Sport Marheting Quarterly, 2005,14.43-53, © 2005 West Virginia University
Fans vs. Spectators: An Analysis of
Those Who Attend Intercollegiate
Football Games
Matthew J. Robinson, Galen T. Trail, Ronald J. Dick, Andrew J. Gillentine
Abstract
During the 2001 college football season more than
40 million individuals attended intercollegiate football
games across the four divisions sponsored by the
National Collegiate Athletic Association (NCAA).
Although it would be easy to classify all of these individuals as either being spectators or fans, that would be
in accurate. Trail, Robinson, Gillentine, and Dick
(2003) developed a model based on the relationship
between motives and points of attachment that classified attendees as either spectators or fans. The purpose
of this study was to use the model to determine how
individuals who attend college football games should
be classified at the four NCAA divisions of college
football. Data were collected from spectators at games
at the the four different NCAA college football divisions. A MANOVA was performed on the motive factors and on the points of attachment factors proposed
by Trail et al., using the division level as the independent variable. Results indicated that there was some support for the Trail et al. model. In general. Division I
attendees have motives and points of attachments that
are consistent with what Trail et al. (2003) classified as
a fan characteristics. Division III attendees have
motives and points of attachment that are consistent
with what Trail et al. (2003) classified as spectator
Matthew J. Robinson, EdD, is an associate professor of
sport management at the University of Delaware. His
research focuses on consumer behavior and brand equity.
Galen T. Trail, PhD, is an associate professor of sport
management at the University of Florida. He conducts
research in consumer behavior and organizational
behavior.
Ronald J. Dick, EdD, is an assistant professor of sport
management at James Madison University. His research
interests include sport marketing and ticketing.
Andrew J. Gillentine, PhD, is an associate professor of
sport management at the University of Miami. His
research interests include organizational behavior and
sport marketing.
"... a fan's consumption behavior may differ from a
spectator's and vice versa: thus each may need to
be marketed to differently."
characteristics. Division II and I-AA have motives and
attachment indicative of both fans and spectators.
Discussion of the results and suggestions for fiiture
research are presented.
Fans vs. Spectators: An Analysis of Those
Who Attend Intercollegiate Football Games
In the United States there are 608 college and universities with National Collegiate Athletic Association
(NCAA) sponsored football programs. In 2001,
40,500,823 individuals attended 3,193 NCAA sponsored intercollegiate football games (NCAA.org
Football Attendance, 2001). At one end of the spectrum was the University of Michigan, an NCAA
Division I-A institution, which had 659,447 individuals
attend six home games. At the other end of the spectrum, Principia College, an NCAA Division III institution, played in front of 500 attendees during its two
home contests. Somewhere in between were schools
like the University of Delaware, an NCAA Division IAA institution, which played in front of more than
102,518 attendees during five games and Slippery Rock
University, an NCAA Division II institution, which
played in front of 23,775 during its five home games
(NCAA Football Attendance, 2001).
While it would be easy to label all of these 40 million
individuals as college football fans, it would be inaccurate, for some may not demonstrate behaviors of a fan,
but may only spectate. Sloan (1989) noted that spectators in the strictest sense are those who merely watch
and observe, but fans are enthusiastic devotees of a
given diversion. Despite the distinction made by Sloan
(1989), Trail, Robinson, Gillentine, and Dick (2003)
suggested that previous researchers had not been consistent in the use of the terms "spectators" and "fans,"
nor had researchers ever delineated between the two in
reporting results. Trail et al. (2003) stated that from a
marketing perspective the distinction may be crucial,
Volume 14 • Number 1 • 2005 • Sport lAarheting Quarterly 43
for a fan's consumption behavior may differ from a
spectator's and vice versa: thus each may need to be
marketed to differently. Trail et al. (2003) set out to
quantify the differences by developing a model that
was based on the relationship between motives (vicarious achievement, acquisition of knowledge, aesthetics,
socialization, drama/eustress, escape, and physical skill)
as measured by the Motivation Scale for Sport
Consumption (MSSC) and points of attachment (the
players, the coach, the community, the sport, the university, the team, and the level of the sport) as measured by the Points of Attachment Index (PAI). The
present study furthers the Trail et al. (2003) work by
using the latent variables presented in the model to
classify those who attend college football games at the
various levels as being spectators, fans, or both.
including a person or group that results in feelings or
sentiments of close attachment" (p. 165-166). Further,
Trail et al. (2000) stated that identification influences
cognitive, affective, and behavioral responses, and
many authors have suggested that team identification is
a key predictor of sport consumption (Cialdini et al.,
1976; Sloan, 1989; Zillmann et al., 1989). Anderson and
Stone (1981) have long argued that sports teams are
symbolic representations of a community and can provide individuals a sense of belonging to that community. Those authors and others (e.g., Wann &
Branscombe, 1993) have solely focused on identification with, or attachment to, a team. Robinson and Trail
(2003) and Matusuoka and Fujimoto (2002) both
argued that individuals may have multiple points of
attachment besides the team and therefore are worthy
of investigating.
