The Effect of Sponsoring: an Experimental Study

Citation:

Flemming Hansen and Lene Scotwin (1994) ,"The Effect of Sponsoring: an Experimental Study", in AP - Asia Pacific Advances in Consumer Research Volume 1, eds. Joseph A. Cote and Siew Meng Leong, Provo, UT : Association for Consumer Research, Pages: 279-287.

Asia Pacific Advances in Consumer Research Volume 1, 1994      Pages 279-287

THE EFFECT OF SPONSORING: AN EXPERIMENTAL STUDY

Flemming Hansen, The Copenhagen Business School

Lene Scotwin, The Copenhagen Business School

BACKGROUND

Sponsoring is a much neglected topic in the study of consumer behaviour. Marketing Abstracts published by the British Market Research Society covering practically all relevant publications, have for the last five years only 6 papers under the heading of "Sponsoring." Of these the majority are papers in business and similar journals and only two appear in academic journals reporting findings from studies of sponsoring.

Sponsoring is among others defined by Meenaghan (1983) as: "Sponsorship can be regarded as the provision of assistance, either financial or in kind, to an activity by a commercial organization for the purpose of achieving commercial objectives." This definition, however, does not explicitly explain the two-sided nature of sponsoring: On the one hand, the sponsors' support of the person or organization being sponsored and, on the other hand, communication relating to this. Thus sponsoring is more than a communication activity it is also a financial act, involving payments to the sponsored event, person, organisation, etc. (In the following "the sponsored.")

This two-sided nature of sponsoring is important. (Scotwin 1993). Costwise, as a market strategy, it implies that both expenses in connection with the direct support of the sponsored one, as well as expenses related to communication in relation to the sponsorship, must be included. It also implies that sponsoring is not only a special kind of advertising, in a sense it is also a part of the total offer the company brings to its market.

Sponsoring has primarily been used in connection with sports-stars, sporting-events and teams, but it is becoming more and more widespread in connection with cultural and other events. The present paper is primarily concerned with sports sponsoring. Interest in sport sponsoring-as measured by revenue and by the prices obtained by sponsored ones-grew during the 80's, but has since levelled off. The growing availability of sponsoring possibilities in other areas than sport may be part of the reason for this.

Sponsoring is widely used. However, because of the problems connected with defining exactly what is involved, available statistics are of a mixed quality. In Denmark and in a few other countries, where statistics have been attempted, it seems that sponsoring amounts to something like 5% of the total advertising expenditures when both contractual and communication costs are included. (Christensen et al. 1991). There are, however, dramatic differences between the use of sponsoring in different sectors and between countries. On the whole, it is more frequently used in connection with consumer goods than with industrial goods. It is, however, widely used in connection with many service products and even the public sector, such as the postal services and Telcom use sponsoring. Seen in relation to advertising expenditures for some companies it is a major expenditure, whereas for others it is a supplementing activity. In a commercial context the communicative aspects of the sponsoring arrangement are the most important ones. The sponsoring of a sports-star resulting in him showing the sponsor's name on his clothes or equipment, is an important aspect of the appearance of the sports star in the media for the sponsor. Similarly with the sponsoring of teams and events the appearance of the name of the sponsor in the media is important for the sponsor. Additionally, however, the sponsor engage in related communication activities, making sure that the sponsoring arrangement is fully utilized. Billboards and signs on sport stadiums, press releases, advertising telling about the sponsorship, the invitation of important customers to sponsored events, are all examples of this.

In evaluating the effects of sponsoring, it is important to realize the complex nature of the sponsoring activities. Each single activity may be looked upon separately or the effects of a total sponsorship programme may be in focus. The kind of measurements that can be used, differ, and the problems involved in isolating the effects from those of other marketing activities differ, and in all cases they are complex. The lack of published research in the area may partly be ascribable to these complications.

Sponsors' motives may be many. Obviously the use of sponsoring as an element in the market strategy, integrating a story and means of communicating it, is important. In evaluating alternative sponsoring arrangements, the potential number of exposures of different kinds is a dominant factor. These commercial considerations, are, however, not always the only ones. At the other extreme the sponsor may provide his contribution solely for idealistic reasons. In Denmark the Carlsberg breweries' sponsoring of the Danish National Historical Museum for more than 100 years, can be seen as a point in case. Then again maybe the founder of the breweries in the 19th century may have had his commercial motives also. He may have found it useful to have the name of the brewery associated with cultural and national elements.

