10.4185/RLCS-2020-1436
Article

THE PERCEPTION OF THE DEBATES AS A DECISION FACTOR IN ELECTORAL BEHAVIOR IN THE GENERAL ELECTIONS OF APRIL 2019

LA PERCEPCIÓN DE LOS DEBATES COMO FACTOR DE DECISIÓN EN EL COMPORTAMIENTO ELECTORAL EN LAS ELECCIONES GENERALES DE ABRIL DE 2019

Nieves Lagares-Diez1
Erika Jaráiz-Gulías1
Paloma Castro-Martínez1

1University of Santiago de Compostela. Spain

ABSTRACT
Introduction. This paper analyses the effects of the electoral debates on citizens’ perceptions about the candidates and on the vote for the main parties in the General Elections of April 2019.
Methodology. Descriptive analysis and generalized linear models with link logit to explain the vote, through the use of the Post-Election Study No. 3248 of the General Elections, 2019 prepared by the Center for Sociological Research (CIS by its acronym in Spanish).
Results. The models show that the perception about which candidate was the most convincing in the electoral debates significantly influences the vote.
Discussion and conclusions. Electoral debates have mainly a reinforcing effect on both the vote and the leadership of the participants in these.

KEYWORDS: electoral debates; political communication; electoral behavior; vote.

RESUMEN
Introducción. En el presente trabajo se analizan los efectos de los debates electorales sobre las percepciones de los ciudadanos acerca de los candidatos y sobre el voto a los principales partidos en las Elecciones Generales de abril de 2019.
Metodología. Análisis descriptivo y modelos lineales generalizados con link logit para explicar el voto, mediante el empleo del Estudio nº 3248 Postelectoral Elecciones Generales, 2019 elaborado por el Centro de Investigaciones Sociológicas (CIS).
Resultados. Los modelos muestran que la percepción acerca de qué candidato resultó el más convincente en los debates electores influye significativamente en el voto.
Discusión y conclusiones. Los debates electorales tienen un efecto principalmente de refuerzo tanto del voto como de los liderazgos de los participantes en estos.

PALABRAS CLAVE: debates electorales; comunicación política; comportamiento electoral; voto.

Received: 20/08/2019.
Accepted: 15/09/2019.
Published: 30/04/2020.

Correspondence:
Nieves Lagares Diez. University of Santiago de Compostela. Spain
mnieves.lagares@usc.es
Erika Jaráiz Gulías. University of Santiago de Compostela. Spain
erika.jaraiz@usc.es
Paloma Castro Martínez. University of Santiago de Compostela. Spain
paloma.castro2@usc.es

How to cite this article / Standard reference
Lagares Diez, N., Jaráiz Gulías, E. & Castro Martínez, P. (2020). The Perception of the Debates as a Decision Factor in Electoral Behavior in the General Elections of April 2019. Revista Latina de Comunicación Social, (76), 39-58. https://www.doi.org/10.4185/RLCS-2020-1436

CONTENTS
1. Introduction. 2.1 The electoral debates and their effects. 2.2 The debates in the General Elections campaign of April 2019: A round trip. 3. Methodology and data. 4. Results. 5.Discussion and conclusions. 6. References.

Translation by Paula González (Universidad Católica Andrés Bello,Venezuela).

1. Introduction

In audience democracies, as Manin defines them (2006), electoral debates are key acts of election campaigns. The particularity of these events lies in the fact that the traditional logic of the campaign, which is based on the relationship between candidates and citizens, is substituted by the direct interaction of the candidates (Castro et al., 2017). This generates, at the same time, the attention of a great part of the electorate, which, when issued through the mass media, usually arrives at times of maximum audience; even to targets that usually are not recipients of political information (as is the case with undecided voters).
Although, the electoral debates are not only part of the media’s information offer as a result of the “mediatization” of contemporary societies, but also respond to the citizens’ information demand. The latter, in the search for their cognitive resources, find in the debates the propitious opportunity to meet the candidates, their programmatic offer, their political positions before a series of issues, as well as their style and aptitudes. In fact, for the public, electoral debates may be the only opportunity to see and judge candidates in a face-to-face context (Schrott, 1990).
Electoral debates turn out to be crucial because, even assuming this active attitude of the electorate, individuals, as rational beings, try to economize resources (such as time) destined to the knowledge of the political offer to shape their preferences and subsequent decision making regarding the participation and orientation of the vote (Downs, 1957). In this sense, the translation of the idea of the “cognitive shortcuts” of the Michigan School to the context of the electoral debates, makes them conceivable as “informative shortcuts” (Castro et al., 2017). If the basic function of an electoral campaign is to provide information, the electoral debates are one of the sources of information of the electoral campaign (Crespo et al., 2011), also fulfilling, in the opinion of Best and Hubbard (1999), three normative functions:

  1. Electoral debates involve viewers in the campaign: they increase their involvement, encouraging voters to show interest in the candidates and the issues they discuss, and their participation in the election.
  2. Electoral debates inform potential voters of the issues: they consider the importance of the various political issues and the alternatives offered as solutions to them, allowing viewers to evaluate the agenda of issues and political positions from the different candidates regarding them.
  3. Electoral debates inform the audience about the political leaders: they reveal the qualifications, the characteristic features, and the political positions about the topics being discussed, allowing the spectators to evaluate the candidates and decide the orientation of their vote.

