1

International Association for Media and Communication Research (IAMCR) 26 th Annual Research Conference, July 20–25, 2008, Stockholm Political Communication Research Convenor: Prof. Dr. Philippe J. Maarek

No Political Ads Please, We Are Swedish!

A Study of Public Opinion on Emerging Political Advertising in Television.

Marie Grusell, University of Gothenburg &

Lars W. Nord, Mid Sweden University

Abstract

This paper discusses public opinion on televised political advertising as an emerging

political communication phenomenon. The National Elections in 2006 may be de-

scribed as the initial breakthrough for political advertising in television in Sweden.

Two major research questions are asked here: Firstly, how can public opinion to-

wards such political advertising be described, and secondly, how may these attitudes

be explained with regard to individual and societal factors?

The results in this study indicate that attitudes towards political advertising in

Sweden are in line with attitudes towards advertising in general. Surprisingly, ideo-

logical beliefs on the individual level seem to be almost irrelevant when trying to

explain existing public opinion. Without being on the political agenda, the political

TV-advertising issue is mainly considered from non-political perspectives. As with

advertising in general it is affected by individual factors and the actual exposure for

such advertising.

This study indicates that the role of political advertising depends on the existing

political culture and the traditions of campaign practices. As a new and emerging

element in the political communication processes, political ads in television are

judged more by non-political considerations than by ideological convictions and po-

litical standpoints.

2

ONCE UPON A TIME, THERE WERE NO TV ADS… For most citizens living in modern democracies, it is probably impossible to imagine

a political campaign completely without political advertising in television. On the

contrary, political ads in television are one of the most distinctive features of election

campaigns in almost every single state with unrestricted freedom of information.

Furthermore, controversial messages in televised spots are often in the centre of pub-

lic and media discourses in the intensive weeks before the Election Day. Finally, a

considerable share of candidate and political party campaign budgets is generally

spent on efforts to produce as effective and persuasive ads as possible, with the inten-

tion to catch voter attention and influence voting decisions in desired directions.

However, Sweden has so far been an outstanding exemption from this rule. In

this country – one of the oldest and democracies in the world and with a Freedom of

Information Act in its constitution since 1766 – political advertising in television has

always been prohibited in the main terrestrial national TV channels. The political

majority in the Swedish parliament has historically perceived televised political ad-

vertising as a negative element in campaign communications. Fears have been ex-

pressed that such ads would jeopardize the quality of public discourse in election

campaigns by offering overwhelmingly emotional, personal and negative messages

without substance. Both left wing and centre-rightwing politicians in the parliament

have supported the critical standpoint on political advertising in broadcast media – in

public as well as in private media (Nord 2008).

Notwithstanding, recent years have seen a gradual transformation process

emerge. The main single explanation is of course structural media development

changes where TV channels based outside Sweden, are offering programmes in

Swedish reaching Swedish households by cable and satellite. This trend started with

London-based private television company TV3 in 1987 and a couple of similar pri-

vate channels have since then operated under similar conditions (Engblom &

Wormbs 2007). However, political advertising in these channels never turned out to

be particularly extensive by different reasons. Initially, because their share of the

national television market in Sweden has been rather modest compared to domestic

terrestrial channels. Secondly, because the foreign-based new channels have had a

focus an entertainment programmes, thus not being perceived by the political parties

3

as the most natural platforms for political messages. Thirdly, because of a remaining

political culture within both commercial TV channels and political advertising in

television stand out as deviations in the political communication processes and prac-

tices.

However, this situation changed even more in the latest National Election cam-

paign in Sweden in September 2006. Due to the planned switch-off of the analogue

television broadcast system in 2007 a majority of Swedish households – equipped

with digital boxes, cable and satellite – were offered a huge number of digitally dis-

tributed television channels in 2006. Faced with this new situation, the political ma-

jority decided to revise its former strict view on political advertising in Swedish TV

channels. Consequently, domestic ‘niche’ channels were allowed to send political

spots. Some of them were basically entertainment-oriented, while others offered

mainly current affairs programmes, documentaries and international news. Therefore,

political advertising in television was allowed in a large number of channels for the

first time during the elections in 2006. Only the single biggest commercial channel,

the private-hybrid TV4, was still not allowed to send political ads. Public service

television in Sweden is by tradition free from all kind of advertisements and financed

almost completely by license fees (Engblom & Wormbs 2007; Nord 2008).

Thus, the National Elections in 2006 may be described as the initial breakthrough

for political advertising in television in Sweden. Being introduced many decades

after the average western democracies, political ads in television still play a minor

role in campaign communication in Sweden compared to political ads in newspapers,

magazines, cinema and outdoor posters. Nevertheless, communication practices and

voter behaviour are not fixed, and it is reasonable to consider the role of televised

political advertising to increase in future election campaigns. Accordingly, such po-

litical ads may be perceived as new phenomena in the Swedish political culture.