Also, personal interest on behalf of the company manager or marketing manager may play a role. Many large European soccer teams are sponsored by individual businessmen and even though the expenses may be tax deductable, it is likely that the means applied could have been used more profitably in other ways. Also sponsoring may be used as a way of extending the personal relationships between management, senior employees, and important people. Inviting guests-and not necessarily solely clients-to sponsored events, is a common phenomenon. In some situations sponsoring may be motivated by the fact that it is the only way in which the company can get access to particular media. In countries with limited access to commercial television, sponsoring may be the only way in which they can get their brand name on the TV-screen. Similarly, companies with limited access to some media-such as tobacco manufacturers-may use sponsoring to reach their audiences.

Finally, in some instances the purpose of the sponsoring may be direct sales. For instance, if a beverage supplier can obtain the sole rights for selling beverages, at a particular event as part of the renumeration for his sponsoring, this may be an important factor in his considerations.

Communication in connection with sponsoring may take different forms. Content-wise it may be limited to the brand name, it may include the brand name and a brief message in the form of a slogan or the like, or it may be broader and, for instance, aimed at modifying the image of the company by communicating relevant image information in connection with the sponsoring messages.

In most instances, however, there are narrow limitations to the complexity of the messages that can be brought. Legal and technical factors, such as limits to the size of the space that can be used exist, and the length of the exposure is often so short that only very brief messages can be perceived.

FIGURE 1

NATIONAL TELEVISION EXPOSURE FOR SPONSORS OF THE US AUTOMOBILE RACES (NASCAR WINSTON CUP SERIES), 1986

EFFECTS AND THEIR MEASUREMENT

The association between the sponsored ones and the products being sponsored, is important for the kind of effects that can be obtained. When Ford motors sponsors Formula 1 races, it is obviously the direct association between what is being sponsored and the products of the manufacturer, which is of importance. When a telephone company sponsors a soccer match, the relation between the company and what is being sponsored, is not close. Effects of the sponsoring may be expected to be very different, depending upon how close an association exists between what is being sponsored and the products of the sponsor. In this connection the use made of sponsoring by companies (such as alcoholic beverages and tobacco manufacturers) with limited access, especially to TV, is also important.

Effects may be discussed at different levels in a general effect hierarchy: exposure, attentive, cognitive and behavourial levels.

Syndicated services are available, (Boedker et al. 1990) providing measures of the extent of the exposure generated by sponsoring. Particularly in connection with television coverage of major events, these estimates are important. Here it is registered how many times a particular facing of a billboard or of a brand name on a sportsman is completely or partially observable on the screen. The extent of each exposure is measured in terms of seconds and the number of viewers of the channel broadcasting is multiplied with the exposure, so that a total measure of exposure in terms of 1000 viewers in 30 seconds results. Such data are, however, rarely published and never in a structured manner. Most frequently they appear in feature articles in newspapers and business magazines. Here the general observation made, is that the extent of the exposure following from sponsoring of major events, such as the Olympic Games, The Soccer World Cup, etc. is extremely large and results in a lower cost per 1000 exposures of 30 seconds, than could be obtained by use of traditional commercials. The comparison of course being difficult the message conveyed through sponsoring is of a much simpler nature than that which can be transmitted through advertising.

Obviously sponsoring as well as advertising can generate attention in terms of brand or company awareness, identifiable in terms of recall and recognition and changes in the same. These effects may also extend to recall and recognition of the sponsoring itself and slogans and themes associated with them. Many studies of recall and recognition of advertising may act as a source of inspiration for studies of these effects of sponsoring, (e.g. Adams et al. 1993, Gardner 1983, Krugman 1986).

Cognitive effects may occur as associations between sponsored ones and sponsor. As suggested earlier, some such associations are very close, such as sponsoring done by manufacturers of sports equipment (e.g. tennis rackets) of actors (e.g. tennis players) or more general, such a particular soccer player's association with a brand name where the simple positive attitudes among soccer fans are expected to influence the evaluation of the products of the sponsor. Findings, however, do not exist.

A more general image effect may also be expected following from elements of the message being communicated in connection with the sponsoring. The limited extent of sponsor messages, however, puts a limit on how complicated such messages can be.