Although these are the functions of the electoral debates, the main effect attributed to them is, besides a greater knowledge of political leaders, the effect of reinforcing the electorate, regardless of whether they can have an impact on undecided citizens (Best and Hubbard, 1999; Yawn and Beatty, 2000; Sierra, 2012; Gallego and Bernárdez, 2017).
Thus, the objective of this work is to find out if the electoral debates held on the occasion of the General Elections of April 2019 have influenced the perceptions that citizens have about the main political leaders who attended these elections, and if, ultimately, they have had some kind of effect on the decision to vote for political parties.

2.1. The electoral debates and their effects

Electoral debates have become fundamental events within the course of electoral campaigns, from their appearance in 1952 in the United States until their conversion into a typically American institution in 1976, years after the famous debates between Nixon and Kennedy (Jamieson and Birdsell, 1988). In the case of Spain, where the Organic Law of the General Electoral Regime (LOREG by its acronym in Spanish) does not require their realization, they are a key event, because, since the first debate between Felipe González and José María Aznar in 1993, and after fifteen years without electoral debates for General Elections, they have been held with continuity in all national election campaigns since 2008, when the face-to-face was played between Mariano Rajoy and José Luis Rodríguez Zapatero.
However, from the traditional face to face, in recent years, four-way debates have been held, as a result of the increase in political pluralism caused by the strong incursion of new political parties on the national scene, primarily Ciudadanos (C´s) and Unidas Podemos. This increase in the number of relevant parties in the state scene, and whose leaders, therefore, participate in the debates, has not only had consequences in the electoral results and the subsequent formation of governments but also represents an added difficulty to the measurement of the effects that such debates may cause on the political behavior of voters.
In this regard, studies on electoral debates can be classified into two large groups (García, 2015): the works that analyze the structure and content of the debates (Weiler, 1989; Benoit and Wells (1996); Benoit and Harthcock, 1999; Benoit and Brazeal, 2002; Herrero and Benoit, 2009) and the research that analyze the effects of the debates on the spectators (Abramowitz, 1978; Bishop et al., 1978; Schrott, 1990; Benoit, McKinney, and Lance Holbert, 2001; Benoit, Hansen, and Verser, 2003; Benoit et al., 2004; Cho and Ha, 2012).
In Spain, research which objective is to demonstrate the impact of the debates on voters, that is, the knowledge and evaluations of the participants, the reinforcement and changes in voting intentions, have been poorly developed (Herranz-Rubio, 2020), proof of them are the works of Lledó (2001), Luengo (2011) and Sierra (2012).
Therefore, the effects of electoral debates, a perspective in which this work is framed, can be attitudinal, that is, how electoral debates can condition the perceptions and assessments of citizens towards political leaders; or behavioral effects, how electoral debates can affect the voting decision of voters (Lledó, 2001).
However, there is a lack of consensus among the previous works. On the one hand, even though televised debates can reach a large part of the population (as evidenced by audience ratings), they are generally only seen by those who already express a great interest in politics and, therefore, are the most informed. According to the limited effects model, according to which electoral campaigns have a minimal effect on the outcome of the elections (Barreiro et al., 2015), the electoral debates would not have any kind of effect beyond the reinforcement of voters’ predispositions, since individuals are reluctant to assimilate new information, especially if they are discordant with previous knowledge, to maintain cognitive coherence (Lledó, 2001). On the contrary, other research defend the existence of some influence of the electoral debates on the citizens’ voting decision (Geer, 1988; Schrott, 1990; Shelley and Hwang, 1991; Lanoue, 1992; Holbrook, 1994 and 1996, and Blais and Boyer, 1996; Best and Hubbard, 1999; Yawn and Beatty, 2000; Lledó, 2001; Benoit et al., 2004; Garrido and Sierra, 2013).
In this way, various authors (Lanoue, 1992; Jamieson and Adasiewicz, 2000; Yawn and Beatty, 2000; Garrido and Sierra, 2015) consider that the main function that electoral debates fulfill, from the point of view of the vote decision, is, in addition to a greater knowledge of the candidates, that of reinforcing the preferences of the electorate, regardless of the incidence or impact they may have on the undecided citizen. In any case, although the only observable effect is the reinforcement, this may have consequences on the electoral results by converting previous provisional inclinations into final votes that, in the absence of the debates, may not have crystallized, manifesting themselves in the form of abstention.
Some authors have pointed out that the cause of the disparity in the empirical results of research is the context in which electoral campaigns are developed. Thus, the most favorable scenario for electoral campaigns and, therefore, electoral debates, to have significant effects on the vote of citizens is that of quarreled elections, in which there is no strong loyalty of the electorate and, therefore, a good part of the citizens are undecided. Thus, the more quarreled the elections are and the greater the percentage of undecided, the greater the effect that the debates may cause. From our point of view, the competitiveness of the elections is not the only decisive factor, but that the constitutive elements of the debate itself also have a certain weight in the effects, without the various factors having any hierarchy, living together in temporality and importance.
In this context, the electoral debates can be decisive as a result of the generation of the perception of which of the candidates has been more and less convincing. This perception about which candidate has been benefited or harmed the most in the debate is due, on the one hand, to the fact that, in the debates, the candidates make an effort to differentiate themselves from the rest, since citizens decide to support a certain candidate through a comparative trial, so they would have no reason to prefer one or another leader if they seemed the same (Isolatus, 2011); and, on the other hand, that the media coverage tends to be expressed using sports terms (fight, winner, loser) and, in this way, the perception of the candidate as a winner can mean an increase in their assessment that, ultimately, can tip the vote of the undecided in one way or another. Besides, debates usually take place in the days near the elections and citizens may not remember everything the participants say, but they can remember who has been the winner and the loser, so that the undecided could base their decision on this opportunity to get information about politicians and their proposals and, therefore, vote for the candidate they perceive as winner (Schrott, 1990).
Likewise, Lledó (2001) affirms that the perception of the winner of the debates is strongly influenced by pre-existing attitudes and preferences, such as the leader’s assessment, party identification or voting intention, among others; or what is the same, the accumulation of previous preferences would act as a barrier when the candidate that is considered the winner is from the opposite party indicated in the vote intention, so that, in any case, the spectators would opt for a tie. Each voter, depending on the “selective perception”, tends to consider the winner the candidate representing the political party for which it feels greater sympathy and, therefore, the candidate who corresponds to the preferred political organization for a greater mass of voters, who shows clear predispositions before the elections, will be the one who appears as the winner of the electoral debate in the polls’ results.
However, the studies that analyze the impact of the winner on the vote are practically non-existent. Abramowitz (1978), in his work on the effect of the American presidential debates of 1976, finds evidence of the effect of persuasion on the rationality of voting decisions, so that voters would adopt their preferred candidates’ positions on various issues; McLeod et al. (1979) identify some impact of several variables related to the debates on the presidential preferences of the voters, so that the respondents who saw Carter as the winner of the debate, showed a greater involvement in the campaigns and a greater degree of affinity with the Democratic Party, in addition to observing changes in voting preferences towards this politician; but it is Schrott (1990) who demonstrates that by selecting a candidate as the winner of the debates, the voters of the East German elections of 1976, 1980, 1982, and 1983, evaluated that politician more positively and its opponents more negatively, increasing the probability of voting for the winning candidate’s party.
Regarding the influence of the debates on the candidate’s image, Popkin et al. (1994) show that attitudes towards incumbent candidates are more stable than attitudes towards challengers. Yawn and Beatty (2000) point out the existence of specific aspects of the candidates’ performance that produce a change of opinion. Nimmo and Savage (1976) argue that these dimensions refer to the political and stylistic roles projected by the candidates to the government, including the political role to those candidate’s acts that are particularly related to his position and aspiration to be a political leader; while the stylistic role would encompass a candidate’s actions that are not directly political, the problem in which he acts to impress voters with his abilities, and the display of personal qualities. On the other hand, Kinder et al. (1980) argue that voters conceive the president prototype in terms of personality and performance. This fictitious president should exhibit personal traits such as honesty, affection, and courage, and his attitude should be characterized by traits such as leadership, diplomacy, and problem-solving skills. But authors argue that more important than this is competition and truth. According to Martel (1983), these features have been particularly important in certain debate strategies.
Our research will try to prove if these previous research approaches are extensible to the debates of the General Elections of April 2019, based on the premise that the influence of debates on electoral behavior is limited.

