Virtual advertising in sports events: does it really work?

Keywords
virtual advertising electronic billboards virtual signage advertising effectiveness

Abstract
This paper examines the effects of virtual advertising in a sports broadcast setting. We analyse the conspicuousness of virtual advertising and match the results with explanatory variables like brand awareness, duration of exposure and frequency of exposure. Furthermore, we measure the role of attitudes towards advertising in general and its impact on attitudes towards virtual advertising of the respondents. Our results indicate that most respondents recognise virtual advertising as such. Advertising effectiveness is driven to a large degree by the frequency of exposure. A positive attitude towards advertising in general leads to a positive attitude towards virtual advertising of the participants.

Matthias Sander
Professor of Marketing, University of Konstanz Box 130, 78457 Konstanz, Germany Tel: 00 49 7531 88 2599 Fax: 00 49 7531 88 3560 Email: matthias.sander@uni-konstanz.de)

RESEARCH PAPER

Claudia Fantapi Altobelli

Professor of Marketing, Helmut Schmidt University University of the Federal Armed Forces Hamburg Holstenhofweg 85, 22043 Hamburg, Germany Tel: +49 40 6541 2772, Fax: +49 40 6541 2090 Email: fantapie@hsu-hh.de

Peer reviewed

Executive summary
Virtual advertising is a technology that allows digitised superimposition of images (i.e. advertising signage) into a television broadcast (Cianfrone et al, 2006; Pyun & Kim, 2004). Little has been known about the effects of this new form of advertising on potential customers, and this paper serves as an exploratory study to deliver more insights into the effectiveness of virtual advertising. We hypothesised that virtual advertising is recognised as such by the recipients. Furthermore, we tried to find out what factors influence the effectiveness of virtual advertising. Variables which can be assumed to influence the measured effectiveness are brand awareness, duration of exposure and frequency of exposure (Tellis, 1997; Longman, 1997; Pieters & Bijmolt, 1997; Turley & Shannon, 2000; Sander, 2004). The impact of these variables was also measured for conventional advertising. We also looked at the role of attitudes towards advertising in general as well as attitudes towards virtual advertising. We hypothesised a positive correlation between attitude towards advertising in

APRIL 2011

International Journal of Sports Marketing & Sponsorship

225

Virtual advertising

general and attitude towards virtual advertising. Our stimulus for this study was an 18-minute video clip from a televised soccer match. Research participants (n=142) were students of a German university. The findings indicate that most of the respondents recognised virtual advertising. Our model gives an excellent explanation for the effect of the variables (brand awareness, duration of exposure and frequency of exposure) affecting the respondents recall performance. The frequency of exposure appears here as the key explanatory variable. Finally, we compared the participants attitudes towards advertising in general with their attitudes towards virtual advertising. We found a highly significant and positive correlation. In conclusion, this investigation provides fundamental data for further research on consumer responses to virtual advertising in sports broadcasts. However, there is only a little conclusive evidence regarding the effectiveness of different kinds of virtual advertising (e.g. 3D animation, animated virtual advertising) so far. More effort has to be made in this direction.

Introduction
Virtual advertising has been in use in the U.S. since the mid 1990s, frequently during sports events (e.g. soccer, baseball, basketball). In Europe, especially due to restrictive regulations, virtual advertising has not so farbeen used very often. Virtual advertising can be defined as the insertion of brands, products, trademarks, logos, slogans and 3D animations by means of digital, computer-supported television signal processing (Cianfrone et al, 2006). This involves overlaying an advertisement into a space in the telecast either over the top of existing ground signage or billboards or in a free space on the field of play or in the crowd (Turner & Cusumano, 2000). The advertising is therefore only visible to the television viewer i.e. the people in the stadium cannot see the imposed image.

The advantages of virtual advertising over traditional advertising are evident. Virtual advertising allows the multiple sale of existing advertising space, and during international sports events in particular, advertisements can be delivered in country-specific broadcasts. Hence, advertising during major international sports events becomes interesting not only for worldwide operating firms, but also for those acting solely at a national level. The implementation of this technology can be differentiated even within a country, at a local or regional level. This permits a targeted group-specific use by tailoring advertisements according to regional consumers buying habits and brand preferences. Moreover, it allows country-specific advertising regulations to be taken into consideration (e.g. for alcoholic beverages or tobacco products). In addition to the existing advertising, it facilitates the use of new advertising space within a certain sports event. These advantages suggest the enormous potential yield of virtual advertising. In 2004 it was estimated that in the U.S. alone, considerably more than $2 billion was being invested in virtual advertising (Pyun, Han & Ha, 2004), and substantial growth rates since then can be assumed. However, little has been known about the effects of this innovative form of advertising on potential customers. This paper has therefore been conceived as an exploratory study to deliver more insights into the effectiveness of virtual advertising and will show future areas of research in this field.

RESEARCH PAPER

Literature review
Although there are a large number of studies that examine various practical and theoretical aspects of sponsorship (see the overview in Walliser, 2003; for more recent studies, Wakefield et al, 2007), so far there have been relatively few studies dealing with the effects of virtual advertising as a related communications instrument. Some studies examine the acceptance of virtual advertising among viewers.

226

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

In general, virtual advertising is accepted by viewers, but with some limitations. The market research institute FORSA investigated the attitudes of German viewers towards advertising in general and specifically towards virtual advertising during soccer matches (FORSA, 1998), with 1,003 respondents interviewed by telephone. The results showed that 31% of respondents felt disturbed by virtual advertising, 16% were indifferent towards it and only 18% did not feel disturbed; 35% had no opinion. As for gender, 40% of males and only 22% of females felt rather distracted. Older people also felt more disturbed (41% of those aged 50-59) than the under 30s (who had hardly any objections to virtual advertising). A study commissioned by GLOBAL SPORTNET was carried out with a 13-member discussion group in Munich (Germany) using the soccer match Benfica Lisbon versus FC Bavaria Munich as well as a video recording of the soccer match Glasgow Rangers versus AC Parma (Sasse & Ludwig, 2002). The testers pointed out the conspicuousness of virtual advertising. Insertions during the match were perceived as disturbing, but it was presumed that viewers would get accustomed to them. Furthermore, the group highlighted certain economic aspects, such as the better accessibility for target groups as well as the multiple marketability of the same advertising space to different customers at different prices. However, it was noted that virtual advertising should be carried out in a discreet, unobtrusive way, as otherwise there might be a danger of viewers moving over to commercial-free pay-TV stations. On the whole, the acceptance of virtual advertising was high among the members of the discussion group. The purpose of the study by Bennett et al (2006) was to analyse the perceptual antecedents of advertising in general through television commercials and virtual advertising to compare the two media vehicles in a sports broadcast setting. They investigated attitudes towards television commercials and virtual advertising among students from a large university in the south-eastern U.S. using Brackett and

