Fantasy sport consumer segmentation: an investigation into the differing consumption modes of fantasy football participants.
Dwyer, Brendan ; Drayer, Joris
Introduction
Sport fandom is one of the preeminent leisure activities in our
society today. However, the sport marketplace has grown to a point
wherein sport fans have numerous opportunities and outlets in which to
spend their limited amounts of time and money. As a result, contemporary
sport consumption has evolved to include several activities including
event attendance, television viewership, and publication subscriptions,
both online and in print. Among these means of sport consumption is
fantasy sport participation.
Defined as an ancillary sport service heavily associated with
real-world sport statistics, fantasy sport participation is primarily an
online activity that is completely customizable, interactive, and
involves nearly every major professional sport, from the National
Football League (NFL) to NASCAR. Recently, the pastime has grown into a
highly popular activity for all types of sport fans. According to the
Fantasy Sport Trade Association (FSTA, 2008), nearly 30 million people
over the age of 12 play fantasy sports within the United States and
Canada. In addition, the FSTA estimates $800 million is spent directly
on fantasy sports products and services each year while an additional
$3.5 billion is spent on media products and services related to the
activity.
For sport marketing researchers, fantasy sport participation also
has the potential to influence several well-researched constructs within
sport consumer behavior inquiry. That is, while the majority of previous
research has focused on an individual's favorite or most-preferred
team (Mahony & Howard, 1998; Mahony & Moorman, 1999; Mahony,
Madrigal, & Howard, 2000; Trail & James, 2001), fantasy sport
participation has the potential to add another layer to contemporary
sport consumption due to its enhanced focus on individual players and
statistics. For instance, a typical fantasy football owner manages 10 to
12 heterogeneous NFL players. Each week this owner competes against an
additional 8 to 10 players. As a result of these combined competitive
interests, this participant may have a curiosity in nearly every NFL
game played each weekend. This phenomenon has the potential to create a
psychological paradox for individuals with a vested interest in their
favorite NFL team. That is, with a limited amount of time and money to
consume NFL football, the widened scope of fantasy football
participation has the potential to dilute one's attachment to their
favorite team. Given this intriguing circumstance, the aim of this study
was to investigate the differing modes of sport consumption exhibited by
fantasy sport participants.
As an activity, however, fantasy sport relies heavily on sport
media consumption as opposed to traditional forms of sport consumption
(i.e., event attendance, merchandise acquisition, etc.). While the
traditional forms still account for a considerable amount of a sport
organization's income, the significance of sport media as a revenue
stream should not be underestimated. In fact, "trends of escalating
consumption via media continue to indicate attendance is becoming less
central to an organization's profitability" (Pritchard &
Funk, 2006, p. 316). Moreover, the gargantuan amount of revenue
generated from television rights fees have driven teams to continually
focus on generating large, committed television audiences. As a result,
quickly evolving technological trends such as high definition
televisions, surround sound stereo, and slow-motion replays have created
a mediated sport product that rivals an in-stadium experience. For
fantasy sport participants, up-to-the-minute scores and statistics,
comprehensive pre-game news and analysis, and league-wide, on-demand
network services such as DirecTV's NFL Sunday Ticket and Major
League Baseball's MLB.TV also add to the allure of mediated sport
consumption.
Thus, to account for the aggressive media consumption habits of
fantasy sport participants (Comeau, 2007; Woodward, 2006) and the
possibility of behavioral changes with regard to a participant's
favorite team, the current study investigated the relationship between
favorite team-specific and fantasy team-specific media consumption.
Fantasy football participants were examined due to the game's
popularity and its designation as the gateway activity to all fantasy
sports (FSTA, 2008). A framework was proposed with four modes of
behavior (see Figure 1): (a) light consumption, (b) favorite
team-dominant consumption, (c) fantasy team-dominant consumption, and
(d) heavy consumption. Each mode holds a different pattern of NFL media
consumption, and in order to validate this proposed framework and
investigate unique differences and similarities in NFL-related product
usage, this study performed group contrasts of related factors of sport
fandom and fantasy participation to determine if the consumption modes
were theoretically distinct. Specifically, the following two research
questions were developed:
RQ1: Among fantasy football participants, do differing high/low
mixes of fantasy team and favorite NFL team consumption identify
distinct modes of NFL media consumption behavior?
RQ2: Based on these modes of behavior, are there any social,
attitudinal, or additional consumption behaviors (attendance-related and
mediated) that differ significantly across the modes?
In addition to answering the research questions, the current
article provides theoretical and practical implications for sport
marketers with respect to the consumption behavior of fantasy sport
participants. Initially, however, the concepts of contemporary sport
consumption, sport media consumption, and fantasy sports are discussed
to lay the groundwork for this study's data analysis and results.
The current study looked to extend and confirm these findings with a
larger sample of fantasy football participants.