2.2. The debates in the General Elections campaign of April 2019: A round trip

The holding of the debates during the election campaign of the General Elections of April 28th, 2019 has not been without controversy. After a first attempt to organize a five-party debate, between the candidates of the PSOE, PP, Unidas Podemos, C’s, and Vox, the electoral board prevents it being held, arguing the breach of the principle of proportionality and thus accepting the resources brought by Esquerra Republicana de Cataluña (ERC), the Partido Nacionalista Vasco (PNV), and Coalición Canaria (CC). RTVE then decides to set the debate date for Tuesday, April 23rd, the same date scheduled for the Atresmedia debate, instead of Monday 22nd, as initially planned.
Given the coincidence of both debates, Pedro Sánchez, who had initially ruled out his participation in the RTVE debate, but accepted the invitation of the private company, claiming that in this debate the Vox leader would be present, announced that he would attend the public corporation’s debate. On the contrary, Pablo Casado and Albert Rivera reject the RTVE proposal, since they had already committed themselves to Atresmedia, while Pablo Iglesias asks that the date be modified, as it finally happens.
Thus, after this controversy, the first of the four-way electoral debates, organized by RTVE, between the main candidates for the Presidency of the Government: Pedro Sánchez (PSOE), Pablo Casado (PP), Pablo Iglesias (Unidas Podemos), and Albert Rivera (C´s) took place on April 22nd, six days before the General Elections were held, in Estudio A1 of Prado del Rey in Madrid. The debate, moderated by the journalist Xavier Fortes and broadcasted live by La 1, 24h, RNE, RTVE.es, in addition to international TVE and Radio Exterior to reach the millions of Spaniards living abroad, had a total duration 1 hour and 26 minutes, divided into four thematic blocks: Economic policies, fiscal, and employment; social policies, welfare state, pensions, and equality; territorial policies; and democratic regeneration and electoral pacts, establishing by lot who would open and close each block. According to the debate’s rules, each candidate would have four and a half minutes per block, of which, one and a half minutes, in which they could not be interrupted, should be dedicated to the presentation of the program, while the other three minutes should be devoted to the debate itself, in the sense of the counter-position of measures and the conduct of attacks and responses between the four candidates; besides an initial one-minute intervention and the known-as “golden minute”, the final message in which the candidates expressly request the vote for the political party they represent.
Finally, the second of the electoral debates between the same political leaders, organized by the communication group Atresmedia and moderated by journalists Ana Pastor and Vicente Vallés, was broadcasted on April 23rd, also in prime time, through the channels Antena 3, La Sexta, Onda Cero, Atresplayer, and the international channel, with a duration of 2 hours and 5 minutes. As a result of it being held the day after the RTVE debate, the Antena 3 one was interpreted by the media in sports terms, as the “return” of the RTVE debate, being called by the organizers themselves as “The decisive debate”.
Although this debate is also articulated in thematic blocks, in this case, three: Electoral Programs, Territorial Model/ Catalonia, and Pacts, it differs from its predecessor in that it would not have time, and in which the moderators would have the possibility to ask direct questions during the debate, besides a quick round to start it, ending, as usual, with the golden minute.