Carrs (2001) model, with its perceptual antecedents (credibility, irritation, entertainment and informativeness) as independent variables. The findings indicate that individual responses to advertising vary according to the type of advertising. The results show that television commercials were perceived as being more entertaining and informative but also more irritating and less credible than virtual advertisements, regardless of their location. However, virtual advertising with logo ads above or below the score display was considered more credible than any other type of advertisement. This means that different locations will trigger different responses. Therefore, the location of virtual advertising is an important consideration for media planning. On the whole, virtual advertising performed well compared to conventional advertising forms. A focus group study carried out at the University of Trier (Germany) investigated how television viewers react to different forms of virtual advertising and the new advertising space. It also examined how the increased quantity of advertising possibly eclipsing the sports event is perceived and to what extent the credibility of television may suffer under virtual techniques (Sasse & Ludwig, 2002). The study was based on the same soccer matches as in the GLOBAL SPORTNET study. A video demonstrating the technical possibilities of virtual advertising was also shown. The results indicate that although viewers accepted virtual advertising per se, they tended to reject virtual advertising on the field. This was because in soccer unlike in sports such as ice hockey or boxing advertising on the field is unusual. Generally, advertising should not divert from the sports event. Insertions next to the goal were rejected because of their size and location. Virtual 3D animations both during the match and at half-time were rejected as well, probably due to the danger of diverting the viewers from the sports event or from other events during half-time. Furthermore, the more closely the advertised brand was associated with sport, the better virtual advertising was assessed. Virtual advertising in programme sectors other than sports was rejected,

RESEARCH PAPER

APRIL 2011

International Journal of Sports Marketing & Sponsorship

227

Virtual advertising

with the argument that this would damage the credibility of television. This, however, would not apply to sport. Generally, it was assumed that virtual advertising would eventually establish itself, provided that it was implemented discreetly and unobtrusively. A second group of studies investigated the effectiveness of virtual advertising. On the whole, virtual advertising is a highly effective communications tool. In a day-after recall test for a representative household sample, Psyma Psychological Market Analysis analysed the acceptance and effectiveness of virtual advertising among German viewers in the aforementioned Benfica Lisbon versus FC Bavaria Munich match (Psyma, 1999). In this recall test virtual billboard advertising achieved higher rates of attention than conventional billboard advertising. Younger television viewers showed higher attention values with regard to virtual advertising than older viewers. Furthermore, the acceptance of virtual advertising among potential customers was very high. RSL Sponsorship, Sport und Leisure carried out a study to assess the acceptance and effectiveness of virtual advertising during the ATP Tennis Tour, for which there were insertions on the tennis courts in Germany, France and Italy (Sasse & Ludwig, 2002). They found that recognition levels for logos inserted on the playing area were much higher than those inserted on the sides of the court 80% of respondents considered virtual advertising at the venue as not disturbing, 45% had a positive impression about the organisations using virtual advertising and 44% indicated that they were positively inclined to buy the products of the advertised brands and firms. A study by Gallup Mexico for Publicidad Virtual compared the effects of virtual advertising and conventional arena advertising on viewers of televised sports (Doyle, 2000): 406 inhabitants of Mexico City aged 18 years and over who had watched a soccer match with virtual advertising on a certain weekend, were interviewed by phone. The central finding of the study was that 60% of the respondents preferred virtual advertising to all other forms of advertising. Virtual advertising was seen as modern, creative and

innovative. A further key result of the study was that virtual advertising would generate a higher perception of brands than all other programme-integrated forms of advertising, like arena signage, advertisements on fences or sports shirt advertising. In a media analysis of five ESPN Sunday Night Baseball games broadcast between 4 May and 3 August 2003, Pyun & Kim (2004) found that virtual advertising on television screens was visible significantly more often than conventional advertising in the stadiums. Assuming that visibility and perception of advertising are necessary conditions for its effectiveness, they expected that virtual advertising would be more effective than conventional advertising. For this purpose, the five games were cut together to make a 10-minute video clip, followed by a written interview to test the recognition rate of the respondents (n=164). The recognition level for virtually advertised brands was 31%, nearly double that of conventionally advertised brands (16%) (Pyun, Han & Ha, 2004). Cianfrone et al (2006) analysed the effects of virtual advertising on consumers recall levels. The study compared the effectiveness of virtual advertising to that of conventional television commercials during a sports broadcast. For this purpose, 110 students of a U.S. university were interviewed by means of a written questionnaire to assess their recall and recognition levels with regard to both virtual and conventional commercials. The questionnaire referred to a 40-minute soccer match that had previously been televised. During the match, different forms of virtual advertising were used, such as virtual advertising on the field, virtual insertion of brands into the score display, virtual insertions on the side boundary lines, virtual advertising on or close to the goal posts and virtual billboard advertising. The match was interrupted by a commercial break with several 30-second commercials. Both unaided and aided recall rates were measured. With unaided recall, commercials were remembered best, followed by virtual advertisements in the score display and virtual billboard advertisements. The high rates for

RESEARCH PAPER

228

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

TABLE 1 Overview of studies on virtual advertising

AUTHOR(S)/ INSTITUTION (YEAR)

RESEARCH DESIGN/ STUDY TYPE

MAIN CONTRIBUTION

FINDINGS/ REMARKS

RSL SPONSORSHIP , SPORT EMPIRICAL AND LEISURE (1997) TV JOURNAL TODAY (1998) PSYMA PSYCHOLOGICAL MARKET ANALYSIS (1999) EMPIRICAL (TELEPHONE INTERVIEWS) EMPIRICAL (DAY-AFTER RECALL TEST)