"... the fan may be more motivated by the achieveRelationships Between Motives and Identification
ment of a favorite team, while the spectator may
Previous research has shown that motives have been
be motivated more by the aesthetics or the skill disassociated with team identification (Fink et al., 2002;
played in a sport."
Motives for Spectating
Deci (1971) suggested that motives are either innate
or learned and generate behaviors because of the satisfaction or enjoyment generated by the activities therein.
Most spectator or fan behaviors fulfill social or psychological needs. Many researchers have suggested potential
motives for sport spectator consumption (Pease &
Zhang, 2001; Sloan, 1989; Trail, Anderson, & Fink,
2000; Wann, 1995; Zillmann, Bryant, & Sapolsky, 1989;
Zillmann & Paulus, 1993) and several have developed
scales for measuring the motives for attending sporting
events or being a fan (Funk, Mahony, Nakazawa, &
Hirakawa, 2001; Funk, Mahony, & Ridinger, 2002;
Kahie, Kambara, & Rose, 1996; Milne & McDonald,
1999; Trail 8f James, 2001; Wann, 1995). We chose to
use the motives proposed by Trail et al. (2000), which
were partially based on theories espoused by Sloan
(1989) and Maslow s (1943) hierarchy of needs, and we
chose the Motivation Scale for Sport Consumption
(MSSC) developed by Trail and James (2001) to measure the specified motives.
Identification and Points of Attachment
Trail et al. (2000) presented the theoretical framework for investigating identification as well. Fink, Trail,
and Anderson (2002) stated that the examination of
team and fan identification is grounded in social identity theory. Tajfel and Turner (1986) contended that
self-concept is composed of personal identity and
social identity. Trail et al. (2000) defined identification
as "an orientation of the self in regard to other objects
Trail, Fink, & Anderson, 2003; Trail & James 2001;
Wann, 1995; Wann, Brewer, & Royalty, 1999; Wann,
Royalty & Rochelle, 2002). At this point though,
research that has examined relationships among individual and specific motives and points of attachment is
sparse (i.e., Robinson & Trail, 2003; Trail et al., 2003).
However, indirect evidence exists to suggest that there
may in fact be relationships between specific motives
and points of attachment (Funk et al., 2001; Funk et
al., 2002).
".. philosophies, budgets, and the offering of scholarships lead to differences in competitive quality
among the four classifications of NCAA college
football and this competitive quality may lead to
whether or not a division has spectators or fans or
both attending."
Model for Distinguishing Between Spectators and Fans
Sloan (1989) contended that fans and spectators
could have different motives for attending. For example,
the fan may be more motivated by the achievement of a
favorite team, while the spectator may be motivated
more by the aesthetics or the skill displayed in a sport.
Fisher and Wakefield (1998) suggested that different
motives and dififerent effects for the individual could
change based on the performance of a team and/or the
outcome of a contest. This implied that an individual,
whose favorite team is not successful, might change the
focus to other aspects rather than dwell on a negative
association. Based on Sloan's (1989) and Fisher and
Wakefield (1998) findings. Trail et al. (2003) successfully tested a model that consisted of three different latent
44 Volume 14 • Number 1 • 2005 • Sport MarHeting Quarterly
Figure f
Fan and Spectator Model (from Trail, Robinson, Gillentine, and Dlch, 2003).
Overarching Motives /^~X
variables derived from tbe MSSC (see Figure 1). One
group, tbe "Overarcbing Motive" variable, consisted of
tbe motives of socialization and escape that are applicable to botb fans and spectators. Tbe second variable,
"Vicarious Acbievement Motive," applied to fans of successful teams and tbe tbird latent variable, labeled
"Spectator Motive," consisted of tbe aesthetic,
drama/eustress, acquisition of knowledge, and appreciation of physical skill motives applied to spectators and
fans of unsuccessful teams.
Tbese tbree latent variables were found to be differentially related to tbe Points of Attacbment derived
from the PAI. Tbe "Vicarious Achievement Motive" was
directly related to a latent variable termed
"Organizational Identification" tbat consisted of the
five points of attacbment: team, coacb, community,
university, and player. Tbe "Spectator Motive" was
directly related to a latent variable labeled "Sport
Identification" tbat consisted of two points of attacbment constructs; type of sport and level of sport.
However, tbe attacbment to tbe player construct did
not represent tbe Organizational Identification latent
variable well. Tbus, for tbe present study we analyzed
attacbment to tbe player separate from tbe
Organizational Identification.
