taking into account the following simple sequences: cognition ! affect or affect ! cognition, we are using either a constructivist or realist approach to perceive the social 12 A. Falkowski and W. Cwalina environment. The primary criterion to differentiate the above approaches is whether perception of affect is direct. According to the constructivist approach, the affect is the result of stimuli processed in cognitive structures. According to the realist approach, the affect is the result of direct perception and is independent of information processing (Zajonc 1980). These two approaches in organizing social reality allow us to look at voter behavior from a much broader perspective, defining the causal relationship between cognition and affect in human functioning in the social world. Therefore, it concerns both the essence of social cognition and its influence on behavior, whose particular case is social behavior. Namely, the political marketing strategy should follow the assumed and empirically tested epistemic background of voter behavior: Do voters recognize the social reality on an affective basis, or is affect a consequence of social cognition such that the images of particular social phenomena have little in common with the real world? Depending on which one of the two following causal relationships has been selected on empirical grounds, a different detailed marketing strategy should be elaborated. On the other hand, one should not neglect the possibility of combining the strategies resulting from realist and constructivist approaches as both can be used in explaining voter behavior. The following section of the paper presents the importance of the studies of the influence of advertising’s perception influence in political image formation in the constructivist and realist approach. These two approaches also allow one to capture cross-cultural differences, which will be demonstrated by the examples of voter behaviors in Poland and the United States. POLITICAL IMAGE FORMATION: A CONSTRUCTIVIST APPROACH A hypothesized sequential model of the influence of advertising on voting behaviors is presented in Figure 1, whereas Figure 2 presents its constructivist variant tested in the research in Poland and the United States. According to the sequential model, the image, operationalized by two factors—Image1 and Image2—influences the general feelings in the first sequence. In the second sequence, after viewing the advertisements, some image reconfiguration as well as change in affect takes place in the voters’ minds. According to the constructivist approach, one can assume both the influence of image on affect within the sequence and the influence of image on image as well as affect on affect between the sequences. To study the image empirically, one can use semantic differentials consisting of a number of adjective scales (e.g., qualified= not qualified, honest=dishonest, trustworthy=untrustworthy) and then with factor analysis generate Image1 and Image2. This method makes it much easier to control the causal relationship between affect and cognition. We employed this method to study the perception of political advertising during the presidential elections in Poland and the United States in 2000. Structural Models of Advertising Influence and Voter Behavior 13 FIGURE 2 Structural equation constructivist sequential model of advertising influence (Cwalina, Falkowski, and Kaid 2005, p. 26). Here we summarize the general results and conclusions; the detailed procedure and materials can be found in Cwalina et al. (2005). The serious candidates for office in those two countries’ elections were Aleksander Kwaśniewski and Andrzej Olechowski (Poland) and George W. Bush and Al Gore (United States). The degrees of goodness of fit of the empirical data to the theoretical sequential model (Figure 2) showed that the constructivist model provided similar explanations of the voting behavior of Kwaśniewski’s and Olechowski’s supporters. The goodness-of-fit indices (v2) for both revealed significant differences between the models and empirical data, leading us to reject the constructivist solution. Somewhat different results were obtained for the candidates in the U.S. presidential election. In the case of the structural constructivist model for Bush, the fit to empirical data was poor. On the other hand, the model for Gore met the goodness-of-fit criterion. The next task is to examine the alternative option, namely the realist approach in organizing political reality in voters’ minds. In this approach the way of creating image is modified by affect. POLITICAL IMAGE FORMATION: A REALIST APPROACH According to the realist approach to research on social cognition, one assumes the following causal sequence: affect ! cognition, that is, affective feeling toward the candidate influences his or her image. This approach is well 14 A. Falkowski and W. Cwalina FIGURE 3 Structural equation realist