Use of the World Wide Web for sport marketing purposes has developed at the same astonishing rate as the Web itself

Use of the World Wide Web for sport marketing purposes has developed at the same astonishing rate as the Web itself. As of June 1997, a search on Yahoo delineated 19,717 sites under the "sports" category. Moreover, these thousands of sports sites translate into millions of sports fans regularly accessing the web for sports information and entertainment Indeed, at one site alone-- IBM reported up to 20 million visitors to their Australian (Tennis) Open site (Callaghan, 1997). Thus the appeal the web is obvious, yet the Web is characterised generally by an inadequate knowledge of the users of sports sites (McDonald, 1997) and, in terms of sport marketing, a lack of objectives and understanding of the medium (Forrest, Pope and Murphy, 1996).

In this paper, we present the results of an exploratory study of one specific sport web site, and discuss the uses of the site from the user’s perspective. We performed this study by using pre-existing market research scales of 1) involvement, in order to identify fans and non-fans; and 2) an adaptation of Schlinger’s Viewer Response Profile to identify site characteristics that would discriminate between these groups. An implicit assumption of this work is that the principal user of a site, and the person for whom it is primarily intended, is the fan.

WEB BACKGROUND

While fifteen percent of users in the United States are reported to have made purchases via the Web (Callaghan, 1997), only 0.5 percent of the publicly available sites are intended for transactional commerce (McDonald, 1997). This has implications for the approach taken by corporations who use the Web on a commercial basis, and it would appear that their intention is primarily to communicate rather than to sell. This is the reported case with sporting team Web sites also, where less than a quarter include a selling function (Forrest, Pope and Murphy, 1996). The same research indicated that most sporting team sites were intended to fulfill a marketing communication role.

Obstacles to using the Web for Marketing Communication

Several obstacles have been identified to using the Web for marketing communication purposes (McDonald, 1997). These include: (1) a lack of intrusiveness, which limits the impact of the message; (2)Advertising form and function limitations, which require the target to actively click on the advertisement itself (i.e. banner) to find more information; radical fragmentation and diversity of site offerings (over 700,000 active sites); and (3) inadequate measurement devices. It is in regard to this last aspect that this research is focused.

Inadequate Measurement Devices

At the outset, it should be made clear that the term "inadequate" in no way refers to an insufficient number of measures available to measure the various and sundry characteristics of the Web-user. Indeed, for such a new marketing communication channel the Web has already developed a bewildering array of methods, measurements and monikers for web users (Murphy and Forrest, 1996). More than a dozen companies, each employing their own methods, constantly assess web-traffic patterns in terms of-- Hits, visitors, clicks, requests, transfers, users, identified users, visits, sessions, ad views, page views, clicks and click-throughs. The driving force behind the predominant means and measures of web use is, of course, advertising. And advertisers are first and foremost interested in quantitative counts of traffic volume. For , in the end, " all advertisers really want to know is how many people visit a Web site in a given period of time and how many of those visitors see a banner and then click on it to view the advertiser's Web page." (Murphy and Forrest, 1996)

Ergo, the means of assessing the use of web sites for marketing communication purposes has concentrated on the users activities once arriving at the site, such as time spent or pages accessed ( Dr‚ ze and Zufreyden, 1997), rather than the reason an individual would choose to access a site and what they were seeking there. And it is these considerations that are most relevant to web site design and content selection. Whereas site traffic counts are important to the media planner, it is the site users motivation and satisfaction that is most relevant to the site’s webmaster and marketing manager.

Toward the Development of Web Measurement Standards

Novak and Hoffman's (1996) New Metrics for New Media define "three distinct levels of analysis for Web advertising measurement: (1) vehicle level, (2) page level, and (3) ad level.

For each level of analysis, there are exposure metrics and interactivity metrics. Exposure metrics are based upon the one-to-many communication model underlying traditional media, and indicate that a visitor has been exposed to (i.e. has had the opportunity to view) a web site, a web page, or an advertisement....Exposure

metrics can be behavioral (i.e. reach and frequency) or cognitive/attitudinal (i.e. recall and recognition) Interactivity metrics are based upon the many-to-many communication model underlying the Web, and indicate the extent to which the visitor actively engages with the Web content or advertisement.. Similarly, interactivity metrics can be behavioral (i.e. duration time) or cognitive/attitudinal (i.e. flow) …. In addition to the behavioral and psychological measures considered above, outcome metrics --(which could be defined in terms of the direct response paradigm (i.e. direct orders, lead and/or traffic generation) must be developed. ( Novak and Hoffman,1996). Pgs21-22,27

Schlinger’s Viewer Response Profile

"In effect, to extract from an attitudinal measure any valid information to be used for decision making, the measure must be selected to fit the product/service, the problem, the objectives/goals of the advertising… there is no single, "magic" measurement that will apply across the board" (McCollum, Spielman, 1979)

There are any number of criteria one could consider when selecting a measurement device to assess the effectiveness of your marketing communications efforts in general and your web sites effectiveness in particular:

Purchase habits in the product category…does the consumer buy one brand or many brands (visit one site or many sites)?