Differences in Four NCAA Divisions of College Football
DeScbriver (1999) stated tbat competitive quality
distinguished professional sport from intercollegiate
sport. It can be argued tbat pbilosophies, budgets, and
tbe offering of scholarships lead to differences in competitive quality among the four classifications of NCAA
college football and tbis competitive quality may lead
to whether or not a division bas spectators or fans or
botb attending.
Volume 14 * Number 1 • 2005 • Sport Marheting Quarterly 45
The more than 1,000 individual institutions that are
members of the NCAA are divided into three divisions:
NCAA Division I, Division II, and Division III, with
each division having a unique philosophy that is the
driving force in creating tbree very distinct environments and levels of competition. Institutions competing
at tbe Division I level aspire for regional and national
atbletic excellence and prominence. Tbey offer athletic
scholarships and strive to finance the atbletic program
with funds generated by tbe atbletic enterprise. These
institutions typically recognize that tbey serve both
internal constituencies as well as tbe general public
{National Collegiate Atbletic Association, 2001). The
average revenue for a Division I atbletic department is
$21,900,000 witb ticket sales being the greatest source
of revenue (Fulks, 2000).
"It was hypothesized that the differences that exist
between the divisions in terms of philosophies,
budgets, and level of competition would lead to
differences in terms of the classification of the
attendees as being either spectators or fans at the
different levels."
In tbe sport of football, Division I is divided into I-A
and I-AA classifications with I-A being viewed as the
bigber level of tbe two. In order for an institution to be
classified as a I-A institution, 60% of its schedule must
be against fellow I-A teams, tbe institution must average at least 17,000 spectators for home games or 20,000
for all games played over the previous four years, and
have a 30,000 stadium seat capacity. NCAA I-A institutions are permitted to offer 85 scholarships for football. Fulks (2000) reported tbat the average budget for
an NCAA Division I-A football program is $5,260,000.
Division I-A football is viewed as the highest level of
competitive quality and it can be supported by the fact
that the majority of tbose individuals who play professional football in the National Football League (NFL)
competed at an NCAA I-A institution.
An NCAA I-AA institution's scbedule must consist of
50% of its games against I-A or I-AA opponents, tbere
are no football attendance or stadium size requirements, and institutions are permitted to offer 63 football scholarships {National Collegiate Athletic
Association, 2001). Fulks {2000} reported the average
operating budget for a Division I-AA football program
was $1,100,000.
NCAA Division II atbletic programs also recognize
tbe dual responsibility of meeting tbe needs of tbe
internal community and tbe general public, but not to
tbe degree of Division I in terms of tbe general public.
Division II institutions permit athletically related
financial aid and believe in offering participation
opportunities to as many students as possible
{National Collegiate Atbietic Association, 2001).
Division II operating budgets are significantly less tban
Division I. The average budget for an entire Division II
athletic department witb football is $1,900,000 {Fulks,
2000). Also at Division II, ticket revenue is not a significant source of revenue because athletic departments
rely on institutional support and student fees as funding sources {Fulks, 2000). In terms of competition.
Division II is not considered as bigb a level of competition as Division I-A or I-AA.
NCAA Division III atbletic programs are financed,
staffed, and controlled tbrougb tbe same general procedures as other departments on campus. The average
budget for a Division III athletic department witb football is $1,000,000 {Fulks, 2000). Division III athletics
place a special emphasis on meeting the needs of internal constituencies sucb as tbe students, alumni, and
institutional personnel rather tban on the general public's entertainment needs. In addition, these institutions
do not offer athletic scholarships to student-athletes
and strive for league and regional excellence ratber tban
national excellence {National Collegiate Athletic
Association, 2001). Attendance at atbletic events is minimal and not viewed as a significant revenue source for
an athletic department. Division III is viewed as the
lowest level of competition witbin tbe NCAA.
"... those attending I-A games were motivated by
vicarious achievement to a greater degree than
those attending games at the other levels."
Tbe purpose of tbis study was to use Figure 1 to help
determine if those attending college football games at
eacb of tbe four levels of tbe NCAA can be classified as
fans and/or spectators. It was bypothesized that tbe differences that exist between the divisions in terms of
pbilosophies, budgets, and level of competition would
lead to differences in terms of tbe classification of tbe
attendees as being either spectators or fans at the different levels. The researchers bypotbesized tbat at the
Division I level tbe attendees would exhibit more fan
tendencies and at the Division II and III levels the
attendees would exhibit more spectator tendencies.
