Of Sports Pages and Bad Attitudes:
An Investigation of the Relationship between
Attention to Sports Media and Attitudes toward Women's Sports
Abstract
Telephone survey data (n = 397) are used to assess the relationship between
attention to sports media and attitudes toward women's sports, predicting that
greater attention to sports media is associated with negative attitudes toward
women's sports. The study draws on sports feminism, the elaboration likelihood
model, and Creedon's powerful, contingent, and contextual effects model. The
author finds little support for the hypotheses and discusses problems with
question wording, social desirability, and a possible change in sports media
content.
Paper submitted by
Lisa M. Weidman
Syracuse University Doctoral Student
225 Cambridge Street
Syracuse, NY 13210
(315) 423-7902
[log in to unmask]
Of Sports Pages and Bad Attitudes
Of Sports Pages and Bad Attitudes:
An Investigation of the Relationship between
Attention to Sports Media and Attitudes toward Women's Sports
Running Head: Of Sports Pages and Bad Attitudes
Paper submitted by
Lisa M. Weidman
Syracuse University Doctoral Student
225 Cambridge Street
Syracuse, NY 13210
(315) 423-7902
[log in to unmask]
Of Sports Pages and Bad Attitudes:
An Investigation of the Relationship between
Attention to Sports Media and Attitudes toward Women's Sports
INTRODUCTION
In the body of literature sometimes called sports feminism, numerous content
analyses of sports media have identified ways in which television sports
broadcasts and newspaper sports pages underrepresent and trivialize women's
sports. These studies have shown, for example, that women's sports are given far
less coverage than men's sports in newspapers (Theberge & Cronk, 1986; Klein,
1988; Alexander, 1994a), in magazines (Lumpkin & Williams, 1991; Rintala &
Birrell, 1984; Duncan & Sayaovong, 1990), and in television broadcasts of sports
events in which men and women compete in identical events (Bryson, 1987; Duncan,
Messner, Williams & Jensen, 1990; Alexander, 1994b). Other studies have revealed
ways in which female athletes and their endeavors have been trivialized,
marginalized, and sexualized by the sports media, especially in television
broadcasts (e.g., Duncan & Hasbrook, 1988; Messner, Duncan & Jensen, 1993;
Daddario, 1994; Halbert & Latimer, 1994).
The assumption that accompanies these studies is that media content that
ignores or trivializes women's sports has a negative influence on the thinking
of those who attend to the media. Jennifer Hargreaves (1994, p. 196) articulates
this assumption when she argues that "the relative neglect of women's sports and
the ways in which they are represented confirm for many people ideas that they
have already internalized from other experiences." She goes on to explain that
"the construction and marginalization of female sports provide a hidden, but
very powerful, message that they are less important than men's sports and that
men are keener to participate and naturally better suited to do so" (p. 196).
These negative messages about women's sports seem to be the concern of all of
the sports media researchers cited above. However, I am unaware of any attempt
to establish empirically a causal relationship between sports media messages and
negative attitudes toward women's sports. This is undoubtedly because such a
relationship would be very difficult to establish due to the interaction between
sports media messages, messages from other sources, and personal views.
Nonetheless, some support for the assumption might be found by comparing the
attitudes of those who rarely, if ever, attend to sports media with the
attitudes of the avid sports fan to see if there are any significant differences
between the two groups. Support might also be found by assessing the correlation
between the amount of attention a person pays to sports media and his or her
attitudes toward women's sports, particulalry if one were to control for certain
characteristics of the audience member that might modify or strengthen this
relationship. The present study attempts to do just that through survey
research. This research springs from three somewhat unrelated theories that I
have woven together because each can be seen to support the assumption under
investigation and because each contributes to an explanation for the
relationship between attention to sports media and attitudes toward women's
sports.
THEORY
Theoretical Perspective
This study was conceived from the perspective of sports feminism, which is a
movement to eradicate discrimination on account of biological sex in sports
(Hargreaves, 1994). Like most feminist theories, which are "grounded in a
concern about, and desire to effect change in, the subjugated status of women"
(Cirksena & Cuklanz, 1992, p. 18), sports feminism calls into question the
patriarchal nature of sport (Brown, 1986).
As with feminist theories, one should acknowledge the various strains of sports
feminism, as Hargreaves (1994) cautions:
Sports feminism is not a unified movement or idea, nor can its
forms easily be characterized as cultural, liberal, orthodox Marxist,
radical, or socialist, so that they would tie in with categories of
feminism found in the general debate about the causes of women's
subordination (p. 26).
However, two of the most pervasive strains of sports feminism, liberal and
separatist, do bear some resemblance to liberal feminism and radical feminism,
respectively. "Liberal sports feminism challenges historically acquired
inequalities in sports between men and women, but it is not a challenge to the
conventional character of modern sports or to the 'essential' nature of modern
capitalism and patriarchy" (Hargreaves, 1994, p. 27). Separatist sports
feminists argue for sex-based separation in terms of education, competition, and
governance (pp. 30-31).
The important contribution of sports feminism to sports sociology and, by
extension, to sports media research, according to Hargreaves (1994),
has been to uncover ways in which men's power over women in
sports has been institutionalized; it has provided a practical and
symbolic challenge to male privilege which has resulted in a general
recognition of gender as a basic category of analysis, and it has
raised consciousness about the complexities and contradictions of
gender relations in sports theory and practice (p. 26).
This study, using gender as a basic category of analysis, attempts to uncover
one way in which men's power over women in sport has been institutionalized
(through sports media messages) while taking into account some of the
complexities of the attitude change process in terms of the personal contexts
and experiences of media audience members.
Two Models of Media Effects
The relationship between mass media messages and audience attitudes has been
the subject of study for many decades. Dating back to the persuasion research of
Harold Lasswell in the 1930s and Carl Hovland in the 1940s and '50s, social
scientists have been trying to build a universal model to explain this
complicated relationship. One of the more recent models is the elaboration
likelihood model, developed by John Cacioppo and Richard Petty (1985). This
model has gained favor in the scholarly community because it is sufficiently
complex to account for a large variety of circumstances and persuasion outcomes.