These circumstances make political ads particularly interesting to analyze, not in

comparison with more mature countries in this aspect, but with regard to public atti-

tudes and perceptions of such a brand-new element in political communication.

This paper is intended to focus on public opinion on televised political advertis-

ing as an emerging political communication phenomenon in a national setting. The

contemporary Swedish political communication context offers an excellent possibil-

4

ity to study how public opinion is shaped in this area. Furthermore, analyzing public

opinion may contribute to a general understanding whether positions on political TV

ads are basically formed by ideological beliefs, socio-demographic conditions or

attitudes towards advertising in general. Consequently, the objective of the paper

initiate a discussion about public opinion on televised political ads in their introduc-

tory phase. Most research in this area has hitherto been focused on analyzing content

and effects on political advertising in countries with long traditions in this field.

However, as campaign practices seem to become more globalized and conform, the

reception and evaluation of campaign tools among citizens and voters in different

phases of evolution should be increasingly interesting to study (Plasser & Plasser

2002; Hallin & Mancini 2004).

MUCH FOCUS ON CONTENT AND EFFECT, LESS ON OPINION Political advertising has to be analyzed within different national political communi-

cation contexts. The majority of previous research has been conducted in the U.S.

and conditions in Europe differ in many ways: most political systems are multiparty

based, election campaigns are shorter and public service media has a stronger posi-

tion in the television markets, just to mention some examples. In all studies of politi-

cal communication studies it is reasonable to consider national characteristics regard-

ing political system, electoral system, media system, political culture and public

opinion (Lijphart 1999; Hallin & Mancini 2004). Most countries in Europe allow

political advertising in television, but restrictions with regard to the number of per-

missible sports, or the appearance in public service broadcast media, vary to a large

extent. Generally speaking, the candidate-oriented campaigning of American elec-

tions has gradually become more common also in a European context.

In most countries in the world, political advertising in television has been a cor-

nerstone in campaign communications since its first appearances in the 1950s in the

US. Due to the important communicative role of political advertising in election

campaigns in most democratic countries, the research in this filed is also one of the

most distinctive features of contemporary political communication studies (Kaid

2004: 155).

5

The definitions of political advertising have varied from descriptions of it as a

more or less market-oriented “promotional device” for political candidates and par-

ties in election campaigns in U.S. elections to a much broader definition. Nowadays,

political advertising is generally described as “any controlled message communicated

through any channel designed to promote the political interests of individuals, par-

ties, groups, governments, or other organizations” (Kaid & Holtz-Bacha 2006:4).

Using this broader definition, political advertising is distinguished from other cam-

paign communication by the source control of the message, and the absence of media

interpretation or framing, and from interpersonal communication by the use of mass

communication channels. Furthermore, the broader definition of the concept applies

to different political marketing campaign contexts, not at least Western Europe

where party controlled public broadcasts are more frequent (Kaid & Holtz-Bacha

1995; Holtz-Bacha 2003).

Political communication research on televised political advertising has mainly

focused on two different aspects: the content and the effects (Kaid 2004: 160; Johns-

ton 2006: 17–24). Content analyses of political advertising in television have tradi-

tionally been dominated by studies comparing issue information and image informa-

tion in the spots. Widespread beliefs that political advertising emphases emotions

and images on the behalf of rational information have not been completely supported

by previous research, which suggests that issue ads dominate election campaigns in

many countries, even if content in modern sports often is blended in a subtle way

(Johnston & Kaid 2002). In recent decades, more research has been focused on

“negative ads”, opponent-focused spots trying to describe him or her in an unfavour-

able way with regard to issue positions, personal character or political agenda (Kaid

& Holtz-Bacha 2006).

The research on the effects of political advertising has been focused on both voter

learning effects and candidate/party evaluation effects. There are contradicting re-

sults from previous studies comparing voter knowledge effects from television news

and televised political advertising. However, it is reasonable to argue that such spots

may contribute to increased issue knowledge, especially when the interactions with

other campaign communications are considered (Holbert et al. 2002). Previous re-

search in the field of candidate/party evaluation confirms that political TV spots mat-

6

ter. They have a significant impact on judgements of candidate likeability and policy

positions (Kaid 2004: 169). More attention has also been paid to the effects of the

increasing number of negative ads. Some results indicate that voters are turned off by

negative advertising and that political cynicism is growing after exposure to such

ads. However, other studies have showed no evidence of effects on the levels of po-

litical trust or political interest (Kaid & Holtz-Bacha 2006).

To conclude, previous research on televised political advertising suggests that

spots are a major force in most political communications systems in most democratic

states. The unique possibility for parties and candidates to control mass disseminated

messages remains the big advantage compared to mediated campaign communica-

tions. Spots may influence voter learning and candidate evaluation, but effects vary

due to political culture factors. Thus, the impact of televised political advertising is

also depending on public perceptions and attitudes of such advertising. This is par-

ticularly true as political advertising has a persuasive intent that casts doubt on its

credibility (Kaid & Holtz-Bacha 2006: 4). Not much research has been done to pro-

vide a theoretical background for political advertising. However, it is reasonable to

believe that the such advertising to a considerable extent can be understood as a

manifestation of political culture, one of several ways through which a political cul-

ture is expressed (Holtz-Bacha 2003: 109–110).