Tbis bypothesis was based on tbe belief that tbose wbo
attend games at tbe bigbest level, Division I-A, would
be motivated more by achievement and would attach
more to tbe bigb profile teams, coaches, communities,
universities, and players. On tbe other hand, those who
attend the lower level games, I-AA, II, and ill, would
be motivated more by tbe spectator motives and attach
more to tbe sport and level of competition. Lastly,
based on the Trail et al. {2003) model, the researchers
bypothesized that tbere would be no differences among
46 Volume 14 • Number 1 • 2005 • Sport MarHeting Quarterly
subscales, three items per subscale) have a seven-point
response format ranging from "Strongly Disagree" (1)
to "Strongly Agree" (7). For this study, the motives
were combined into the higher level variables depicted
by the Trail et al. (2003) model: Overarching Motive
(escape and social). Spectator Motive (skill, aesthetics,
drama, and knowledge), and the Vicarious
Achievement Motive.
The Point of Attachment Index has seven subscales
that focus on identification with I) the players, 2) the
coach, 3) the community, 4) the sport, 5) the university,
6) the team, and 7) the level of the sport (e.g., college,
Method
not professional). Each subscale had three items and
thus
the PAI has 21 items total, with a seven-point
Data were collected from spectators at four intercolleresponse
format ranging from "Strongly Disagree" (1)
giate football games (Division I-A, «=191; Division IAA, tt=221; Division II, n=229; and Division III, n=220) to "Strongly Agree" (7). The PAI had shown good reliability and construct validity previously (Robinson &
during the 2001 -2002 academic year. Out of 910 quesTrail, 2003; Trail et al, 2003), with one exception as
tionnaires distributed, 861 usable questionnaires were
noted above. The identification with player subscale did
returned, for a return rate of 94%. The selection of the
schools for the study was based on the geographic prox- not load well on the Organizational Identity factor.
Again, the subscales were combined into tbe variables
imity to the researchers. The data were collected at one
depicted by the Trail et al. (2003) model with the excepgame for each of the divisions. AU four of the schools
tion of identifreation with player, which was maintained
selected had winning seasons the year of the collection,
as a separate variable. Thus, Organizational
all of the schools had a tradition of success in football
Identification was represented by four subscales (team,
(e.g., postseason competition, Ail-American players),
coach, community, and university) and Sport
and all of the teams were coached by individuals who
Identifreation was represented by two subscales (level
had an above average winning percentage as a head
and
sport).
coach.
attendees at the different division level games on the
Overarching Motive variable. On the Vicarious
Achievement variable and the Organizational
Identification variable (minus attachment to player),
the researchers expected that attendees at the Division
I-A level would score higher tban the other three levels.
On tbe Spectator Motives variable, the Sport
Identification variable, and tbe Attachment to Player
variable, researchers expected that the attendees at the
Division I-A level would score lower than the otber
three levels.
Instrumentation
The questionnaire was comprised of two scales: the
Motivation Scale for Sport Consumption (MSSC) and
the Point of Attachment Index (PAI). The MSSC consists of seven subscales (vicarious achievement, acquisition of knowledge, aesthetics, socialization,
drama/eustress, escape, and physical skill) and has
shown adequate psychometric properties in previous
use (Fink et al., 2002; Robinson & Trail, 2003; Trail et
al., 2003; Trail & James, 2001). The 21 items (seven
Procedures
Undergraduate sport management students assisted
with the distribution and collection of tbe questionnaire at each contest. The students were trained in data
collection methods prior to the data collection.
Students were placed at randomly selected entrances to
seating areas in the football stadiums. This was done to
help ensure more varied representation among spectators (i.e., not all students, not all season ticket holders,
not all donorSj etc.). The questionnaires along with a
Table f
Between Subjects Effects
_2
Power
Source
DV
df
Division
Level
Achievement
Overarching Motive
Spectator Motive
Organizational
Identification
Player Identification
Sport Identification
3
3
3
3
37.26
23.75
13.96
54.68
.000
.000
.000
.000
.115
.077
.047
.161
1.000
1.000
1.000
1.000
3
3
37.26
17.29
.000
.000
.117
.057
1.000
1.000
P
Volume 14 • Number I • 2005 • Sport Marheting Quarterly 47
Table 2
Means and (Standara Deviations) ofOVs by Level (DIA. DIAA. DII. Dili)
DIA
DIAA
DII
Dili
Sample
6.11
4.83
(1.49)
5.20
5.25
(0.92)
5.00
(1.43)
(1.32)
(1.40)
5.82
5.24
4.97
5.40
5.34
Motives a b c d f
fo.8O)
(1.09)
(1.15)
(1.06)
(1.08)
3. Spectator
Motives a b c e f
5.72
fo.78)
5.37
(0.80)
5.22
(1.02)
5.61
(0.89)
5.47
(0.90)
4. Organizational
5.47
4.49
4.16
4.58
4.64
Identification a b c d f
(o.81)
(1.10)
(1.17)
(1.15)
(i.l7)
5. Player
Identification a b c d e
3.40
(^ 53)
2.93
(1.47)
4.12
3.72
(1.61)
4.36
(1.66)
(1.68)
4.91
4.78
4.60
5.41
4.91
(1.17)
(1.42)
(1.20)
(1.18)
Item
Motives
1. Achievement
Motives ^ ^ c f
2. Overarching
6. Sport
Identification ^ ^ f
(1.18)
Note. ^ indicates differences between DIA and DIAA at p < .05, b indicates differences between DIA and DII at p < .05,
indicates differences between DIA and Dili at p < .05, ^ indicates differences between DIAA and DII at p < .05, ^ indicates
differences hetween DIAA and Dili at p < .05, 'indicates differences between DII and Dill at p < .05
clipboard and pencil were distributed to the spectators
prior to the game and the respondents were asked to
complete the scales measuring the motives and identification. The students collected the questionnaires when
completed. The purpose of the study, as well as the
instructions for completion of the survey, was included
in the questionnaire.