The elaboration likelihood model (ELM) is primarily designed to explain the
cognitive process through which attitudes yield to persuasive messages (Petty &
Priester, 1994). Though the sports media messages that have been critiqued by
sports feminists are not necessarily intended to persuade, the ELM is useful
here because it does account for persuasion that occurs when the audience member
is not actively considering an attitude change and because it acknowledges a
number of variables that can intervene in the persuasion process.
The elaboration likelihood model identifies two "routes to persuasion" (Petty &
Priester, 1994, p. 98). The first, or "central route," to persuasion is direct
and "involves effortful cognitive activity whereby the person draws upon prior
experience and knowledge in order to carefully scrutinize all of the information
relevant to determining the central merits of the position advocated" (p. 98).
This process is said to be highly elaborated. The second, or "peripheral route,"
involves much less elaboration, as it does not require effortful evaluation of
the persuasive communication. Variables that interact in the persuasion process
include the person's motivation to process the message, the person's ability to
process the message, the person's initial attitude, and the quality of the
argument. Messages are processed peripherally "when a person's motivation or
ability to process the issue-relevant information is low" (p. 101).
This model, particularly the peripheral route, which is more susceptible to
subtle cues, is useful in explaining how repeated exposure to sports media
messages that glorify men's sports and demean women's sports might influence the
attitudes of those who receive these messages. It is also useful in that it
takes into account cognitive states (low interest, low comprehension) and
message characteristics (unconvincing, unclear) that might interfere with
attitude change. This could help to explain why some heavy users of sports media
might not have negative attitudes toward women's sports and vice versa. However,
as mentioned above, the elaboration likelihood model is really intended to
explain the process by which a single message may lead to a specific attitude
change, not how widely held social conceptions, such as sex roles, are
repeatedly reproduced and reinforced by the mass media.
Thus, for further insight into the relationship being investigated, we turn to
the powerful, contingent and contextual effects model proposed by Pamela J.
Creedon. This model "builds on the accumulated knowledge of the earlier
[effects] models" (Creedon, 1994, p. 12) and, to compensate for their inability
to reliably predict media effects, Creedon's model "integrates critical insights
about context" (p. 12), which have emerged from critical cultural and feminist
theories. "Both traditions [cultural theories and feminist theories] suggest
that to understand media effects one must understand the cultural contexts in
which the messages are received" (p. 12).
Thus, the powerful, contingent and contextual effects model states that the
media are powerful, but their power is contingent on many contextual factors,
such as race, ethnicity, class, gender, sexual orientation, ideologies, and
political, social, and economic structures. This model "accounts for the fact
that different individuals view the world differently based on the way that they
have experienced it" (p. 12). This model can be applied to attitude change
resulting from repeated exposure to a message, as well as to change resulting
from a single exposure. The model contributes to our expectations about the
relationship between negative media messages and negative attitudes in that it
cautions us not to expect such messages to affect all people in the same way or
to the same extent.
Hypotheses
The present study attempts to take into account some of the contextual factors
identified in Creedon's model and in the writings of the cultural and feminist
theorists while still trying to identify general patterns in the media exposure
and attitude equation, which would provide some support for the assumption of
sports feminists regarding the role of the media in reinforcing negative views
of women's sports and female athletes. To that end, I propose three hypotheses.
The first predicts a general relationship between attention to sports media and
attitudes toward women's sports.
H1) The more a person attends to sports media, the more
unfavorable are that person's attitudes toward women's sports.
This hypothesis broadly tests the sports feminism proposition that the negative
messages regarding women's sports that were identified in the content analyses
discussed earlier (that women's sports are uninteresting, unimportant, and
inferior to men's sports and that female athletes need not be taken seriously)
are pervasive and that they have a persuasive effect on audience members,
causing them to have negative attitudes toward women's sports. Specifically,
this hypothesis predicts that increased exposure to such messages will be
associated with increasingly negative views of women's sports. If such a
relationship exists, the hypothesis should be supported by the survey data.
The independent variable in Hypothesis 1 is attention to sports media. This
variable refers to the amount of time a person spends in an average week
attending to sports media; that is, reading the sports pages of a newspaper,
listening to sports news or live sports broadcasts on the radio, and watching
sports news or broadcasts of sporting events on television (both broadcast and
cable).
The dependent variable is attitudes toward women's sports. This variable is a
composite, or index, of several specific attitudes toward women's sports. These
are: the importance of women's sports as compared to men's sports; whether
women's sports are exciting; whether women enjoy sports competition; and whether
women are good athletes.
The remaining hypotheses involve the same independent and dependent variables
as the first hypothesis, but they also incorporate some of the context variables
advocated by the powerful, contingent and contextual effects model, specifically
those that may enhance or diminish the relationship between attention to sports
media and attitudes toward women's sports. These hypotheses predict that the
correlation will be more pronounced among people who have never played
competitive sports and people who do not have a close female family member who
plays competitive sports because, for people who have them, these personal
experiences could serve to negate or mitigate the effects of negative media
messages. These hypotheses and their theoretical linkages are as follows:
H2) Heavy users of sports media who have never played
competitive sports will have attitudes toward women's sports that
are less favorable than those of medium and light users of sports
media who have played competitive sports.
H3) Heavy users of sports media who do not have a close
female family member who plays competitive sports will have
attitudes toward women's sports that are less favorable than those
of medium and light users of sports media who do have a close
female family member who plays sports.