Consequently, this paper is intended to focus on how public opinion on political

advertising in television develops when such advertising is introduced in a political

communication culture without previous experience in this field. In the following

part, a brief background of Sweden – with regard to political communication and

public opinion on advertising – is offered.

COMMERCIALIZED MEDIA, NON-COMMERCIALIZED POLITICS Swedish democracy is based upon a multiparty parliamentarian system, where the

party traditionally has been more important than the candidate in national elections.

The election system is strictly proportional, even if recent political reforms have of-

fered voters the option of choosing a single candidate on the party ballot. Further-

more, national, regional and local elections are held on the same day, which, in es-

7

sence, means fewer elections where campaign strategies can be developed and prac-

tices can be improved (Nord 2006).

As political advertisements in television generally have been prohibited, one of

the most distinctive features of modern political campaigns has to a large extent been

absent in the Swedish election context. Some political spots have of course reached

the Swedish viewers through cables and satellites from TV stations based abroad, but

they have hitherto a marginal role in political communication aspects. Additionally,

no free broadcasting time is made available to political parties on either television or

radio during the election campaign. The fact that not even public service companies

allow such political party presentations is unique to Sweden. Instead, the tradition in

public service media is to have journalist-led questioning and special programs with

party leaders during the final weeks before the elections, and a final debate between

the party leaders two days before Election Day. Accordingly, the most effective di-

rect channels of communication with citizens are excluded from political campaigns

in Sweden. Thus, the marginal role of televised political advertising, and the absence

of party programs in media, put Swedish news media in a very strong position with

regard to the agenda setting and the framing of political campaign activities (Nord &

Strömbäck 2006).

At the same time, advertising in general has become more important in Sweden.

This is mainly because of the recent expansion of commercialized broadcast media,

and the occurrence of new media. In the late 1980´s, there was only a small amount

of radio advertising, almost no TV-advertising, and hardly anyone knew about the

Internet. However, during the last decades advertising in media practically has dou-

bled (Gustafsson 2005). These days radio, TV and the Internet are looked upon as

well-known media carriers and they reach a large audience every day (Mediebarome-

tern 2007). Consequently, the advertising market in Sweden is characterized by dra-

matic changes in a very brief time.

From a historic perspective the overall advertising tradition in Sweden can be de-

scribed as one-sided. Traditionally, advertising has mainly been distributed by

printed media, and it is only during the last two decades there has been a structural

change in the domestic advertising market. However, this change has been major.

Nowadays, the advertising market has several new operators; TV, radio, Internet,

8

direct mail etc. The advertising market with the most significant economic growth

currently is advertising on the Internet.1

SWEDISH ADVERTISING OPINIONS IN GENERAL Previous research has indicated that our relationship with advertisements can be de-

scribed as complex and ambivalent, like a love-hate relationship (O’Donohoe, 2001;

Scipione, 1997). We like some advertisements, but love to hate others. Results indi-

cate that we prefer advertisements in morning newspapers, at the cinema and out-

doors, while advertisements distributed through email and mobile phones are not at

all welcome (Grusell 2008). A more profound analysis indicates that mass media

context adds dimensions to the public’s attitude towards advertising. Where and how

we receive the advertisement is of crucial importance to how we perceive it. The

public’s view of advertising seems to be linked to the mass media context, i.e. the

context in which the advertisement is placed. Well-received advertising could be

described as being by personal choice, while depreciated advertising can be de-

scribed as being forced upon us by mass media in general. We like advertising when

we feel we have chosen to be exposed to it, but dislike it when it interferes with our

primary mass media usage.

But what are the factors that define the public’s view of advertising in mass me-

dia? Three major inputs can be observed here; society, properties of advertisement

in mass media and personal characteristics. The individual’s life status and interests

define the foundation for how an advertisement is received, and society in interaction

with advertising and the properties of the media’s form and content impact its accep-

tance. This relationship is naturally complex, which leads to the conclusion that the

public’s view of advertising in general is a result of the interaction between several

different factors.

Previous studies of the Swedish publics’ view of advertising have indicated that

the public’s different attitudes regarding advertisements can be explained by indi-

viduals’ ideological points of view and mass media usage habits (Grusell

2007/2008). This result indicates that if a person has a more liberal and accepting

1 http://www.irm-media.se/irm/(ycd5wl4500iwtj55jqwlh1ui)/default.aspx

9

attitude towards the commercial market, he or she may also accept advertisements to

a larger extent. Increased acceptance of advertisements may also occur among indi-

viduals who are often exposed to them. There is also an age difference; older people

are more accepting of morning newspaper advertisement, while younger people are

more accepting of TV commercials. Similar results have been noted in U.S. studies.