Frequency of purchase (site visits)?

Number of brands in the category ( alternative sites)?

Market share and atmosphere of competition surrounding the advertised brand (web-site); and brand loyalties in the category.

Objectives of the advertising? Is the advertiser trying to convert competitive brand users; increase consumption among current users; or simply bring about a change in the way people view the product, service, industry or corporation?

Alternatively, one must define the essential function of the site itself. Is it more promotional (of products/services that are primarily promoted and solely distributed, purchased and consumed off-line); transactional ( products/services are primarily promoted and solely distributed, purchased and consumed on-line) or informational (in-depth information about the product/service is distributed on-line and the product /service is promoted, purchased and consumed on and/or off-line).

To date, the problem of how the marketer evaluates marketing communications has been addressed in the television advertising industry (for example, Wells, Leavitt and McConville, 1971; Schlinger, 1979). Such means are appropriate for examination of Web communications in light of the propensity of advertisers to use the Web in a similar way to television (McDonald, 1997), probably due to the need for a familiar operating environment.

By integrating and interpreting Schlinger’s (1979) Viewer Response Profile (VRP) with Novak and Hoffman’s Web Metrics the researcher can begin to access the web users web-site experiences from a multi-dimensional perspective. Wherein, a webusers VRP can be regarded as cognitive/attitudinal measure of outcome metrics. To wit:

Entertainment: pleasure, enjoyment and fun;

Confusion: difficulty in following the message;

Relevant News: has it provided useful information;

Brand Reinforcement: of existing positive attitudes;

Empathy: a feeling of vicarious participation;

Familiarity: in terms of being unusual or different;

Alienation: irrelevance and irritation.

Under testing the scale exhibited predictive validity (52% of variance explained). Co-efficient alpha was not reported, though item-factor loadings ranged form 0.33 to 0.88. Additionally, the scale showed convergent validity with the Wells, Leavitt and McConville (1971) reaction profile (78% explanation of variance when jointly factor analysed).

It is our contention that sporting team sites are intended for use by sporting team fans rather than the casual Web "surfer". This contention has some, albeit limited, support by implication from the literature (for example, Forrest, Pope and Murphy, 1996). By definition, the fan is someone who feels a sense of involvement with the sporting team in question. Involvement is an internal state of arousal in response to a stimulus of some sort (Zaichkowsky, 1985; Andrews et al, 1990; Mittal, 1995; Poiesz and Cees, 1995), but there has been debate over the nature of the stimulus (Rothschild, 1984; Andrews et al, 1990; Poiesz and Cees, 1995). Certainly it is agreed that involvement possesses intensity, direction and persistence - all measurable properties, but whether the stimulus for the response is the product itself or the actual decision is still debated.

Four basic research areas have opened into the area of involvement: attention/processing; personal/situational; audience/process; and enduring/product involvement (Rothschild, 1984; Andrews et al, 1990; Celuch and Slama, 1993; Poiesz and Cees, 1995). These form the nomological network, or pattern of theoretical constructs that define the domain. In the involvement construct the nomological network runs from antecedents (personal needs, goals and characteristics; and situational and decisional factors) to involvement (intensity, direction and persistence) to consequences (search, information processing and persuasion). Related constructs are opportunity to process and ability to process (Andrews et al, 1990; Zaichkowsky, 1994).

With more than one particular research strategy available for the construct, interaction between the various elements has been noted (Laczniak and Muehling, 1990; Gainer, 1993). There is, however, good reason to maintain separation of the consequences of involvement and the research methods involved. As noted by Zaichkowsky (1994), product involvement is not suitable for measuring advertising involvement, because it is measuring a different stimulus. This is not to dispute that product involvement influences advertising message involvement, as demonstrated by Laczniak and Muehling (1990).

The original Product Involvement Inventory (PII) was developed by Laczniak (1985). McQuarrie and Munson (1986) argue that involvement is a multi-dimensional construct and that the original PII was confounded by attitudinal variables. They presented a Revised PII and later a Revised RPII (1991). This last version captures two elements of involvement: interest and importance, while still allowing for a total involvement score. When subjected to tests of internal consistency, their new scale consistently exhibited alphas of greater than 0.80.

H1: As involvement increases, the individual will attach more importance to information than to entertainment functions of sporting team Web sites.

H2: As interest increases, the individual will attach more importance to information than to entertainment functions of sporting team Web sites.

H3: As importance increases, the individual will attach more importance to information than to entertainment functions of sporting team Web sites.