These hypotheses are based in part on theories of selective processes (e.g.,
Klapper, 1960), which state that "people tend to expose themselves to those mass
communications that are in accord with their existing attitudes and interests"
(Klapper, 1960, p. 19) and that "reinforcement is or may be abetted by
predispositions" (pp. 18-19). Additionally, the elaboration likelihood model
posits that one of the factors that can promote or inhibit persuasion is a
person's initial attitude. Thus, if a person already has favorable attitudes
toward women's sports based on his or her personal experiences, that person's
attitudes are less likely to be shaped or changed by negative media portrayals
of women's sports and vice versa.[1]
These two hypotheses are also based on active audience theories and their
conceptions of resistant readings. In other words, these hypotheses acknowledge
that people who have participated in organized, competitive sports and people
whose mother, sister, wife, or daughter participates in competitive sports may
filter out or resist the negative messages about women's sports said to permeate
the sports media. Their first-hand experience with competitive sports and with
women's and girls' participation in competitive sports may mitigate the
influence that the sports media have (assuming, again, that they do perpetually
represent women's sports negatively). Thus, a more negative set of attitudes
toward women's sports might be expected among people who have no personal
experience as athletes and among those who do not have a close female family
member who participates in competitive sports.
Finally, biological sex, level of education, and income are three more context
variables that might enhance or diminish the strength of the relationship
between attention to sports media and attitudes toward women's sports. These
were used as control variables in the statistical test of Hypothesis 1.
METHOD
Survey research was used to test these hypotheses. The researcher contributed a
number of items to a larger survey written by eight graduate students and the
instructor of an advanced survey research methods class. The study, a telephone
survey of adults living in a mid-sized city in the northeastern United States,
was conducted over a 14-day period[2] in the fall of 1996.
A CD-ROM telephone directory (SelectPhone, Northeast, 1997, first quarter) was
used to develop the sampling frame for the study. All telephone numbers within
the local calling area of the city of . . . were initially selected from the
CD-ROM database. From these, all business and duplicate telephone numbers were
eliminated, producing a list of 197,000 residential telephone numbers in the
city and its surrounding suburbs. This list was used as the sampling frame.
These 197,000 numbers were randomly ordered, based on a computer-generated list
of random numbers, and a sample of 1,150 numbers was selected and divided into
groups of 50 telephone numbers to create 26 replicates of 50 numbers each.
Further randomization within each household was achieved through the use of the
Kish method of respondent selection. This method was also used to help ensure
that older people and males would not be under-represented in the study.
A week prior to the start of the official survey, a pre-test of the instrument
was conducted on a Saturday morning and afternoon. The graduate students in the
survey research course interviewed 25 respondents drawn from a random sample of
(. . .)-area residential telephone numbers. The same interviewing techniques
that would be used in the actual survey were used in the pre-test. The data
collected from the pretest were examined, and based on this information, some
items on the instrument were eliminated and others were reworded. In addition,
new codes were created for open-ended items and items with an "other, specify"
category.
The revised instrument contained 120 items, reflecting the varied research
interests of the eight members of the survey research class. Eighteen of the
items on the instrument were used in the present study.
The interviewers for the actual survey were 44 graduate students of the (. . .)
School at (. . .) University. Fifteen of these interviewers were provided with
two hours of training before their first four-hour interviewing shift and an
additional hour of training subsequent to their first shift. The other 25 were
provided with 80 minutes of training prior to their first shift and no
subsequent training, though they did receive ongoing coaching from shift
supervisors. All interviewers were also provided with training manuals which
explained how to make initial contact with respondents, how to ask survey
questions in a consistent manner, how to persuade reluctant respondents, and how
to convert refusals. The manuals also included an explanation of acceptable
prompts and feedback comments that interviewers were allowed to use.
The eight graduate students in the survey research class served as supervisors
during data collection. The supervisors managed the paperwork, evaluated
interviewers, gave interviewers telephone numbers to call, verified 10% of all
completed questionnaires, and conducted interviews as time permitted.
Supervisors also made sure that at least one attempt was made to convert every
refusal into a completed survey.
After two days of interviewing, a coding and analysis committee reviewed the
first 101 completed instruments for additional coding possibilities and created
new codes as necessary. After the two-week survey period, the supervisors worked
together to code all of the completed instruments in accordance with the new
coding categories. When this task was completed, supervisors independently
entered the data into SPSS databases, resulting in six separate databases, which
were merged by the coding and analysis committee. Frequencies and descriptive
statistics were run on each variable, the data were reviewed, and obvious errors
in the database were corrected. These three steps were then repeated in order to
catch errors not noticed in the first iteration. In addition, a 10% sample of
each person's data entry work was examined and cleaned by the coding and
analysis committee.
Operational Definitions of Variables
Attention to sports media, the independent variable in all three hypotheses,
was measured by respondents' answers to three questions regarding their use of
sports media. These questions were worded as follows:
"How many days per week do you listen to live sports or sports
news on the radio?"
"How many days a week do you watch live sports or sports news on
TV?"
"How many days a week do you read sports?"
This last item was asked toward the end of a series of questions regarding the
frequency with which respondents read different sections of the newspaper; thus,
the wording of the question made sense in the context in which it was asked.
This series of questions was introduced as follows: "Now I would like to ask you
about different parts of the newspaper. Please tell me, in an average week, how
many days do you read . . ." The first item simply completed the question with
"world news." Respondents who said that they read the world news section of the
paper at least once a week were then asked, "On an average day, about how much
time do you spend reading world news?" They were then asked the same two
questions about national news, state news, and local news before being asked
about the sports pages.
Respondents' answers to these three days-per-week items were summed to create
an index of attention to sports media. The range, then, for this index was zero
(for those who said they do not attend to sports content delivered by any of the
three media) to 21 (for those who said they attend to all three media for sports
information every day). This ratio level index, though perhaps not as precise as
an index of minutes per week attending to sports media, was thought to work well
for three reasons: a) it gives a general idea of the attention that respondents
pay to sports media, b) it is sufficiently broad to reveal variance in the data,
and c) it is probably more accurate than a measure of minutes per week because
it eliminates the error involved with respondents trying to calculate an average
number of minutes for an activity that is generally not performed for an equal
number of minutes each day.