Research on advertising as a phenomenon indicates that younger people in general

are more accepting of advertising than older people (Shavitt et al. 1998).

To sum up, Sweden may be perceived as a country where a more commercialized

media system in recent decades has increased the amount of advertising exposed to

the public. Printed media dominance in this aspect has been replaced by an increas-

ing role for broadcast media and new media. The public attitudes on advertising in

general are to a large extent shaped by individual ideological beliefs and age, as well

as personal media use habits.

THE STUDY The objective of this paper is to analyze public opinion on political advertising in

television as a new communication element in Swedish election campaigns. Two

major research questions are asked: firstly, how can public opinion towards such

political advertising be described, and secondly, how can this public opinion be ex-

plained with regard to individual and societal factors? As no previous research in this

field exists in Sweden, this study may be considered basically as a pilot study.

Methodologically, the data in this study is based on a national mail survey, De-

mobarometer 2006. The survey was in field during August – November 2006. In the

survey a representative sample of citizens was asked about their perceptions of po-

litical communication in Sweden, for example about their exposure for, and opinion

on, political advertisements in television (see Appendix). Questionnaires and return

envelopes were sent to a random sample of 2000 individuals between 16 and 80

years of age, and living in Sweden. The net response rate was 50 percent, even

though that was less than desirable, an analysis indicates that the respondents are

representative with regard to gender and age. With regard to educational level, indi-

viduals with higher education are somewhat overrepresented.

10

The opinion results in this paper are based upon five different statements in the

survey: Political TV-advertising gives me guidelines regarding which party that suits

me best, political TV-advertising gives me an oversimplified picture of party stand-

points, political TV-advertising influences my voting decision on Election Day, po-

litical TV-advertising appeals too much to emotions and neglect facts, and political

TV-advertising is too negative in its tone towards political opponents. There is a 5-

point respond scale from ‘I disagree completely’ to ‘I agree completely’. There is

also a possibility to choose the ‘I don’t know’-alternative. The statements are pre-

sented independently as well as thru an index. In the analysis the index is referred to

as the opinions index (introduced in table 5). To measure the reliability of the index

we have calculated the Cronbach's Alpha, one of the standard measures of reliability.

Cronbach's Alpha can be interpreted as a correlation of the index with another index

that could be constructed to measure the same activities. There is no standard for

how high Cronbach's Alpha should be, but results over .80 is considered high quality

(Carmines & Zeller, 1979). In this study the Cronbach's Alpha is .9181.

RESULTS: SWEDISH PUBLIC OPINION ON POLITICAL ADVERTISING IN TELEVISION Nowadays, Swedes are constantly exposed to different forms of advertising. The

development of advertising in Sweden has changed dramatically during a relatively

brief period. At the present time different forms of media are seen as natural channels

for advertisement distribution and reach a major audience on a daily basis. Currently,

the traditional TV-advertising reaches a substantially larger audience than other me-

dia advertising (see figure 1 in appendix). Consequently, a majority of the Swedish

public is exposed to TV-advertising during an average day (Mediebarometern 2007).

As noticed previously, political advertising in television is a new phenomenon in

Sweden. Thus, an initial relevant question to ask is whether the public actually has

been exposed to such political advertising in Swedish television. The results below

indicate that less than half of the public has been exposed to this new form of adver-

tising (tab. 1). Furthermore, slightly less then every third person has been exposed

some time, while only 14 percent of he public says that they have been exposed sev-

eral times. These results may be quite reasonable to understand, and to interpret: the

11

televisions channels that at the present time have the possibility and the permission to

broadcast this form of political TV-advertising do not reach the entire television au-

dience. Accordingly, the Swedish public in general has limited personal experience

and knowledge of political advertising in television. Notwithstanding, this result ba-

sically offers an excellent opportunity to explore the nature and origin of public opin-

ion on a new political communication phenomenon in Sweden.

(Table 1 about here)

The overall opinion about advertising could be described as sceptical; a majority of

the public does not believe that advertising is trustworthy. In this paper the aim is to

examine the public opinion on political advertising in television. In order to examine

opinion such opinion five different statements, mentioned in the previous chapter,

were introduced for the respondents to evaluate in the survey.

As noted in table 2 (below), the public judgements of the five-presented state-

ments indicate that the public’s overall attitude towards political TV-advertising is in

line with the general public opinion on advertising (Grusell 2008). Not surprisingly,

a majority of the Swedes do not have the impression that they are actually and per-

sonally influenced by political advertising in television.

Furthermore, the result shows that nearly four out of ten Swedes believe that po-

litical TV-advertising gives them an oversimplified picture of party standpoints. In

addition, very few Swedes believe that political advertising in television influences

their voting decision on Election Day. There is also a widespread belief that such

advertising appeal too much to emotions and neglect facts. Two out of ten Swedes

also think political advertising in television is too negative in its tone towards politi-

cal opponents. Obviously, Swedish citizens in general do not feel that they are influ-

enced by this new form of political advertising. However, the results also indicate

that there is a large group of respondents that has answered “don’t know”. In the five

different statements this category varies between 27 to 41 percent units. This is a

result that probably may be associated to the fact that political advertising in televi-

12

sion is a new phenomenon, and that there is a considerably large group of the public

that has not been exposed to this advertising form.

