Content-Type: text/html

Has the salary gap closed?
A survey of men and women managers at U.S. television stations



Jennifer Greer, assistant professor
University of Nevada-Reno, Reynolds School of Journalism

& Keisha Williams, associate producer
WFAA-TV, Dallas


Contact: Jennifer Greer
Mail Stop 310
University of Nevada-Reno
Reno, NV 89557

Office: (702) 784-4191
Home: (702) 746-3892
<a href="/cgi-bin/wa?LOGON=A3%3Dind9909B%26L%3DAEJMC%26E%3D7BIT%26P%3D1281771%26B%3D--%26T%3DTEXT%252FPLAIN%3B%2520charset%3DUS-ASCII" >[log in to unmask]</a>



Submitted to the Radio-Television Journalism division of the Association for
Education in Journalism and Mass Communication for the 1999 annual conference.
Abstract

Has the salary gap closed? A survey of men and women managers at U.S. television
stations

A survey of 169 general managers, general sales managers, news directors, and
program managers at the nation's television stations found that while more women
have reached the industry's top ranks, they still report lower salaries, number
of benefits, and feelings of authority than male managers. However, when
personal (gender, education, and age) and job characteristics (including market
size and job title) were entered into a regression analysis, gender was a
significant predictor only for salary.
      In 1972, Stone reported that only two of 398 respondents in a survey of
news directors in the Radio-Television News Directors Association were female.
Two decades later, women were beginning to crack the glass ceiling. A 1996
survey found that women held 24.1% of the news director positions in television
(Media Reports to Women, 1997).  Joyce Reed, a news director at KWTV in Oklahoma
City, said in 1995, "We have a long, long way to go before we dominate
newsrooms, but we've come a long way since 1976 when I was the only girl in the
newsroom" (quoted in Prato, 1996, p.48).  But despite the improvement in the
last 25 years, women still hold less than a quarter of television the news
director positions.
     The limited number of women in top television management has affected
society in many ways. Within the stations, the lack of women in management
deprives employees of the diverse strengths that women bring as managers (Eagly
& Johnson, 1990; Herkelmann et al., 1993; Gottepe & Schneier, 1989). Walsh
(1996) contends that the under-representation of women in the television
industry "deprives women of positive reinforcement, deprives children of female
role models, and deprives the nation of key intellectual and emotional feedback"
(p. 4).
     Others contend that dearth of women in management is reflected in content,
leading to an inaccurate portrayal of women and society as a whole (Hayes,
1979). In many markets, weather and health top the list of most important
issues, but crime and justice stories dominate newscasts (Fitzgerald, 1997).
Ziegler and White (1990) contend that women's issues, including as health,
gender-bias, and domestic violence, are not being explained on television in a
realistic manner. With more women at the table when decisions are being made,
areas such as education and discrimination might be discussed more often
(McCartney, 1997).  Epstein's 1978 survey of female television reporters
suggests that women are more likely than men to focus on topics like health
care, family welfare, education, rape, and child abuse, just to name a few.
Equal representation in media management is essential for diversity in news
coverage. Fox president Mitch Stern said in 1997 that without equal
representation, "we would be very one-dimensional" (quoted in Littleton, p. 32).
     The growing number of female television managers is a start in diversifying
the industry.  But even as women are moving into top positions, they are faced
with other barriers.  Hernandez (1996) contends women in television management
now are being discriminated against in areas such as "salary, access to jobs,
access to information, harassment, and negative attitudes" (p. 12).  Others
contend that while more women are being given prestigious job titles, their
comments are not being taken seriously and they are not being allowed to make
the decisions that matter (Chaffins, Forbes, Fuqua, & Cangemi, 1995).
     Past research has examined how many women are working in the television
industry and climbing the ranks to executive positions. Fewer studies have
examined how women compare to their male counterparts once they reach the top.
The few studies that go beyond looking at sheer numbers have concentrated on
salary gaps.  And most studies have looked at one specific job title (reporter,
anchor, or news director) or at all women in the industry in general.  This
study aims to fill that gap by surveying top television executives in four
different job categories (general manager, general sales manager, news director,
program director) about their compensation and other experiences on the job. Now
that more women are heading departments in the television industry, are they
making the decisions that affect their stations as a whole?  Are these women
receiving salaries and benefits equal to their male counterparts?  These are the
key issues that this study will address by surveying managers at stations in
varying market sizes across the nation. Further, the study examines whether
salary, benefits and feelings of authority are related to characteristics other
than gender, including market size, education, length of time in current
position, length of time in the industry, age and ethnicity.
Literature review
Women in management
     As recently as 1995, women only held 5% of all top-level corporate
positions (Kaufman-Rosen, 1995). In the Fortune 500 companies, women hold 2% of
senior executive positions, 3.6% of board directorships, and 1.7% of corporate
officerships (Chaffins et al., 1995).  Another study examining Fortune 1000
industrial and Fortune 500 service companies found that 95% of the managers were
men and nearly 97% were white (Castro & Furchtgott-Roth, 1997). While women are
nearly nonexistent at the very top levels of the corporate world, they fare
better when all management positions are examined. Linden (1994) estimates that
women hold about 40% of all management positions.
     Researchers say women face some of the most blatant forms of discrimination
with gaps in salary, benefits, and feelings of authority (Hernandez, 1996).  In
many companies, the highest-paid women has the 20th highest salary in the
organization as a whole (Castro & Furchtgott-Roth, 1997). In general, women in
middle management were said to make about $.66 for every $1 made by men
(Solomon, 1990).  Morrison and Von Glinow (1990) found similar results,
reporting that most females in top management earn 42% less than males in the
same position. But recent studies show those gaps closing. During the past 15
years, the salary of women in upper-level positions has increased 18.3% compared
to an average 1.7% salary increase by their male colleagues (Harris, 1995).
     In the cable industry for example, the average female manager's salary was
$50,376 in 1997 while the average male salary was $59,354. That 15% salary gap
was down 2% from the previous year. Women supervisors fared worse, however, with
average salaries about 25% lower than their male counterparts (Media Report to
Women, 1997). That same year, women with degrees and job tenure of one to three
years earned 30% less than their male counterparts in the cable industry (Media
Report to Women, 1997).
     In general, the gap between the genders in compensation holds even when men
and women have the same educational backgrounds.  Over the past two decades,