"In terms of points of attachment, it is
evident again that differences also exist among
the divisions"
Data Analysis
A MANOVA was performed using the GLM procedure in SPSS 10.5. There was one independent variable
(division level) and six dependent variables
(Overarching Motives, Spectator Motives, Vicarious
Achievement Motive, Organizational Identification,
Sport Identification, and Player Identification). This
allowed researchers to test whether mean differences
among the division levels on a combination of dependent variables existed (Tabachnick 8c Fidell, 1989) for
both the motives and the points of attachment.
Results
For the MANOVA, with the use of the Wilk's criterion, the combined dependent variables were affected by
division level, F(18, 2410) = 19.40, p < .OOL The results
reflected a moderate association between the division
level and the combined DVs (G^ = .12).
The test of between-subjects effects indicated that
attendees differed by division level on all six of the
dependent variables (Table 1). The post-hoc
(Bonferroni) tests indicated that those attending the
Division I-A game (M = 6.1) were motivated by vicarious achievement to a much greater extent than those at
either the I-AA (M = 5.0), DII (M ^ 5.2), or Dili (M =
4.8) levels. Differences existed between the latter levels
also (Table 2). Level of division explained 11.5% of the
variance in vicarious achievement. Difrerences in the
overarching motive were also apparent, although not as
a great (O^ = .08), as DI-A differed from both all other
levels again (Table 2). Relative to the spectator motives,
Division I-A differed from I-AA and II, but not
Division III and the differences were less (O^ = .05).
The test of between-subjects effects also indicated
that attendees differed by division level on the three
aspects of points of attachment (Table 2). Division
48 Volume 14 • Number 1 • 2005 • Sport Marheting Quarteriy
levei explained the most variance in organizational
identification (G = .16), a moderate amount of variance in player identifreation (O = .12), and also
explained a small amount of variance in the sport
identifreation (O^ = .06). The post-hoc (Bonferroni)
tests indicated that those attending the Division I-A
game (M = 5.5) scored higher on the organizational
identifreation than any of the other divisions DI-AA
(M = 4.5), DII (M = 4.2), Dili (M = 4.6), which all differed also, except for I-AA and III. The results relative
to the sport identification variable were slightly different though. Attendees at the Division III school (M =
5.4) were more highly identifred with the sport/level
than those at the other schools DI-AA (M = 4.8), DII
(M = 4.6), DI-A (M = 4.9). Relative to player attachment, again Division III attendees scored highest (M =
4.4), followed by DII (M = 4.1), DI-A (M - 3.4), and
DI-AA (M = 2.9), with all differences being signifrcant
except between DII and Dill.
Discussion
It is apparent from the results that differences do
exist among the divisions based on the motive variables and the points of attachment variables. However,
although there was some support for the model that
Trail et al. (2003) proposed, there were aspects that
were not supported.
Specifrcally, the model was supported by results that
indicated that those attending I-A games were motivated by vicarious achievement to a greater degree than
those attending games at the other levels. In general, the
trend was as expected (Figure 2), except that those
attendees at the Division III level indicated slightly
higher levels of vicarious achievement than those at the
I-AA and II levels. Fink et al. (2002) defrned vicarious
achievement as the need for social prestige, self-esteem,
and sense of empowerment for an individual through
his or her association with a successful team. These
aspects are probably what motivate an individual to
become a fan of a successful Division I-A college football team. Division I-A fans may also be practicing the
process of basking-in-reflected glory (BIRGing) that has
been defrned by Wann and Branscombe (1990) as an
individuals desire to increase their association with
successful others. It appears the fans at the Division I
level are motivated to attend in order to live vicariously
through the achievements of their favorite team. The
high degree of vicarious achievement can be attributed
to the fact tbat the average fan never was, nor will be, a
scholarship athlete. Furthermore, the fan never had
Figure 2
Mean Motives (Achievement. Overarching, Spectator) by Football Division Level.