ACCEPTANCE AND EFFECTIVENESS VERY POSITIVE ATTITUDE OF VIRTUAL ADVERTISING DURING TOWARDS VIRTUAL ADVERTISING ATP TENNIS TOUR ATTITUDE TOWARDS ADVERTISING IN GENERAL AND TOWARDS VIRTUAL ADVERTISING NEGATIVE ATTITUDE TOWARDS VIRTUAL ADVERTISING, ESPECIALLY AMONG MALE AND OLDER RESPONDENTS

EFFECTIVENESS OF VIRTUAL HIGHER ATTENTION RATES FOR VIRTUAL ADVERTISING IN A SOCCER MATCH BILLBOARD ADVERTISING THAN CONVENTIONAL BILLBOARD ADVERTISING; HIGH ACCEPTANCE OF VIRTUAL ADVERTISING ACCEPTANCE AND ATTITUDE TOWARDS VIRTUAL ADVERTISING ATTITUDE TOWARDS VIRTUAL ADVERTISING ACCEPTANCE OF VIRTUAL ADVERTISING RECOGNITION RATES FOR VIRTUALLY AND CONVENTIONALLY ADVERTISED BRANDS POSITIVE ATTITUDE TOWARDS VIRTUAL ADVERTISING BUT DISCREET, NOT TOO OBTRUSIVE, IMPLEMENTATION NECESSARY VERY POSITIVE ATTITUDE TOWARDS VIRTUAL ADVERTISING AND HIGH ACCEPTANCE SOME POSITIONS FOR VIRTUAL ADVERTISING IN A SPORTS EVENT REJECTED VERY HIGH RECOGNITION RATES FOR VIRTUALLY ADVERTISED BRANDS

GLOBAL SPORTNET (1999) DISCUSSION GROUP

GALLUP MEXICO (1999)

EMPIRICAL (TELEPHONE INTERVIEWS) DISCUSSION GROUP MEDIA ANALYSIS

RESEARCH PAPER

UNIVERSITY OF TRIER (2001) PYUN & KIM (2004)

CIANFRONE ET AL (2006)

LABORATORY STUDY

EFFECTS OF VIRTUAL ADVERTISING VIRTUAL ADVERTISING AT LEAST AS AND TELEVISION COMMERCIALS EFFICIENT AS TELEVISION COMMERCIALS ON CONSUMER RECALL LEVEL EFFECTS OF ADVERTISING TYPE (VIRTUAL/CONVENTIONAL) AND ANTECEDENTS OF ATTITUDE TOWARDS ADVERTISING IN GENERAL ON INDIVIDUAL RESPONSES TO ADVERTISING INDIVIDUAL RESPONSES TO ADVERTISING VARY ACCORDING TO THE TYPE OF ADVERTISING

BENNETT ET AL (2006)

LABORATORY STUDY

commercials, however, were partly due to the fact that some brands had been advertised both in commercials and virtually. With aided recall, commercials recall scored highest, again followed by virtual advertisements in the score display. Virtual advertising on or close to the goal posts and virtual advertising on the field performed worst. The results showed that virtual advertising, in comparison with commercials, came off well in terms of recall levels. One must also take into account that apart from the permanent

virtual advertisements in the score display commercials were visible in one unit for a considerably longer time than virtual advertisements. Virtual advertising proved to anchor the advertised brands very well in consumers minds and may be regarded as at least as effective as conventional television commercials. Table 1 gives a summary of the studies mentioned, the aspects analysed and their main findings.

APRIL 2011

International Journal of Sports Marketing & Sponsorship

229

Virtual advertising

Hypothoses
None of the studies mentioned in the literature review analysed whether television viewers can recognise virtual advertising, in the sense that they can differentiate explicitly between virtual advertisements and conventional advertisements. To achieve valid results with regard to the acceptance and perception of virtual advertising, it is necessary that viewers do not confuse virtual advertisements with other forms of advertising. Hence, the first hypothesis is: H1: Virtual advertising is recognised by a majority of television viewers. As previous studies in the field of traditional sports advertising and sponsoring have shown, the effectiveness of advertising depends on a multitude of variables (Wakefield et al, 2007; Harshaw & Turner, 1999; Olivier & Kraak, 1997; Moore, Pickett & Grove, 1999; Turley & Shannon, 2000; Grohs, Wagner & Vsetecka, 2004). The studies suggest that, among others, variables like exposure time, exposure frequency and the prior awareness degree of the brands being advertised will determine the recall of these brands, i.e. these variables seem to have a significant impact on advertising effectiveness in a sports broadcast setting (Sander, 2004; Deimel, 1992; Schumann, 1987; Tellis, 1997; Walliser, 1997b). This applies to conventional advertising as well as to virtual advertising. As shown in other studies (e.g. Sandler & Shani, 1989; Deimel, 1993; Walliser, 1997a; Hackforth, 1989), these variables will presumably have a positive impact on recall and recognition rates. In the studies of Psyma (1999), Pyun & Kim (2004) and Cianfrone et al (2006), the recall and recognition rates vary, depending on the types of advertisements. However, these studies do not explicitly take into account the mentioned explanatory variables. Therefore, our second hypothesis is:
RESEARCH PAPER

H2: Exposure time, exposure frequency and prior brand awareness have a positive impact on the recall of the brands advertised. The attitude of recipients towards virtual advertising is also relevant. As shown in the literature review, some studies showed a very positive attitude towards virtual advertising (e.g. Psyma, 1999; Doyle, 2000), while other studies presented a more critical perception of this new form of advertising (e.g. FORSA, 1998; Sasse & Ludwig, 2002). There is a large body of studies dealing with attitudes towards advertising in general (e.g. Wang et al, 2002; Petrovici & Marinov, 2007). It is particularly interesting to ascertain whether people with a positive attitude towards advertising in general also have a positive attitude towards virtual advertising and vice-versa. Valid results in this context are still lacking. Hence, the third hypothesis is: H3: Attitude towards advertising in general is positively correlated with attitude towards virtual advertising.