Occasionally, direct sales effects can also be observed. Among companies having used sponsoring, many stories circulate of the unique effects obtained in special situations. For instance an almost unknown household product manufacturer sponsoring an extreme outsider in an international boat race, where the sponsored one turns out to be the winner, may dramatically influence the sale of the product. Even though there are many such examples, no attempts have, however, been made to systemize these experiences and in most cases the effects of sponsoring, are difficult to identify, because so many other factors, with or without the control of the marketer, influence sales.

In the following we shall briefly introduce some of the published practical and theoretical findings on the effects of sponsoring.

About exposure effects, considerable evidence is available as proprietary information among major sponsors. Data from the syndicated services mentioned before have, however, never been made available in a systematic fashion and their potential as an independent variable has not been utilized in published works. An example has, however, been made available by (Ross 1988) who shows the effects of major sponsors of US motive automotive activity in 1986. Here exposure time and an estimate of its value is shown (Figure 1).

Similarly, recall and recognition measures are known by most major sponsors. Examples of such measures have been published. (Crone et al. (1999), Wehenzahl et al. (1985), Otker (1986), Parker (1989), Pavelchak (1988), Reth (1986), Ryssel et al. (1988), Sandle (1989), Sleigh (1990), Vinding (1999), Wright (1988)). An example of such a study is shown in figure 2. It appears that, whereas unaided recall is limited apart from M¦LK (milk: the main sponsor) aided recall reaches significant -and highly variedClevels.

FIGURE 2

RECALL OF ADVERTISING SIGNS AT A TELECASTED SOCCER MATCH--DENMARK 1989

On the whole, available findings show significant effects in terms of brand as well as sponsor recall and recognition. Systematic attempts to relate these effects to parameters of importance are non-existent, apart from studies relating the effects to the extent of the exposure.

Therefore, in spite of the existence of more detailed findings among sponsoring companies, the only conclusion to be drawn from published studies is that recall and recognition and therefore awareness to a significant degree is influenced by sponsoring.

Attempts to quantify image effects and association generated by sponsoring have been made also. Published (and unpublished) work suggest that it is very difficult to quantify such effects. Whereas recall and recognition effects are common, few and, rarely important, image changes have been proven. Thus, Ocker et al (1986) studying effects of the sponsoring done by Philips in connection with the world soccer games, shows dramatic awareness changes, but only few and minute image effects.

Also sales effects have been observed and, as suggested earlier, many cases have shown remarkable results. However, they only prove that effects occur-then say very little about he conditions that facilities such effects.

THEORETICAL BACKGROUND

In the consumer behaviour literature no specific theoretical formulations exist aimed atstudying effects of sponsoring. However, a number of theoretical constructs may help to understand how sponsoring works. Some of these will be introduced briefly here.

Researchers relying upon effect hierarchy models (Ray 1973) suggest that the effect of commercial messages can be seen as a hierarchy ranging from exposure over attention, awareness, attitude (image) change to behaviourial effects. As suggested in the preceding section the little published evidence available relies on measures relations to different levels in such a hierarchy. On the whole it is expected that thinking in lines of an effect hierarchy model is useful in an attempt to understand the effects of sponsoring.

However, some caution must be introduced. Generally sponsor messages are simple, often limited to only brand or company names, and rarely touching upon more than one aspect of the product. For that reason, one would expect that mere exposure effects could result. (Zajonz (1968) Hansen (1984) and Obermiller (1988)). Here the increased familiarity with the brand name generated through sponsoring in itself should generate preferences for the brand.

Also theories relating to associative learning give useful suggestions as to the way in which sponsoring works. Whatever associations the sponsored one generates with the consumer, the establishment of a link between the sponsored one may result in some of the same associations being connected with the sponsor. This may apply in a positive as well as in a negative sense. If a particular actor is very positively evaluated and admired this may generate a positive attitude being associated with sponsor. Particularly such effects are likely when the connection between what is being sponsored and sponsor is a natural one. The tennis pro, who sponsors a particular brand of tennis racket, may transfer some of the professionalism, that the viewers associate him to the racket. On the other hand negative associations may be generated also. The consumer regarding a sport like boxing as inhuman, sweaty and unpleasant, may generate negative associations with the manufacturer of sweat shirts sponsoring such events.

Insight into how sponsoring works may also be gained, departing in theoretical formulations regarding aroused sets, evoked sets, choice sets, etc. (Howard and Sheth (1969), Urban and Houser (1979), etc.) The limit extent of the sponsoring message may restrict the effect primarily to building of brand awareness, without attitudinal related effects. This may, however, be important to the marketer to the extent that this awareness results in his brand occurring more frequently in consumers' evoked sets in buying situations.