(Table 2 about here)

But what is the significance of exposure? Previous advertising studies have shown

that more advertising exposure leads to a more accepting standpoint (Grusell 2008).

Thus, it is reasonable to ask whether a similar pattern will be discovered when ana-

lyzing political advertising in television. The results indicate that the overall conclu-

sion is the same; more exposure results in a more accepting and favorable opinion.

Exposure to this new form of advertising has not discouraged public opinion (tab. 3).

The result also confirms that individuals that have been exposed several times are

inclined to have an opinion about political TV-advertising, a result that can be com-

pared to individuals that never have been exposed. In this non-exposure group the

majority of the respondents have chosen the possibility to answer; “Don’t know”.

Interestingly, we note that even if the individual respondent has not been exposed to

political advertising in television, he or she may have well-articulated opinions on

the new phenomena. This result may be perceived as contradictory, but may proba-

bly be related to the fact that advertising nowadays peer see is a common segment in

the everyday life. Accordingly, latent predispositions may exist about this new ele-

ment of political communication.

(Table 3 about here)

As indicated above, the public attitudes towards political advertising in television

vary with regard to exposure for such advertising. However, it is reasonable to ask

whether other variables may influence the individual’s attitudes. Contemporary re-

search studies of public opinion on advertising indicate that such opinion in general

differs depending on a variety of individual and societal factors. Possible factors that

may affect such attitudes are for example gender, age, education level and ideologi-

cal values (Grusell 2008). However, the question may be raised whether these factors

13

also have a significant influence when analyzing political advertising in television as

a new phenomenon, as is the case of Sweden. Additionally, it may be relevant to ask

whether public trust in politicians and public interest for community affairs matter

when trying to explain the existing public attitudes towards this new form of adver-

tising. Consequently, an ‘opinion index’ was constructed in order to get a broader

perspective of the publics’ opinion of political TV-advertising in television.

Previous research has indicated that there are small gender differences when ana-

lyzing public opinion on traditional advertising in Sweden. However, when political

TV-advertising is considered substantial differences are obvious; males tends to an-

swer more often then women that they are neutral, and they are also less negative

then women. When political advertising in television is analyzed, females are gener-

ally more negative than men in Sweden (tab. 4, below).

Age has often proved to be significant when traditional advertising has been stud-

ied. Younger people generally have a more accepting view on advertising then older

generations. However, an interesting difference may be noted when media use is

considered; younger tends to use commercial media more often then older people do.

Therefore it’s fair to assume that age and exposure are interacting in this case. How-

ever the same pattern is discovered when political advertising is analyzed. Thus,

younger people are more positive towards this form of new advertising then older

generations.

The education factor has been slightly contradictory in traditional advertising

research. Generally, high-educated people tend to have more negative attitudes then

low educated citizens. However, there are exceptions to this rule: advertising on the

Internet, advertising at the movies, and out-door advertising are advertising forms

where high educated are more positive than low educated (Grusell 2008). However,

when analyzing political advertising in television the pattern is clear; high-educated

people are more negative than low educated people. High educated tends to choose

the neutral category as well as being less positive then other groups.

(Table 4 about here)

14

As stated in the introduction part of the paper, advertising in general is in many ways

a thoroughly debated and disputed social phenomenon. Accordingly, it is reasonable

to assume that the individual’s view of society from a political ideology dimension is

of importance when public opinion on advertising is considered. Previous studies of

traditional advertising have also indicated that there is a significant difference in this

aspect. People with a more liberal opinion on the commercial market tend to have a

greater acceptance of advertising (cf. Grusell 2008; Grusell 2007; Elfving 2005).

Consequently, it would be reasonable to predict the same results when analyzing

political advertising in television. However, this is not the case; on the contrary, there

are no significant relations between attitudes toward political advertising in televi-

sion and the individual’s adherence to political ideology (tab. 5). Furthermore, other

political factors such as public trust and public interest in community affairs also

prove to be of less importance in this aspect.

(Table 5 about here) One possible explanation for this result may be that opinions regarding political ad-

vertising in television have not been expressed by the political parties to the same

extent that have party opinions on traditional advertising. This advertising form is a

new phenomenon in the Swedish society and as such it hasn’t yet been clearly asso-

ciated to specific and publicly declared political party standpoints.

EXPLAINING POLITICAL TV AD OPINIONS IN SWEDEN – A”MULTI-

VARIATE” REGRESSION ANALYSIS

As previously noted, a series of possible factors may prove to be relevant when the

emerging public opinion on political advertising in Swedish television is analyzed. In

order to develop the analysis one step further we will now examine how gender, age,

education, dimensions of ideological values, public trust in politicians, public interest

for community affairs and exposure to political advertising in television relate to

each other.