6.2
6.0 '
\
5.8 "
>
5.6 •
5.4 '
/
/
5.2 •
5.0 •
g
Vicarious Achieve
4.8 •
Overarching Motives
4.6
Spectator Motives
Division I-A
Divsion I-AA
Division II
Division III
Division Level
Volume 14 • Number 1 • 2005 • Sport Marheting Quarterly 49
Figure 3
Mean Identification (Piayer. Organizational. Sport) by football Division Level.
Player ID
Sport ID
3.0
Organizational ID
2.5
Division I-A
Divsion I-AA
Division II
Division II!
Division Level
have a chance to play for the team and coach he or she
may have idolized as a youth, and quite probably never
played in front of 100,000 people in a stadium and
millions on television. Therefore these fans gain a sense
of empowerment and increase their self-esteem by
being associated with this level of competition. This is
further supported by the fact that those who attended
the the lower levels of competition had significantly
lower levels of vicarious achievement. At these levels of
competition the tendency to bask In the reflected glory
is probably less due to the lack of prestige and visibility
of these programs. For example, Division HI players are
non-scholarship athletes who play in front of smaller
crowds with limited media coverage and there is not as
much at stake financially based on the outcomes of
games.
Based on Trail et al.'s (2003) model, and the support
for the differences in vicarious achievement, the
researchers were quite surprised when the anticipated
differences were unexpected and the trend for the
Spectator Motive variable was not supported (Figure
2). We expected that those attending Division III games
would score more highly on this combined variable
than those attending Division I-A games, evidenced by
a negative trend through the four levels. We assumed
that because of the lack of visibility and prestige associated with the lower division schools, attendees would
have to be motivated to come by other aspects such as
the aesthetic qualities of the game, the appreciation for
the physical skills of the athletes, and the drama of the
event. This was somewhat supported as the Division III
attendees scored higher on this variable than either
Division II attendees or Division I-AA attendees.
However, the researchers did not expect that the
Division I attendees would score higher than all three
of the other divisions. The researchers anticipated that
because these individuals would be high on vicarious
achievement, as they were, they would not be drawn by
motives such as aesthetics or drama. The old adage
about winning ugly, apparently does not apply. These
individuals wanted their team to win and win pretty.
We also thought that drama would not be a high
motive because typically people motivated by vicarious
achievement do not want to be in doubt of a win. They
want to win big, with the outcome never in doubt.
The hypothesis about the Overarching Motives not
differing across divisions was also not supported. Based
on Trail et al.'s (2003) model, the researchers anticipated that because these motives were supposed to be relevant to both fans and spectators alike, there would be
50 Volume 14 • Number 1 * 2005 • Sport Marheting Quarterly
no differences among attendees. As seen in Figure 2,
Division I-A attendees were significantly more social
and escapist than those who attended games at the
other levels. This makes little sense, as typically all individuals, Vi'hether they are fan or spectator, want to get
away and socialize when they go to a game. Oliver
(1999) does note that there are individuals, whom he
calls self-isolationists, v^rho attend and have no interest
in the social aspects of the event, but yet are still highly
involved with the team and game. Perhaps those attendees at the Division II game were self-isolationists.
In terms of points of attachment, it is evident again
that differences also exist among the divisions. As noted
by Trail et al. (2003), vicarious achievement is typically
highly correlated with attachment to the team, coach,
community, and university. This was readily apparent
for those who attended Division I games. Those individuals seem to attend games for the purpose of supporting their favorite team, which is coached by their
favorite coach, who has been hired by the fan's favorite
university in their favorite community. The hypothesis
that attendees at Division I games would be significantly
higher than their counterparts at the lower division
games was supported, and the trend was in the expected
direction. The exception was again the fans at the
Division III school, which scored a little higher than
expected (Figure 3).
We hypothesized that Division III attendees would be
more highly attached to the players than those at the
other three levels, and this was the case. However, the
trend did not follow the expected path, as we anticipated
that as the level of institution increased the player identification would decrease (Figure 3). We felt that at the
Division III level, a greater percentage of the crowd
would have some association with a specific player than
at the other levels (parents attending to see a son play or
an individual attending to watch a classmate), thus elevating the mean score of the responses. This was apparent. However, this cannot be the reason for the mean
score for attendees at the Division I level being greater
than those at Division I-AA. At the DI-A level, it is more
likely that player attachment is due to the visibility and
prestige of a star player, rather than the specific association of parent to player, or player to friend. This visibility was the rationale for the inclusion of attachment to
player in the Organizational Identity construct in Trail et
al.'s (2003) model. Perhaps attachment to player only fits
within this construct when there are star players on the
team: players who generate vicarious achievement for a
substantial number of spectators, players such as
Michael Jordan or Walter Payton.
Based on Trail et al.'s (2003) model, spectators who
attend games because of aesthetic or dramatic qualities
of the game (not vicarious achievement) should have
different points of attachment (i.e., attachment to the
sport or the level of sport), rather than achievement
driven points of attachment. This was somewhat supported in that Division III attendees were higher on
Sport Identification than those at the other levels.