The dependent variable for each of the hypotheses, attitudes toward women's
sports, was measured by respondents' answers to four survey items. These items
involved statements about women's sports followed by Likert scale response
options. The statements used in the study were based on attitudes said to be
engendered by the negative and nonexistent media coverage of women's sports,
particularly those articulated by Hargreaves (p. 196, 1994), as discussed
earlier. The items were introduced on the survey as follows:
For the next several questions, please consider your answers in
terms of the sports world and indicate the degree to which you agree
or disagree with each statement. Please say whether you strongly
agree, agree, are neutral, disagree or strongly disagree after I read
each statement.
The items appeared on the instrument in the following order.
"In general, women's sports are not as important as men's
sports."
"In general, women's sports are not exciting."
"Most women don't enjoy athletic competition."
"Most women are not good athletes."
After reading each statement, interviewers read a prompt ("Would you say you .
. .") and the list of response options. Responses were initially coded 5 for
strongly agree, 4 for agree, 3 for neutral, 2 for disagree, and 1 for strongly
disagree. Because the items were all phrased negatively, the most negative
attitudes were indicated by the highest numbersD5 for strongly agree and 4 for
agree. This runs counter to our common sense understanding of positive and
negative, so the responses were recoded in reverse order so that the most
negative attitudes were given low scores (1 for strongly agree, 2 for agree) and
the most positive attitudes were given high scores (4 for disagree and 5 for
strongly disagree with the negative statements). "Neutral" responses continued
to be coded as 3.
A note about the wording of the first attitude item: this item involves a
direct comparison of women's sports to men's sports for two reasons. First, I
wanted to give context to the word "important" so that respondents would not
think of the importance of women's sports in comparison to life-or-death issues,
such as war and famine, or to things needed for survival, such as food and
shelter, which would lead most respondents to disagree with the statement, and
2) I wanted to know if a comparison to men's sports would bring out more
agreement than statements that did not involve such a comparison. I did not
phrase all four items this way, though, because the assumption and hypotheses
being tested pertain specifically to the media portrayal of women's sports, not
of women's sports relative to men's sports, which is only one aspect of the
portrayal.
An attempt was made to combine the four attitude items into an additive index.
However, the reliability of this index was not high enough to justify its
use[3], so each hypothesis test was run four times, once for each attitude item.
The second hypothesis predicts the same negative relationship between attention
to sports media and attitudes toward women's sports as predicted in Hypothesis
1, but the prediction involves another independent variable: whether the
respondent has ever played competitive sports. The hypothesis predicts an
interaction between the respondent's personal experience with sports and the
attention he or she pays to sports media (the other independent variable), such
that people who have never played competitive sports and who pay a great deal of
attention to sports media will view women's sports more negatively than will
those who have played competitive sports and who pay little attention to the
sports media. Attention to sports media and attitudes toward women's sports were
measured as described above. Personal involvement in competitive sports was
determined by a survey item that read as follows:
"Do you now or have you ever played competitive sports, besides
in gym class?"
Respondents were categorized according to whether they said yes or no to this
question.
Similarly, Hypothesis 3 predicts an interaction between two independent
variables, attention to sports media and the experience of having a female
family member who participates in competitive sports. Attention to sports media
and attitudes toward women's sports were measured as described above. The other
independent variable, being related to a female athlete, was measured by a
survey item that asked:
"Do you have a close female family member who plays competitive
sports?"
As in Hypothesis 2, respondents were categorized by their response to this
yes-or-no question.
Statistics Used in the Analysis
Partial correlation coefficients and Analysis of Variance (ANOVA) were used to
test the hypotheses. Prior to running these statistical tests, however, the
reliability of each of the indices was assessed. The alpha for the index of
attention to sports media was .72, which was considered sufficiently high to
justify using this index in the statistical tests. As mentioned previously, the
alpha for the index of attitudes toward women's sports was rather low, only .59.
Thus, the hypotheses were tested on each of the attitude items separately,
rather than on the additive index of the four items.
RESULTS
The survey yielded 397 completed interviews and sixteen partial interviews. The
procedural response rate for the survey was 79%, and the demographic make-up of
the survey sample closely matched that of the local community with two important
exceptions: the sample contained more women than the community (58.7% as opposed
to 51.98%), and the sample contained more college educated people than the
community as a whole (77.6% of the sample had some college or more, whereas only
50.69% of the community has some college or more, according to 1990 U.S. Census
data). The average age of respondents was 46 years, and the average household
income was approximately $40,000.
Survey respondents said they pay relatively little attention to sports media.
The means for each of the three sports media use items was very low (1.7 days
per week for sports on the radio, 2.6 days per week for sports on TV, and 2.3
days per week for reading the sports section of the newspaper) (see Table 1).
These low means are partly due to a bimodal distribution wherein a large
percentage said they never attend to any sports media and another group, a
considerably smaller percentage, said they read, watch or listen to the sports
media every day. For example, 106 people said they never watch live sports or
sports news on TV, and 71 said they watch live sports or sports news on TV seven
days a week.
For the most part, respondents disagreed with the survey items regarding
attitudes toward women's sports. Over 75% of respondents disagreed or strongly
disagreed that women's sports are not exciting, that women don't enjoy athletic
competition, and that women are not good athletes. Somewhat more variability was
found in responses to the first item, "Women's sports are not as important as
men's sports" (see Table 1), perhaps because of the direct comparison to men's
sports, but even on this item, over 60% of respondents disagreed or strongly
disagreed with the statement. These descriptive statistics foreshadow the
outcomes of the statistical tests, which found very little support for the
hypotheses.
As predicted, the partial correlation tests revealed negative relationships
between attention to sports media and three of the attitudes toward women's
sports, indicating that, to some degree at least, the more attention respondents
paid to sports media, the more negative their attitudes toward women's sports.
However, the relationship was positive for the attitude item that read, "Women
don't enjoy sports competition," implying that the more attention respondents
paid to the sports media, the more likely they were to disagree with this
statement. This is the opposite of what was predicted by Hypothesis 1.