Review of Literature
Contemporary Sport Consumption
Previous sport management literature has categorized sport
consumption into participation in competitive, nature-related, and
fitness activities as well as spectatorship in the form of event
attendance, television viewership, and reading of sport publications
(Shohlan & Kahle, 1996). The distinction between the various forms
of spectatorship is important as some of the most highly involved sport
fans rarely attend games. Given the enhanced accessibility via
televisual and electronic media communication, these fans continue to
practice the traditions associated with being an avid supporter, and,
thus, require the same amount of attention as event attendees.
In addition, within the last two decades, the amount of televised
and new media sport programming has exploded. For instance, in Bryant,
Brown, and Cummins' (2004) week-long analysis of broadcast and
basic cable programming during June 2004, 532 sports programs were
listed, adding up to 38,675 minutes, or nearly 645 hours, of sport
content. Given that there are only 168 hours in a week, it is safe to
say that sport consumers have numerous viewing choices. In addition, the
recent flood of sports-related websites epitomizes the legitimacy of
sport in the realm of new media services (Boyle & Haynes, 2003).
After search engines, the most frequently visited sites on the web were
those that offered some kind of entertainment and sports (Ferguson &
Perce, 2000). Furthermore, online betting and fantasy sports are two of
the fastest growing areas in terms of interactivity, sports, and the
Internet (Boyle & Haynes, 2003). Interestingly, previous definitions
of sport consumption have failed to include any form of interaction with
sport as a component of consumption, and with the sudden increase in
social media applications such as Facebook, message boards, Twitter, and
blogs, sports fans have the ability to actively interact with sport
products at a level unknown to them just a decade ago (Seo & Green,
2008). Fantasy sport participation, though not defined as a social media
activity, contains a strong social component (Farquhar & Meeds,
2007), and, thus, should be added to this group of interactive forms of
sport consumption.
Overall, the rapid growth of televised and new media sport content
has created several additional means for sport consumption (Sullivan,
2006). For sport marketers and consumer behaviorists, this has created
an additional avenue to forecast sport consumer behavior through highly
developed attitudes. In 2006, researchers Pritchard and Funk
investigated the symbiotic and substitution relationship between media
use and event attendance. According to the authors, the most interesting
facet of the study was the information provided about the media-dominant
consumer stating that these patrons are "more likely to purchase
team-related merchandise, view media advertising and promotions, and are
as involved with the sport as the 'heavy' consumer" (p.
316). Given the previously mentioned connection between fantasy sports
and media, the current
study aimed to examine an important portion of the media-dominant
sport population.
Fantasy Sport Participation
What was once considered a niche hobby for statistical fanatics,
the activity of fantasy sport participation has blossomed into a
lucrative and highly influential industry that has encapsulated the most
active of sport fans. For instance, a recent survey of participants
conducted by Ipsos Public Affairs indicated that fantasy players are
stronger consumers of the major product and service categories than the
average sport fan and the general population, as a whole (Fisher, 2008).
In addition to this robust consumption behavior, the average fantasy
participant continually represents corporate America's most-coveted
demographic (Fisher, 2008; FSTA, 2008). As a result, the industry has
come to demand serious consideration from sport marketing practitioners
and researchers, alike.
Unfortunately, the scholarly literature in the area of fantasy
sports is still underdeveloped (Lomax, 2006). Previous studies have
examined gambling associations, masculinity issues, and communication
(Bernhard & Eade, 2005; Davis & Duncan, 2006; Shipman, 2001). In
2007, researchers Farquhar and Meeds identified the following
motivational factors for fantasy sport participation: surveillance,
arousal, entertainment, escape, and social interaction. The study
concluded that two perceived gratifications of participating in fantasy
sports, arousal and surveillance, led highly involved participants to
believe they "got more out of fantasy sports when they put in more
time and money" (Farquhar & Meeds, p. 1217).
Researchers Drayer, Shapiro, Dwyer, Morse, and White (2010)
examined the relationship between fantasy football participation and the
consumption of NFL products and services. Following a theoretical
framework focused on the attitude and behavior relationship, the authors
proposed that fantasy participants created new perceptions of the NFL
through fantasy football. The redefined NFL broadened their consumption
behavior of associated products and services, and the continually
evolving outcomes of NFL competition influenced both a
participant's favorite team norms and fantasy-specific perceptions.
In all, the authors determined that fantasy sport participation has
created a new, highly engaged sport fan with a redefined interest in
professional football. In addition, fantasy sport-related consumption
behavior has the potential to be more diverse and league-wide as opposed
to singularly team-specific (Drayer et al., 2010).
Methods
Survey Development and Research Design
Survey items were developed based on a comprehensive literature
review of consumer behavior antecedents, media consumption, and fantasy
sport research, namely the qualitative results from Drayer et al.
(2010). After items were created, the questionnaire was analyzed by
three independent investigators. These evaluators, from a mid-sized
university in the southern United States, suggested minor alterations to
wording and question order. In all, the instrument was deemed
satisfactory and suitable for the intended inquiry. The following
sections highlight the selection of items and rationale for inclusion.