Also theories relating to involvement (Celsi (1988) Krugman (1965) Lastovica et al. (1978) etc.) may provide a useful point of departure for theoretical speculations regarding the effects of sponsoring. Here we may be concerned with involvement with what is being sponsored and/or involvement with sponsor and his product. To the extent that the consumer is involved with a particular sport, sponsoring this may increase the effectiveness compared with the sponsoring of other events. There may, however, be a limit to this. In a particular situation with a person watching a game, the game in itself may become so exciting that all attention is distracted from anything but the the game and very little mental capacity is available for processing information concerning billboards, signs on equipment and the like. Similarly, the more involved the consumer is with a particular product or company, the more effective the sponsoring by this company may be.

In recent years several researchers (e.g. Appel (1987), Bates et al. (1990) , Cohen et al. (19701, Goldberg (1987)) have concerned themselves with the extent to which the environment in which a message is being brought, influences the reception of the message. Particularly moods have been studied and, in several instances, a relationship between positive moods generated by the programme environment and the effect of the advertising has been established. Since sponsoring by its nature tends to occur in positively evaluated environments, such mood effects may play a role in the explanation of the effectiveness of sponsoring.

Similarly traditional work on source effects may be relevant. In the classical work on this, the trustworthiness and expertise (Hovlund et al. (1968) as well as the perceived intention of the source is critical. With sponsoring, expertise may occur to the extent that the sponsor is involved with products directly related to the sponsored activity such as, car manufacturers sponsoring car races, manufacturers of sport shoes, sponsoring athleltics, etc. Additionally trustworthiness ascribed to the sponsored may account for part of the effectiveness of the messages. Finally, the appearance of the sponsor's message on billboards etc., may act as overheard conversation where no intentions are ascribed and cognitive defences are low.

These different theoretical formulations-and possibly others-do not necessarily rule out each other, but, rather complement each other, in attempts to generate understanding in the way in which sponsoring works. In the following, a study is described where hypotheses are tested which are derived from primarily these theoretical foundations.

BACKGROUND FOR THE PROJECT

The study to be presented here has its background in a major research project concerning sponsoring and its effect, conducted at the Copenhagen Business School: Marketing Department.

Early results of this project have been published in Orla Nielsen, (1990). Following these publications, a continued project, partly sponsored by 15 companies involved in sports sponsoring and partly by the Danish Research Academy, was started. The overall project included a study of the problems and experiences to be identified among the sponsoring companies and other companies involved in sports sponsoring.

Reports from this and other earlier phases of the project can be found in Scotwin (1993).

In parallel with this, a study of the-limited-available literature was completed. Also included was a qualitative study of consumers' view of sponsoring using focus group interviews and observation of TV viewing audiences.

The last part of the project is the experimental study to be reported in the following.

"EXPERIMENTAL HYPOTHESIS"

Basically it is expected that sponsoring, like advertising, generates effects at all levels of the effect hierarchy. More specifically one should expect:

1a. Recall and recognition of messages of a magnitude similar to that which advertising exposure of similar duration and intensity generates.

1b. Image changes like those generated by advertising, but less expressed than effects measured in terms of recognition and recall.

2. Different kinds of sponsoring are to have different effects. Varying with the intensity of the presentation and the clearness of the association between sponsor and sponsored, it is expected that effects of sponsoring vary.

3. Apart from effects following from variations in target group sizeCthat sponsoring is the more effective the more involved the viewer is in what is being sponsored.

4. The extent to which consumers are familiar with different brands and company names, facilitates their perception and remembrance of exposure to these through sponsoring. That is, sponsoring for more known brands and companies is better recalled and recognized than sponsor messages for less known sponsors. However, there is an upper limit to this relationship. Companies may be so well known that the additional brand and/or company awareness, generated by sponsoring, is limited. For that reason it is expected that, relatively speaking, less known companies gain more in terms of recall and recognition than more known companies.

5. Based on studies of source effect that a delayed-so-called sleeper effect-will occur as a result of sponsorship information.