sequential model of advertising influence (Cwalina, Falkowski, and Kaid 2005, p. 33). supported by Zajonc’s (1980) cognitive theory of affects, which demonstrated, on the basis of a number of empirical studies, that an affective reaction evoked by the perceived object appears first and is only then followed by cognitive elements. Therefore, it is quite reasonable that the affect plays an important role in forming the image of politicians. In terms of the sequential model of advertising influence presented in Figure 2, the arrows connecting the image with the affective feelings should be reversed as presented in Figure 3. In the case of the model for both Kwaśniewski and Olechowski, the goodness-of-fit coefficients showed a statistically significant difference between the realist model and empirical data. In contrast, an opposite result was obtained for each of the candidates in the U.S. presidential election. Here, the structural realist models were a good fit for both Bush and Gore. COMPARATIVE ANALYSIS Figure 4 presents a comparison of the accuracy of the constructivist and realist sequential models expressed by fitting v2 parameters. The results of empirically testing the causal relationships between affect and image in structural models for particular countries and particular candidates are quite unequivocal. The degrees of the models’ goodness of fit presented in Figure 4 point to the slightly higher explanatory power of the realist approach, compared to the constructivist approach to explaining voting behaviors. This means that affects form a politician’s image in a significant Structural Models of Advertising Influence and Voter Behavior 15 FIGURE 4 Constructivist and realist structural sequential models fitting parameters. way. That is, affective attitude toward a politician is the primary reaction of voters when encountering a specific candidate. Only on this basis do voters make cognitive appraisals and attribute the characteristics of which the politician’s campaign image consists. Therefore, from the marketing point of view, the key task is to evoke a positive affective attitude toward the promoted politician, and only subsequently should one suggest to voters that these affects are connected with the politician’s particular attributes. The results of the research conducted in Poland and the United States suggest that there are different patterns of advertising’s influence on citizen voting behaviors in these two countries. In the case of Poland, this process is better described by realist models, assuming that the key factor for influencing voting behaviors is evoking positive emotions toward a candidate and then providing voters with a justification for such an affect. In the United States, on the other hand, both approaches allow one to describe, in a similar way, voting behaviors. It suggests that in the case of such established democracies as the United States, voters follow and analyze more carefully messages from presidential candidates’ campaigns than they do in evolving democracies, such as Poland. A MODEL OF VOTER’S CHOICE BEHAVIOR Multivariate models are an advanced step in the development of voter behavior analysis that are created across the ‘‘old’’ paradigms (Johnston, Pattie, and Allsopp 1988; Newman and Sheth 1985). These models have been subject to empirical tests during elections at various levels and in different countries and seem to offer a better chance to explain the foundations of voter behavior. 16 A. Falkowski and W. Cwalina One proposal that offers promising theoretical and applicational solutions is Newman and Sheth’s (1985) voter’s choice behavior model (see also Newman 1999b). According to this model, the following seven domains are assumed to guide voter behavior: (1) issues and policies–refers to a list of salient issues and policies including economics, foreign, and social policies as well as leadership characteristics; (2) social imagery–refers to all relevant segments of the voting population likely to be supportive of the candidates being studied; (3) emotional feelings–represent the voters’ emotional attitude toward the candidates; (4) candidate image–refers to the candidate’s image based on personality traits; (5) current events–refers to issues and policies that develop during the course of the campaign; (6) personal events–refers to situations in the personal life of the candidate; and (7) epistemic issues–refers to those aspects of the candidate that would provide the perceived satisfaction of voters’ curiosity, knowledge, and exploratory needs. These cognitive domains are presented in Figure 5. This model includes an implicit assumption that all these domains are independent from each other and, potentially, equally important for determining voting decisions. It determines, then, a particular approach to the empirical analysis of data: regression analysis or discriminant analysis. However, it is well documented in psychological