Media consumption measures: Fantasy team and favorite NFL team. In
line with previous leisure research (Backman & Crompton, 1991;
Pritchard & Funk, 2006), this study employed an orthogonal research
design. That is, this framework proposes four types of media consumption
formed as a function of two factors--fantasy team and favorite team
media consumption. Six statements regarding the frequency of sport media
consumption relative to both one's fantasy team and favorite team
were used to assess an individual's behavior. Given the breadth of
NFL programming and based on previous sport media consumption studies
(Drayer et al., 2010; Pritchard & Funk, 2006), respondents were
asked to indicate the number of hours per week, from zero to more than
20, spent following both teams through several mediums, including
newspapers, magazines, the Internet, radio, event programming, and sport
journalism programming.
Based on the results of these 12 questions, a principal component
analysis (PCA) with promax rotation was implemented in order to
determine the number of dimensions and obtain a more complete
understanding of the underlying structure of the data. A PCA is
generally used when the research purpose is data reduction (Tabachnick
& Fidell, 2007). Thus, the analysis was run to refine the 12 items,
provide parsimony, and validate the proposed framework by establishing
two distinct factors. Once again in unison with previous media and
consumption studies (Backman & Crompton, 1991; Pritchard & Funk,
2006), median scores for each factor were then used to construct the
four groups of differing consumption.
Other contrast measures. After constructing the four different
modes of consumption, three group contrasts (ANOVA, MANOVA, chi-squared
tests) were instituted to determine the distinctiveness of each mode.
First, a refined version of Mahony, Madrigal, and Howard's (2000)
Psychological Commitment to Team (PCT) scale was used to assess a
participant's loyalty toward their favorite NFL team. The concept
of psychological commitment commonly represents the attitudinal
component of loyalty (Backman & Crompton, 1991; Pritchard, Havitz,
& Howard, 1999). The PCT is a psychometrically sound instrument that
specifically emphasizes the resistance of changing preference toward a
particular professional sport team (Mahony et al., 2000). Given a
fantasy participant's enhanced focus on a group of heterogeneous
individual NFL players and limited amount of time and money to consume
sport, it was logical to assume differing levels of fantasy football
consumption may affect one's loyalty to their favorite team. Thus,
the PCT scale was added to compare and contrast team loyalty differences
between the four distinct consumption modes and to provide additional
validation of the proposed framework.
Next, five statements evaluating interest in the individual players
that make up the respondent's fantasy team were used to measure
fantasy player attachment. Attachment is a well-established and heavily
researched antecedent of sport consumption (Robinson & Trail, 2005).
However, the majority of previous sport spectator research with regard
to attachment has primarily focused on teams, not individual players.
Therefore, five statements were derived to measure attachment to
individual fantasy football players. The five statements were developed
following a review of attachment literature and based on the Drayer et
al. (2010) results. Given that this was the initial use of this scale,
the dimensionality, internal consistency, and convergent validity were
assessed. Similar to the inclusion of the PCT scale, the distinctiveness
of each consumption mode with regard to this construct provided further
attitudinal contrasts between groups and validation of the proposed
framework.
Ten items were then used to assess the frequency of NFL gameday
consumption. The findings of Drayer et al. (2010) suggested that fantasy
football participation resulted in enhanced mediated sport consumption
through a redefined NFL where individual player statistics via real-time
updates were as important as traditional game outcomes. In addition,
previous research with regard to fantasy sport motives indicated that
social interaction and the need to communicate with competitors was a
significant connection point (Farquhar & Meeds, 2007). Therefore,
respondents were asked to indicate, on a five-point Likert-type scale,
the frequency in which they consume the following products and services
on gamedays: (1) pre-game shows, (2) post-game shows, (3) Monday Night
Football (MNF), (4) Sunday Night Football (SNF), and (5) other sport
journalism shows. In addition, participants were asked to reveal the
frequency in which they: (6) attended a game, (7) went to a
bar/restaurant to specifically follow the NFL, (8) checked the Internet
for scores and statistics, and (9) used a cell phone to communicate with
fantasy league members, including the use of (10) text messaging services. Lastly, demographic contrasts such as age, income, education,
and years participated were interpreted through a chi-squared analysis.
Ultimately, these items were analyzed in order to discover the
distinctiveness of each consumption mode and validate the proposed
framework.
Participants and Procedures
Data were collected online from individuals who visited two popular
fantasy sports web sites (ESPN.com and CBSsports.com). These sites were
selected due to their popularity and level of interaction between
participants. Currently, ESPN.com and CBSsports.com rank second and
third in terms of unique fantasy football users with 3.37 million and
3.24 million participants, respectively (Nielsen Media, 2006). Fantasy
football participants were surveyed due to the game's popularity
and its designation as the gateway activity to all fantasy sports (FSTA,
2008). Specifically, fantasy football message boards were utilized to
attract respondents.