6. Effects to increase with the number of sponsor messages (for the same sponsor) to which the respondent is exposed.

EXPERIMENTAL DESIGN

Sponsor information may occur in many different media, it may take different forms and it may be received in many different situations. To study the effect of sponsoring experimentally, one has to limit oneself along all of these dimensions. In the present study it has been decided to concentrate on messages broadcast on TV and the sponsoring messages used are given three different forms, e.g.:

Sponsoring of programmes

Sponsoring in games shows

Sponsoring of individual actors

FIGURE 3

EXPERIMENTAL DESIGN

Based upon experience with advertising research, a forced exposure design is used (Ackerbaum et al. (1967) and Grun et al. (1983)). To compare with the effects measured following advertising messages, advertising for the products is included in the design of the experiment. The experiment was conducted in Copenhagen in the winter of 1991 using 221 marketing students as subjects. The students were randomly assigned to four different experimental groups, each of which followed the same general design (with the experimental variations being described in the following).

1.  Pre-measures.

2.  Forced exposure.

3.  Post-measures.

4.  Delayed post-measures.

The respondents were instructed that they were participating in a study concerned with commercial TV. This was a natural suggestion in Denmark at the time, when commercial TV had recently been introduced. A few respondents indicated, when being interviewed following the session, that they were aware that the project was really concerned with measuring the effects of sponsoring. Those respondents were deleted from further tabulations.

The experimental variations were introduced in an approx. 30 min. TV programme designed for the study.

For the experiment five companies were selected.

* Unibank-a recently formed coalition of major national banks

* Toms-a chocolate bar manufacturer with "Toms" as its major brand name.

* Texaco-a distributor of oil products

* Rsdvad-a local manufacturer of stainless steel household products, normally not involved in national advertising and sponsoring activities at the same level as the three others

* Master Foods-represented with Mars Bars

Television was chosen, since in Denmark this has been the primary medium of concern for sponsors and with commercial television recently introduced, it also offered the possibility of comparing the use of advertising with sponsoring.

The television programme consisted of four elements.

1. A brief news section.

2. An advertising block.

3. An entertainment programme, which in two cases were games shows with gifts to be won, otherwise a plain talk show.

4. A sports programme.

For each of the four companies, different ways of communicating were studied. A traditional 30 seconds commercial was used. In the talk show sponsors were introduced as programme sponsors, whereas in the games show they were introduced as gift givers. Finally, Unibank appeared as the major sports sponsor in the last session in which also the Mars brand name appeared. The total design is shown in figure 3.

With this design, different combinations of exposure are introduced:

1. One exposure: In terms of a commercial (Texaco and Rsdvad in group 2 and Toms in group 3).

2. One exposure as an indirect sponsor (gift sponsor): (Toms in group 2, Rsdvad and Texaco in group 3 and Mars in all four groups.

3. One exposure as a direct sponsor (programme sponsor). (Unibank in group 1, Texaco and Rsdvad in group 4).

4. Two exposures, one in the advertising block and one as indirect sponsor (Toms in group 1, Texaco and Rsdvad in group 4).

5. Two exposures, one in the advertising block and one as direct sponsor (Unibank in group 3).

6. Two exposures, one as direct and one as indirect sponsor (Unibank in group 2).

7. Three exposures, one in the advertising block, one as direct and one as indirect sponsor (Unibank in group 1). The respondents received from 1-3 exposures with the nature and combination of exposures varied.

TABLE 1

AIDED AND UNAIDED RECALL OF TOMS, TEXACO, RADVAD AND UNIVANK SPONTS AND SPONSORSHIP IN PERCENTAGE FOR EACH OF THE SEVEN EXPOSURE COMBINATIONS

Measurements were obtained on self-administered questionnaires. Subjects indicating that they were aware of the special concern with sponsoring were deleted.

The pre-measurement, in addition to demographic factors, such as sex, age, family etc., included questions as to what sporting events the respondents watched on TV, which sports they were actually involved with and what kind of TV programmes they were watching. Also included were measurements of their attitudes to the new commercial (TV2) channel, to sponsored programmes and their awareness of 19 different companies, out of which 5 were the ones included in the experiment. The measurements were obtained on 4-point interval scales.

Finally, respondents rated 19 companies on 5 image statements: (1)" a company having products of high quality" (2) "a company concerned about the environment" (3) "a company having well educated and good employees," (4) "a company doing a lot for sport and culture" and finally (5) "a company with good management." All of these were measured on 5-point scales.

The immediate post-measurement included the same attitude measurements and in addition to this aided and unaided recall of both TV programmes, sponsoring messages and advertisements were included.

At the delayed post-measure, the measurements obtained immediately after the session were repeated. This measurement took place two weeks after the first measurement.