Our preliminary analysis indicated that exposure to political TV-advertising led

to a more accepting attitude. Furthermore, we noted that women were more negative

15

than men, that younger were more positive than older, and that high educated were

more negative towards political TV-advertising then low educated. We have also

concluded that there were no significant relations between dimensions of ideological

values, public trust in politicians and the individual’s interest for community affairs.

However, in an effort to explain the relative importance of different factors, we have

to analyze them using a multiple regression analysis.

Consequently, we present a regression model (in table 6 below) of the output of

the different factors. The overall impression of this model is that it confirms the re-

sults of the previous analysis. Thus, gender influences attitudes towards political

advertising in television. Women tend to be more negative towards this new element

of political advertising. There is also an age relation. Younger persons are likely to

have more accepting attitudes towards political advertising in television then older.

This result is expected; younger persons perceive advertising as an integrated part of

the everyday life, while older persons are more unaccustomed to advertising in this

aspect.

(Table 6 about here)

However, the multivariate model offers one surprise; the effect for exposure.

When the attitudes towards political TV-advertising are analyzed the exposure for

such ads explains more than any other factor in the model. Individual and societal

factors are of less importance when trying to explain public attitudes towards politi-

cal advertising in television in Sweden. In the last part of this paper, this result will

be further discussed.

WHEN POLITICAL COMMUNICATIONS BECOME NON-POLITICAL…

The objective of this paper was to analyze public opinion on political advertising in

television as a new communication element in Swedish election campaigns. Two

major research questions were asked regarding public opinion towards political ad-

vertising in television, and the possible explanations for this public opinion.

16

The results in this study indicate that attitudes towards political advertising in

Sweden are in line with attitudes towards advertising in general. Thus, most Swedes

deny that they are influenced by televised political ads when making voting decisions

or informing themselves on party policy positions. Still, a large number of the elec-

torate has not yet been exposed to such ads and has no declared opinion in this mat-

ter. Therefore, it is reasonable to believe that more persistent opinions may be devel-

oped in future elections campaigns.

When the existing opinions on political advertising in television are further ana-

lyzed, the exposure for such advertising stands out as the single most important fac-

tor in explaining these opinions. The more people are exposed to this kind of ads, the

more likely they are to express positive opinions about them. Age and gender matter,

as young people and men are slightly more positive in their attitudes as is the case

with advertising opinions in general in Sweden. However, no other factor seems to

be as important in this aspect as is the exposure for political advertising in television.

Surprisingly, ideological beliefs on the individual level seem to be almost irrele-

vant when trying to explain existing public opinion. In studies of attitudes towards

advertising in general ideological predispositions proved to be decisive, with right-

wing partisans being more positive and left wing believers expressing more critical

attitudes. However, when advertising attitudes in a more political context were exam-

ined, political ideology seemed to lose its importance as an explanatory factor.

Paradoxically, political predispositions seem to matter more when general adver-

tising opinions are considered, compared to when political advertising attitudes are

examined. This should probably not be interpreted as if political advertising in Swe-

den is a politically indifferent topic. On the contrary, it could rather serve as an ex-

ample of the nature of opinion formation processes regarding new elements of politi-

cal communication, where strong ideological positions not yet have been taken.

As such an emerging phenomenon, political advertising in television does not

engage partisans representing different political views. Without being on the political

agenda, political TV-advertising is mainly considered from non-political perspec-

tives. As with advertising in general, it is affected by individual factors and the actual

exposure for such advertising. To put it simply, attitudes towards political advertising

17

in television in Sweden is much more about advertising than it is about politics, at

least in the current stages, which were analyzed in this study.

Thus, as long as political advertising in television remains as a politically ‘cold

case’ in Sweden, individual factors may prove to be more decisive when analyzing

public opinion in this area. Consequently, individual perceptions of the role of adver-

tising in contemporary societies and the personal acceptance of commercialization

processes within the media may be of great importance when explaining public atti-

tudes. Sooner or later, the issue may be politicized in a more traditional way, but

until then political advertising in television is probably evaluated in accordance with

the criteria for public judgements of advertising in general.

To conclude, political advertising in television is definitely one of the main char-

acteristics of elections campaigns in most modern democracies. As such, political

spots are in the centre of public discourse and function as an essential element of the

political struggle between incumbents and their opponents. However, as this study

indicates, the central role of political advertising is depending on the existing politi-

cal culture and the traditions of campaign practices. As a new and emerging element

in the political communication processes, political ads in television are judged more

by non-political considerations than by ideological convictions and political stand-

points. Thus, the Swedish experience confirms that political ads have to be perceived

as politically significant in order to be a central part of the political communication.