[FIGURE 1 OMITTED]
Of the 2,536 individuals who viewed the initial postings, 509 began
the survey. Of these, 36 were excluded from analyses because they
indicated that they were less than 18 years old, and 167 were excluded
because they failed to complete the questionnaire. The 306 respondents
who remained represented a 12.1% completion rate. The sample examined in
this study was younger than previously studied samples, but was still
representative of the fantasy sport demographic with regard to gender,
education, income, and ethnicity (FSTA, 2008). In addition, the use of
message boards to solicit participants may be considered a limitation of
the study due to the type of participant that uses these platforms. That
is, while previous research has determined that self-selected
respondents participate because of ease, accessibility, and online
status (Walsh, Kiesler, Sproull, & Hesse, 1992), the perception of
highly involved users may represent a sampling bias. Further research
into the exact typology is required and additional research on fantasy
football participants should utilize other solicitation means.
Demographic characteristics are depicted in Table 1.
Analysis and Results
Using the responses from the 306 fantasy football participants, a
principal component PCA with promax rotation was performed on the 12
mediated sport consumption items. In line with our hypothesis, two
factors, favorite team (factor 1) and fantasy team consumption (factor
2), were retained based on a variety of criteria. Specifically, the
Kaiser criterion, which considers all eigenvalues greater than one as
common factors (Table 2; Kaiser, 1970), a Scree-Plot test, and
interpretability suggested two factors (Tabachnick & Fidell, 2007).
However, following scale refinement procedure and consistent with
previous research (Tabachnick & Fidell, 2007; Drayer et al., 2010),
six items were removed due to poor factor loadings (less than of equal
to) 0.3). The items were newspaper, magazine, and radio consumption for
both fantasy team and favorite team. Item deletion improved the internal
consistency of the model evident with the strong alpha coefficients for
both factors (favorite team, [alpha] = 0.83; fantasy team, [alpha] =
0.85). The item deletion, however, may signify an interesting finding
for sport marketers and media companies. According to the FSTA (2008),
over 98% of all fantasy sport leagues reside completely online, and it
appears this residency has kept participants online. That is, the
inconsistent consumption rates of print publications (e.g., magazines
and newspapers) may suggest fantasy participants are consuming the
sizeable amount of fantasy sport content currently available on the web
in a more consistent manner. Similarly, the industries of radio
broadcasting and newspapers, in general, have struggled since the late
1990s and the rapid accent of Internet usage (Freire, 2007; Morton,
2007). This may explain some of these unanticipated results. Regardless,
the remaining items and factors, including loadings and alphas, are
detailed in Table 2. With reliable factors in hand, the next step was to
identify the (high/low mix) modes proposed in Figure 1.
Research Question 1
Each of the four modes of consumption was produced as a function of
the two factors (Tabachnick & Fidell, 2007). Similar to previous
orthogonal research and work on consumer patronage (Backman &
Crompton, 1991; Mahony et al., 2000; Pritchard & Funk, 2006), four
modes of consumption were plotted. Median factor scores for
fantasy-dominant and favorite team-dominant media usage identified high-
and low-consumption mixes for each mode.
Once the modes were identified, the next step determined whether
they constituted the four unique forms of consumption proposed in the
framework displayed in Figure 1. To begin with, MANOVA results shown in
Table 3 examined whether the median-mix procedure formed distinct
behavioral groupings. Previous consumer behavior studies have used a
similar median-mix approach to understand consumption (Backman &
Shinew, 1994; Pritchard & Funk, 2006; Warrington & Shim, 2000),
and this procedure was performed as an internal validity check of the
proposed framework. The results suggest average fantasy team media usage
for light consumption and favorite team-dominant spectators ranged from
less than one hour to two hours per week, which was significantly less
(F = 68.8, p < .001) than heavy and fantasy-dominant fans (3 to 12
hours). Favorite team media use also differed across the groups (F =
130.1, p < .001), with follow-up Scheffe t tests demonstrating that
the favorite team-dominant and heavy groups consistently reported
greater media use with respect to the participant's favorite NFL
team. In addition, the results indicated consumption of televised
programming, both event and sport journalism, were significantly
different for individual groups as well. Ultimately, the results in
Table 3 show that the high/low median mix did generate forms of
consumption consistent with Figure 1.
Research Question 2
Next, a series of analyses were conducted to determine if any
social, theoretical, or gameday consumption behaviors differed across
the groups mentioned above. Previous research has suggested that fantasy
participants show a level of attachment to the heterogeneous group of
players that make up their fantasy team (Drayer et al., in press;
Farquhar & Meeds, 2007). Consequently, a measure of player
attachment was used to compare the modes of consumption (see Table 4).