FINDINGS

Major findings regarding aided and unaided recall of sports and sponsorhsips are shown in table 1. Several observations relevant to the hypothesis formulated earlier can be made.

In the present kind of forced exposure experiment unaided recall is very high and the recall obtained by providing aid is consequently not dramatically larger. Also the pattern in the findings is the same regardless of whether one is looking at aided or unaided recall. On the whole, the various sponsor treatments taken alone and in combination with advertising, result in high aided and unaided recall measures of the same magnitude. Thus the sponsor measurements create message recall of a magnitude similar to that of advertising-(hypotheses 1a).

From the findings (last column) it can also be seen that there are no systematic variations in the effect following from the number of exposures (hypothesis 5). However, it must be remembered that recall measures-regardless of the effectiveness of the messages-can be expected to differ in magnitude because of the different awareness of brand and company names. If direct sponsoring alone is disregarded (this treatment is used only by Unibank which is very well known and in the period of the experiment had heavy exposure in the media following the merger of the different banks going into the coalition named Unibank), two or more exposures result in higher recall than a single exposure in all other instances.

In the way in which sponsoring is communicated it also seems that direct sponsoring taken alone or in combination with advertising, is more efficient than indirect sponsorship (hypothesis 2).

In the soccer game appearing on the TV programme, Mars was sponsoring one of the teams and their brand name appeared on the shirts of the players. Unibank was sponsoring the entire sports programme. Here one would expect the more involved (interested) viewers to be more attentive than the less interested ones. Table 2 shows how unaided brand recall varies between respondents being more or less interested in football.

It appears that for Mars, more interested (involved) viewers have significantly higher recall than those not involved. That the same is not the case for Unibank may be ascribable to the overall high level of recall for this company.

More evidence in favour of the involvement hypotheses appears in table 3. For Mars, recall varies, depending on the way in which people evaluate the programme. Those doing so more positively have significantly higher recall than those evaluating the programme neutrally or negatively.

TABLE 2

UNAIDED RECALL (% OF RESPONDENTS RECALLING) OF SPONSOR OF PROGRAMME (UNIBANK) AND OF SPONSOR WITH BRAND NAME (MARS) APPEARING ON PLAYERS' SHIRTS

TABLE 3

PERCENTAGE OF RESPONDENTS UNAIDED RECALLING THE MARS BRAND NAME APPEARING ON THE ACTOR IN THE SPORTS PROGRAMME DEPENDING ON THEIR EVALUATION OF THE PROGRAMME SHOWN

TABLE 4

INITIAL COMPANY AWARENESS AND AVERAGE UNAIDED RECALL FOLLOWING EXPERIMENTAL TREATMENTS

In table 4 the relationship between the initial brand/company knowledge and unaided recall is shown. It can be seen how increased brand/company awareness results in higher recall (Hypothesis 4).

Comparison between pre- and post-measures of brand and company awareness show only significant differences for Rsdvad, the initially least known company among those in the experiment. The fact that the other companies initially are known by most of the respondents may account for this.

Aggregated images of the four sponsors as measured along the five dimensions are shown in table 5. These are computed for the pre-, the first and the second post measures separately. Here there are no significant changes for either of the four products in any of the conditions. If sponsoring and sponsoring in combination with advertising is to result in marked attitude changes at least more than the few experimental exposures involved here are required. Consequently the possible existence of a delayed (sleeper) effect cannot be expected either.

When, however, attention is directed towards each of the individual five image dimensions, a few significant changes occur in the answers to the statement relating to the company's involvement in the particular activity being sponsored. In table 6 this is shown with significant differences underlined. (p<0.01).

The preceding data analysis has been relatively crude and an attempt to identify significant effects in addition to those already reported, was made with the use of logistic regression analysis based upon the data at the individual level: i.e. based on individual changes in brand/company awareness and attitudes. These analyses confirm the findings already presented, but they do not give further insight into the relationship between the independent variables and the measurement of the effect of the sponsoring.

The findings can be summarized as follows:

All the tested sponsoring messages generate attention at levels suggesting that sponsoring meaningfully can be applied as marketing communication. The experiments also show that sponsoring may be a useful tool in segmented marketing: general interest in the area sponsored as well as positive evaluation of the sponsored event generate increased awareness of the sponsoring. High initial brand/company awareness generates higher sponsoring awareness, but relatively speaking sponsoring seems more efficient for lesser known companies/brands. A second major conclusion is that, whereas marked effects in terms of recall can be measured, it seems that attitude changes are less expressed. This seen on the background of the very simple nature of most sponsoring messages suggests that sponsoring is more useful in situations where there is a need for generating increased awareness, than in situations where specific information has to be provided or where the attempt is to modify attitudes relating to the product.