literature that cognitive and emotional elements should be treated not as separate, but as interactive factors (Falkowski and Cwalina 1999; Cwalina et al. 2000, 2005). Therefore, one can pose a question regarding possible causal relationships among the set of variables previously treated as distinct and independent, and this concerns, in particular, the proper placement in this causal chain of a voter’s emotional attitudes toward candidates. FIGURE 5 Model of voter’s choice behavior (Newman and Sheth 1985, p. 179). Structural Models of Advertising Influence and Voter Behavior 17 STRUCTURAL MODELS OF VOTER’S CHOICE BEHAVIOR: CONSTRUCTIVIST AND REALIST APPROACHES The causal relationship between the image and affect can be determined for the model of voter’s choice behavior in a manner similar to that for the sequential model of the influence of advertising on voting behavior. This reinterpretation of the model of voter’s choice behavior requires describing the causal structure between the cognitive domains and affect components, which then allows analysis of vote behaviors within the constructivist and realist approaches (Cwalina, Falkowski, and Newman 2008, 2010). The original model developed by Newman and Sheth (1985) does not include the function of the media and election polls as important factors that mediate and oftentimes directly influence voter behavior. Therefore, the model should be supplemented by the media, understood as an additional cognitive domain which reflects voters’ beliefs or imaginations concerning the importance of the media for political elections in general. The structure of causal relationships is intuitively obvious here: the media influences cognitive domains that, in turn, form voter’s emotions toward a candidate. The assumption for this model, namely that emotions influence directly voter intention, is based on the results of empirical research which proved, unequivocally, a very strong relation between these two variables (Cwalina et al. 2000). It seems then that emotional feelings are a good predictor of voting intentions, which means that a voter chooses the candidate who is ‘‘warmer’’ and evokes more positive emotions. FIGURE 6 Structural constructivist model of voter behavior (Cwalina et al. 2004, p. 13). 18 A. Falkowski and W. Cwalina Such an approach assumes the constructivist approach in organizing political reality by voters, where cognitive domains influence the development of the affect toward a politician; it is a similar relationship to that occurring in the case of political advertising influence, where the image of a politician influences the affect toward him or her (compare Figure 2 and Figure 6). However, the relationships between them can also be reversed and one can test the realist approach, where the affect determines the way of organizing cognitive domains in voters. A question arises then how the voters develop in their minds the internal structure in which affect and cognition intertwine. In order to answer this question, one needs to test empirically the structural models of voter behavior in the constructivist and realist approaches. The procedure of testing structural models of voting behavior in the constructivist and realist approaches was presented by Falkowski and Cwalina (2002), Cwalina, Falkowski, Newman, and Verčič (2004), and Cwalina et al. (2008, 2010). Here, we summarize those results. CONSTRUCTIVIST MODEL OF VOTER BEHAVIOR Although the research on voter behavior examining the structural model and following the constructivist approach was conducted in Polish, American, and Slovenian samples, the research following the realist approach includes only Polish and American samples, limiting our comparisons to those two countries. There are many causal combinations that can be derived from the six cognitive domains, the media, and affect (emotional feelings). Thorough testing of these potential combinations revealed that the model best fitted to voter behaviors was the one depicted in Figure 6. In this model, Media 1 and Media 2 resulted from a principal component analyses of the responses obtained from the media questions in a survey of voters’ preferences. (See Cwalina et al. 2008, for a description of this survey and question set.) Based on the best fitting model depicted in Figure 6, we conducted a comparative analysis of voter behavior in Poland and the United States using the constructivist approach. The essence of the model is that the affect toward a politician is the result of cognitive domains developed in voters’ minds as a result of media influence. In both countries the research was conducted during the 2000 presidential campaigns. In Poland there were only two candidates who were seriously considered in that election: Aleksander Kwaśniewski, the incumbent, and Andrzej Olechowski, the challenger; in the United States the two major party presidential candidates were George W. Bush and Al Gore. A