Furthermore, the correlations were not significant on any of the items except
for the item that read, "Women's sports are not exciting." The correlation
between attention to sports media and this attitude was weak (-.11, and ranging
from -.09 to -.11 when controlling for sex, education and income), but it was
significant at the 95% confidence level, even when controlling for sex,
education, income, and most combinations thereof (see Table 3).
For the analyses of variance, values for "attention to sports media" were
consolidated into low, medium, and high attention categories in order to
maximize cell size. The ANOVAs revealed no significant differences between the
attitudes of people who pay low, medium, or high attention to sports media
(Tables 4 through 7). In fact, the mean attitude scores of each of these groups
(low, medium and high attenders) were very similar.
Having played competitive sports did not seem to affect a person's attitudes
toward women's sports, as there were no significant differences between the
attitude scores of those who have played competitive sports and those who
haven't. Nor did having a female family member involved in competitive sports
seem to affect a person's attitudes toward women's sports, as no significant
differences were found between those who have a female athlete in the family and
those who don't.
Interestingly, the only aspect of the ANOVAs to achieve significance (p < .05)
was the three-way interaction of attention to sports media, personal experience
playing competitive sports, and having a female family member who plays
competitive sports on the attitude that women's sports are not exciting (see
Table 5). Why this attitude and none of the others achieved some significance in
both statistical tests is unclear. However, the consistent, albeit weak,
findings of both tests do lend some support for the hypothesis that increased
exposure to sports media is associated with less favorable attitudes toward
women's sports, specifically that women's sports are not exciting.
DISCUSSION
This study was conceived from the perspective of sports feminism, a movement
which seeks to eradicate sex-based discrimination[4] in the sports world. This
study attempted to find support for the assumption or proposition that the
sports media portrayal of women's sports (presumed to be incomplete and
unflattering, based on prior research findings) negatively affects audience
members' attitudes toward women's sports. This study also attempts to take into
account some of the complexities of the attitude change process in terms of the
personal contexts of audience members, as suggested by Creedon (1994).
The hypotheses being tested stated that there is a negative linear relationship
between attention to sports media and attitudes toward women's sports; in other
words, the more attention a person pays to sports media, the more negative their
attitudes toward women's sports will be. Two secondary hypotheses stated that
this relationship would be stronger for people who have no personal experience
playing competitive sports and for people who do not have a female family member
who plays competitive sports, as these people would have more limited real-life
exposure to women in athletics.
Though the findings of this study do not strongly support the hypotheses, they
do indicate that the assumption being tested may have some merit. For at least
one of the attitude items tested (the view that women's sports are not
exciting), the relationship between attention to sports media and attitude
toward women's sports was negative, and though not particularly strong, the
correlation was statistically significant.
A number of circumstances have conspired to prevent a confident pronouncement
that sports media messages do lead to negative attitudes toward women's sports.
First of all, this study, being a survey, could never fully support an effects
hypothesis; it could merely show a correlation between the variables.
Secondly, support for this assumption was only found in relation to one of four
attitude items. In order to make a strong argument for a correlation between
increased sports media attention and lowered esteem for women's sports, we would
need to see strong support for a number of attitude items or for a reliable
index of attitude items.
Thirdly, this study was probably not the best possible test of such a
correlation. The wording of the interviewers' introduction to the attitude items
or of the attitude items themselves may not have elicited respondents' true
attitudes toward women's competitive sports. In particular, the items may have
been too vague. Respondents may have thought of city league softball or other
recreational sports when the statements about "women's sports" were read to
them, whereas I had professional and semi-professional (i. e., college
scholarship) sports in mind when writing the statements. It's also possible that
some respondents answered for society as a whole, i.e., how our society views
women's sports, rather than offer their own personal opinions about women's
sports. of the attitude items may also have led respondents to answer
In addition, the negative phrasing and the sweeping generalities of the
attitude items, coupled with the fact that over half of the interviewers were
women, may have caused a social desirability confound. Perhaps male respondents
were reluctant to agreeDparticularly to a female interviewerDthat women's sports
are not important, exciting, et cetera, for fear of sounding like a chauvinist,
even if they did agree. Even female respondents may have been loathe to agree
with such statements, regardless of how they view women's sports, simply because
the statements were phrased so negatively and the generalizations were so broad.
Furthermore, this sample, which was clustered so disproportionately around low
attention and very positive attitudes, may not have contained the variability
necessary to plot a linear relationship. As mentioned above, respondents to this
survey reported fairly low attention to the sports media, and a majority
disagreed in response to each of the attitude items. Thus, one could argue that
heavy users of sports media and even people with negative attitudes toward
women's sports were underrepresented in the survey sample.
Of course, the findings of this study leave the door wide open for the argument
that the assumption is simply not true, either because the sports media do not
present (or no longer present) a barrage of negative messages about women's
sports or because even if they do, this has no effect on the attitudes of
audience members. Some might argue that coverage of women's sports has increased
in quantity and improved in quality in recent years, such that female athletes
and their contests are treated with as much respect and nearly as much adulation
as male athletes and their contests. Some of the author's own recent research
supports this argument (Weidman, 1997). In fact, this seems to be the most
plausible explanation for the generally positive attitudes toward women's sports
indicated in the survey, and such changes in the coverage of women's sports
would also help to explain why the predicted relationship was only weakly
supported in this study.
However, it still seems reasonable to think that when and if sports writers and
broadcasters do trivialize or marginalize women's sports, younger audiences in
particular get the message that women's sports aren't very important or popular
and that female athletes should not be taken seriously. I would argue that even
the minimal support found here is reason enough to investigate this assumption
further. Future research may involve surveys with attitude items that are more
specific and emphasize that the interviewer is requesting the respondent's
personal opinion, not that of society as a whole. Such items should be phrased
both positively and negatively in order to determine whether a social
desirability confound is at work. It might also be appropriate to include
attitudes not included in this study, such as whether or not it is appropriate
or attractive for women to play sports.