TABLE 5

AVERAGE IMAGE EVALUATION FOR THE FOUR SPONSORS IN THE SEVEN DIFFERENT EXPOSURE COMBINATIONS

TABLE 6

AVERAGE IMAGE EVALUATION ON FIVE INDIVIDUAL IMAGE DIMENSIONS BEFORE AND AFTER EXPOSURE

FURTHER ISSUES

The experiment suggests a number of problems, it might be useful to look into in the future. First it should be possible to study directly the extent to which the variations in sponsoring recall result in changes in brand/company awareness and attitudes. Also the fact that attitudinal changes could be observed when dimensions were measured relating to the sponsoring, suggests that the development of special scales for the identification of attitudinal effects of sponsoring might be a fruitful endeavour.

Similarly, sleeper effects at the brand awareness or company awareness level could also be looked for (although they did not appear in the present study). However, the complex nature of sponsoring in terms of the many different ways in which messages can be brought across, suggests that many aspects of sponsoring and its effects need qualitative studies and more intensive interview techniques than those that can be used in experimental studies of the kind attempted here. Possibly such studies in parallel with further experimental work may improve our insight into the way in which sponsoring influences consumer behaviour.

LITTERATURLISTE

Aschenbaum, A.A., Hailey, R.L. & Gatty, R.; On-Air vs. In-Home Testing of TV Commercials, p.15-19, in Journal of Advertising Research, vol. 7, number 4, December, 1967

Adams, A., Mehrota, S. & Van Auken, S.; Reliability of Forced-Exposure Television Copytesting, p. 29-32, in Journal of Advertising Research, vol. 23, number 3, June/July 1983.

Appel, Valentine; Editorial Environment and Advertising Effectiveness, p.11-16, in Journal of Advertising Research, August/September, 1987.

Axelrod, Joel, N.; Induced Moods and Attitudes toward Products, p. 19-24, in Journal of Advertising Research, 1963, 3, June.

Batra R. & Stayman D.M.; The role of Mood in Advertising Effectiveness, p. 203-215, in Journal of Consumer Research, vol. 17, number 2, September 1990.

Burke, M.C. & Edell, J.A.; Ad Reactions Over Time: Capturing Changes in the Real World, p.114-118, in Journal of Consumer Research, vol. 13, June, 1986.

B°dker, M. & Larsen, S.V.; Metoder til effektmsling og deres anvendelse ved sportssponsering, og Msling af sportssponeringens effekt-fire eksisterende metoder, kap. 11 og 12, i Nielsen, Orla (red.); Sponsering, K°benhavn 1990.

B°dker, M., Grove, M., Kristensen, J. & Larsen, S.V.; Sportssponseringens mslsµtninger og effekthierarkimodellen, kap. 8, i Nielsen, Orla (red.); Sponsering, K°benhavn 1990.

Celsi, R.L. & Olson, J.C.; The Role of Involvement in Attention and Comprehension Processes, p.210-224, in Journal of Consumer Research, vol. 15, September, 1988.

Christensen, C.F. & Kjµr-Hansen, U.; De danske reklameomkostninger 1988, K°benhavn 1991.

Cohen J.B. & Areni, C.S.; Affect and Consumer Behavior, p.188-241, in Robertson, T.S. & Kassarjian, H.H.; Handbook of Consumer Behavior, New Jersey, 1991.

Gardner, Meryl, P.; Advertising Effects on Attributes Recalled and Criteria Used for Brand Evaluation, p.310-318, Journal of Consumer Research, vol. 10, December 1983.

Goldberg, M.E. & Gorn, G.J.; Happy and Sad TV Programs: How They Affect Reactions to Commercials, p.387-403, in Journal of Consumer Research, vol. 14, December, 1987.

Green, P.E. & Schaffer, C.M.; Ad Copy Testing, p.73-80, in Journal of Advertising Research, vol. 23, number 5, October/November, 1983.

Gross, A.C. & Traylor, M.B. and Shuman, P.J.; Corporate Sponsorship of Art and Sports Events in North America, p.9-13, in Esomar Congress 1987.