3. Methodology and data

To find out, in the first place, whether the debates have any effect on the perceptions of citizens about political leaders or their decision to vote and, secondly, if the debates effectively constitute a conditioning element of the vote to the main parties (Partido Popular (PP), Partido Socialista Obrero Español (PSOE), Ciudadanos (C’s), and Unidas Podemos (UP)) in the last General Elections of April 2019, an initial analysis is carried out from a descriptive perspective, prelude to a generalized linear regression model with link logit function to explain the vote to each of the political formations. For our research, we have used Post-Electoral Study No. 3248* General Elections 2019 prepared by the Center for Sociological Research (CIS). This study has a battery of questions related to electoral debates, which the CIS has already used in numerous previous studies.
For our analysis, we have selected the most relevant variables, some of which have been necessary to recode. Next, they are related:

* The Post-Electoral Study No. 3248 General Elections 2019 was conducted between May 10th and May 15th, 2019 to a total sample of 5,943 individuals, selected by a multi-stage sampling procedure, stratified by clusters, the sample error being, for a confidence level of 95.5% (two sigma) and, for the whole of the sample and in the case of simple random sampling.

4. Results

Electoral debates are considered the star events of the electoral campaigns (Sierra, 2012; Carrera, 2015; García, 2015), not only because they represent a favorable opportunity to contrast the different positions and the various measures proposed by the candidates for the presidency in an election, but also for their great reach. Proof of this is the high ratings that are recorded after the broadcast of each of these programs. The RTVE debate had 8,886,000 viewers on average (43.8% screen share), which means that 31.1% of the population saw at least one minute of the debate, reaching in the golden minute 9,629,000 spectators (48.7%). For its part, the Atresmedia debate achieved even higher levels of audience: 9,447,000 viewers (on average) followed the debate (48.7% share), which implies that 37.9% of the population saw, at least, one minute of the debate, exceeding ten million viewers in the golden minute (10,963,000 viewers, that is, 51.6% share) (Barlovento, 2019a, 2019b).

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.
graph1
Graphic 1. Follow up of the debates.

In line with the official audience data, the CIS post-election survey data confirms the wide scope of the electoral debates. 35.5% of the population saw both debates, and around 10% saw one of them (11.2% saw that of RTVE and 9.1%, that of Atresmedia). 23.1% of citizens, although they did not see them, claim to have had references from them (Graph 1).
Besides, of 55.8% of respondents who claim to have seen one or both debates, the majority saw both of them entirely (34%) or both of them partially (26.4%), although 9.9% of the respondents saw the one of RTVE completely and 11.7% partially, and 6.8% saw the one of Atresmedia completely and 10.1% partially (Graph 2).

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.
graph2
Graphic 2. Type of debates follow-up.

As to which of the four candidates who presented themselves to both debates turned out to be the most compelling (Graph 3), the great majority of the spectators consider that it was Pablo Iglesias (34.7%), being Pedro Sánchez and Albert Rivera well below (more than 15% and 18% of respondents think that Pablo Iglesias proved more compelling than Pedro Sánchez and Albert Rivera, respectively), but, above all, Pablo Casado, since only 5.6% of the citizens who watched the debates consider that he has been the most compelling candidate in these televised events.

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.
graph3
Graphic 3. Most convincing candidate.

Nonetheless, not only the perception of the candidate as a winner is related with a higher assessment of him (chart 1), as literature reflects, but, also, the mere fact of being a viewer of the electoral debates (in contrast with those who have not seen it regardless of the political predispositions of citizens, is related to a better assessment of political leaders (Chart 2).
As seen in Chart 2, therefore, it seems that the public of the electoral debates of April 2019 values candidates better than those who did not see them, although it could also be the case that they see the debates because they value better the candidates than those who do not see them. In any case, the relationship between both variables is confirmed by applying the Kolmogorov-Smirnov test of unilateral homogeneity to groups of two populations, in this case, to the audience of the debate and the non-viewers of the debate. Based on the test results, for the levels of significance normally established (α = 0.01, α = 0.05 and α = 0.10), when comparing respondents who have seen the debate with those who have not seen it, not only are significant differences found between both groups, but the assessment of each political leader who has participated in both electoral debates is better if the interviewees have seen these programs as opposed to those who have not seen them.

Chart 1. Comparison of the Assessment of political leaders with the Assessment of political leaders by most Compelling candidate.
table1
Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.

Chart 2. Evaluation of political leaders by the viewing of the debates.
table2
Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.

As we have already seen, in the existing literature there is usually a certain consensus around the idea that citizens will think that the most compelling candidate is the one who represents the party they feel closest to or is closest to their ideas and, in the same way, for which they have thought to vote in the elections (Lledó, 2001). However, in the General Elections of April 2019, this assumption does not seem to be true in all cases (tables 3 and 4), although it is for most respondents. 91.7% of the voters who declare sympathy for Unidas Podemos and 86.9% of those who finally voted for this political confluence consider that the most compelling candidate was Pablo Iglesias, but 30.2% of those who affirm to feel closer to PSOE and 32.6% of its voters think so, too. On the other hand, besides the 42.4% of citizens who express sympathy for the PP and 34.7% of the voters of this political party who consider Pablo Casado to be the most compelling candidate, the 23.2% that feel closer to Vox share this feeling as well as the 17.8% of those who voted for him. Finally, 68.1% of C´s supporters and 61.5% of its voters declare that Albert Rivera was the most compelling candidate, as is the case of 25.9% of respondents who feel closer to PP and 22.4% of the voters of this party, and 42.7% of those who feel closer to Vox and 40.2% of its voters (in this regard, remember that Vox did not have its representative in the debate).