However, factorial validity, consistency, and convergent validity were
assessed first. In order to investigate the dimensionality of the
construct, a PCA was conducted on the attachment scores. One factor was
discovered with factor loadings greater than .70 for each of the five
items. In addition, the attachment construct scores were deemed
internally consistent with an Alpha coefficient of .78 and convergently
valid with an Average Variance Extracted (AVE) score of .56. With
respect to research question two, mean scores for the attachment
construct indicated differences across the groups and further validate
the proposed framework. Specifically, fantasy-dominant consumers
indicated stronger levels of attachment to individual players than light
and favorite team-dominant participants (F = 9.9, p < .001). While
the results are theoretically valid and reliable, this was the first use
of the scale; therefore, it may be a potential limitation of the study
and further research is required in this area.
With respect to team loyalty, the mean commitment scores of Mahony
et al.'s (2000) PCT scale were examined, and the results showed
significant differences across the groups. Favorite team-dominant and
heavy consumers indicated higher levels of psychological commitment
toward their favorite NFL team than fantasy-dominant and light
participants (F = 11.1, p < .001).
In addition to the theoretical measures applied above, a series of
gameday behaviors were analyzed to determine any differences among
fantasy participants. The results indicated that 9 of the 11 consumption
activities examined were significantly different between
fantasy-dominant/heavy consumers and favorite team-dominant/light
consumers. To illustrate, fantasy-dominant/heavy consumers indicated
watching a significantly greater amount of the five televised programs
examined (pre-and post-game shows, SNF, MNF, and cable sports channels).
Additionally, these same consumers signified going to a restaurant or
bar more often to watch NFL games and spent more time communicating with
league members via phone calls and text messaging. Lastly and similar
with previous research (Drayer et al., 2010, the results indicated that
there were no significant differences across the groups with regard to
event attendance (F = 3.7, p = .167).
Finally, even though the attitudinal contrasts characterized the
modes in a theoretically consistent manner, further descriptive
contrasts were conducted. Demographic accounts are displayed in Table 5.
Results here note no significant difference across the groups in terms
of age of the participants (F = 2.2, p < .086), level of education
([chi square][21] = 20.9, p = .464), and household income ([chi
square][15] = 20.1, p = .168). However, an important discovery indicates
that heavy and fantasy-dominant consumers have more years of experience
playing fantasy football (F = 8.2, p < .001). Given the consumptive behavior of these two groups, this finding may speak to the potential of
fantasy football as a predictor of NFL consumption, for it appears
league-related consumption increases with years of experience.
Discussion
Due to the interactive qualities of fantasy sport and its place
within an online environment, fantasy sport users represent an important
portion of the media dominant sport consumer population. However, even
within this population, it appears that fantasy participants differ
significantly in terms of product usage and media consumption. The
present study provided an in-depth look into the differences (as well as
the similarities) among fantasy sport participants based on varying
levels of consumption and has provided sport marketers with important
information about this lucrative segment of consumers. The following
section details some of the key discoveries and provides insight for
future studies in this area.
Managerial Implications
Not surprisingly, favorite team-dominant participants indicated
that their psychological commitment remained with their favorite team.
They supported their team by spending time watching more programming
related to their favorite team instead of their fantasy team. However,
an important finding in this study is that participants that were
considered heavy users maintained their commitment to their favorite
team as opposed to the players on their fantasy team. They spent more
time watching programming related to their favorite team and reported
higher levels of psychological commitment to their favorite team than
fantasy-dominant participants. In other words, when given a choice,
heavy consumers will still choose to associate most strongly with their
favorite team instead of their fantasy team. This result, combined with
the finding that each type of consumer has similar levels of event
attendance, should reassure the NFL and its teams that fantasy football
is not negatively impacting fans' attitudes toward their favorite
team.
Additionally, the findings of the current study suggest that fans
who are more engaged in fantasy football watch more NFL games. While the
favorite-team dominant consumer may only watch for a single three-hour
block on Sunday afternoon, the fantasy-dominant consumer has an interest
in games throughout the league and therefore watches significantly more
NFL programming (see Table 3). Therefore, not only is fantasy football
not negatively impacting fans' attitudes toward their favorite
team, but it also serves as a tool to increase television viewership,
which should incentivize the NFL to promote fantasy football
participation to its fans.
An intriguing outcome of the current study centers on the
similarities between groups with regard to gameday consumption habits.
That is, with respect to television viewership and Internet and
telecommunications usage, the fantasy-dominant consumer is highly
similar to the heavy consumer. The group contrast results for nearly
each form of gameday consumption classified these two consumer segments
together. Interestingly, the favorite team-dominant and light consumers
were also often grouped together. This, perhaps, indicates the
importance of fantasy football participation as an advanced marketing
vehicle for the NFL. An enhanced interest in a group of NFL players as
opposed to a singular team appears to increase media consumption on a
variety of levels. Specifically, the findings of the current study
should incentivize the NFL to utilize different means of communicating
with its fan base, particularly the most highly dedicated fans and the
fantasy-dominant consumers.