women have been strongly represented in graduate training programs, earning more
than 50% of all master's degrees since the early 1980s (Castro &
Furchtgott-Roth, 1997).  Despite the increase in education, few women have made
it to the very top of the salary ranks.  A 1994 Business Week study reported
"that women with degrees from Top 20 business schools earned 12% less in their
first year than men with comparable qualifications and took longer to move into
top management" (Castro & Furchtgott-Roth, 1997, p. 24).  In these high-level
management positions, women with equal education are usually said to start at
similar salaries as their male counterparts.  But as 10 years pass, men end up
making at least 20% more than women (Morrison & Von Glinow, 1990).
     The gaps between men and women managers also extend to benefits. Price
Waterhouse and Sheffield Consulting Group report that women receive smaller and
cheaper company cars and less medical insurance. "It appears that companies are
substantially under-rewarding their female management staff" (Management, 1997,
p. 36).  Jennie Langley, director of corporate affairs at Carter Holt Harvey,
suggests:  "We need to be smarter about getting what we want.  It's too easy to
get hooked up in the dollars, comparing ourselves with the blokes" (quoted in
Management, 1997, p.37).  Langley says women need to start using their benefit
packages to develop themselves personally.  Instead of agreeing to golf
memberships, performance bonuses, and shares, women need to focus on their
needs, such as time to study, child care benefits and taking courses to better
them as managers (Management, 1997).
     While gaps in salary and benefits can be easily measured, perhaps the most
damaging gap - women's perceived self-worth to their organization D is harder to
quantify. "Most females are now given mere 'token' positions in companies with
only the appearance of power and prestige rather than being blatantly excluded
from such positions, as was the case mere decades ago" (Chaffins, et al., 1995,
p. 380). Morrison and Glinow (1990) report that women fill one-third of
management positions, but they are not given the same power as their male
counterparts.
Women in Television
     In the 1960s, Stone found that one-fourth of all broadcasting jobs were
held by women, but nearly all were traffic, continuity, or secretarial positions
(Stone, 1973-74). Following the amendment of Title VII of Civil Rights Act to
prohibit discrimination by gender as well as by color, creed, or national
origin, women began to gain ground in the industry. By 1970, a survey found that
45% of the workers in the nation's newsrooms were women and that 94% of news
directors said they would hire a woman reporter (Stone, 1973-74).  Just as
supervisors were becoming more accepting of women in the newsroom, more women
were preparing to work in the field.  Creedon (1989) calls this the "gender
switch," with women outnumbering men in journalism programs since 1977.
     Weaver, Howard, and Wilhoit's 1992 study of journalists found women gaining
ground across all media. In 1971, 20.3% of working journalists were women. By
1992, women accounted for 34% of all journalists.  However, television had the
lowest percentage of women, with women accounting for 24.8% of all television
journalists.  In contrast, 44.1% of journalists at weekly newspapers and 45.9%
of the journalists at news magazines were women (Weaver, et al. 1992). While
Gerbner found in 1993 that 63% of those delivering news were men (cited in Walsh
1996, p.190), Stone reported just four years later that half of all television
news reporters and anchors in the United States were women. But Stone wondered
whether those women held equal status within their profession when the cameras
were off.
     Rakow and Kranich (1991), in examining how women serve as presenters of
news, found "a gendered division between 'serious, important' news that is
overwhelmingly masculine and 'human interest, lifestyle' news that is more
likely to be the purview of women reporters and readers" (p. 11). Gerbner
summarized this trend: "Overall, the powers in television want their men like
their wine - aged - and their women like their fruit - fresh and sweet" (quoted
in Walsh, 1996, p. 4).
     Some contend the television industry holds different evaluation standards
for men and women.  Ferri and Keller (1986) cited a female anchor in their
survey who "noted that they are often judged by their appearance, while their
male counterparts are judged more for their work skills" (p.467).  In a
follow-up study, Ferri (1988) concluded that women anchors perceive more
barriers than male anchors, including overemphasis placed on physical
appearance, overcoming gender stereotypes, and differential treatment in hiring
process based on gender (Ferri & Keller, 1986). A recent survey by Engstrom and
Ferri (1998) evaluated gender role expectations and found appearance and
balancing career and family as problems for many women in the television
industry.  Women said their primary career barriers dealt with the way others
perceived them.
     Another barrier many women in television say they continue to face is
balancing family and career. Burks and Stone (1993) examined career perceptions
of men and women television news directors and found that women news directors
were more concerned with the lack of quality time at home.  Many women news
directors said their family obligations were impeding their career progress,
leading the researchers to conclude that the under-representation of women in
upper-management might because women don't want to risk their family time for a
career. Weaver and Wilhoit (1996) reported that women had more difficulty
balancing personal and professional lives even though there were signs that this
barrier was beginning to change.
Women in Television Management
     While women have made strides in the television industry in recent decades,
they remain underrepresented in management positions.  Women now account for
about one-fourth of the news directors at U.S. commercial TV stations, and the
number of women news directors is increasing by about 5% a year (Media Report to
Women, 1997). Stone predicts that women will hold half of all news director
positions by 2001.  In the early 1970s, women were more likely to be executives
at independent stations and in small markets, rather than at affiliates of the
major networks.  The gap was closing by 1994, when 18.2% of female all news
directors worked at independent stations, compared to 12.8% of male news
directors (Media Report to Women, 1997).  Women seem to be faring particularly
well at Fox.  In 1997, Fox became the first network to hire women as general
managers in their stations in the top three markets, and women general managers
headed seven of the 22 Fox affiliates (Littleton, 1997).
     But some contend that contend that numbers don't equal power.  According to
Prato (1995), "more women than ever are taking over the management of local
television newsrooms, but this trend doesn't mean female news executives have as
much power in the business as men do" (p. 48).  Burks and Stone (1993) argue
that women's participation in top-level media management is lowest when compared
to any other industry.  And Hernandez (1996) asserts that women in the
television management are more likely to hold positions in areas such as
financing, rather than areas that directly affect the news content.
Research Questions
     The literature cited above suggests the following research questions:
     RQ1: How do women in television management compare with their male
     counterparts in job demographics (as measured by numbers, salary, benefits,
     and feelings of authority)?