Nevertheless, an obvious limitation of this pilot study is of course the focus on

one single country, which restricts the possibilities of generalizing in relation to the

findings. However, the theoretical approach used here when explaining advertising

attitudes could be developed further in a comparative research framework, discussing

and analyzing different factors explaining public opinion on politically advertising

on television in different stages of development. Another methodological problem in

this study was the relatively high number of respondents in the survey that claimed to

be rather indifferent to political advertising in TV.

Finally, this paper argues for more systematic research in this field in the future.

The adoption of global campaign practices in different national election settings im-

poses a huge challenge for contemporary political communication research. Conse-

quently, the introduction of televised political advertising in Sweden could serve as

18

an interesting example of a situation where international trends and national contexts

meet, with results that are not in line with common experiences and are not easy to

predict.

19

References Carmines, E. G., & Zeller, R.A (1979) Reliability and validity assessment. Newbury Park, CA: Sage Publ.

Engblom, L.Å. & Wormbs, N. (2007), Radio och TV efter monopolet, Stockholm: Ekerlids.

Grusell, M. (2007), Advertising? Yes Please, But Only When It’s My Choice. Journal of Media Business Studies, vol. 4:3.

Grusell, M. (2008), Reklam – en objuden gäst? Allmänhetens uppfattningar om reklam i morgonpress och tv. Institutionen för Journalistik och masskommunikation, Göte-borgs universitet

Gustafsson, K-E. (2005), Reklamens makt över medierna. Stockholm: SNS förlag.

Hallin, D. C. & Mancini, P. (2004), Comparing Media Systems. Three Models of Me-dia and Politics, Cambridge: Cambridge University Press.

Holbert, R.L. et al. (2002), The Role of Communication in an Age of Television, Jour-nalism Quarterly, 55, pp. 106–115.

Holtz-Bacha, C. (2003), Political Advertising in Election Campaigns, In Maarek, P.J. & Wolfsfeld, G. (ed.) Political Communication in a New Era, London: Routledge.

Johnston, A. & Kaid, L.L. (2002) Image Ads and Issue Ads in Presidential Advertis-ing: Using Videostyle to Explore Stylistic Differences in Televised Political Ads from 1952 to 2000, Journal of Communication, 52, pp. 281–300.

Johnston, A. (2006), Methodologies for the Study of Political Advertising, In Kaid, L.L. & Holtz-Bacha, C. (ed.) The Sage Handbook of Political Advertising, Thousand Oaks: Sage Publ.

Kaid, L.L. & Holtz-Bacha, C. (ed.) (2006), The Sage Handbook of Political Advertis-ing, Thousand Oaks: Sage Publ.

Kaid, L.L. (2004), Political Advertising, In Kaid, L.L. (ed.): Handbook of Political Communication Research, Mahwah: Lawrence Erlbaum.

Lijphart, A. (1999), Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries, New Haven:Yale University Press.

Mediebarometern 2007 (2008), Göteborg: Nordicom, Göteborgs universitet.

Nord, L. W. (2008), Medier utan politik. En studie av de svenska riksdagspartiernas syn på dagspress, radio och TV, Stockholm: Santérus.

20

Nord, L.W. & Strömbäck, J. (2006), Do Politicians Lead the Tango? A Study of the Relationship between Swedish Journalists and their Political Sources in the Context of Election Campaigns, European Journal of Communication Vol. 21, No. 2, pp. 147-164.

Nord, L.W. (2006), Still the Middle Way: A Study of Political Communication Prac-tices in Swedish Election Campaigns, The Harvard International Journal of Press/Politics, Vol. 11, No. 1, pp. 64-76.

O’Donohoe, S. (2001), Living with ambivalence. Attitudes to advertising in postmod-ern times. Marketing Theory, vol. 1:1.

Plasser, F with Plasser,G. (2002), Global Political Campaining. A Worldwide Analysis of Campaign Professionals and Their Practices, Westport: Praeger.

Scipione, P.A. (1997) Too much or too little? Public perceptions of advertising expen-ditures. Journal of Advertising Research, vol. 37:3.

Shavitt, S.; Lowrey, P. & Haefner, J. (1998) ‘Public attitudes toward advertising: more favourable than you might think.’ I Journal of Advertising Research, vol. 38:4.

Trotzig, E. (2007) Reklam. I Carlsson, U. (red.) Mediesverige 2007. Göteborg: Nordi-com, Göteborgs universitet.

21

Appendix

Questions in the survey:

I) Have you seen political advertisements in television during the election campaign?