However, as mentioned previously, there is a tendency for
fantasy-dominant consumers to have stronger levels of attachment to
individual players as opposed to their favorite team. While this group
of consumers is still in the minority, the results also showed that
fantasy-dominant consumers have played fantasy football for more years
than favorite team-dominant consumers. Perhaps as consumers play fantasy
football, they are slowly and continuously redefining how they consume
the NFL toward a fantasy-dominant perspective. A longitudinal study is
needed to examine whether or not psychological commitment and
consumption shifts over time from favorite team-dominant to
fantasy-dominant. In the end, the NFL should embrace both populations as
they each show a strong tendency to consume at high levels.
The fantasy-dominant consumer should also be more attractive to
restaurants, bars, football-related websites, and cellular phone
providers due their significantly higher levels of consumption.
Businesses in these industries should consider catering their message to
the fantasy-dominant consumer. While this study focused only on mediated
consumption, future studies should examine the differences in
consumption of other products and services to see if potential exists to
market specifically toward one group of consumers over the other. The
qualitative study by Drayer et al. (2010) suggested that fantasy
participation does not lead to purchases of NFL-related products. Thus,
there is potential for practitioners to bridge the gap between fantasy
sport participation and other forms of traditional sport consumption.
However, additional research with regard to this specific population of
sport fans is required.
Ultimately, this study shows that the advent of fantasy football
has created distinct groups of NFL fans that have vastly differently
attitudes and behaviors. The NFL and its constituents would be wise to
try to understand these populations as well as they can in order to
maximize the profit-potential of each group. Finally, this study also
focused specifically on existing fantasy players. Future research should
examine the attitudes and behaviors of these distinct groups of fantasy
players with NFL fans that do not play fantasy football.
References
Backman, S. J., & Crompton, J. L. (1991). Using a loyalty
matrix to differentiate between high, spurious, latent and low loyalty
participants in two leisure services. Journal of Park and Recreation
Administration, 9(2), 1-17.
Comeau, T. O. (2007). Fantasy football participation and media
usage. Unpublished doctoral dissertation, University of Missouri -
Columbia.
Drayer, J., Shapiro, S. L., Dwyer, B., Morse, A. L., & White,
J. (2010). The effects of fantasy football participation on NFL
consumption: A qualitative analysis. Sport Management Review, 13,
129-141.
Fantasy Sports Trade Association. (2008, July 7). Fantasy sports
industry grows to an $800 million industry with 29.9 million players.
Chicago: Jeff Thomas. Retrieved July 8, 2008, from
http://www.fsta.org/news/pressreleases/
Farquhar, L. K., & Meeds, R. (2007). Types of fantasy sports
users and their motivations. Journal of Computer-Mediated Communication,
12, 12081228.
Fisher, E., (2008, November 17). Study: Fantasy players spend big.
Street & Smith's SportsBusiness Journal, 11(29), 1-2. Retrieved
from http://www.sportsbusinessjournal.com/article/60598
Freire, A. (2007). Remediating radio: Audio streaming, music
recommendation and the discourse of radioness. Radio Journal:
International Studies in Broadcast & Audio Media, 5(2/3), 97-112.
Fullerton, S., & Dodge, R. H. (1995). An application of market
segmentation in a sports marketing arena: We all can't be Greg
Norman. Sport Marketing Quarterly, 4(3), 43-61.
Haley, R. I. (1968). Benefit segmentation: A decision-oriented
research tool. Journal of Marketing, 32, 30-35.
Kaiser, H. (1970). A second generation little jiffy. Psychometrika,
35(4), 401415.
Mahony, D. F., & Howard, D. R. (1998). The impact of attitudes
on the behavioral intentions of sport spectators. International Sports
Journal, 2(2), 96-110.
Mahony, D. F., & Moorman, A. M. (1999). The impact of attitudes
on intentions to watch professional basketball teams on television.
Sport Management Review, 2, 43-66.
Mahony, D. F., & Moorman, A. M. (2000). The relationship
between the attitudes of professional sport fans and their intentions to
watch televised games. Sport Marketing Quarterly, 9(3), 131-139.
Mahony, D. F., Madrigal, R., & Howard, D. R. (2000). Using the
Psychological Commitment to Team (PCT) scale to segment customers based
on loyalty. Sport Marketing Quarterly, 9(1), 15-25.
Morton, J. (2007). Facing the future. American Journalism Review,
29(2), 68. Retrieved from Academic Search Complete database.
Mullin, B. J., Hardy, S., & Sutton, W. A. (2007). Sport
marketing. Champaign, IL: Human Kinetics.
Nielsen Media (2006). Nielsen net ratings usage analysis. Retrieved
from the Fantasy Sport Association's Fantasy Football Market
Report.
Peterson, R. A., & Sharpe, L. K. (1973). Market segmentation:
Product usage patterns and psychographic configurations. Journal of
Business Research, 1(1), 11-20.