     RQ2: If differences are found between the genders on the variables of
     salary, benefits, and feelings of authority, are these primarily predicted
     by gender? What other personal (age, ethnicity, education) and professional
     variables (job title, market size, length of time in current position,
     length of time in the industry) significantly predict variance in salary,
     benefits, and feelings of authority?

Method

     A mail survey was used to gather information on male and female television
managers. Dillman's Total Design Method (1983) was used in an effort to increase
response rate. The investigator sent subjects a detailed cover letter explaining
the importance of the study and ensuring confidentiality. Along with the cover
letter and the questionnaire, sent out in May 1998, the investigator sent
self-addressed, stamped envelopes.  A follow-up reminder postcard was mailed in
early June to television managers who had not responded to the first mailer.  In
late June, a second questionnaire and cover letter was sent to the managers who
had not responded. first mailer.
Population and Sample
     The population was all managers at 211 Television Designated Market Areas
(DMAs) identified in the 1996-97 Television and Cable Factbook. The population
will include only the four major television affiliates ABC, CBS, NBC, and FOX.
These affiliates were chosen because these affiliates typically provide some
type of local news.  Therefore, each would have a news director in addition to
the other three executive positions surveyed.  Further these four affiliates
also have a longer history of providing news and are fairly consistent
throughout all markets.
     Twenty-five markets were randomly selected, representing more than 10% of
the population represented. The markets listed in the DMA book were ranked from
one as the largest to 211 as the smallest. To draw a representative sample, the
researcher divided the 211 markets into three groups: one to 50 were considered
large markets, 51 to 100 were considered medium markets, and 101 to 211 were
considered small markets. A former news director in a large market suggested the
categorization of market size.  A table of random numbers, corresponding to
market ranks, was used in selection of television markets for the study. The
sample included eight stations in the large and medium market sizes and nine
stations in the small markets was drawn.  One additional market was drawn in the
smaller category because these markets tend to have fewer major affiliates and
fewer managers at each station.
     Throughout the selection process, the investigator evaluated the sample to
make sure it was representative of the population, in terms of geography as well
as size. After evaluating the markets selected randomly, the investigator
discovered that East Coast markets were over-represented and virtually no
western markets were included.  Therefore two large and two medium markets in
the eastern region were replaced with four West Coast markets to provide a
better geographic representation.  The following states were represented in the
study: California, Georgia, Ohio, Colorado, Pennsylvania, Florida, Indiana,
Wisconsin, Illinois, South Carolina, Oregon, Mississippi, Texas, and Alaska.
City names are not included to protect confidentiality of the respondents.
     The mail survey was distributed to four managers at each station: the
general manager, general sales manager, news director, and program director.
These positions were chosen because of the authority they possess.
Investigators called the stations selected one to two weeks before sending out
the first mailer to collect the correct names and verify titles of the managers
at the station. With four stations and four managers in each of 25 markets, the
sample drawn potentially could have been 400 managers.  But after calling each
of the 25 markets, the investigator's survey list was reduced from 400 potential
respondents to 335 because not all markets had the four major affiliates or
someone in each of the four targeted positions. Based on information obtained
from the stations, the sample of 335 included 244 men and 91 women managers.
Eighty-three surveys were mailed out to managers at ABC affiliates; 83 to CBS
stations; 94 to NBC stations, and 75 FOX affiliates. Surveys were mailed out to
88 general managers, 88 general sales managers, 77 news directors, and 82
program managers.
Variables
     The independent variables in this study are gender, education level, length
of time in current position, length of time in industry, age, ethnicity, and
size of market.  These variables were measured by single multiple-choice items
on the questionnaire.  Gender was measured with a single item, with men coded as
1 and women as 2. Education level was measured by asking the subjects a
multiple-choice question with the following choices: (1) Some high school; (2)