detailed presentation of the research sample and procedure is discussed in Cwalina et al. (2004) and Cwalina et al. (2008). The goodness of fit of the constructivist model for the Polish empirical data was poor for both Polish candidates, which rules out that solution. Structural Models of Advertising Influence and Voter Behavior 19 However, in the United States, the goodness of fit was very good for both candidates. This finding gives rise to speculation that the evolving democracy in Poland has not yet developed to the point that voters can organize reality using the constructivist approach, which was evident in American voters’ behaviors. By conducting analyses following the realist approach, we have a means of answering this question. The realist approach assumes that it is the affect that influences cognitive domains. REALIST MODEL OF VOTER BEHAVIOR According to the realist approach, cognitive domains are formed in voters’ minds on the basis of affect toward the candidate, which formed in response to the media. Figure 7 presents the best goodness-of-fit model from those tested applying the realist approach. This model assumes a mutual interaction between media and cognitive domains, which is moderated by affect or emotional feelings. COMPARATIVE ANALYSIS Figure 8 shows the fitting parameter v2 for the best models under each approach, which facilitates comparison of the two. FIGURE 7 Structural realist model of voter behavior (Cwalina, Falkowski, and Newman 2010, p. 356). 20 A. Falkowski and W. Cwalina FIGURE 8 Constructivist and realist structural models fitting parameters (Cwalina, Falkowski, and Newman 2010, p. 365). These results show that from the statistical point of view, the goodness of fit (v2) of the constructive model is poor for Poland, but acceptable for the United States. On the other hand, the realist model is acceptable for both candidates in the United States, but only for one in Poland: Olechowski. Based on these results, the realist paradigm explains voter behavior better than the constructivist one, which means that cognition of the candidate is already ‘‘colored’’ by affect. In other words, the key factor influencing voting behavior is evoking positive emotions toward the candidate and then providing voters with a justification for such affect. That means that some of the cognitive domains directly influencing voter behavior are already distorted by emotional feelings. In the United States, however, one can say that both approaches are relevant. This means that in an established democracy like the United States, voters already have learned to analyze messages from presidential campaigns cognitively. These American voters are more resistant to the unconscious power of affect than are voters in evolving democracies, such as Poland. Obviously such conclusions should be treated as a hypothesis for further testing during comparative analysis on the basis of other countries both with well-developed and developing democracy. THE REALIST VERSUS CONSTRUCTIVIST APPROACH TO ORGANIZING POLITICAL REALITY: IMPLICATIONS FOR FURTHER RESEARCH In the sequential model, voter behavior is influenced by advertising, while in the structural model of voter’s choice behavior is influenced by the Structural Models of Advertising Influence and Voter Behavior 21 broad cognitive concept of organizing political reality in its media, cognitive, and emotional dimensions, based on the assumption of causal relationships among its elements. Such an analysis affords a much broader research perspective in the sphere of political behaviors as well as in the broad context of human social functioning. This can be presented as follows. In the broad context of human social behavior, such cause-effect analysis allows researchers to test two approaches to explaining human cognitive and affective functioning: the constructivist and realist one. The results presented here provide a strong test for the specificity of social cognition of the surrounding political reality. Our findings have demonstrated that the realistic approach better explains human behavior in the social environment (see Figures 4 and 8). The slight advantage of the realist paradigm over the constructivist one is consistent with the Zajonc’s (1980) cognitive theory of affect explaining why emotions precede cognition and Wattenberg’s (1987) statement that although most of the voters know nothing about particular politicians, they still have strong feelings toward them. These results are also consistent with other research into human behavior. In the psychological theory of decisions, people apply heuristics to understand the events they experience (Kahneman, Slovic, and Tversky 1982). Such heuristics distort reality, however, and do not always lead to correct predictions. From this one can infer that one of the elements distorting perceived events is affect preceding cognition. Causal relationships applied in the sequential