• Yes, several times

• Yes, at least once

• No, not at all

II) To what extent do you agree with the following statements about political adver-tisements in television? (1=disagree completely, 5=agree completely, don’t know-option)

• Political TV-advertising gives me guidelines regarding which party that suits me best

• Political TV-advertising gives me an oversimplified picture of party standpoints

• Political TV-advertising influences my voting decision on Election Day

• Political TV-advertising appeals too much to emotions and neglect facts

• Political TV-advertising is too negative in its tone towards political opponents

Figure 1

Figure 1 Part of the public exposed to advertising in different media an average weekday 2007 (percent)

Advertising in:

2

16

18

27

28

52

58

0 10 20 30 40 50 60 70

Mobilphone

Newspaper or magazine

Evening paper

Internet

Radio

Ad in the paper

Tv

Source: Mediebarometern 2007 (Nordicom-Sverige)

Tables 1- 6

Table 1 Exposure for political advertising in TV (percent) Exposure for political TV-advertising Yes, several times 14 Yes, some time 32 No, never 54 Percent 100 Number of responses 942 Commentary: Data is based on the number of people who has an opinion about political TV-advertising Source: Democracy Barometer, The Centre for Political Communication Research, Mid Sweden University.

22

23

Table 2 Statements of political TV-advertising (percent)

Political TV-advertising gives

me guidelines regarding which party that suits

me best

Political TV-advertising gives me an

oversimplified picture of party

standpoints

Political TV-advertising influ-ences my voting

decision on Election Day

Political TV-advertising ap-

peals to much to emotions and neglect facts

Political TV-advertising is

too negative in its tone towards

political opponents

Agree 4 39 4 36 23 Neutral 12 12 8 15 21 Disagree 55 19 60 14 16 Don’t know 28 31 27 36 41

Percent 99 100 99 101 101 Number of responses

908 912 911 906 910

Commentary: Data is based on the number of people who has an opinion about political TV-advertising Source: Democracy Barometer, The Centre for Political Communication Research, Mid Sweden University.

Table 3 Statements of political TV-advertising by exposure (percent)

Political TV-advertising gives

me guidelines regarding which party that suits

me best.

Political TV-advertising gives me an oversimpli-

fied picture of party standpoints.

Political TV-advertising influ-ences my voting decision on Elec-

tion Day

Political TV-advertising ap-

peals to much to emotions and neglect facts

Political TV-advertising is too

negative in its tone towards

political opponents

Exposure for political TV-advertising

Agr

ee

Indi

ffer

ent

Dis

agre

e D

on’t

know

A

gree

Indi

ffer

ent

Dis

agre

e D

on’t

know

A

gree

Indi

ffer

ent

Dis

agre

e D

on’t

know

A

gree

Indi

ffer

ent

Dis

agre

e D

on’t

know

A

gree

Indi

ffer

ent

Dis

agre

e D

on’t

know

Yes, several times 15 26 25 2 17 25 17 3 16 25 18 3 11 28 19 3 18 25 18 4

Yes, some time 41 42 42 11 32 45 44 13 39 40 59 10 36 46 43 15 38 41 43 20

No, never 44 32 32 86 51 30 39 84 44 35 23 88 54 26 38 82 45 34 40 76

Percent 100 100 99 99 100 100 100 100 99 100 100 101 100 100 100 100 101 100 101 100

Number of responses

499 112 40 258 168 108 353 278 549 75 39 243 121 134 323 325 141 189 205 371

Commentary: Data is based on the number of people who has an opinion about political TV-advertising Source: Democracy Barometer, The Centre for Political Communication Research, Mid Sweden University.

24

Table 4 Attitudes of political TV-advertising by population segments (percent)

Gender AGE Education level Woman Men 18-29 30-49 50-77 Low Medium High

Positive 31 34 39 33 29 33 39 24 Neutral 30 38 35 33 33 32 31 37 Negative 39 29 26 34 39 35 30 39 Percent 100 101 100 100 101 100 100 100 Number of responses 480 405 171 292 420 171 366 339 Commentary: Data is based on the number of people who has an opinion about political TV-advertising Source: Democracy Barometer, The Centre for Political Communication Research, Mid Sweden University. Table 5 Attitudes of political TV-advertising by dimensions of political ideology,

politician trust, interest for community affairs (percent) Public trust in politicians Public interest for

community affairs Ideological values

Little Some A lot Not interested

Some interest

A lot of interest Left Right Others

Haven’t made up

there mind or a blank

vote Opinions index Positive 28 33 38 26 36 31 33 34 29 29 Indifferent 41 30 28 29 32 35 33 35 42 31 Negative 31 36 33 45 32 34 34 31 29 40 Percent 100 99 99 100 100 100 100 100 100 100 Number of responses

288 440 158 86 289 512 268 329 38 228

Commentary: Data is based on the number of people who has an opinion about political TV-advertising Source: Democracy Barometer, The Centre for Political Communication Research, Mid Sweden University.

Table 6 Explaining public attitudes towards political advertising in television (standardized regression coefficients)

Model 1 Opinion Index

Gender -,086 ** Age -,073 * Education ,081 ** Public trust in politicians -,027 Public interest for community affairs -,053 Ideological values ,002 Exposure for political TV-advertising ,354 *** R² 14,9 Number of responses 858

Commentary: Significance levels =* p<.05, ** p<.01 and *** p<.001. People who has an opinion about political TV-advertising Source: Democracy barometer, Centre for Political Communication Research, Mid Sweden University.

25