Chart 3. Most compelling candidate by Party for which people feel closer or is closer to their ideas.
table3

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.

Chart 4. Most convincing candidate by Vote Recall in the 2019 General Elections.
table4

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.

In short, besides the fact that the majority of the interviewees consider the electoral debate’s winner the party leader for whom they feel sympathy or for whom they voted in the elections, a significant percentage, however, perceive that the winner may be the candidate of a party that is not the one they feel closest to or for whom they finally did not vote, but, generally, it is the head of the list of a party located in the same ideological group. That is, voters can rate better the performance of the candidate of a party that they do not feel very close to or for which they decided not to vote, but referring to this displacement as an intra-block movement.
In Benoit’s opinion (2007a), candidates should discourage party supporters from deserting or voting for their opponent, tempt the viewers who feel closer to the rival parties to defect and attract the candidates from independent parties or third parties. As Xavier Fortes himself explains in the RTVE electoral debate:
“We are in a very open electoral scenario and you (addressing the candidates participating in the debate) have to look, not only for the parish but also for the undecided”.
Undecided that, according to the data of the CIS Pre-Electoral General Elections 2019 Macro-Barometer Study No. 3242 of March 2019, at that time, accounted for 41.6% of the electorate.
By observing, in a descriptive way, if the debates have caused some kind of effect on the voting decision of the electors, we can confirm, according to the literature, that the reinforcement is the main effect produced by the electoral debates since 92.7% of the spectators affirm that the viewing of the debates has not changed their voting intention. Although, an important 7.1% of viewers affirm that they did produce a change in their voting intention, which would make a significant difference compared to the figures that are usually estimated in most of the Anglo-Saxon literature, between 1 and 4% (Jamieson and Adasiewicz, 2000).
Of the 7.1% of the interviewees who affirm that the electoral debates have modified their voting decision, in more than half of the citizens (51.1%), the debates have had a conversion effect, that is, the voters have finally voted for a party for which they had no plans to vote before watching the debates. This a very relevant fact since the conversion means the most traumatic effect for voters because the change of vote means a bigger effort for the electorate than their activation or deactivation (i Coma, 2008). However, 12.7% of respondents acknowledge that they thought not to vote and that they finally voted, that is, the debates have caused a greater effect of activation than deactivation since only 2.5% of respondents argue that debates discouraged them from going to the ballot boxes.

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.
graph4
Graphic 4. Change of voting intention after the debates.

Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.
graph5
Graphic 5. The way in which your vote changed after the debate.

Known the effects that the viewing of the debates have caused in the political behavior of citizens, with the aim of corroborating if the electoral debates are a conditioning element of the vote to the main parties in the General Elections of April 2019, below, a model for the vote is presented to each of the main parties: PP, PSOE, C’s, and Unidas Podemos (Chart 5), in which there were initially introduced, as explicative factors, the variables: sex (although this has not been significant for any model); age, which has only been significant in the case of the PP; the level of studies, which is not significant in the cases of the PP nor the PSOE; the employment situation, which is not significant for any of the four analyzed political formations; the ideological self-assessment, which is not significant for C´s; the sympathy or closeness to the political party in question and, as a representative variable of the electoral debates, the most compelling candidate in these televised events.
Starting with the voting model for the PP, it is significant, for a level of significance α = 0.05, age, ideological self-assessment, sympathy towards this political party, and the fact that Pablo Casado turned out to be the most compelling candidate in the electoral debates. So, odds** of voting for the PP increases 1.022 times for each year that the age of the individual increases; it is increased 1,606 times as the respondent is placed in a more right position on the scale of ideological self-assessment; and it increases 179.733 and 2.969 times when a citizen declares to feel sympathy for the PP and affirms that Pablo Casado was the most compelling candidate in the electoral debates, respectively.

** Chart 5 shows the Beta coefficients together with the typical errors and the level of significance represented by asterisks, so that: , but the standardized Beta coefficients are discussed in the text. The odds represent the ratio of the relationship between the probability of success of the event and the probability of failure of the event, in this case, the fact of voting for the political party in question.

The generalized linear model for the case of the PSOE, reveals that the vote towards this political party is explained through ideological self-assessment, sympathy towards this political organization, and the fact that Pedro Sánchez turned out to be the most compelling candidate in the debates. Thus, the odds of voting for the PSOE increases 0.749 times as the individual positions himself further to the left on the ideological self-assessment scale; it increases 285.110 each time a citizen declares to feel closer to the PSOE, and it increases 4.244 times for each voter who believes that Pedro Sánchez was the most compelling candidate in the televised contests.
As for C´s, their vote is not explained through ideological self-assessment, but through the respondents’ level of education and, again, for feeling sympathy for this political party and considering Albert Rivera as the most compelling candidate. In this way, the odds of voting for C’s increases 1.333 times as the level of education of the electorate increases, and increases 325.330 and 14.265 times for each citizen who declares to feel sympathy for this political organization and who thinks that Albert Rivera was the most compelling candidate in the electoral debates, respectively.
Finally, the vote to Unidas Podemos can be explained by the variables: level of studies, ideological self-assessment, and by the fact of feeling close to this political confluence and considering Pablo Iglesias the most compelling candidate in the debates. As in the case of C’s, the odds of voting for Unidas Podemos increases 1.283 times as the level of studies of the respondents increases, in addition to increasing 0.558 times as citizens are in a position more to the left on the scale of ideological self-assessment, and 129.735 and 4.467 times for each voter who declares to feel sympathy for Unidas Podemos and for each voter who perceives Pablo Iglesias as the most compelling candidate in the electoral debates, respectively.