Future research might also be conducted in another market or at another time of
year in an attempt to include more medium and heavy sports media users in the
study, though I am not convinced that the market or the season are to blame for
the relative lack of attention to sports media among survey respondents. The
sports media offerings available in the survey area via television, radio, and
newspapers were plentiful and diverse during the two weeks of data collection;
the extremely popular university football team was playing locally once a week,
and professional and college football, major league baseball, and major league
hockey games from around the country were being broadcast and reported on
regularly at the time of the survey (the last week of September and the first
week of October).
Finally, other research methods should be employed in future investigations of
this problem. Experiments may get closer to establishing a causal linkage
between attention to sports media and attitudes toward women's sports, and
additional content analyses could assess whether the sports media are currently
sending out messages that trivialize and otherwise degrade female athletes and
women's sports. A longitudinal content analysis, in particular, might reveal
whether the sports media have changed over time in terms of how they portray
female athletes and women's sports.
TABLE 1. Means and standard deviations for sports media use, attitudes toward
women's sports, personal involvement in sports items, and demographic variables.
Variables
Mean
Std. Deviation
N
How many days per week do you listen to live sports or sports news on the
radio?a
1.70
2.57
391
How many days a week do you watch live sports or sports news on TV?a
2.60
2.56
393
How many days a week do you read sports?a
2.31
2.94
388
Attention to sports indexb
6.60
6.45
378
In general, women's sports are not as important as men's sports.c
3.42
1.25
394
In general, women's sports are not exciting.c
3.79
.97
391
Most women don't enjoy athletic competition.c
3.81
.98
386
Most women are not good athletes.c
4.03
.88
396
a Responses were coded from 0 to 7 days a week.
b The sum of scores from three sports media use questions. Cronbach's
alpha = .72
c Responses were coded: 1 = strongly agree, 2 = agree, 3 = neutral, 4 =
disagree, and 5 = strongly disagree.
TABLE 2. Percentages for sex and personal involvement with competitive sports
questions.
Variables
%
Sex
Male
Female
41.3
58.7
100%
(N = 397)
Do you now or have you ever played competitive sports, besides in gym
class?
Yes
No
53.7
46.3
100%
(N = 397)
Do you have a close female family member who plays competitive sports?
Yes
No
39.4
60.6
100%
(N = 396)
TABLE 3. Partial correlation coefficients for attention to sports media and
attitudes toward women's sports.
Variables correlated with attitudes toward women's sports
Control variables
Zero-order correlation coefficient
Partial correlation coefficient
"Women's sports are not as important as men's sports."
Attention to sports media*
Sex
Education
Income
Sex, education
Sex, income
Education, income
Sex, education, income
-.04
-.08
-.04
-.02
-.08
-.07
-.02
-.07
"Women's sports are not exciting."
Attention to sports media*
Sex
Education
Income
Sex, education
Sex, income
Education, income
Sex, education, income
-.11a
-.10a
-.11a
-.10a
-.10a
-.09
-.10a
-.09a
"Women don't enjoy sports competition."
Attention to sports media*
Sex
Education
Income
Sex, education
Sex, income
Education, income
Sex, education, income
.07
.06
.07
.07
.06
.05
.07
.06
"Women are not good athletes."
Attention to sports media*
Sex
Education
Income
Sex, education
Sex, income
Education, income
Sex, education, income
-.02
-.01
-.02
-.02
-.01
-.02
-.02
-.02
*Responses were coded from 0 to 21.
a p < .05
TABLE 4. Three-Way Analysis of Variance of attention to sports media and
personal experience with competitive sports on attitudes toward women's sports,
specifically whether women's sports are as important as men's sports.
Women's sports are not as important as men's.
Main effects and interactions
Means
Std. Devs.
F
sig
Main effect of attention to sports media
Low
Medium
High
3.42
3.40
3.43
.01
-.02
.01
.03
ns
Main effect of having played sports
Yes
No
3.40
3.44
.03
-.02
.18
ns
Main effect of having a close female family member who plays sports
Yes
No
3.45
3.40
-.02
.02
.10
ns
Interaction between attention to sports media and having played sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
3.55
3.32
3.42
3.38
3.28
3.74
2.12
ns
Interaction between attention to sports media and having a female family member
who plays competitive sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
3.52
3.38
3.49
3.33
3.38
3.47
.05
ns
Three-way interactions between attention to sports media, having played sports,
and having a female family member who plays sports
Low, yes, yes
Low, no, yes
Low, yes, no
Low, no, no
Medium, yes, yes
Medium, no, yes
Medium, yes, no
Medium, no, no
High, yes, yes
High, no, yes
High, yes, no
High, no, no
3.50
3.55
3.59
3.25
3.41
3.64
3.43
3.28
3.23
3.86
3.33
3.68
.02
ns
TABLE 5. Three-Way Analysis of Variance of attention to sports media and
personal experience with competitive sports on attitudes toward women's sports.
Women's sports are not exciting.
Main effects and interactions
Means
Std. Devs.
F
sig
Main effect of attention to sports media
Low
Medium
High
3.90
3.76
3.71
.11
-.04
.09
1.43
ns
Main effect of having played competitive sports
Yes
No
3.75
3.84
-.04
.05
.50
ns
Main effect of having a close female family member who plays competitive sports
Yes
No
3.81
3.78
.01
-.01
.39
ns
Interaction between attention to sports media and having played sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
3.86
3.94
3.69
3.82
3.71
3.69
.24
ns
Interaction between attention to sports media and having a female family member
who plays sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
4.05
3.84
3.73
3.77
3.69
3.72
.05
ns
Three-way interactions between attention to sports media, having played sports,
and having a female family member who plays sports
Low, yes, yes
Low, no, yes
Low, yes, no
Low, no, no
Medium, yes, yes
Medium, no, yes
Medium, yes, no
Medium, no, no
High, yes, yes
High, no, yes
High, yes, no
High, no, no
4.04
4.05
3.74
3.90
3.78
3.64
3.57
3.89
3.57
4.07
3.88
3.50
3.28
p < .05
TABLE 6. Three-Way Analysis of Variance of attention to sports media and
personal experience with competitive sports on attitudes toward women's sports,
specifically whether women enjoy athletic competition.