However, the trend of decreasing sport attachment, as
the division level increased, was not supported (Figure
3). Those attending at the lowest division were hypothesized to be attached to football in general and college
football in particular, as was supported, but Division IA attendees were bigger fans of the sport and level than
those at the Division I-AA and II levels, which was not
consistent with the model.
Marketing Strategies
It appears important that the marketing directors
develop marketing plans specific to the division. In the
case of Division I programs, it is important to highlight the successes by which the fans seek to have their
vicarious achievement motives met, while also incorporating the points of attachment of the team, the
coach, community, and university. It can be argued
that the strengths of vicarious achievement through
the team and coach are already evident. NCAA
Division I athletic departments have capitalized on
spectators' and fans' attachment with a team and university through the sale of licensed products. Licensed
products offer a fan the opportunity to demonstrate
his or her attachment to a team and/or university. It is
not surprising to see that Division I-A generates more
revenue than Division I-AA, 11, and III. This may be
traced back to the stronger attachment, the greater
media exposure, and the larger number of fans. In the
case of coaches, Division I-A coaches are in demand
for speaking engagements within the community and
surrounding areas, and in some cases command significant fees to speak to corporations and are asked to
serve as endorsers for products. This may facilitate the
strong attachment fans have for a coach.
Other strategies that could be incorporated would be
creating a hall of fame area that would include team
pictures, awards, etc. from the past as well as interactive
booths to replay highlights from past team successes.
Although a star player may be the centerpiece of a single season marketing effort, the sport marketer must
not lose track that it is the team and university attachment that is most significant. Traditions should be
emphasized within the marketing efforts and for programs looking to increase attachment, potential traditions should be identified and highlighted in marketing
efforts. When individuals identify too strongly with a
player, the university may lose those fans once a player
no longer plays for the university.
Volume 14 • Number 1 • 2005 • Sport MarHeting Quarterly 51
In the case of the lower divisions, it is the players to
which individuals attach. In that situation, the marketers should offer special benefits towards parents and
friends who are attending to see a specific player play.
The challenge for Division III marketers is to try to
prevent defections by the friends and family of individual players after the players leave the program.
Marketers need to implement strategies to transfer the
allegiance of these attendees from the players to the
team and/or university. Potential ideas include offering
benefits for former players and families at games, initiating an active booster club, and communicating to
past spectators through a newsletter, webpage, or email
list.
Fulks, D. L. (2000). Revenues and expenses of divisions I and ll intercollegiate
athletic programs: Financial trends and relationships - 1999. Overland
Park, KS: National Gollegiate Athletic Association.
Funk, D. C., Mahony, D. F.. Nakazawa, M., & Hirakawa, S. (2001).
Development of the sport interest inventory (SII): Implications for
measuring unique consumer motives at team sporting events.
International Journal of Sports Marketing & Sponsorship, 3, 291 -316.
Funk, D. G., Mahony, D. ¥., & Ridinger, L L. (2002). Characterizing consumer motivation as individual difference factors: Augmenting the sport
interest inventory (SII) to explain level of sport. Sport Marketing
Quarterly, JI, 33-43.
Kahle, L. R., Kambara, K. M., & Rose, G. M. (1996). A functional model of
fan attendance motivations for college football. Sport Marketing
Quarterly, 5(4), 5]-60.
Maslow, A. H. (1943). A theory of human motivaXion. Psychobgkal Review,
50, 370-396.
Matusuoka, H. and Fujimoto, J. (2002). Foci offan's psychological commitment. Paper presented at the meeting of the North American Society of
Sport Management, Ganmore, AB: Canada.
Milne, G. R. & McDonald, M. A. (1999). Sport marketing: Managing the
There were limitations to the study, the largest being
exchange process. Sudbury, MA: Jones and Bartlett Publishers.
NGAA Football Attendance. (2001). NCAA football attendance: 2001 sumthat financial and geographic restrictions led to data
mary of all divisions. Retrieved December 8,2002, from
being collected at only one site for each of the four divihttp://www.ncaa.org/
sions and at those sites the sections were not randomNational Collegiate Athletic Association. Manual. (20010). The NCAA
ized. This does limit the results in terms of the
Manual - 2000. Indianapolis, IN: Author.
Oliver, R.L. (1999). Whence consumer loyalty?/ourna/o/MaritcftVig, 63,
generalization of results and making conclusive stateSpecial Issue, 33 44.
ments. In addition, future research needs to test Trail et
Pease, D. G. & Zhang,). J, (2001). Socio motivational factors affecting specal.'s (2003) model more extensively, as they noted.
tator attendance at professional basketball games. International Journal
However, testing the model to determine if it in fact
of Sport Management, 2,3i -59.