model and in the structural model of voter behavior allow one to conduct cross-cultural analyses. The studies presented here comparing voter behaviors in Poland and the United States reached the same conclusion by testing both the sequential model of the influence of political advertising on voting behavior as well as the structural model of voter behavior in the two countries. The analyses of structural equation methodology used in the research showed that the realist approach worked better than the constructivist one for Poland, and both were applicable to the United States. Such a result allows one to conclude that Poles are more emotional and less rational than Americans in their voter behavior. We postulate that this results from well-established democratic processes in the United States that allow American voters to conduct more deliberate calculation and a more cognitive analysis of their political environment. In Poland, however, the democratic processes are not as fully developed yet. One may assume that they apply, to a large extent, to older voters; therefore, additional analyses would have to be performed that would constitute additional evidence that political maturity or conditioning makes older voters more resistant or slower to adjust to new processes. Our conclusion that Polish voters’ political behaviors are controlled to a larger extent by emotions (Cwalina and Falkowski 2008) is based on a comparative analysis that clearly demonstrates that the degrees of the goodness 22 A. Falkowski and W. Cwalina of fit of data to the constructivist structural model of voter behavior are better for the American voters. Further studies conducted in other countries on voter behavior are also very promising. Such a research concept focused on conducting cross-cultural comparisons has been proposed by Cwalina et al. (2008). The analysis of voter behaviors in Poland, the United States, and Slovenia proposed there may constitute a quantitative approach toward defining the level of advancement of various democratic processes in different countries by comparing the constructivist and realist paradigms in voter behavior. Similar cross-cultural research on democratization processes could also be conducted within the sequential model of political advertising’s influence. In addition to the studies presented here showing the differences in receptivity to political advertising in Poland and the United States, another study applies the sequential model to compare voting behavior models for France and Germany (Cwalina et al. 2000). The research presented here and using the methodology of structural equations shows the specificity of emotional and cognitive mechanisms in voter behavior. Using the example of a country with well-developed and established democratic processes—namely the Unites States—and one where democracy is still developing, one may observe various ways of initiating those mechanisms. In Poland, voter behavior is still shaped by emotions that influence the way of experiencing political events defined by cognitive domains. In the United States, however, voter behavior is shaped by political events not distorted by emotions. It is only by experiencing these events that one develops some affect toward a politician. PRACTICAL IMPLICATIONS FOR MARKETING STRATEGY The research on voter behavior presented here may be an introduction to research on international marketing and has a number of practical implications for campaign managers taking decisions and actions in established democracies such as the United States and in evolving democracies, such as Poland. The research results on sequential processes in the perception of political advertisements suggests that Polish voters are more sensitive to marketing actions than American voters. This can be explained by the fact that Polish people are much less familiar with the strategies of influence used during the making of political advertising than Americans are. Another consideration that may account for some of the differences in the reactions of Polish and U.S. citizens is the content of the advertising. Content analyses of advertising in the two countries have shown that, in general, U.S. political advertising is more likely to rely on logical proof, whereas political spots in Poland make more use of emotional proof (Kaid 1999). This could explain some of the findings that suggest that Polish voter responses can be explained more fully by the Structural Models of Advertising Influence and Voter Behavior 23 realist approach, with its emphasis on affect, than the constructivist. Further analyses of marketing consequences resulting from adapting a particular approach to understanding voting behaviors confirm the important differences between Poland and the United States. Attempts to select important attributes of the images that influence the forming of emotions toward a politician (constructivist approach) as well as attempts