Women don't enjoy athletic competition.
Main effects and interactions
Means
Std. Devs.
F
sig
Main effect of attention to sports media
Low
Medium
High
3.80
3.69
3.94
-.02
-.13
.12
1.72
ns
Main effect of having played competitive sports
Yes
No
3.85
3.80
.00
.00
.03
ns
Main effect of having a close female family member who plays competitive sports
Yes
No
3.82
3.83
.02
-.02
.03
ns
Interaction between attention to sports media and having played competitive
sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
3.82
3.79
3.72
3.67
3.93
3.98
.04
ns
Interaction between attention to sports media and having a female family member
who plays sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
3.91
3.75
3.68
3.70
3.85
4.03
1.03
ns
Three-way interactions between attention to sports media, having played sports,
and having a female family member who plays sports
Low, yes, yes
Low, no, yes
Low, yes, no
Low, no, no
Medium, yes, yes
Medium, no, yes
Medium, yes, no
Medium, no, no
High, yes, yes
High, no, yes
High, yes, no
High, no, no
4.04
3.75
3.68
3.80
3.78
3.50
3.65
3.74
3.80
4.00
4.08
3.96
1.11
ns
TABLE 7. Three-Way Analysis of Variance of attention to sports media and
personal experience with competitive sports on attitudes toward women's sports,
specifically whether women enjoy athletic competition.
Women are not good athletes.
Main effects and interactions
Means
Std. Devs.
F
sig
Main effect of attention to sports media
Low
Medium
High
4.02
3.98
4.05
.00
-.04
.03
.11
ns
Main effect of having played competitive sports
Yes
No
4.08
3.96
-.04
.03
1.83
ns
Main effect of having a close female family member who plays competitive sports
Yes
No
3.98
4.05
.05
-.06
1.06
ns
Interaction between attention to sports media and having played competitive
sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
4.06
3.99
4.13
3.84
4.06
4.05
1.30
ns
Interaction between attention to sports media and having a female family member
who plays sports
Low and yes
Low and no
Medium and yes
Medium and no
High and yes
High and no
4.07
4.00
3.83
4.09
4.02
4.09
1.49
ns
Three-way interactions between attention to sports media, having played sports,
and having a female family member who plays sports
Low, yes, yes
Low, no, yes
Low, yes, no
Low, no, no
Medium, yes, yes
Medium, no, yes
Medium, yes, no
Medium, no, no
High, yes, yes
High, no, yes
High, yes, no
High, no, no
4.00
4.15
4.10
3.93
3.89
3.71
4.43
3.89
4.00
4.07
4.13
4.04
.09
ns
References
Alexander, S. (1994a). Newspaper coverage of athletics as a function of
gender. Women's Studies International Forum, 17, 6, 655-662.
Alexander, S. (1994b). Gender bias in British television coverage of major
athletic championships. Women's Studies International Forum, 17, 6,
647-654.
Berube, M. S., et al, (Eds.). (1982). American Heritage dictionary (2nd
ed.). Boston: Houghton Mifflin Company.
Brown, B. A. (1986). Sport science and the myth of women's sport.
Resources for Feminist Research, 15, 3, 31-33.
Bryson, L. (1987). Sport and the maintenance of masculine hegemony.
Women's Studies International Forum, 10, 4, 349-360.
Cacioppo, J. T., & Petty, R. E. (1985). Central and peripheral routes to
persuasion: The role of message repetition. In L. F. Alwitt & A. A.
Mitchell (Eds.), Psychological processes and advertising effects: Theory,
research and application. Hillsdale, NJ: Lawrence Erlbaum Associates.
Cirksena, K. & Cuklanz, L. (1992). Male is to female as _____ is to _____:
a guided tour of five feminist frameworks for communication studies. In L.
Rakow (Ed.), Women making meaning. New York: Routledge.
Cohen, G. L. (1993). Women in sport: Issues and controversies. Newbury
Park, CA: Sage.
Connell, R. W. (1990). An iron man: The body and some contradictions of
hegemonic masculinity. In M. A. Messner and D. F. Sabo (Eds.), Sport, men
and the gender order: Critical feminist perspectives (pp. 83-95).
Champaign, IL: Human Kinetics.
Creedon, P. J. (1993). The challenge of re-visioning gender values. In P.
J. Creedon (Ed.), Women in mass communication (pp. 3-23). Newbury Park,
CA: Sage.
Creedon, P. J. (1994). Women, media and sport: Creating and reflecting
gender values. In P. J. Creedon (Ed.). Women, media and sport: Challenging
gender values (pp. 3-27). Thousand Oaks, CA: Sage.
Daddario, G. (1994). Chilly scenes of the 1992 Winter Games: The mass
media and the marginalization of female athletes. Sociology of Sport
Journal, 11, 275-288.
DeFrantz, A. L. (1993). The Olympic Games: Our birthright to sports. In G.
L. Cohen, (Ed.), Women in sport: Issues and controversies, pp. 185-192.
Newbury Park, CA: Sage.
DiIorio, J. A. (1989). Feminism, gender, and the ethnographic study of
sport. Arena Review, 13, 1, 49-60.
Duncan, M. C. (1990). Sports photographs and sexual difference: Images of
women and men in the 1984 and 1988 Olympic Games. Sociology of Sport
Journal, 7, 22-43.
Duncan, M. C., & Hasbrook, C. A. (1988). Denial of power in televised
women's sports. Sociology of Sport Journal, 5, 1, 1-21.
Duncan, M. C., & Sayaovong, A. (1990). Photographic images and gender in
Sports Illustrated for Kids. Play & Culture, 3, 91-116.