completed high school; (3) some college; (4) graduated from college; (5) some
graduate work, no degree; and (6) completed advanced degree(s).  Length of time
in current position was measured by four choices from (1) less than a year to
(4) five years or more. Length of time in the television industry was measured
by five choices from (1) less than a year to (5) 16 years or more. Age was a
single-item measure asking respondents to fill in their age in years, and six
categories of ethnicity were listed. Size of market was determined by a code
number on the questionnaire returned by respondents. For analysis, markets were
broken into three groups as described above with 3 as the largest market and 1
being the smallest markets. Finally, job title was, again, determined by a code
number on the questionnaire that indicated which of the four management titles
the respondent held. For analysis, the three non-general manager job titles were
collapsed into one group (1) and general managers, who were expected to earn
more as the station's top officials, were classified as (2)
     The dependent variables measured are numbers of men and women, salaries,
employment benefits, feelings of authority.  The salary question was measured
with 12 response options ranging from (1) $29,000 to (12) $160,000 or more.  The
employment benefits question also was a multiple-choice question, where subjects
were asked to mark all of the benefits that apply to their employment package.
These 13 choices included company car, expense account, golf memberships,
additional insurance, and salary bonuses. Respondents were asked to list all
benefits not on the list. A total benefits score was computed for each
respondent by summing the number of benefits selected.
     The dependent variable of feelings of authority was measured by combining
Steel and Mento's (1987) Perceived Decision-Making Influence scale and Kohl's
(1989) Self-Perceived Influence scale. The five-item perceived decision-making
influence scale (Steel & Mento, 1987) assesses employees' perceptions of the
degree of influence they have over decisions that affect them or their work.
This scale reports reliability coefficients for the measure ranging between .85
and .90.  Steel and Mento's scale correlates with ratings of job satisfaction
and supervisory performance.  In the original scale (1987), subjects were asked
five questions based on this Likert-type scale. Responses were measured on a
7-point agree-disagree rating scale. For the current study, only minor wording
changes were made to make the questionnaire relevant to television managers.
For example, "Within my work group the people most affected by decisions
frequently participate in making the decisions," was changed to "within my
station..."
     The Self-Perceived Influence scale (Kohl 1989) measures how much an
employee feels he or she influences decisions in an organization.  Reliability
coefficients for perceived influence scale in past studies ranged from .93 to
.86. For the current study, three out of five of the items in the original
scale
were used. The three questions were used because they focused more on
participation and influence of the person's opinions and views. Again, the only
modification was in wording. Responses were measured on a 7-point Likert scale
from 1 (very small) to 7 (very large).  A sample item was "How much weight does
your fellow management team give to your opinions?"
     After the responses were collected, factor analysis was performed on the
feeling of authority scale created for this survey. The analysis revealed a
single-factor solution (Eigenvalue = 5.23, explaining 65.40% of the variance),
and the eight-item scale was reliable at .92.  Reliability showed that removing
one item of the eight-item scale would have improved the scale's reliability to
.93.  However, the high initial reliability and single-factor solution
indicated
that all eight-items should be retained for the measure.  To compute an
authority score, the responses to the eight decision-making influence items were
averaged for each respondent. The higher the score, the higher the feelings of
authority the subject possessed.
        Finally, a measure of job satisfaction was included in the survey.  Job
satisfaction was measured by using Downs and Hazen's (1977) Communication
Satisfaction Questionnaire (CSQ).  Subjects were asked three questions about
their level of job satisfaction.  The first question, "How satisfied are you
with your job?", was measured on a 7-point Likert scale from 1 (Very Satisfied)
to 7 (Very dissatisfied).  The second question, "In the past 6 months, how has
your level of job satisfaction changed?", was measured on a 5-point Likert-type
scale from Improved significantly to Decreased Significantly. This measure was
included for descriptive purposes only and was not used as a dependent variable
in analysis. It will be used in the discussion of the analysis.
Findings