Chart 5. Voting models.
table5
Source: self-made based on data from the CIS Post-Electoral Study No. 3247. 2019 General Elections.

5. Discussions and conclusions

After the motion of censure to the government of Mariano Rajoy filed by the PSOE and the impossibility of approving the budgets, Pedro Sánchez calls early elections for April 28th, 2019, with the threat of the emergence of Vox, a far-right party, in the Congress of Deputies. In this context, the electoral campaign is held, within which two electoral debates take place between four of the main candidates for the presidency of the government.
Despite the exceptional nature of these elections, they could not be considered as quarrels, because in response to the pre-election polls published by different media, it seemed clear that the PSOE would win the elections, leaving the PP in second place. The only possible doubt was that C’s finally surpassed Unidas Podemos, as most of the surveys indicated. A situation that would not encourage electoral campaigns and, therefore, electoral debates, to have decisive effects on the orientation of the citizens’ vote.
Based on the results obtained in the voting models for the main parties, we could affirm, firstly, that the sociodemographic variables that, for the Columbia School, defined the orientation of the citizens’ vote, have little influence on the explanation of the vote of the General Elections of April 2019, because only age remains a constitutive factor of the vote to the PP, increasing the probability of voting for this political party as the age of the individual increases; and the level of studies is an explanatory factor for the cases of the new political formations: C’s and Unidas Podemos, so that as the level of studies of the respondents increases, the probability of voting for both parties increases. This could be because more educated people would be the ones who would be more willing to inform themselves about the new actors that have broken into the political landscape.
Secondly, the psychological components of the vote defined by the Michigan School still enjoy a considerable weight in the explanation of the vote. In fact, sympathy, a term with which we define the feeling of closeness or proximity to the ideas of a political party, is the variable with the most explanatory power of the vote to the four analyzed parties, as is usual in most models of electoral behavior; and the ideological self-assessment is also a determining factor in the orientation of the vote for the cases of the PP, the PSOE, and Unidas Podemos in the expected sense, that is, the probability of voting to the PP increases as citizens get closer to the right, and the probability of voting for the PSOE or Unidas Podemos increases as individuals are placed in positions further to the left on the scale of ideological self-assessment. However, it is remarkable that ideological self-assessment is not a significant variable to explain the vote to C’s, perhaps because of the desire, at that time, of the leaders of this political party to place the orange formation in the center of the ideological spectrum, giving a hinge party image that could agree with the right and the left.
Finally, the inclusion of electoral debates in the voting models has been carried out through the variable related to the most compelling candidate, since the mere viewing of the debates, concerning those citizens who have not been spectators of these, is not significant when explaining the vote to the parties whose representatives participated in these televised events. In this regard, it is also noteworthy that, by introducing the variable assessment of political leaders, the most compelling candidate variable ceases to be significant and this can only be understood as that the performance of candidates in electoral debates implies a constitutive and reinforcing element of the leaders themselves, affirmation that is confirmed when we observed that the assessments of leaders significantly increase when the citizens have been spectators of the debates as opposed to those that have not been.
Therefore, what is relevant is not only that the assessments of leaders are increased as a result of the viewing of the debates, but the fact that, in the four voting models, for the four parties, the identification of the leader as the most compelling candidate in the debate implies an increase in the probability of voting for the political party in question. That is, it seems clear that the debates and the reading that citizens make of them have a direct effect on the electoral behavior of those who see them, regardless of the effect, widely demonstrated, of other classic political variables, such as sympathy or ideological self-assessment.
Therefore, although electoral debates pose a risk to the image of political leaders and the resulting electoral results, failure to attend an electoral debate can have even more damaging consequences for leaders than receiving attacks from political rivals. Perhaps, this effect has been observed by Mariano Rajoy’s advisors after the 2015 General Elections, since in the previous four-way debate, he was replaced by Soraya Sáenz de Santamaría, while he decided to go to the four-way debate before the 2016 General Elections.
Although this study confirms the limited effects model, according to which electoral campaigns have a minimal effect on the outcome of the elections and, therefore, electoral debates would not have any kind of effect beyond the reinforcement of the voters’ predispositions; It also highlights the transformation of the transition from a bipartisan system to a multiparty system beyond its consequences on the election results and the subsequent conformation of government, by making visible the change that supposes in the perceptions of the spectators of the debates about the candidate who is more compelling, in the sense that the candidate is no longer identified as the representative of the political party for which citizens feel more sympathy or for who they finally vote, but the consideration that a candidate is the most compelling may not be due to political predispositions or the most genuine act of political behavior: the vote, as long as the political leader is the representative of a political formation that is in the same ideological bloc.