Duncan, M. C., Messner, M. A., Williams, L., & Jensen, K. (1990). Gender
Stereotyping in Televised Sports. (Available from the Amateur Athletic
Foundation of Los Angeles, 2141 West Adams Boulevard, Los Angeles, CA
90018).
Eitzen, D. S. & Zinn, M. B. (1989). The de-athleticization of women: The
naming and gender marking of collegiate sport teams, Sociology of Sport
Journal, 6, 362-370.
Eitzen, D. S. & Sage, G. H. (1993). Sociology of North American sport.
Dubuque, IA: Wm. C. Brown Communications, Inc.
Gitlin, T. (1980). The whole world is watching. Berkeley, CA: University
of California Press.
Gramsci, A. (1971). Selections from the prison notebooks by Antonio Gramsci
. Q. Hoare & G. Nowell-Smith (Eds. and Trans.). New York: International
Publishers.
Graydon, J. (1983). But it's more than a game: It's an institution:
Feminist perspectives on sport. Feminist Review, 13, 5-16.
Halbert, C. & Latimer, M. (1994). "Battling" gendered language: An
analysis of the language used by sports commentators in a televised coed
tennis tournament. Sociology of Sport Journal, 11, 298-308.
Hargreaves, J. A. (1994). Sporting females: Critical issues in the history
and sociology of women's sports. London: Routledge.
Harris, J. F. (1991) Nobody knows her name: The depiction of the Black
female athlete in national sports telecasts. Unpublished master's thesis,
University of Texas, Austin.
Kane, M. J. & Greendorfer, S. L. (1994). The media's role in accommodating
and resisting stereotyped images of women in sport. In P. J. Creedon
(Ed.), Women, media and sport (pp. 28-44). Thousand Oaks, CA: Sage.
Klapper, J. T. (1960). The effects of mass communication. New York: Free
Press.
Klein, M.-L. (1988). Women in the discourse of sport reports.
International Review for the Sociology of Sport, 23, 2, 139-151.
Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people's choice:
How the voter makes up his mind in a presidential campaign. Duell, Sloan
and Pearce.
Lumpkin, A. & Williams, L. D. (1991). An analysis of Sports Illustrated
feature articles, 1954-1987. Sociology of Sport Journal, 8, 16-32.
Mandell, R. D. (1976). The first modern Olympics. Berkeley and Los
Angeles, CA: University of California Press.
Messner, M. A. (1988). Sports and male domination: The female athlete as
contested ideological terrain. Sociology of Sport Journal, 5, 197-211.
Messner, M. A., Duncan, M. C., & Jensen, K. (1993). Separating the men
from the girls: The gendered language of televised sports. Gender and
Society, 7, 1, 121-137.
Monmush, B. (Ed.). (1995). Television and video almanac. New York: Quigley
Publishing Company.
Petty, R. E. & Priester, J. R. (1994). Mass media attitude change:
Implications of the elaboration likelihood model of persuasion. In J.
Bryant & D. Zillman (Eds.), Media effects: Advances in theory and research.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Rakow, L. F. (1986). Rethinking gender research in communication. Journal
of Communication, 36, 4, 11-26.
Rintala, J. & Birrell, S. (1984). Fair treatment for the active female: A
content analysis of Young Athlete magazine. Sociology of Sport, 1, 231-250.
Rohrbaugh, J. B. (1979) Femininity on the line. Psychology Today, 13, 3,
30-33, 36, 40, 42.
Roloff, M. E. & Solomon, D. H. (1989) Sex typing, sports interests, and
relational harmony. In Wenner, L. (Ed.) Media, sports, and society (pp.
290-311). Newbury Park, Calif.: Sage.
Sabo, D. & Jansen, S. C. (1992). Images of men in sport media: The social
reproduction of gender order. In S. Craig (Ed.), Men, masculinity and the
media (pp. 169-184). Newbury Park, CA: Sage.
Shifflett, B. & Revelle, R. (1994). Gender equity in sports media
coverage: A review of the NCAA News. Journal of Sport and Social Issues,
18, 2, 144-150.
Theberge, N. & Cronk, A. (1986). Work routines in newspaper sports
departments and the coverage of women's sports. Sociology of Sport
Journal, 3, 195-203.
Tuchman, G. (1978). Introduction: The symbolic annihilation of women by
the mass media. In G. Tuchman, A. Daniels, & J. Benet (Eds.), Hearth and
home: Images of women in the mass media (pp. 3-38). New York: Oxford
University Press.
Tuchman, G. (1981). The symbolic annihilation of women. In S. Cohen & J.
Young (Eds.), The manufacture of news: Deviance, social problems, and the
mass media (pp. 169-185). Beverly Hills, CA: Sage.
Weidman, L. (1997, May). They called a game: The language television
sportscasters use to describe men's and women's NCAA basketball games.
Paper presented at the 47th Annual Conference of the International
Communication Association, Montreal, Quebec.
Wenner, L. A. (1989). Media,sports, and society: The research agenda. In
L. A. Wenner (Ed.), Media, sports, and society. Newbury Park, CA: Sage.
Williams, C. L., Lawrence, G. & Rowe, D. (1985). Women and sport: A lost
ideal. Women's Studies International Forum, 8, 6, 639-645.
Williams, R. (1991). Base and superstructure in Marxist cultural theory.
In C. Mukerji & M. Schudson (Eds.), Rethinking popular culture (pp.
407-423). Berkeley, CA: University of California Press.
[1] Such a person might also pay little attention to sports media because of the
negative messages about women's sports, which would be in keeping with
Hypothesis 1.
[2] Interviews were conducted from 9:00 A.M. to 9:00 P.M. Mondays through
Saturdays and from 1:00 P.M. to 9:00 P.M. on Sundays.
[3] Cronbach's alpha was .59.
[4] Sports feminism also seeks to eradicate discrimination based on race, class
and sexual orientation, but the eradication of sex-based discrimination is
thought by some to be its primary goal.
|