     After the two mailings, 169 managers of the 335 in the sample returned
completed questionnaires, for a response rate of 50.4%.  Respondents ranged in
age from 25 to 73, with about 50 percent of respondents falling between the ages
of 37 and 48. One hundred and seventeen (69.2%) of the subjects were male and 51
(30.2%) of the respondents were female.  For ethnicity, 157 (92.9%) identified
themselves as white, two (1.2%) as African-American, five (3.0%) as Hispanic,
and two (1.2%) as other. Because only nine respondents in the sample identified
themselves as minorities, ethnicity was dropped in all further analysis.
     Of the 169, 49 (28.9%) managers returned the survey in the top 50 markets,
70 (41.4%) managers returned the survey in the medium-sized markets, and 49
(28.9%) managers returned the survey in the small markets.  Thirty-five (20.7%)
general managers, 42 (24.9%) general sales managers, 42 (25.0%) news directors,
and 49 (29.2%) program managers returned surveys. Forty-four (26.0%) of the
managers who responded were from ABC affiliates; 42 (24.9%) were from CBS; 46
(27.4%) from NBC; and 36 (21.4%) from the FOX affiliate.
     RQ1: How do women in television management compare with their male
     counterparts in job demographics (numbers, salary, benefits, and feelings
     of authority)?
     When examined by both ratios of the sample selected (men = 244, 72.8%;
women = 91, 27.2%) and actual respondents (men =117, 69.2%; women = 50, 30.2%)
men outnumbered women more than 2 to 1 in television these management positions.
For salaries, men (M = 7.86, corresponding to the $80,000 to $89,999 salary
range) reported significantly higher average salaries than women (M = 5.17,
corresponding to the $60,000 to $69,999 range; t = 4.61, df = 151, p < .001).
Further, men (M = 4.54) reported having significantly more benefits than women
(M = 3.61; t = 2.09, df  = 155, p < .038).  Finally, men (M = 5.78) reported
significantly higher average feelings of authority than women (M = 5.28; t =
3.14, df  = 163, p < .002).
  Results for the dependent variables of salary, benefits, and feelings of
authority also were examined by dividing the variables into high, medium, and
low categories and comparing by gender on a crosstabluation analysis.  While no
significant differences emerged in this type of analysis for benefits,
differences were found for salary and feelings of authority.
  For salary, those with salaries below $70,000 were classified as low, those
with salaries between $70,000 and $119,999 were classified as medium, and
managers with salaries above $120,000 were classified as highly paid. A
Chi-Sqaure test showed that men were significantly more likely than expected to
be in the medium and high salary groups, while women were over-represented
(compared to expected counts) in the low salary classification. The results are
presented in Table 1-1.
     Table 1-1: Crosstabulation of Salary by Gender
               observed count (expected count)
Gender
Low salary
Medium salary
High salary
Men
30 (38.5)
40 (37.1)
37 (31.5)
Women
25 (16.5)
13 (15.9)
 8 (13.5)
     (2 = 10.2, df=2, p < .01

        For feelings of authority, those with average scores below 5.5 (out of a
possible 7.0) were classified as low, those who scored between 5.5 and 6.0 were
classified as medium, and those with scores above 6.0 were classified as having
high feelings of authority. While the numbers observed in the medium category
were very close to the expected values, men again were significantly
over-represented in the high category and women were over-represented in the low
category. The results are presented in Table 1-2.
     Table 1-1: Crosstabulation of Feelings of Authority by Gender
               observed count (expected count)
Gender
Low authority
Medium authority
High authority
Men
28 (36.2)
37 (36.2)
50 (42.5)
Women
24 (15.8)
15 (15.8)
11 (18.5)
     (2 = 10.59, df=2, p < .01

     RQ2: Are the variances in salary, benefits, and feelings of authority
     being driven primarily by gender or by other personal (age, education) and
     professional (job title, market size, length of time in current position,
     length of time in industry)?

     Because significant differences were consistently found between the genders
on salary and authority (and to a lesser extent benefits), multiple linear
regression analysis was used to discover if gender or some other factor was the
primary predictor of these observed differences. Stepwise analyses were run
first on personal characteristics only (gender, age, education) then on all
personal and professional variables combined.
               Table 2-1 Regression Analysis for Salary
                                        Betas (N = 149)
Predictor variables
Regression 1
Regression 2
Gender (being female)
-.249**
   -.217***
Age
  .249**
  .206**
Education
.189*
.045



Job title (general manager)

     .341***
Market size (large markets)

    .353***
Length of time in position

 -.198**
Length of time in industry

.099



R2
.197
.441
Adjusted R2
.180
.421
                * p < .05, ** p < .01, *** p < .001
        As Table 2-1 indicates, both personal and professional variables significantly
predict changes in salary. When personal characteristics were run independently,
gender loaded first as the strongest predictor of change in salary. As expected
based on previous studies and the analyses in Research Question 1, being female
is negatively related with higher salaries. In other words, women managers are
likely to earn less than their male counterparts. Education was the next biggest
predictor in the equation, with those with higher education more likely to earn
higher salaries. Finally, age was also a significant predictor. Older managers
are likely to earn more, according to the model. The combined personal
demographics model explained about 19.7% of the total variance in salaries.
        To examine the effect of professional factors as well, the four job-related
variables were included in the regression analysis. Table 2-1 shows that job
title predicted the largest portion of the variance in salary, with general
managers likely to earn higher salaries. Market size was the next in predictive
value, with those in larger markets likely to earn more. In the combined
analysis, gender was the third largest predictor of salary. Next came age,
followed by length of time in the position, which interestingly had a negative
relationship with salaries. This data indicates that new hires make more than
those who have held their management position longer. Neither length of time in
the industry or education were significant predictors of salary in the combined
analysis.
               Table 2-2 Regression Analysis for Benefits
                                        Betas (N = 152)
Predictor variables
Regression 1
Regression 2
Gender (being female)
-.151+
-.083
Age
  .030
-.004
Education
..103
  .075