to find argumentation for evoked feelings (realist approach) lead, in the case of the two countries, to slightly different conclusions. Therefore, the best way to influence voting behavior in Poland is to form positive emotions toward the candidates regardless of the argumentation in the messages from candidate’s campaign. In the United States, on the other hand, it can be said that both approaches, the realist and constructivist, are relevant. Therefore, the best way to influence American voters is to present logical argumentation in the messages of the candidate that, subsequently, shape emotional attitude toward the candidate. This same conclusion was reached by testing the model of voter choice behavior. The analyses have demonstrated that the realist approach worked better than the constructive paradigm for Poland, and both are applicable for the United States. Such a result, again, allows one to state that Poles are more emotional and less rational than Americans in their voter behavior. The presented division into constructivist and realistic paradigms in the research on voter behaviors may better demonstrate the differences between traditional and evolving democracies. From the marketing point of view, political consultants face an important challenge: to grasp the structure of the factors influencing the support for a candidate-employer. The results presented here suggest that this is very individualized and, probably, changes over time. Therefore, in conducting campaigns for various politicians, consultants must make every effort each time to understand the specificity of voter behaviors toward the candidate and the role of situational factors (media and polls, current events, etc.). Moreover, the research presented here points to the important elements on which one must concentrate in order to conduct an effective electoral campaign. The models of voter behavior developed for ‘‘one’s own’’ candidates may not have any application when analyzing opponents. Political opponents ‘‘activate’’ a distinctly different pattern of factors influencing their support. That requires a separate, individualized approach from political marketing specialists. Therefore, no universal ways of running such campaigns should be developed because different cognitive domains of different candidates are relevant for voting preferences. The promotional campaign of candidates for state offices should therefore be developed individually for each candidate. The individualization of political campaigns also concerns the political marketing consultants who wish to work in a foreign country. They must be prepared for a certain modifications in their own ways of working. Such an individually developed strategy can be prepared on the basis of an 24 A. Falkowski and W. Cwalina analysis of cause-effect relations using the methodology of structural equations presented in this article. From the practical perspective, the presented analyses allow one to measure precisely and monitor emotional attitude toward both candidates and their images. In this way the proposed methodology may be a useful tool for planning and conducting an election campaign. In this context it seems that both approaches presented here, realist and constructivist, may provide complementary sources of information about meeting the goals of a political campaign. REFERENCES Abelson, R. P., D. R. Kinder, M. D. Peters, and S. T. Fiske. (1982). Affective and semantic components in political person perception. Journal of Personality and Social Psychology, 42(4), 619–630. Bruner, Jerome S. (1973). Beyond the information given. Studies in the psychology of knowing. New York: W. W. Norton and Company. Buck, Ross. (1988). Human motivation and emotion. New York: John Wiley. Cwalina, W. and A. Falkowski. (2000). Psychological mechanisms of political persuasion: The influence of political advertising on voting behavior. Polish Psychological Bulletin, 31(3), 203–222. Cwalina, W. and A. Falkowski. (2003). Advertising and the image of politicians. National elections in Poland, France, and Germany. In Flemming Hansen and Lars B. Christensen (Eds.), Branding and advertising (pp. 205–231). Copenhagen: Copenhagen Business School Press. Cwalina, W. and A. Falkowski. (2008). Constructivist mind: False memory, freedom, and democracy. Journal of Political Marketing, 7(3=4), 239–255. Cwalina, W., A. Falkowski, and L. L. Kaid. (2000). Role of advertising in forming the image of politicians: Comparative analysis of Poland, France, and Germany. Media Psychology, 2(2), 119–146. Cwalina, W., A. Falkowski, and L. L. Kaid. (2005). Advertising and the image of politicians in evolving and established democracies: Comparative study of the Polish and the U.S. presidential elections in 2000. Journal of Political Marketing, 4(2=3), 19–44. Cwalina, W., A. Falkowski, and B. I. Newman. (2008). A cross-cultural theory of voter behavior. New York: Haworth Press=Taylor & Francis Group. Cwalina, W., A. Falkowski, and B. I. Newman. (2010). Towards the development of a cross-cultural model of voter behavior: Comparative analysis of Poland and the US. European Journal of Marketing, 44(3=4), 351–368. Cwalina, W., A. Falkowski, B. I. Newman, and D. Verčič. (2004). Models of voter behavior in traditional and evolving democracies: Comparative analysis of Poland, Slovenia, and U.S. Journal of Political Marketing, 3(2), 7–30. Diamond, Edwin and Stephen Bates. (1992). The spot: The rise of political advertising on television. Cambridge: MIT Press. Falkowski, A. and W. Cwalina. (1999). Methodology of constructing effective political advertising: An empirical study of the Polish presidential election in 1995. In Structural Models of Advertising Influence and Voter Behavior 25 Bruce I. Newman (Ed.), Handbook of political marketing (pp. 283–304). Thousand Oaks, CA: Sage Publications. Falkowski, A. and W. Cwalina. (2002). Structural models of voter behavior in the 2000 Polish presidential election. Journal of Political Marketing, 1(2=3), 137–158. Fodor, Jerry A. (1991). A theory of content and other essays. Cambridge, MA: MIT Press. Gibson, James J. (1979). The ecological approach to perception. Boston: Houghton Mifflin. Johnston, Ron J., Charles J. Pattie, and J. Graham Allsopp. (1988). A nation dividing? The electoral map of Great Britain 1979–1987. London: Longman. Kahneman, Daniel, Paul Slovic, and Amos Tversky. (1982). Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press. Kaid, Lynda Lee. (1999). Comparing and contrasting the styles and effects of political advertising in European democracies. In Lynda Lee Kaid (Ed.), Television and politics in evolving European democracies (pp. 219–236). Commack, NY: Nova Science Publishers. Kaid, Lynda Lee and Mike Chanslor. (1995). Changing candidate images: The effects of political advertising. In Kenneth L. Hacker (Ed.), Candidate images in presidential election (pp. 83–97). Westport, CT: Praeger. Kaid, Lynda Lee and Christina Holtz-Bacha (Eds.). (1995). Political advertising in Western democracies: Parties and candidates on television. Thousand Oaks, CA: Sage Publications. Kern, Montague. (1989). 30-second politics: Political advertising in the eighties. New York: Praeger. Neisser, Ulric. (1967). Cognitive psychology. New York: Appleton-Century-Crofts. Newman, Bruce I. (1999a). The mass marketing of politics: Democracy in an age of manufactured images. Thousand Oaks, CA: Sage Publications. Newman, Bruce I. (1999b). A predictive model of voter behavior: The repositioning of Bill Clinton. In Bruce I. Newman (Ed.), Handbook of political marketing (pp. 259–282). Thousand Oaks, CA: Sage Publications. Newman, B. I. and J. N. Sheth. (1985). A model of primary voter behavior. Journal of Consumer Research, 12(2), 178–187. Pinker, Steven. (1999). How the mind works. New York: W. W. Norton & Company. Reed, Edward S. (1996). Encountering the world: Toward an ecological psychology. New York: Oxford University Press. Schwartz, Tony. (1973). The responsive chord. Garden City, NY: Anchor Press. Scott, Alwyn. (1995). Stairway to the mind. New York: Springer-Verlag. Wattenberg, M. P. (1987). The hollow realignment: Partisan change in a candidatecentered era. Public Opinion Quarterly, 51(1), 58–74. Zajonc, R. B. (1980). Feeling and thinking: Preferences needs no inferences. American Psychologist, 35(2), 151–175. AUTHOR NOTES Andrzej Falkowski is a Professor of Psychology and Marketing, and head of the Department of Marketing Psychology at the Warsaw School of Social Sciences and Humanities in Poland. He has been a Fulbright Scholar at the 26 A. Falkowski and W. Cwalina University of Michigan. His publications include numerous articles in consumer behavior, political marketing, and cognitive psychology journals, as well as chapters and books. He is co-author of A Cross-Cultural Theory of Voter Behavior (Haworth Press, 2007) and Political Marketing: Theoretical and Strategic Foundations (M.E. Sharpe, 2011). He is advisory editor of the Handbook of Political Marketing, and editorial board member of the Journal of Political Marketing. Wojciech Cwalina is Professor in the Department of Marketing Psychology at the Warsaw School of Social Sciences and Humanities, Poland. His research interests include political marketing, social psychology, and media psychology. He is a member of the editorial board for Psychologia Społeczna (Social Psychology) and the Journal of Political Marketing. He is the author of Television Political Advertising (TN KUL, Poland, 2000), co-author of Political Marketing: Psychological Perspective (GWP, Poland, 2005), A Cross-Cultural Theory of Voter Behavior (Haworth Press, 2007), Political Marketing: Theoretical and Strategic Foundations (M.E. Sharpe, 2011), and numerous articles (e.g., in Media Psychology; Journal of Political Marketing, European Journal of Marketing, Journalism Studies) and book chapters.