Data and methodology

In this study, students of a German university were interviewed about the effectiveness of virtual advertising. The interviews were based on a nonrandom sample of participants, as is customary for exploratory studies (e.g. Cianfrone et al, 2006; Brand & Greenburg, 1994). An 18-minute video clip of the soccer match Hertha BSC Berlin versus VfB Stuttgart was chosen, showing both conventional and virtual advertising. Conventional advertising was broadcast in the form of perimeter advertising billboards and sports shirt advertising, while virtual advertising was shown as virtual billboards on the right and left hand sides of the goal. The virtual billboards were visible only for the clip viewers and not for the audience in the stadium.

230

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

TABLE 2 Structure of the sample

SAMPLE TOTAL N=142 GENDER INTEREST IN SOCCER BROADCASTS a
a = not specified: 1.5%

MALE 90.8% HIGH 34.5% MEDIUM 23.9%

FEMALE 9.2% LOW 40.1%

The students were shown the match clip, and a written standardised interview followed. A total of 142 undergraduate und graduate students participated in the study, pooled from several business management classes. We split the participants into an unaided recall group (n=71) and an aided recall group (n=71) to measure possible differences in the awareness levels of the branded products. All questionnaires could be evaluated. The survey recorded participants gender and their general interest in soccer match broadcasts on a 3-point scale (high, medium, low ). Gender-related differences in the perception of (conventional) advertising have already been found in other studies regarding the effectiveness of advertising (e.g. Brunel & Nelson, 2003; Sander, 2004); thus, it seemed reasonable to include this in the test design. We included interest in soccer match broadcasts in our variable set as we assumed that a pronounced interest in this direction may influence the effectiveness of advertising (Walliser, 1997a; Meenaghan, 2001). Because of a low variance in the student sample, we did not explicitly include age as a variable in the analyses. The structure of the sample is shown in Table 2. Additionally, the questionnaire comprised questions about attitudes towards advertising in general and towards virtual advertising in particular, using some perceptual antecedents of the Brackett and Carr (2001) model. We used items with a positive phrasing as well as reversed items. Attitudes towards advertising in general were measured with the items I find advertising disturbing, Advertising offers interesting information about products and brands, Advertising creates new jobs and Advertising is entertaining. Attitudes

towards virtual advertising were measured with the items Virtual advertising is an interesting new form of advertising, Virtual advertising irritates television viewers, By virtual advertising playing areas can be better marketed and Virtual advertising should be prohibited. For this purpose, we used 5-point multiattribute Likert scales (Churchill, 1979; Converse & Presser, 1986). Possible answers were offered on a scale of one to five, from 1 (I totally agree) to 5 (I do not agree at all). Furthermore, in a preliminary study we assessed prior brand awareness of the same respondents, given that different levels of brand awareness may influence the effectiveness of the advertising for those brands. This effect has been measured in other sports event studies with conventional advertisements (Walliser, 1997a; Sander, 2004). A total of five brands/organisations were advertised during the soccer match, with different, partly multiple, forms of advertising. This rather low number of brands advertised is probably due to the fact that the match was part of the qualifying round, with a limited number of television viewers. Advertising was delivered as advertising on the shirts of the players (Arcor and EnBW), on the shirt of the referee and his assistants (Dekra) and as billboard advertising (Premiere and betandwin.de). Billboard advertising for both brands was made on physical billboards alongside the pitch and virtually on the right and left hand sides of the goal. In addition to the questions, we performed a media analysis of the match. First, we measured the duration of exposure of the individual brands in the video clip (in seconds). The duration of exposure is considered as a typical determinant of the recall performance for

RESEARCH PAPER

APRIL 2011

International Journal of Sports Marketing & Sponsorship

231

Virtual advertising

sports advertising (Pieters & Bijmolt, 1997; Nebenzahl & Hornik, 1985). In accordance with findings of the psychology of perceptions and common practice in comparable studies, durations of less than a second were not considered within this survey, for it is assumed to have no effectiveness at all (Taub & Abrams, 1971). As a rule, it is presumed that one second represents the lower limit from which point onwards a (more or less) conscious perception may take place. Beside the duration of exposure, the frequency of exposure of the individual brand advertisements was measured, given that effectiveness might indeed vary, depending on whether a brand is shown more rarely but longer, or more frequently but for a shorter duration (Longmann, 1997; Tellis, 1997; Turley and Shannon, 2000). We measured frequency as the number of times the brand name was clearly legible by the viewers for more than one second. We could not take into account the design of the billboards in the test as none of the billboards used was particularly eye-catching. Furthermore we could not include relatedness (like Wakefield et al, 2007) due to the fact that only one of the five advertised brands has some relationship to sports.

Analyses and results

In the questionnaire we explained the character of virtual advertising and asked the respondents whether they had recognised any kind of virtual advertising in the video clip. The respondents could choose between different locations of virtual advertising: virtual billboard advertising along the pitch, virtual advertising on the ground, virtual goal-side billboard advertising and virtual shirt advertising. The fact that virtual advertising had been used in the soccer match was recognised by 77.5% of participants and 92.7% of these did assign virtual advertising correctly, i.e. they recognised it as goalside billboards. Therefore, H1 is supported. Further analysis showed that there is no significant difference

in the perception of virtual advertising with regard to gender and interest in soccer match broadcasts. A detailed insight into the effectiveness of different kinds of advertisements can be obtained by regarding the recall levels of these advertisements for the individual brands. Table 3 displays the aided and unaided recall rates for the individual brands, depending on the kind of advertising for these brands. The participants could choose here between sports shirt advertising, billboard advertising along the pitch, goal-side billboard advertising and other, multiple selections being possible (and correct in the case of Premiere and betandwin.de). As it is common practice, the aided recall questionnaire not only contained questions concerning awareness of the brands featured during the soccer match, but also included dummy brands, i.e. brands not advertised in the match (Sander, 2004; Deimel, 1992; Schumann, 1987). Four respondents who ticked these brands in the questionnaire were excluded from further consideration to guarantee a high validity of the results. Overall, the recall levels are very high. This may be due to the fact that only five brands were advertised. Also, the differences in aided and unaided recall rates for the same advertisement are rather small in many cases (for the betandwin.de billboard the unaided recall level is even a bit higher than the aided recall level). Aside the small number of brands advertised, this result may be explained by the fact that the questionnaire was administered to the participants directly after being exposed to the match clip. Nevertheless, as shown in Table 3, the recall values of the individual brands vary to a great extent. For example, the brand Dekra was recalled by only half as many respondents as the brand Arcor. To explain this effect, an OLS regression was conducted. The following results refer to the aided recall. Calculations with the unaided recall produced comparable results, due to the fact that there are only small differences between aided and unaided recall rates. We estimated the following model:

RESEARCH PAPER

232

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

TABLE 3 Recall performance in dependence of modality of advertising

BETANDWIN.DE BETANDWIN.DE ACTUAL GOAL-SIDE BILLBOARD BILLBOARD ADVERTISING ADVERTISING ALONG THE (VIRTUAL PITCH ADVERTISING) AIDED RECALL LEVEL UNAIDED RECALL LEVEL 82.1 % 82.9 % 71.6 % 52.9 % PREMIERE ACTUAL BILLBOARD ADVERTISING ALONG THE PITCH 41.8 % 37.1 % PREMIERE GOAL-SIDE BILLBOARD ADVERTISING (VIRTUAL ADVERTISING) 64.2 % 58.6 % DEKRA SPORTS SHIRT ADVERTISING ENBW SPORTS SHIRT ADVERTISING ARCOR SPORTS SHIRT ADVERTISING

44.8 % 42.9 %

74.6 % 74.3 %

82.1 % 78.6 %

TABLE 4 Frequency of exposure, duration of exposure and awareness of the advertised brands
BETANDWIN.DE BETANDWIN.DE ACTUAL GOAL-SIDE BILLBOARD BILLBOARD ADVERTISING ADVERTISING ALONG THE (VIRTUAL PITCH ADVERTISING) FREQUENCY OF EXPOSURE DURATION OF EXPOSURE IN SECONDS PRIOR BRAND AWARENESS LEVEL 42 557 21 83 PREMIERE ACTUAL BILLBOARD ADVERTISING ALONG THE PITCH PREMIERE GOAL-SIDE BILLBOARD ADVERTISING (VIRTUAL ADVERTISING) 15 71 DEKRA SPORTS SHIRT ADVERTISING ENBW SPORTS SHIRT ADVERTISING ARCOR SPORTS SHIRT ADVERTISING

RESEARCH PAPER

4 23

12 38

20 79

93.5%

99.0%

99.0%

58.5%

99.0%

a: Variable excluded due to poor legibility or duration of exposure < 1 sec.

RECALLij = a0 + a1DURij + a2FREQij + a3AWAREi + eij with: RECALLij: aided recall rate of brand i according to the modality of advertising j, a0: absolute term, DURij: duration of exposure to brand i according to the modality of advertising j, FREQij: frequency of exposure to brand i according to the modality of advertising j, AWAREi: prior brand awareness of brand i, and eij: error term for brand i and modality of advertising j.

Table 4 shows the observations for the duration and frequency of exposure for each brand, differentiated according to the modality of advertising and prior brand awareness rates of the brands. Billboard advertising along the pitch for the brand Premiere was no longer taken into account here, as the billboard was either hardly legible or the duration of exposure was under the assumed limit of one second. This explains the rather low awareness levels in comparison to other brands and advertising forms (see Table 3). We assume that the measured recall level can be explained by irradiation effects as Premiere was the main sponsor of the match. The low

APRIL 2011

International Journal of Sports Marketing & Sponsorship

233

Virtual advertising

TABLE 5 Results of the OLS regression

r2 = 0.942 (r2ADJ = 0.884) F-VALUE = 16.179 SIGNIFICANCE = 0.023 EXPLAINED VARIABLE AIDED RECALL
* = HIGHLY SIGNIFICANT

EXPLANATORY VARIABLE BRAND AWARENESS LEVEL FREQUENCY OF EXPOSURE DURATION OF EXPOSURE

BETA-VALUE - 0.272 1.686 - 0.963

SIGNIFICANCE 0.149 0.014* 0.06

frequency of exposure for the brand Dekra is due to the fact that only the referee and his two assistants were wearing the sports shirt advertising for this brand, whereas the brand EnBW was advertised by all members of the team VfB Stuttgart and the brand Arcor by all members of the team Hertha BSC Berlin. Our model gives an excellent explanation for the effect of the variables affecting the respondents recall performance. An r2 of 0.942 (r2adj=0.884) indicates a very good model fit. The whole model is highly significant (p=0.023). The autocorrelation test of the independent variables showed negligible correlation. The frequency of exposure appears as an outstanding explanatory variable (p=0.014). There is obviously a strong connection between the recall effectiveness and the frequency of exposure of advertised brands. This result is in line with other studies (Grohs, Wagner & Vsetecka, 2004; Tellis, 1997; Sander, 2004). The importance of this variable, compared to the other independent variables, becomes evident in terms of the beta value, which is clearly higher than with the other variables (the sign is of no relevance here). Of less significance is the duration of exposure (p=0.06). Furthermore, the sign is negative. At first this might be surprising, but it is consistent with the results from other studies (Sander, 2004). Obviously, it is the frequency and not the duration of exposure that is essential for the recall performance. This indicates that recipients do perceive the advertised brand during a longer exposure, but, once perceived, their attention shifts immediately back to the sports

event. Prior brand awareness as an explanatory variable proved to be non-significant (p=0.149). The small variance of these variables might be essential here, as nearly all brands showed very high awareness rates of the respondents (see Table 4). Therefore, H2 is confirmed with regard to the frequency of exposure and is rejected with regard to brand awareness and the duration of exposure. The results of the OLS regression are summarised in Table 5. In order to gain further information about the cognitive structures of the participants, we asked the participants to assess their attitude towards advertising in general as well as towards virtual advertising. The participants attitude towards advertising in general is summarised in Table 6, showing that the internal validity of the scale as a whole was unexpectedly low (Alpha = 0.42). Thus, we focus on the respondents attitudes towards the single items. The respondents critical attitude towards advertising in general is in line with the findings of other studies (Barnes, 1982; Witkowski & Kellner, 1998), with only the item Advertising creates new jobs displaying a strongly positive attitude towards advertising. The feedback regarding attitudes towards virtual advertising (see Table 7) was better. A clear majority agrees that by means of virtual advertising, advertising space can be better marketed. Nearly three-quarters of all respondents disapprove of a ban on this form of advertising. Equally, a majority is of the opinion that virtual advertising does not have an irritating effect. This finding, however,