Job title (general manager)

     .296***
Market size (large markets)

.042
Length of time in position

-.072
Length of time in industry

.066



R2
.023
.088
Adjusted R2
.016
.082
                +  p < .1; * p < .05, ** p < .01, *** p < .001
        As Table 2-2 indicates, only gender came close to being a significant
predictor for benefit variance in the personal demographics regression model.
However, this predictor was very week (t= -1.87, p < .06). The total variance
explained by the personal demographics model was a slight 2.3%, all of which was
accounted for by change in gender. When job variables were added to the
equation, only job title was a significant predictor of the number of benefits.
Being a general manager, as expected, leads to more benefits. The total variance
in benefits explained by the combined personal and job demographics regression
model was 8.8%, all of which was accounted for by job title.
          Table 2-3 Regression Analysis for Feelings of Authority
                                        Betas (N = 159)
Predictor variables
Regression 1
Regression 2
Gender (being female)
    -.213***
-.117
Age
.056
-.113
Education
.101
  .092



Job title (general manager)

      .352***
Market size (large markets)

 .018
Length of time in position

-.047
Length of time in industry

 .207*



R2
.046
.197
Adjusted R2
.039
.159
                 * p < .05, ** p < .01, *** p < .001
        Gender again played a weak role in feelings of authority among managers. When
personal characteristics were run alone, gender was the only significant
predictor. The total variance explained by the personal demographics model was
4.6%, all due to gender. Being female predicted lower scores on feelings of
authority. When job variables were added in, job title again was the dominant
predictor of variance in benefits in the model, followed by length of time in
the industry. Gender was not a significant predictor in the combined model. The
total variance in benefits explained by the combined personal and job
demographics regression model was 19.7%, most of which was accounted for by job
title and length of time in the industry.
  It was not surprising that job title accounted for significant differences in
all three variables, given the fact that the four managers surveyed were not
really at equal levels in their organizations. General managers are at the level
above the other three positions surveyed for this project. It would be expected,
therefore, that managers in these positions would be better compensated and feel
more powerful at their stations. Of the 35 general managers who returned
surveys, 31 were men and only 4 were women. Of the 132 in the other three
positions, 85 were men and 47 were women. Therefore, to better compare mangers
at the same levels, analyses for all three dependent variables were repeated for
only the 132 managers on the "second tier" and job title was dropped as an
independent variable. The results are presented in Table 2-4.
          Table 2-3 Regression Analysis for All Dependent Variables
General managers removed
                                        Betas (N = 129)
Predictor variables
Salary
Benefits
Authority
Gender (being female)
    -.226**
-.045
-.109
Age
   .033*
-.039
-.118
Education
.090
.139
  .075




Market size (large markets)
      .391***
.116
 .068
Length of time in position
   .202**
.026
-.118
Length of time in industry
.097
.054
   .176+