Pritchard, M. P., & Funk D. C. (2006). Symbiosis and
substitution in spectator sport. Journal of Sport Management, 20,
299-321.
Pritchard, M. P., Havitz, M. E., & Howard, D. R. (1999).
Analyzing the commitment loyalty link in service contexts. Journal of
the Academy of Marketing Science, 27(3), 333-348.
Robinson, M. J., & Trail, G. T. (2005). Relationships among
spectator gender, motives, points of attachment, and sport preference.
Journal of Sport Management, 19(1), 58-80.
Seo, W. J., & Green, B. C. (2008). Development of the
motivation scale for sport online consumption. Journal of Sport
Management, 22(1), 82-109.
Shohlan, A., & Kahle, L. (1996). Spectators, viewers, readers:
Communication and consumption in sport marketing. Sport Marketing
Quarterly, 5(2), 11-20.
Smith, W. R. (1956). Product differentiation and market
segmentation as alternative marketing strategies. Journal of Marketing,
21, 3-8.
Sutton, W. A., McDonald, M. A., & Milne, G. R. (1997).
Escalating your fan base. Athletic Management, February/March, 3-5.
Tabachnick, B.G., & Fidell, L.S. (2007). Using multivariate
statistics (5th ed.). Boston, MA: Pearson Education.
Trail, G. T., & James, J. D. (2001). The motivation scale for
sport consumption: Assessment of the scale's psychometric properties. Journal of Sport Behavior, 24(1), 108-127.
Walsh, J., Hesse, B., Kiesler, S., & Sproull, L. (1992).
Self-selected and randomly selected respondents in a computer network
survey. Public Opinion Quarterly, 56, 241-244.
Wind, Y. (1978). Issues and advances in segmentation research.
Journal of Marketing Research, 15, 317-337.
Woodward, D. (2005). A whole new ballgame: How fantasy sports has
evolved in the mass media. Unpublished doctoral dissertation, University
of Texas at Arlington.
Brendan Dwyer, PhD, is assistant director of student services and
outreach in the Center for Sport Leadership at Virginia Commonwealth
University. His research interests include issues related to sport
marketing and sport consumer behavior, specifically sport consumer
behavior of fantasy sport participants, issues in intercollegiate
athletics, and the financial management of college athletics.
Joris Drayer, PhD, is an assistant professor of sport and
recreation management at Temple University. His research interests
include topics related to ticketing in both the primary and secondary
ticket markets, as well as consumer behavior, particularly in relation
to fantasy sports participation.
Table 1.
Demographic Characteristics of the Study Sample (N=306)
% of % of
Variable Respondents Variable Respondents
Gender Education
Male 95.1% Less than High School 3.9%
Female 4.9% High School Graduate 12.1%
Age Some College 31.4%
18-24 34.0% College Graduate 33.7%
25-34 37.9% Technical School 3.6%
35-44 19.9% Graduate School 10.5%
Over 44 8.2% Rather Not Say 1.6%
Other 3.3%
Ethnicity Income
African American 1.3% Less than $25,000 11.4%
Asian 2.9% $25,000--$49,999 20.6%
Caucasian 88.9% $50,000--$74,999 19.3%
Hispanic 2.6% $75,000--$99,999 17.3%
Multiracial 2.0% $100,000 or more 16.0%
Rather Not Say 1.0% Rather Not Say 15.4%
Other 1.3%
Table 2.
Factor Analytic Results of the Consumption Measures (N=306)
Dimensions of Mediated Sport Consumption Factor 1 Factor 2
Fantasy Team Measures (a = 0.85)
# of hours following fantasy team via the 0.125 0.870
Internet
# of hours following fantasy team's live 0.143 0.874
games
# of hours following fantasy team via 0.395 0.763
televised programming
Favorite Team Measures (a = 0.83)
# of hours watching favorite team's live 0.766 0.230
games
# of hours following favorite team via 0.906 0.200
televised programming
# of hours following favorite team via 0.879 0.139
the Internet
Eigenvalue 3.311 1.276
% Variance Explained 55.191 21.274
Table 3.
Modes of Behavior (N=306)
Favorite
Light Fantasy Team
Consumption Behavior Consumption Dominant Dominant
(hrs/week) (n =79) (n =93) (n =55)
Favorite NFL Team *
Live Games 1.4 (cd) 1.8 (cd) 4.8 (abd)
Televised Programming 0.4 (cd) 0.7 (cd) 4.8 (abd)
Internet 0.8 (cd) 0.8 (cd) 6.2 (ab)
Fantasy Team *
Live Games 1.8 (bd) 7.6 (ac) 2.0 (bd)
Televised Programming 04 (bcd) 2.8 (acd) 1.2 (abd)
Internet 2.2 (bd) 7.6 (acd) 1.8 (bd)
Heavy
Consumption Behavior Consumption
(hrs/week) (n=79) F p
Favorite NFL Team * 130.148 .001
Live Games 6.6 (abc) 47.193 .001
Televised Programming 7.0 (abc) 138.289 .001
Internet 8.8 (ab) 133.476 .001
Fantasy Team * 68.771 .001
Live Games 10.0 (ac) 83.663 .001
Televised Programming 8.2 (abc) 96.105 .001
Internet 11.8 (abc) 80.148 .001
Post hoc Scheffe tests: (a) different (p < .05) from light group mean,
(b) different (p < .05) from fantasy-dominant group mean, (c)
different (p < .05) from favorite team-group mean, and (d) different
(p < .05) from heavy group mean.