RESEARCH PAPER

234

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

TABLE 6 Respondents attitudes towards advertising in general

ITEM I FIND ADVERTISING DISTURBING ADVERTISING OFFERS INTERESTING INFORMATION ABOUT PRODUCTS AND BRANDS a ADVERTISING CREATES NEW JOBS ADVERTISING IS ENTERTAINING a = NOT SPECIFIED: 0.7% CRONBACHS ALPHA = 0.42 I RATHER AGREE (1+2) 50.7 % 17.6 % 70.9 % 20.4 % I RATHER DISAGREE (3+4+5) 49.3 % 81.7 % 29.1 % 79.6 %

TABLE 7 Respondents attitudes towards virtual advertising

ITEM VIRTUAL ADVERTISING IS AN INTERESTING, NEW FORM OF ADVERTISINGa VIRTUAL ADVERTISING IRRITATES TELEVISION VIEWERSa BY VIRTUAL ADVERTISING PLAYING AREAS CAN BE BETTER MARKETEDa VIRTUAL ADVERTISING SHOULD BE PROHIBITEDb
a= NOT SPECIFIED: 2.1 % b = NOT SPECIFIED: 2.8 %

I RATHER AGREE (1+2) 37.3 % 42.3 % 62.7 % 23.3 %

I RATHER DISAGREE (3+4+5) 60.6 % 55.6 % 35.2 % 73.9 %

RESEARCH PAPER

CRONBACHS ALPHA = 0.69

should not be generalised, as the kind of virtual advertising certainly plays a significant role here. Internal validity of this scale as a whole was much higher (Alpha=0.69). Furthermore, we controlled for the sequence effects by varying the sequence of the items, both for the attitude towards advertising in general and for the attitude towards virtual advertising. However, t-test statistics showed no significant differences between groups (with the only exception of the item Advertising is entertaining, p=0.03). Thus sequence effects can be neglected. We also verified the sensitivity of results with regard to those respondents who checked a mid-position (=3) on the Likert scales for the individual items. Following the hypothesis that these are the undecided or lazy persons choosing the middle position to

simplify matters, an exclusion of these respondents may seem reasonable. Analysis shows, however, that in this case too, the results hold and majorities do not change for the individual items, with only one exception (Virtual advertising is an interesting new form of advertising: 49.1% rather agree, 48.1% rather disagree, 2.8% not specified). Hence, on the whole, the results are very firm. As we postulated that attitudes towards advertising in general might influence attitudes towards virtual advertising, we also verified the relationship of the participants attitudes towards advertising in general and towards virtual advertising by means of correlation analysis. The correlation is highly significant (p=0.000) and, as expected, positive (r=0.40). Therefore, the results support H3.

APRIL 2011

International Journal of Sports Marketing & Sponsorship

235

Virtual advertising

Discussion
The results are consistent with previous studies but allow more differentiated and extensive insights. A main finding is the fact that the duration of exposure does not play an essential role, whereas the frequency of exposure has a significant influence on the effectiveness of advertising. For advertising decisions, this suggests implementing virtual advertisements in frequent alteration in order to achieve stronger attentiveness effects. However, consideration must be given to limitations, which sometimes prohibit the use of additional advertising space for virtual advertising on frequently visible sites (FIFA, 1999). Despite a positive tendency in the assessment of virtual advertising, implementing this new form of advertising too obtrusively is perceived as rather disturbing. Our investigation indicates that more than 40% of the participants are irritated by virtual advertising. A main reason for this may be because this communications tool is hardly known yet. Above all, virtual advertising should not divert from the sports event, so as not to jeopardise acceptance by television viewers. As clearly shown in other studies with conventional advertising (Babin & Sheri, 1996; Russell, 2002; Sander, 2004, Walliser, 1997b), prior brand awareness also tends to have a positive effect on brand recall. As mentioned, the non-significance of this variable in this study is probably due to the small variance of this variable. We expect that in virtual advertising, limited display options will cause high and lasting advertising effects, especially in the case of sufficient brand awareness. Hence, virtual advertising seems to be not very suitable for the introduction of (new) brands. That television viewers recognise virtual advertising and have an overall positive attitude towards virtual advertising (especially in comparison to conventional forms of advertising) makes this communications tool interesting for advertisers. Thus shifting advertising budgets partly to this new form of advertising seems reasonable in order to achieve a more effective mix of communications instruments.

Limitations and future research

Our study cannot claim representativeness, as it was conceived as an exploratory analysis. While convenience samples are acceptable in pilot testing or when they represent a significant part of the focal population, in most cases the results cannot be extended due to poor external validity (for a discussion of the limitations of students samples see Lamb/Stem, 1980; Morgan, 1979). In our case, as our research purpose is exploratory, our sample provides valuable insights into the effects of virtual advertising. Further studies should broaden the sample to represent the whole range of viewers of sports broadcasts. Similar to most other studies, as mentioned above, our study is a laboratory study, where a certain distortion occurs due to the inherent monitoring effect (Lynch, 1982; Winer, 1999). Nevertheless, as the aim of our study was camouflaged, we can assume that the internal validity was guaranteed (Schram, 2005). Furthermore, valid conclusions can be drawn from the comparison between the effectiveness of conventional and virtual advertising, as both forms of advertising were affected in equal measure by the laboratory situation. So far little research has been done regarding the effectiveness of different kinds of virtual advertising (3D animation, animated virtual advertising, etc.). In our study we focused solely on the effects of virtual goal-side billboard advertising on consumer response, as this was the only form of virtual advertising used in the match. Thus, further research needs to be done to include different types of virtual advertising. With respect to future research, it must be stated that only a few insights have been gained so far into the effectiveness of varied locations of virtual advertising. The study of Cianfrone et al (2006) already suggested that different locations may have different advertising effects. The frequency and/or duration of exposure are certainly of importance here as well. However, this can only be presumed, as in their study in contrast to our study these variables were not explicitly investigated. As clearly shown in