R2
.285
.042
.031
Adjusted R2
.260
.009
.023
                 +  p <.1; * p < .05, ** p < .01, *** p < .001
  Looking only at the respondents who were general sales managers, news
directors, and program managers (three relatively equal management titles)
significant predictors were identified only for salary. Market size was the
strongest predictor of salaries, followed by gender and length of time in the
position. No significant predictors were identified for benefits. For authority,
only length of time in the industry produced a weak effect. Only the salary
model predicted any real variance, with 28.5% of variance in salary explained by
the model.
Discussion
Summary
     The first research question evaluated how women in television management
compare with their male counterparts on salary, benefits and feelings of
authority.  The results showed that, overall, men outnumbered women by more than
50% and men averaged at least $20,000 more than women in average salary. The
data also revealed men had significantly higher average benefits and feelings of
authority overall. Dividing respondents into categories of low, medium, and high
on salary, benefits, and authority confirmed these findings. For both salary and
benefits, women were over-represented in the low groups, while men were
over-represented in the high groups. No differences, however, were found for the
number of benefits held by TV managers.
     Once differences were discovered in the genders, regression analysis was
used to determine what variables contributed to these differences. In each case,
gender was found to be a significant predictor when personal characteristics
were examined alone. But when the personal characteristics were examined in
conjunction with job-related factors, gender remained a strong predictor only
for salary. Consistently, the analysis showed that the variable with the largest
predictive value was job title. As would be expected, being a general manager D
the manager above the other three included in the study D was most predictive of
higher salaries, benefits, and feelings of authority. Next for salary in
predictive weight was, was market size. Stations in larger markets typically
have more resources and can afford to pay all employees at higher rates than
medium and small markets. One curious finding on the job-related variables is
that length of time in the industry was not a significant predictor for salary
or benefits, only for authority. Further, length of time in the current position
was negatively related to salary. These finding suggest that the industry is not
raising salaries of veterans as quickly as salaries for new managers. As is the
case in other industries, stations probably have to pay newcomers more to
attract them than they have to pay to retain current managers.
     Removing the general managers from the analysis, clear predictors were
found only for salary, where market size, gender and length of time in the
position played key roles. No significant predictors, including gender, were
discovered for benefits and authority among the lower-tier of top television
management.
Implications
     While these job demographics are important variables, it should be noted
that gender still was a significant predictor for salary, even with all seven
factors controlled for. Further, gender was a significant predictor for salary
looking among managers working at comparable levels. This paints a clear picture
that women in top-level television management continue to struggle for equality.
Over the past decade, more women have moved into these positions, but they
continue to battle for equal compensation. Women fare better in benefits, but
benefits alone will not keep women in these television management positions. The
question that the industry must address is whether women will want to stay in
the industry given the inequities. Perhaps women will become frustrated and
leave for other industries.  Attracting women to top management positions is
futile if steps aren't taken to retain the women that already fill the
positions.
     Some evidence for this position comes from Weaver, Howard, and Wilhoit's
1992 survey of journalists. While women were virtually equal in numbers to men
when they were just starting out in broadcast journalism, as the years progress
women's percentage of overall workers decreased tremendously. Of all television
workers who had been in the industry four years or less, women accounted for
45%. But women in the industry 20 plus years dropped to 24% of all workers with
the same tenure (Weaver, Howard, & Wilhoit, 1992).
     On a positive note, the results indicate that women by and large feel as
powerful as male managers in the decisions made by the management team. Women
may not be making as much as their male counterparts, but they do feel their
voices are being heard. These feelings of efficacy should not be discounted.
Often it is frustrations over role, not salary inequities that force people from
their jobs. Women who see themselves as having the authority to make decisions
might overlook pay gaps and feel happy in their jobs.
     As a check of this proposition, a single item was included in the current
survey, asking respondents how satisfied they were with their jobs. Of 168
respondents who answered that question, 159 (94.6%) said they were somewhat
satisfied (21 responses), satisfied (71 responses), or very satisfied (67
responses). No one checked the neutral box, and only 9 said they were somewhat
dissatisfied (4 responses), dissatisfied (3 responses), or very dissatisfied (2
responses). Because there was so little variance in the responses, no
differences were likely to occur based on gender of respondent. As expected,
mean scores for men (1.92) and women (1.94) were virtually identical. What the
job satisfaction scores, as basic as they are, suggest is that while clear
salary differences emerge between men and women, the genders are equally
satisfied in their jobs.
Limitations
     Although all efforts were made to include a large number of women in the
sample and a response rate of more than 50% was achieved, it must be noted that
only 51 women managers returned. Caution is urged in interpreting the data
analysis because of this small number. This especially is true in the final
regression analysis, which included only 47 women and independent six variables.
The significant lack of findings for gender on benefits and feelings of
authority might stem from that low number.
     The few minority respondents posed a similar problem. While it is clear
that there are very few minority managers in the industry, the limited response
from minorities (nine returned surveys) meant that race could not be a variable
in analysis. Future studies that wish to include race should perhaps
deliberately over-represent minority managers in their sample, although this
data is hard to obtain before the actual surveys are mailed out.
     Caution also is urged in interpreting the reported lower feelings of
authority among women managers found in research question one. Differences in
this construct do not necessarily mean that women are powerless in their
management positions. An alternative explanation is that perhaps men and women
actually have the same authority, but they perceive their authority in different
ways. But that explanation should not mitigate the importance of the findings.
Does it matter if a person only feels less powerful even when in reality she has
the same authority as her male counterparts? A perception of powerfulness is
still a problem if employee morale and retention are the issues of concern.
Further, the responses from this population suggested that most women in the
markets surveyed held the position of program director. Future research should
look at what type of voice this position has from the view of the management to
help explain how women mangers contribute to station decisions.
     Finally, the survey instrument was not designed to rule out rival
explanations for the findings. Even though gender is a significant predictor in
the variance in salary (and to a lesser extent benefits and authority when only
personal factors are considered), other factors associated with gender may be
the real cause of the differences. For example, women may have lower salaries
because they don't change jobs (the fastest route to a raise in the TV industry)
as much as men. Research suggests that men and woman tend to have different
patterns when it comes to moving for a job. Perhaps women are more hesitant to
relocate once they've established themselves in a community. Staying at one
station, as women might be prone to do, typically means raises in small
increments over the years. Future research, especially longitudinal studies,
could examine career patterns of women and men managers, seeing if differences
emerge in how often they change stations and why they choose to leave. In-depth
interviews with current and past television managers also would further our
understanding of gender-related issues not measure by the mail questionnaire
method.
Conclusion
     The results of this study should send a message to television stations and
to the industry as a whole.  Over the past few decades, the industry has done a
better job of hiring women and promoting them into the top levels of management.
But are these new hires going to stay in the industry?  Simply attracting a
sufficient number of women is not the solution if they grow dissatisfied with
their jobs and leave. Given the lag in salaries, and to a lesser extent
benefits, as a function of gender, stations clearly are not compensating their
female managers equally for their work. Perhaps more important is the relative
powerlessness women managers feel in their positions compared to their male
counterparts. Stations must not only compensate women equally to make them feel
valued, they must also keep women actively involved in the daily decision-making
process. Although the vast majority of women surveyed say they feel satisfied
with their jobs, these things currently are not happening. Because of its
influence on society, the television industry should strive to present as
accurate reflection of society as possible, including women and women's issues.
The inclusion - and retention - of women in television management might better
enable the industry to serve this role.
 References