Robinson, M. R. & Trail, G. T. (2003). Motives and points of attachment:
does distinguish fans from spectators was not an
Differences between college football, men's basketball, and women's basketunequivocal success as support was mixed. Future studball spectators. Paper presented at the meeting of the North American
ies can be conducted to test the model to determine if
Society of Sport Management, Ithaca, NY.
the model holds in other environments (e.g., profesSloan, L R. (1989), The motives of sports fans. In I.H. Goldstein (Ed.).
sional sport, international competitions) and with other
Sports games and play: Social and psychological vievi^points {2"^ ed. pp.
175-240). Hillsdale, NI: Lawrence Eribaum Associates.
techniques. The present research does indicate that colTabachnick, B. G.. & Fidell, L. S. (1989). Using multivariate statistics (2"^*
lege marketers need to recognize that attendees do difed). NY: HarperCollins Publishers.
fer to some extent across divisions and perhaps the fans Tajfel, H. & Turner, |. C. (1986). The social identify theory of intergroup
are more prevalent at the NCAA Division I level and
behavior In S. Wordel & W. Austen (Eds.), Psychology of intergroup relations (pp 7-24). Chicago, IL: Nelson-Hall.
spectators are more prevalent at the lower levels.
Trail,
G. T, Anderson, D. F., & Fink,). S. (2000). A theoretical model of
However, all marketers should do their own market
sport spectator consumption behavior. International Journal of Sport
research to determine the characteristics of their own
Management, 1, 154-180.
attendees before creating marketing plans.
Trail, G. T, Fink, |. S., & Anderson, D. E (2003). Sport spectator consumption behavior. Sport Marketing Quarterly, 12, 8-16.
Trail, G.T.,& James, I. D. (2001). The motivation scale for sport consumption: Assessment of the scale's psychometric properties. Journal of Sport
Anderson, D. E & Stone, G. P. (1981). Sport: A search for community. In S.
Behavior, 24, 108-127.
L. Greendorfer & A. Yiannakis (Eds.), Sociology of sport: Diverse perspecTrail, G. T, Robinson, M., Gillentine, A., & Dick, R. (2003). Motives and
tives (pp. 164-172). West Point, NY: Leisure Press.
points of attachment: Fans versus spectators. Sport Marketing Quarterly,
Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman, S., &
12.217-227.
Sloan, L. R. (1976). Basking in reflected glory: Three (football) field
Wann, D. L. (1995). Preliminary validation of the sport fan motivation
studies. Journal of Personality and Social Psychology, 39, 406-415.
scale. The Journal of Sport & Social Issues, 19, 377-396.
Ded, E. L. (1971). Effects of externally mediated rewards on intrinsic motiWann, D. L, & Branscombe, N. R. (1990). Die-hard and fair-weather fans:
vation. Journal of Personality and Social Psychology, 18, 105-U5.
Effects of identification on BlRGing and CORFing tendencies. Journal of
Deschriver, T. D. (1999). Factors affecting spectator attendance at NCAA
Sport and Social Issues, 14, 103-117.
division II football games. International Sports Journal, 4,55-65.
Wann, D, L., & Branscombe, N. R. (1993). Sports fans: Measuring degree of
Fink, I. S., Trail, G. T., & Anderson, D. F. (2002). An examination of team
identification with their team. International Journal of Sport Psychology,
identification: Which motives are most salient to its existence?
24, 1-17.
International Sports Journal, 6(Summer), 195-207.
Wann, D. L., Brewer, K. R., & Royalty, I. L. (1999). Sport fan motivation:
Fisher, R. ]., & Wakefield, K. (1998). Factors Leading to Group Identification: A
Relationships with team identification and emotional reactions to sportField Study of Winners and Losers. Psychology & Marketing, i5( 1), 23-40.
ing events. International Sports Journal, 3(Summer), 8-18.
Conclusion
References
52 Volume 14 • Number 1 • 2005 • Sport MarHeting Quarterly
Wann, D. L, Royalty, J., & Rochelle, A. R. (2002). Using motivation and
team identification to predict sport fans' emotional responses to team
performance. Journal of Sport Behavior, 25, 207-215.
Zillmann, D., Bryant,)., & Sapolsky, B. S. (1989}. Enjoyment from sport
spectatorship. In J.H. Goldstein (Ed.), Sports, games, and play: Social &
psychological viewpoints. (2nd ed., pp. 241-278). Hillsdale, N.J: Lawrence
Erlbaum Associates.
Zillmann, D., & Paulus, P. B. (1993). Spectators: Reactions to sports events
and effects on athletic performance. In R. N. Singer, M. Murphey, & L.
K. Tennant (Eds.), Handbook of Research on Sport Psychology (pp. 600619). New York: Macmillan Publishing Company.
Volume 14 • Number 1 • 2005 • Sport MarHeting Quarteriy 53