Bibliographic references

  1. Abramowitz AI. (1978). The Impact of a Presidential Debate on Voter Rationality. American Journal of Political Science, (22), 680-680. http://doi.org/10.2307/2110467
  2. Barreiro XL, Pereira M, García G. (2015). Los efectos sobre el voto de la campaña electoral en las elecciones europeas de 2014 en España. Revista Española de Ciencia Política, (39), 67-93.
  3. Benoit WL, Wells WT. (1996). Candidates in conflict: Persuasive attack and defense in the 1992 presidential debates. University of Alabama Press.
  4. Benoit WL, Harthcock A. (1999). Functions of Great Debates: Acclaims, Attacks, and Defenses in the 1960 Presidential Debates. Communication Monographs, 66(4), 341-357. http://doi.org/10.1080/03637759909376484
  5. Benoit WL, McKinney MS, Holbert RL. (2001). Beyond Learning and Persona: Extending the Scope of Presidential Debate Effects. Communication Monograph, 68(3), 259-273. http://doi.org/10.1080/03637750128060
  6. Benoit WL, Brazeal LM. (2002). A Functional Analysis of the 1988 Bush-Dukakis Presidential Debates. Argumentary and Advocacy, 38(4), 219-233. http://doi.org/10.1080/00028533.2002.11821569
  7. Benoit WL, Hansen G, Verser R. (2003). A Meta-Analysis of the Effects of Viewing U.S. Presidential Debates. Comunication Monographs, 70(4), 355-350. http://doi.org/10.1080/0363775032000179133
  8. Benoit WL, Hansen G, Verser R. (2004). Presidential Debate Watching, Issue Knowledge, Character Evaluation, and Vote Choice. Human Communication Research, 30(1), 335-350. http://doi.org/10.1111/j.1468-2958.2004.tb00727.x
  9. Benoit WL. (2007). Communication in Political Campaigns. Peter Lang.
  10. Best SJ, Hubbard C. (1999). Maximizing Minimal Effects: The Impact of Early Primary Season Debates on Voter Preferences. American Politics Quarterly, 27(4), 450-467. http://doi.org/10.1177/1532673X99027004004
  11. Bishop GF, Meadow RG, Jackson-Beeck M. (Eds.) (1978). The Presidential Debates. Praeguer.
  12. Blais A, Boyer M. (1996). Assesing the Impact of Televised Debates: The Case of the 1988 Canadian Election. Bristish Journal of Political Science, (26), 143-164. http://doi.org/10.1017/S0007123400000405
  13. Carrera V. (2015). La red de confianza de los debates electorales y la mente de los candidatos. Revista Marco, (1), 39-61.
  14. Castro P, Maneiro E, Peso A. (2017). Análisis de los debates electorales: el procés como master frame. En Rivera JM, Montabes J, Lagares N (Eds.), Cataluña en proceso: las elecciones autonómicas de 2015, (pp. 213-235). Tirant lo Blanch.
  15. Cho J, Ha Y. (2012). On the Communicative Underpinnings of Campaigns Effects: Presidential Debates, Citizen Communication, and Polarization in Evaluations of Candidates. Political Communication, 29(2), 184-2014. http://doi.org/10.1080/10584609.2012.671233
  16. Crespo I, Garrido A, Riorda M, Carletta I. (2011). Manual de Comunicación Política y estrategias de campaña. Candidatos, medios y electorales en una nueva era. Biblios.
  17. Downs A. (1957). An Economic Theory of Democracy. John Wiley.
  18. Gallego M, Bernárdez A. (2017). Influencia y repercusión mediática de los debates “cara a cara” celebrados ante las elecciones generales de 2008 en España: José Luís Rodríguez Zapatero (PSOE) Vs. Mariano Rajoy (PP). Vivat Academia. Revista de Comunicación, (141), 139-154. http://doi.org/10.15178/va.2017.141.139-154
  19. García J. (2015). La cobertura mediática de los debates electorales en España. Revista Española de Ciencia Política, (38), 135-161.
  20. Garrido A, Sierra J. (forthcoming). Los efectos de los debates electorales sobre la decisión de voto: Un estudio a través de las encuestas del CIS. Universidad de Murcia.
  21. Geer JG. (1988). The Effects of Presidential Debates on the Electorate´s Preferences for Candidates. American Political Quarterly, 16(4), 486-501. http://doi.org/10.1177/004478088016004005
  22. Herranz-Rubio C. (2020). La producción científica sobre los debates electorales en España: Análisis bibliométrico (1993-2018). Revista Mediterránea de Comunicación/Mediterranean Journal of Communication, 11(1). http://doi.org/10.14198/MEDCOM2020.11.1.2
  23. Herrero JC, Benoit WL. (2009). Análisis funcional de los debates en las elecciones generales de 2008. Zer-Revista de Estudios de Comunicación, 14(27), 61-81.
  24. Holbrook TM. (1994). The Behavioral Consequences of Vice-Presidential Debates: Does the Undercard Have any Punch? American Politics Quarterly, 22(4), 469-482. http://doi.org/10.1177/1532673X9402200404
  25. Holbrook TM. (1996). Do Campaigns Matter? Sagei.
  26. Coma FM. (2008). ¿Por qué importan las campañas electorales? CIS.
  27. Informe Barlovento: Debate Electoral 22 Abril 2019.
  28. Informe Barlovento: Debate Electoral 23 Abril 2019.
  29. Isolatus P. (2011). Analyzing Presidential Debates. Functional Theory and Finnish Political Communication Culture. Nordicom Review, 32(1), 31-43.
  30. Jamieson KH, Adasiewicz C. (2000). What can voters learn from election debates. En S. Coleman (Ed.). Televised election debates. International perspectives, (pp. 25-42). MacMillan.
  31. Jamieson KH, Birdsell DS. (2000). Presidential Debates: The Challenge of Creating an Informed Electorate. Oxford, Reino Unido: Oxford University Press.
  32. Kinder DR, Peters MD, Abelson RP, Fiske ST. (1980). Presidential Prototypes. Political Behavior, (2), 315-337.
  33. Lledó P. (2001). La influencia de los debates electorales sobre la decisión de voto: el caso de mayo de 1993. Revista Española de Ciencia Política, (5), 143-170.
  34. Luengo Ó . (2011). Debates electorales en televisión: una aproximación preliminar a sus efectos inmediatos. Revista Española de Ciencia Política, (25), 91-96.
  35. Lanoue JD. (1992). One that Made a Difference. Cognitive, Consistency, Political, Knowledge, and the 1980 Presidential Debate. Public Opinion Quarterly, 56(2), 168-184. http://doi.org/10.1086/269309
  36. Manin B. (2006). Los principios del gobierno representativo. Alianza Editorial.
  37. Martel M. (1983). Political Campaign Debates: Images, Strategies, and Tactics. Longman.
  38. McLeod JM, Durall JA, Ziemke DA, Bybee CR. (1979). Reactions of Young and Older Voters: Expanding the Context of Effects. En: Kraus S (Ed.). The Great Debates: Carter vs. Ford, 1976. Indiana University Press.
  39. Nimmo D, Savage RL. (1976). Candidates and their images: Concepts, Methods, and Findings. Goodyear.
  40. Popkin SL. (1994). The Reasoning Voter. University of Chicago Press.
  41. Shelley MC, Hwang H. (1991). The Mass Media and Public Opinion Polls in the 1988 Presidential Election. Trends, Accuracy, Consistency and Events. American Political Quarterly, 19(1), 59-79. http://doi.org/10.1177/1532673X9101900104
  42. Schrott P. (1990). Electoral Consequences of Winning Debates. Public Opinion Quarterly, 54(4), 567-585. http://doi.org/10.1086/269228
  43. Sierra J. (2012). Planteamientos sobre los efectos de los debates electorales en televisión: aplicación al caso español. XXII World Congress of Politican Science. Madrid, España.
  44. Weiler M. (1989). The 1988 Electoral Debates and Debate Theory. Argumentation and Advocacy, 25(4), 214-219. http://doi.org/10.1080/00028533.1989.11951401
  45. Yawn M, Beatty B. (2000). Debate-induced Opinion Change: What Matters? American Politics Quarterly, 28(2), 270-285. http://doi.org/10.1177/1532673X00028002007