Beasley, M., & Silver, S. (1977). Women in media:  A documentary source
     book. Washington, DC: Women's Institute for Freedom of the Press.

Castro, I., & Furchtgott-Roth, D. (1997, February). Should women be worried
     about the glass ceiling in the workplace?  Insight on the News, 13, 24-28.

Chaffins, S., Forbes, M., Furqua, H., & Cangemi, J. (1995). The glass
     ceiling: Are women where they should be? Education, 115(3), 380-387.

Creedon, P. J. (1989). The challenge of re-visiting gender values. In P. J.
     Creedon (Ed.), Women in mass communication: Challenging gender values.
     Newbury, CA: Sage Publications.

Dillman, D. (1983). Mail and other self-administered questionnaires. In P.
     H. Rossi, J. D. Wright, & A. B. Anderson (Eds.), Handbook of survey
     research. New York: Academic Press.

Downs, C.W., & Hazen, M. (1977). A factor analytic study of communication
     satisfaction. Journal of Business Communication, 14, 63-73.

Eagly, A., & Johnson, B. (1990). Gender and leadership style: A
     meta-analysis. Psychological Bulletin, 108, 223-256.

Engstrom, E., & Ferri, A. (1998). Gender differences in the perceptions of
     television news anchors' career barriers. Las Vegas: University of Nevada,
     Las Vegas.

Ferri, A. J., & Keller, J. (1986). Perceived career barriers for female
     television news anchors. Journalism Quarterly, 63, 463-467.

Fitzgerald, M. (1997). Local TV news lacks substance. Editor & Publisher,
     130(21), 8-10.

Gottepe, J., & Schneier, C. (1989). Role of sex and gender roles, and
     attraction in predicting emergent leaders. Journal of Applied Psychology,
74
     , 165-167.

Harris, D. (1995, January). 16th annual salary survey. Working Women, 20,
     25-34.

Hayes, E. (1979). Women in management:  An analysis of attitudes toward
     women in television management.  Unpublished master's thesis, Southern
     Illinois University at Carbondale.

Herkelmann, K., Dennison, T., Branham, R., Bush, D., Pope, K., & Cangemi,
     J. (1993, Fall). Women in transition: Choices and conflicts. Education,
     114(1), 127-133.

Hernandez, D. (1996, April 20). Obstacles facing women journalist. Editor &
     Publisher, 129(16), 75.

Kaufman-Rosen, L. (1995, March 27). Holes in the glass ceiling theory.
     Newsweek, 125, 24-26.

Kohl, A. (1989, July). Determinants of influence in organization buying: A
     contingency approach. Journal of Marketing, 53, 50-65.

Linden, D.W. (1994, July 4). The class of 1965. Forbes.

Littleton, C. (1997, May 19). Women find room at top at fox stations.
     Broadcasting & Cable, 127, 32(2).

Management. (1997, May). Women close the gap negotiating a top salary. 44,
     36.

McCartney, J. (1997, June). Newslite: The tendency of network newscasts to
     favor entertainment and tabloid stories. American Journalism Review, 19,
     18-26.

Media Reports to Women. (1997, Summer). 1990's bring mixed results for
     women, minorities in broadcasting, 23, 1.

Media Report to Women. (1997, Winter). Women gaining as TV news directors,
     could be one third of them by 2001, 25, 10.

Morrison, A.M., & Von Glinow, M.A. (1990). Women and minorities in
     management. American Psychologist, 45, 200-208.

Prato, L. (1996, November). Women move up in TV newsroom. American
     Journalism Review, 18, 48.

Rakow, L. F., & Kranich, K. (1991). Woman as sign in television news.
     Journal of Communication, 41(1), 8-25.

Steel, R., & Mento, A.(1987). The participation-performance controversy
     reconsidered:  Subordinate competence as a mitigating factor. Group &
     Organization Studies, 12, 411-423.

Stone, V. (1973-74). Attitudes toward television newswomen. Journal of
     Broadcasting, 18(1), 49-62.

Walsh, M. (1996, March 9). Television's women. America, 174(8), 4.

Weaver, D., Howard, R., & Wilhoit, G. (1992). The American journalist in
     the 1990s. Arlington, VA: The Freedom Forum.

Ziegler, D., & White, A. (1990, Spring). Women and minorities on network
     news. Journal of Broadcasting & Electronic Media, 34, 215.