RESEARCH PAPER

236

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

the present study, it is the frequency of exposure that obviously plays an eminent role in the effectiveness of virtual advertising. This depends, among other factors, on the location of virtual advertisements. We could not measure the impact of an integrated communications strategy on the effects of virtual advertising because other communications instruments were missing in our video clip. However, the highest recall values in the study of Cianfrone et al (2006) were obtained in the case where virtual advertising is part of an integrated communications strategy (e.g. it is presented in connection with television commercials). This result is not surprising, and its reasons can be found as with integrated communication and cross-media advertising in learning theory (Olivier & Kraak, 1997; Solomon & Englis, 1994; Rothschild & Gaidis, 1981). This implies that companies intending to use virtual advertising within sports broadcasts should combine it with other forms of advertising (e.g. television commercials or programme sponsoring) in order to achieve a lasting effect among viewers. Further research has to confirm this interaction effect.

individual countries within the European Union vary, thus hampering the widespread use of this type of advertising (Matthies, 2004). Harmonising these regulations might support the application of this fascinating advertising technique.
2011 International Marketing Reports

References
Babin, L. & Sheri, C. (1996) Viewers Recognition of Brands Placed within a Film. International Journal of Advertising 15 (2), 140-151. Barnes, M. (1982) Public Attitudes to Advertising. Journal of Advertising 1 (2), 119-128. Bennett, G., Ferreira, M., Tsuji, Y., Siders, R. & Cianfrone, B. (2006) Analysing the Effects of Advertising Type and Antecedents on Attitude Towards Advertising in Sport Comparison of Television Commercials and Virtual Advertising in a Sports Broadcast Setting. International Journal of Sports Marketing & Sponsorship 8 (1), 62-81. Brackett, L. K. & Carr, B. N. Jr. (2001) Cyberspace Advertising vs. Other Media: Consumer vs Mature Student Attitudes. Journal of Advertising Research 41 (5), 23-32. Brand, J. E. & Greenburg, B. S. (1994) Commercials in the Classroom: The Impact of Channel One Advertising. Journal of Advertising Research 34 (1), 18 -27. Brunel, F. F. & Nelson, M. R. (2003) Message Order Effects and Gender Differences in Advertising Persuasion. Journal of Advertising Research 43 (3), 330-341. Cianfrone, B., Bennett, G., Siders, R. & Tsuji, Y. (2006) Virtual Advertising and Brand Awareness. International Journal of Sport Management and Marketing 1 (4), 289-310. Churchill, G. A. (1979) A Paradigm for Developing Better Measures of Marketing Constructs. Journal of Marketing Research 16 (February), 64-73. Converse, J. M. & Presser, S. (1986) Survey Questions: Handcrafting the Standardised Questionnaire. Newbury Park, CA. Deimel, K. (1993) Erinnerungswirkung der Sportwerbung. Ergebnisse einer empirischen Studie (Recall and Recognition Effects of Sports Advertising. Results of an Empirical Study). Marketing ZFP 15 (1), 5-14.

RESEARCH PAPER

Conclusion
Our study provides deeper insights into consumer responses to virtual advertising in sports broadcasts. For the first time, consumer responses to virtual advertising are analysed together with a media analysis of a sports broadcast. As a result, we found a good model fit. This study provides initial evidence that virtual advertising can be an effective communications tool for advertisers, sponsors and broadcasters. The (still) overall reserved employment of virtual advertising compared to traditional forms of advertising, even in the U.S., may be particularly attributed to restrictive regulations (Turner & Cusumano, 2000; McEvilly, 1998). This also seems to be the main reason for the hitherto relatively infrequent use of virtual advertising in Europe. Regulatory frameworks for virtual advertising in the

APRIL 2011

International Journal of Sports Marketing & Sponsorship

237

Virtual advertising

Deimel, K. (1992) Wirkungen der Sportwerbung (Effectiveness of Sports Advertising), Frankfurt a. M. Doyle, A. (2000) Use of Virtual TV Ads Expected to Grow Rapidly. Retrieved from: www.infocomm.org/Newsnetwork/ index.cfm?objectID=B9CCD11B-CB08-11D4A09800D0B7913DE7&method=display&iPart=6 FIFA (1999) Regulations for the Use of Virtual Advertising. Zurich, FIFA. FORSA (1998) TV Today Das Fernsehbarometer: Die Trends im deutschen Fernsehen. Schwerpunkt: Werbung im Fernsehen. Studie im Auftrag von TV Today (TV Today The TV Barometer: Trends in German TV with focus on TV Commercials. Study Commissioned by TV Today.) Grohs, R., Wagner, U. & Vsetecka, S. (2004) Assessing the Effectiveness of Sport Sponsorships An Empirical Examination. Schmalenbach Business Review 56 (4), 119-138.

Moore, J. N., Pickett, G. M. & Grove, S. J. (1999) The Impact of a Video Screen and Rotational-Signage Systems on Satisfaction and Advertising Recognition. Journal of Services Marketing 13 (6), 453-468. Morgan, F. (1979) Students in Marketing Research: Surrogates vs. Role-players. Journal of the Academy of Marketing Science 7 (3), 255-264. Nebenzahl, I. & Hornik, J. (1985) An Experimental Study of the Effectiveness of Commercial Billboards in Televised Sports Arenas. International Journal of Advertising 4 (1), 27-36. Olivier, A. J. & Kraak, E. M. (1997) Sponsorship Effectiveness. What Is Driving Consumer Response?, Proceedings of the 210th ESOMAR Seminar, New Ways For Optimising Integrated Communications. Session III: Sponsorship, Paris, April 16-18. Petrovici, D. & Marinov, M. (2007) Determinants and Antecedents of General Attitudes Towards Advertising. A Study of Two EU Accession Countries. European Journal of Marketing 41 (3/4), 307-326. Pieters, R. G. & Bijmolt, T. H. (1997) Consumer Memory for Television Advertising: A Field Study of Duration, Serial Position and Competition Effects. Journal of Consumer Research 23 (4), 362-372. Psyma (1999), Virtuelle Bandenwerbung Day After Recall Test, Benfica Lissabon vs. Bayern Mnchen 10/08/1999 (Virtual Billboard Advertising Day After Recall Test, Benfica Lisbon vs Bavaria Munich). Studiennummer: 1068994, Nrnberg. Pyun, D. Y., Han, J. & Ha, J. H. (2004) Attitudes and Effectiveness toward/of Virtual Advertising on Major League Baseball. Paper presented at the Annual Conference for the Sport Marketing Association, November, Memphis, TN. Pyun, D. Y. & Kim, J. (2004) An Examination of Virtual Advertising Exposure on Major League Baseball: Comparing to In-Stadium Advertising Exposure by a Content Analysis. Journal of Korean Sport Research 15 (1), 683-694. Rothschild, M. L. & Gaidis, W. C. (1981) Behavioral Learning Theory: Its Relevance to Marketing and Promotions. Journal of Marketing 45 (Spring), 70-78. Russell, C. A. (2002) Investigating the Effectiveness of Product Placements in Television Shows: The Role of Modality and Plot Connection Congruence on Brand Memory and Attitude. Journal of Consumer Research 29 (12), 306-318. Sander, M. (2004) Wirkungen von Drehbanden als innovative Form der Bandenwerbung (Effects of Rotational Billboards as Innovative Form of Billboard Advertising). Marketing ZFP 26 (3), 199-213.