METHOD

Sample

Instrument and Operationalisation

Involvement with the Brisbane Broncos Rugby League Football team was measured using McQuarrie and Munson’s Revised RPII (1991). Response to the Broncos’ web site was measured using Schlinger’s Viewer Response Profile (1979). In order to adapt the instrument to a) the Australian domain, and b) the Internet, the questionnaires were submitted to an expert panel of three marketing academics with Australian-American and Interactive Marketing experience.

One change was made to the Revised RPII scale to allow for cultural differences. Item 5, with the differing constructs of "dull" and "neat" was altered to read "dull" and "fine". The Viewer Response Profile was altered to replace the words "commercial" to "site", and "character(s)" to "image(s)". A copy of the final, revised instrument is presented at Appendix 1.

After collection, and prior to analysis, data were subjected to item reliability tests. The RPII scale demonstrated high reliability (Cronbach’s a = 0.95) overall as well as in the sub-sets of Importance and Interest (Cronbach’s a = 0.91 and 0.94 respectively). The VRP did not perform as well overall (Cronbach’s a = 0.73 for the total scale). Sub-sets performed as follows: Entertainment (Cronbach’s a = 0.89); Confusion (Cronbach’s a = 0.78); Relevant News (Cronbach’s a = 0.72); Empathy (Cronbach’s a = 0.80); Alienation (Cronbach’s a = 0.70). Brand Reinforcement could not be subjected to reliability analysis as it consisted of only two items. Familiarity achieved a reliability less than the 0.70 recommended by Nunally (1978) and now accepted as the norm in the social sciences (Spector, 1992). It was therefore deleted from the data set prior to analysis. A further analysis of Confusion showed that Cronbach’s a could be raised to 0.82 on deletion of item 8, and in the case of Empathy, Cronbach’s a was raised to 0.86 by the removal of item 19. These two items were removed form the data set prior to analysis.

Treatment of data

Data were subjected to a forward, stepwise, discriminant analysis to establish differences in response profile values between high and low involvement groups. High involvement was defined as being those individuals recording an involvement with the Broncos higher than the mean for the total sample. This was also applied in the Involvement sub-sets of Importance and Interest. Means were: Total Involvement = 3.7125; Importance = 3.3; Interest = 4.125.

FINDINGS

Table 1 Stepwise Discriminant Analysis Comparing High and Low Bronco Involvement Respondents

Step

Variables entered

Standardised

coefficients

Wilks’ lambda

Brand reinforcement

Relevant News

Alienation

Entertainment

-1.02

1.03

0.61

-0.35

0.66*

0.58*

0.44*

0.42*

Wilks’ lambda = 0.42; Chi square = 30.76; df = 4; p < .0001

Canonical correlation = 0.76

Grouped cases correctly classified = 90.0%

*All lambda coefficients have p values < .0001

Table 2 Stepwise Discriminant Analysis Comparing High and Low Bronco Interest Respondents

Step

Variables entered

Standardised

coefficients

Wilks’ lambda

Brand reinforcement

Entertainment

Relevant News

Confusion

-1.05

-0.71

1.03

-0.59

0.64*

0.42**

0.49*

0.42**

Wilks’ lambda = 0.35; Chi square = 37.71; df = 4; p < .0001

Canonical correlation = 0.81

Grouped cases correctly classified = 90.0%

* Lambda coefficients have p values < .001

** Lambda coefficients have p values < .05

Table 3 Stepwise Discriminant Analysis Comparing High and Low Bronco Importance Respondents

Step

Variables entered

Standardised

coefficients

Wilks’ lambda

Brand reinforcement

Relevant News

Alienation

1.05

-0.84

-0.65

0.82*

0.64**

0.61***

Wilks’ lambda = 0.51; Chi square = 24.80; df = 3; p < .0001

Canonical correlation = 0.70

Grouped cases correctly classified = 82.5%

* Lambda coefficients have p values < .001

** Lambda coefficients have p values < .01

*** Lambda coefficients have p values < .05

Appendix 1: Survey Instrument

Please answer these questions about the Brisbane Broncos:

To me the Brisbane Broncos are (circle the number closest to your response):

1. important 1 2 3 4 5 6 7 unimportant

2. irrelevant 1 2 3 4 5 6 7 relevant

3. mean a lot do not mean

to me 1 2 3 4 5 6 7 a lot to me

4. unexciting 1 2 3 4 5 6 7 exciting

5. dull 1 2 3 4 5 6 7 fine

6. matter to me 1 2 3 4 5 6 7 don’t matter to me

7. fun 1 2 3 4 5 6 7 not fun

8. appealing 1 2 3 4 5 6 7 unappealing

9. boring 1 2 3 4 5 6 7 interesting

10. of no concern 1 2 3 4 5 6 7 of concern to me

Please respond to the following statements about the Broncos’ web site by circling the number that best indicates your response:

1. The site is lots of fun