AUTHORS

Nieves Lagares Diez: Professor of Political Science and Administration at the University of Santiago de Compostela and an expert in the study of the formation and organization of political parties, leadership, and political marketing. Among her most recent publications are the books published alogside J. M. Rivera and J. Montabes in 2016, Cataluña en proceso: las elecciones autonómicas de 2015 (Valencia: Tirant lo Blanch); and together with C. Ortega and P. Oñate in 2019, Las elecciones autonómicas de 2015 y 2016 (Madrid: CIS), and the article “Condiciones y razones del procés” with E. Jaráiz and X. L. Barreiro (2017), Aucaria, 19, 141-170.
mnieves.lagares@usc.es
Orcid ID: https://orcid.org/0000-0001-5304-0581

Erika Jaráiz Gulías: Doctor in Political Science and a professor in the Department of Political Science and Sociology of the USC since 2008. She belongs to the Political Research Team of the same university, listed as a Competitive Reference Group by the Xunta de Galicia. She has presented numerous communications at national and international conferences. She has experience in electoral campaigns and has been Manager of the Commonwealth of the Inter-municipal area of Vigo (2007-2014) and Chief of Staff of the Secretariat of Analysis and Projection of the Presidency of the Xunta de Galicia (2005-2007). She has participated in numerous contracts and research projects, some of which she has directed as principal researcher, among which are the “Realization of a practical guidance guide for conducting studies of demand analysis and satisfaction surveys (2014 )” and the preparation of the“ Citizen Perception Report: The Quality of Public Services IX ”(2014 and 2015)”, both for the Spanish Agency for Evaluation and Quality of Services (AEVAL by its acronym in Spanish). She usually teaches in several Masters, both her own and official ones, among which the ones that stand out are the Master’s Degree in Marketing, Consulting and Political Communication of the USC, in which she is Secretary of the Academic Commission, and the Master’s Degree in Leadership and Public Management (INAP), where she also coordinates the TFM module. She has also received methodological training at the Essex Summer School in Social Science Data Analysis and Collection and the Universitat Pompeu Fabra, among others. She is the Writing Secretary of the Journal of Political and Sociological Research.
erika.jaraiz@usc.es
Orcid ID: https://orcid.org/0000-0003-2382-6713

Paloma Castro Martínez: Paloma Castro Martínez is a senior research support technician in the Political Research Team of the University of Santiago de Compostela, listed as a Competitive Reference Group by the Xunta de Galicia. Currently, she is pursuing a Ph.D. program in Political Marketing, actors and institutions in contemporary societies in the same university. Her most recent publications are: “The protest and the vote. How much of protest is in the vote to the new parties?”, written in collaboration with María Pereira and Adrián García, for the book Las elecciones autonómicas de 2015 y 2016 by N. Lagares, C. Ortega, and P. Oñate (Eds.). Madrid. CIS, 2019; “Media agenda and political agenda”, written with Antón Losada, and “Analysis of the electoral debates: the process as a master frame”, together with Elba Maneiro and Alejandro Peso, for the book Cataluña en proceso, by J.M. Rivera, J. Montabes, and N. Lagares (Eds.). Valencia. Tirant lo Blanch, 2017. She also has methodological training after completing the Master in Statistical Techniques at the University of Santiago de Compostela.
paloma.castro2@usc.es
Orcid ID: https://orcid.org/0000-0001-9453-7770