* Main effects MANOVA design, with follow-up ANOVA tests.
Table 4.
Attitudinal and Behavioral Contrasts (N=306)
Favorite
Light Fantasy Team
Consumption Behavior Consumption Dominant Dominant
(frequency, 1-5) (n =79) (n =93) (n =55)
Attachment to Players 3.5 (ac) 2.9 (b)
Psychological Commitment
to Team 3.7 (cd) 3.6 (cd) 4.0 (ab)
Gameday Consumption
Pre-Game Shows 3.2 (bd) 4.1 (ac) 3.3 (bd)
Post-Game Shows 3.0 (bd) 3.8 (ac) 3.2 (bd)
Sunday Night Football 3.9 (bd) 4.6 (ac) 4.1 (bd)
Monday Night Football 4.1 (bd) 4.6 (ac) 4.1 (bd)
Sports News Channels 3.2 (bd) 4.1 (ac) 3.3 (bd)
Event Attendance 1.4 1.6 1.9
Bar 2.0 (bd) 3.3 (ac) 2.4 (bd)
Internet 4.1 (bd) 4.7 (ac) 4.2 (bd)
Phone Calls 2.0 (bd) 2.9 (ac) 2.1 (bd)
Text Messaging 1.7 (bd) 2.7 (ac) 1.8 (bd)
Heavy
Consumption Behavior Consumption
(frequency, 1-5) (n=79) F p
Attachment to Players 3.2 9.795 .001
Psychological Commitment
to Team 4.0 (ab) 11.055 .001
Gameday Consumption
Pre-Game Shows 4.4 (ac) 20.072 .001
Post-Game Shows 3.9 (ac) 10.863 .001
Sunday Night Football 4.6 (ac) 10.232 .001
Monday Night Football 4.6 (ac) 7.245 .001
Sports News Channels 4.3 (ac) 16.935 .001
Event Attendance 1.6 3.678 .167
Bar 3.4 (ac) 23.156 .001
Internet 4.7 (ac) 10.739 .001
Phone Calls 3.0 (ac) 12.665 .001
Text Messaging 3.0 (ac) 18.231 .001
Post hoc Scheffe tests: (a) different (p < .05) from light group mean,
(b) different (p < .05) from fantasy-dominant group mean, (c) different
(p < .05) from favorite team-group mean, and (d) different (p < .05)
from heavy group mean.
Table 5.
Demographic Contrasts (N=306)
Favorite
Light Fantasy Team
Consumption Dominant Dominant
Demographic Factors (n =79) (n =93) (n =55)
Age (Years) 32.1 29.8 28.7
Number of Years Played 4.3 (bd) 6.7 (ac) 4.39 (bd)
Education *
Less than High School 5 (3) 3 (4) 1 (2)
High School Graduate 6 (10) 14 (11) 6 (7)
Some College 28 (25) 20 (30) 18 (17)
College Graduate 25 (27) 37 (32) 18 (18)
Technical School 2 (3) 5 (3) 1 (2)
Graduate School 9 (8) 11 (10) 8 (6)
Income *
Less than $25,000 10 (9) 9 (11) 7 (6)
$25,000-$49,999 18 (15) 15 (20) 8 (11)
$50,000-$74,999 13 (14) 19 (19) 12 (10)
$75,000-$99,999 8 (13) 20 (17) 9 (9)
$100,000 or more 14 (12) 21 (16) 8 (8)
Heavy
Consumption F;
Demographic Factors (n=79) [chi square] (df) p
Age (Years) 28.4 2.221 .086
Number of Years Played 6.3 (ac) 8.197 .001
Education * 20.924 (21) .464
Less than High School 3 (3)
High School Graduate 11 (9)
Some College 30 (24)
College Graduate 23 (26)
Technical School 3 (3)
Graduate School 4 (8)
Income * 20.109 (15) .168
Less than $25,000 9 (9)
$25,000-$49,999 22 (17)
$50,000-$74,999 15 (15)
$75,000-$99,999 16 (14)
$100,000 or more 6 (13)
Post hoc Scheffe tests: (a) different (p < .05) from light group mean,
(b) different (p < .05) from fantasy-dominant group mean, (c)
different (p < .05) from favorite team-group mean, and (d) different
(p < .05) from heavy group mean.
* Cross-tabulated actual and (expected) counts shown: likelihood
ratios reported for y? tests.