RESEARCH PAPER

Hackforth, J. (1989) Zwischen Bandenwerbung und Bandenwirkung Erste Ergebnisse einer Studie zur EURO 88 (Effects of Billboard Advertising First Results of a Study for EURO 88), in Sport- und Kultursponsoring, Hermanns, A. (Ed.). Mnchen, 100-111. Harshaw, C. E. & Turner, E. T. (1999) Assessing the Recognition of Perimeter Advertising Signage by Television Viewers of NASCAR Winston Cup Events. Sports Marketing Quarterly 8 (4), 35-40. Lamb, C. W. & Stem, D. E. (1980) An Evaluation of Students as Surrogates in Marketing Studies. Advances in Consumer Research, 7, Olson, J. C. (Ed). Ann Arbor, 796-799. Longman, K. A. (1997) If Not Effective Frequency, Then What? Journal of Advertising Research 37 (4), 44-50. Lynch, J. G. Jr. (1982) On the External Validity of Experiments in Consumer Research. Journal of Consumer Research 9 (December), 225-239. Matthies, A. (2004) Virtuelle Werbung Rechtliche Fragen bei der bertragung von Sportereignissen im Fernsehen (Virtual Advertising Legal Aspects for Broadcasting of Sports Events in TV). Hamburg. McEvilly, T. (1998) Virtual Advertising in Sports Venues and the Federal Lanham Act 43(a): Revolutionary Technology Creates Controversial Advertising Medium. Seton Hall Journal of Sport Law 8 (2), 603-627. Meenaghan, T. (2001) Understanding Sport Sponsorship Effects. Psychology & Marketing 18 (2), 95-122.

238

International Journal of Sports Marketing & Sponsorship

APRIL 2011

Virtual advertising

Sandler, D. M. & Shani, D. (1989) Olympic Sponsorship vs. Ambush Marketing: Who Gets The Gold? Journal of Advertising Research 29 (4), 9-14. Sasse, N. & Ludwig, S. (2002) Virtuelle Werbung im Sport Akzeptanz einer neuen Werbeform (Virtual Advertising in Sports Events Acceptance of a New Form of Advertising) in Sport und neue Mrkte: Innovation Expansion Investition, Trosien, G. & Dinkel, M. (Eds). Butzbach-Griedel, 191-200. Schram, A. (2005) Artificiality: The Tension between Internal and External Validity in Economic Experiments. Journal of Economic Methodology 12 (2), 225-237. Schumann, F. (1987) Und am Rand steigt der Bekanntheitsgrad? Die Wirkungen der Bandenwerbung bei zwei ausgesuchten Spielen der Fuball EM 1984 (And along the Pitch Increases Brand Awareness? Effectiveness of Billboard Advertising in Two Soccer Matches of the EM 1984) in Sportmedien & Mediensport. Wirkung, Nutzung, Inhalte der Sportberichterstattung, Hackforth, J. (Ed.). Berlin, 57-80. Solomon, M. R. & Englis, B. G. (1994) The Big Picture: Product Complementarity and Integrated Communications. Journal of Advertising Research 34 (1), 57-64. Taub, H. A. & Abrams, C. (1971) Effects of Target Value and Exposure Duration on Recall in a Visual Search Task. Journal of Applied Psychology 55 (4), 393-398. Tellis, G. (1997) Effective Frequency: One Exposure or Three Factors? Journal of Advertising Research 37 (4), 75-80. Turley, L. W. & Shannon, J. R. (2000) The Impact and Effectiveness of Advertisements in a Sports Arena. Journal of Services Marketing 14 (4), 323-336.

Turner, P. & Cusumano, S. (2000) Virtual Advertising: Legal Implications for Sport. Sport Management Review 3, 47-70. Wakefield, K. L., Becker-Olsen, K. & Cornwell, T. B. (2007) I Spy a Sponsor. The Effects of Sponsorship Level, Prominence, Relatedness and Cueing on Recall Accuracy. Journal of Advertising 36 (4), 61-74. Walliser, B. (2003) An International Review of Sponsorship Research: Extension and Update. International Journal of Advertising 22, 5-40. Walliser, B. (1997a) What Sponsorship Can Learn From Outdoor Advertising, Proceedings of the 210th ESOMAR Seminar, New Ways For Optimising Integrated Communications. Session III: Sponsorship, Paris, April 16-18. Walliser, B. (1997b) ber den Zusammenhang zwischen Markenbekanntheit und Wiedererkennung bei der Bandenwerbung (About the Relationship of Brand Awareness and Recognition in Billboard Advertising). Marketing ZFP 19 (1), 43-52. Wang, C., Zhang, P., Choi, R. & DEredita, M. (2002) Understanding Consumers Attitude Toward Advertising. Eighth Americas Conference on Information Systems 2002, 11431148. Winer, R. S. (1999) Experimentation in the 21st Century: The Importance of External Validity. Journal of the Academy of Marketing Science 27, 349-358. Witkowski, T. H. & Kellner, J. (1998) Convergent, Contrasting and Country-Specific Attitudes toward Television Advertising in Germany and the United States. Journal of Business Research 42 (2), 167-174.

RESEARCH PAPER

APRIL 2011

International Journal of Sports Marketing & Sponsorship

239

Copyright of International Journal of Sports Marketing & Sponsorship is the property of International Marketing Reports Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.