Grove, M. & Kristensen, J.; Metoder til msling af eksponering og deres anvendelse ved sportssponering, og Metoder til msling af opmµrksomhed og deres anvendelse ved sportssponsering, kap. 9 og 10, i Nielsen, Orla (red.); Sponsering, K°benhavn 1990.

Hansen, Flemming; Towards an Alternative Theory of the Advertising Communication Process, p.69-80, in International Journal of Research in Marketing, 1984, 1.

Krugman, Herbert, E; Low Recall and High Recognition of Advertising, p.79-86, in Journal of Advertising Research, 26, 1986.

Krugman, Herbert, E.; The Measurement of Advertising Involvement, p. 583-596, in Public Opinion Quarterly, vol. 30, 1967.

Krugman, Herbert, E.; The Impact of Television Advertising: Learning without Involvement, p.349-356, Public Opinion Quarterly, vol. 29, 1965.

Krugman, Herbert, E.; Memory without Recall, Exposure without Perception, p.7-12, in Journal of Advertising Research, vol. 17, number 4, August, 1977.

Lastovica, J.L. & Gardner, D.M.; Low Involvement Versus High Involvement Cognitive Structures, p.87-92, in Advances in Consumer Research, vol.5, 1978.

Marketing Abstracts 1989-93, British Market Research Society.

Meenaghan John, A.; Commercial Sponsorship, Bradford 1983.

Nebenzahl, I. & Hornik, H.; An Experimental Study of the Effectiveness of Commercial Billboards in Televised Sports Arenas, p.27-36, in International Journal of Advertising, 1985, 4.

Nielsen, Orla (red); Sponsering, K°benhavn 1990.

Obermiller, Carl; Varieties of Mere Exposure: The Effects of Processing Style and Repetition on Effective Response, p.17-30, in The Journal of Consumer Research, vol. 12, June, 1985.

Otker, T. & Hayes P.; Judging the Efficiency of Sponsorship: Experience from the 1986 World Soccer Cup p.562-591, in The 40th ESOMAR Marketing Research Congress, Montreux 1987.

Parker, Ken; Measuring the Impact of Sponsorship, p.44-48, in Admap, May, 1989.

Parker, Ken; Measuring Effectiveness: Is there more to Sponsorship than a Warm Glow, p.28-30, in Admap, November, 1990.

Pavelchak, M.A., Antil, J.H. & Munch, J.M.; The Super Bowl: An Investigation into the Relationship Among Program Context, Emotional Experience, and Ad Recall, p.360-367, in Journal of Consumer Research, vol. 15, December, 1988.

Ray, Michael, L.; Marketing Communication and The Hierarchy-of-Effects, p.147-176, in Clarke, P. (red.); New Models for Mass Communication Research, Beverly Hills, 1973.

Roth, Peter: Sportswerbung-Grundlagen, Strategien, Fallbeispiele, Verlag Moderner Industrie, Vesttyskland, 1986.

Ryssel, C. & Stammingener, E.; Sponsoring World-Class Tennis Players, p.110-116, in European Research, May, 1988.

Sandler, D.M. & Shani, D.; Olympic Sponsorship vs. "Ambush" Marketing: Who gets the Gold, p.9-14, in Journal of Advertising Research, August/September, 1989.

Scotwin, Lene; Sportssponsering-Institut for Afsµtnings°konomi, Copenhagen, feb.1993.

Simkins, Jean; Sponsorship-EUI Special Report No. 86, London 1980.

Sleight, Steve; Sponsorship-what it is and how to use it; Cambridge 1989.

S°ndergaard, J. & Vedel, H.; Definition af sponserig, Sponsering som handlingsparameter, og Skal virksomheden sponsere? Kap. 2, 3 og 4, i Nielsen, Orla (red.); Sponsering, K°benhavn 1990.

Turk, Peter, B.; Effective Frequency Report: Its Use and Evaluation by Major Agency Media Department Executives, in Journal of Advertising Research, 28, 1988-89.

Vinding, Niels; Hvad fsr en sponsor for sine penge, s. 10-12, i Markedsf°ring, number 16, 1990.

Wright, Roger; Measuring awareness of British football sponsorship, p. 104-108, in European Research, May, 1988.

Zajonc, Robert, B.; Attitudinal Effects of Mere Exposure, p. 1-27, in Journal of Personality and Social Psychology, Monograph Supplement, vol. 9, number 2, part 2, june, 1968.

----------------------------------------