|
HOW WOMEN ARE FARING AS THE DUST SETTLES:
THE EFFECT OF GENDER ON JOURNALISM/MASS COMMUNICATION
EVALUATIONS IN A COMMUNICATION CONTEXT
Prepared by:
Therese L. Lueck
Associate Professor
Richard E. Caplan
Associate Professor
Kathleen L. Endres
Professor
School of Communication
The University of Akron
Akron, OH 44325-1003
Ph: 330-972-7954
FAX: 330-972-8045
Submitted to:
Commission on the Status of Women
Association for Education in Journalism and Mass Communication
1996
HOW WOMEN ARE FARING AS THE DUST SETTLES:
THE EFFECT OF GENDER ON JOURNALISM/MASS COMMUNICATION
EVALUATIONS IN A COMMUNICATION CONTEXT
Abstract
Researchers tested course evaluation results for journalism/mass
communication at a midwestern university's school of communication. Both
standardized evaluations and the indexes of culturally desirable gender traits
showed high ranking for women instructors. Instructors were not matched for
variables such as experience. Interaction effects between same-sex instructors
and students were not found; however two-way interactions between instructor sex
and type of course were observed.
How Women Are Faring as the Dust Settles:
The Effect of Gender on Journalism/Mass Communication Evaluations
in a Communication Context
Introduction
Student evaluations of faculty teaching constitute a widely
accepted method of assessing courses and the quality of teaching in higher
education. Although they are relied upon for personnel and merit decisions, much
controversy continues to surround the student teaching evaluation as a
measurement of teaching effectiveness.
Much of this controversy centers on potential bias, with important
aspects being the effect of sex and sex-role perceptions. This study was
designed to investigate the effect of these variables on course evaluations. It
is the contention of these researchers that students have reactions and
expectations of instructors based on the sex of the instructor and,
additionally, that student expectations for journalism/mass communication
instructors may be conditioned by a communication context. Researchers attempted
to chart potential for student perceptions to bias an instrument otherwise
presumed to be an objective measure of teaching effectiveness.
Amid a climate of consolidation of academic programs, evaluations
take on a heightened significance. With this consolidation often placing them
into communication departments, journalism and mass communication instructors
must become aware of how they are being evaluated. In umbrella communication
departments, journalism/mass communication instructors and their coursework may
exist under the aegis of policies designed for and procedures established in a
broadly based communication environment. This situation has far-reaching
implications not only for evaluations but for resources, salaries, policy-driven
decisions, potentially course content and ultimately the fit and value of
journalism/mass communication courses and instructors. In this environment of
consolidation, women, as relative new-comers to the journalism/mass
communication faculty, particularly need to be wary of the tools used to assess
such information, what they measure and how they are used.
To gauge difference and effects of sex and sex-role on course
evaluations of journalism/mass communication instructors in this communication
context, researchers conducted their naturalistic study at a midwestern
university at which the dust has settled: Its journalism and mass communication
courses are taught from within the umbrella communication department.
Literature Review
Higher education in the United States in the latter 1990s exists in
a climate of economically driven and legislatively managed consolidations.
Within this context, journalism/mass communication courses and programs have
been merged with broadly based schools of communication. In these situations,
women have generally maintained optimism. For instance, when the Ohio State
University proposed the journalism-communication merger, Pamela Shoemaker,
then-director of the school of journalism, stated that when the dust settled,
the school would probably offer fewer journalism courses but that the students
would have more flexibility in structuring their programs of study (Magill,
1994, p. 21). Journalism/mass communication professionals have been watching
this merger phenomenon closely. As have her two predecessors, Shoemaker, current
president of the Association for Education in Journalism & Mass Communication,
monitors the situation. In retaining her optimism, she explores alliances and
umbrella organizations on the professional educators' level (Shoemaker, 1996, p.
2). With increasing consolidations, journalism/mass communication educators will
have more interaction with and within speech-driven communication organizations.
How journalism/mass communication educators fit with these organizations may be
a reflection of how they are evaluated and valued at the school level that
integrates journalism/mass communication with communication.
Teaching performance is an important part of this fit, and it
remains an important consideration in retention, tenure, promotion and merit
decisions. How to measure and evaluate teaching effectiveness has long been
debated in a variety of disciplines (Cashin, 1988; Arreola, 1989). According to
a survey of 453 department heads (Centra, 1977), the chair's evaluation had the
greatest weight in determining teaching effectiveness. Tied for second were
colleagues' opinions and systematic student ratings.
Of these three, student ratings are often accepted without thought
of factors apart from teaching performance that might influence student
responses. Research has explored any number of variables affecting the ratings
(Feldman, 1983; Feldman, 1986; Marsh, 1984; Marsh and Ware, 1982).
Sex and sex-role have been among the factors examined in
determining if student ratings are valid measures of teaching performance.
Validity here is defined: "Does the test measure what it is supposed to?"
(Cashin, 1988). In the course of research, questions of bias have surfaced
(Swim, Borgida, Maruyama & Myers, 1989), or factors that negatively or
positively have an impact on the results of the instrument (see Marsh, 1984, pp.
707-754). While the research is not conclusive, a large number of studies have
found that the sex of the instructor may influence ratings.
For purposes of this study, researchers relied upon definitions
from Basow (1992): "Sex is a biological term; people are termed either male or
female depending on their sex organs and genes. In contrast, gender is a
psychological and culture term [referring] to society's evaluation of behavior
as masculine or feminine" (p. 2). Basow, in quoting from Ashmore & Del Boca
(1979, p. 222), explains that the term gender, or sex-role, stereotype refers
to "those structured sets of beliefs about the personal attributes of women and
men" (p. 3). She also pointed out, "[S]ocial desirability of masculine and
feminine traits is related to gender stereotyping" (pp. 6-7).
Sex and gender have been studied as variables that make a
difference in the evaluation process. Many studies have found male instructors
rated higher than women teachers. Kierstead, D'Agostino and Dill (1988), for
example, found that both male and female students consistently rated their
female instructor lower than the male teacher; the respondents of both sexes
expected female instructors to fulfill a different standard of behavior than
their male teachers. "If female instructors want to obtain high student ratings,
they must be not only highly competent with regard to factors directly related
to teaching but also careful to act in accordance with traditional sex role
expectations," the researchers explained. "In particular, our results indicated
that male and female instructors will earn equal SRIs [Student Rating
Instruments] for equal professional work only if the women also display
stereotypically feminine behavior" (pp. 342-344). This negative rating of female
instructors by students of both sexes tended to support earlier studies (e.g.,
Lombardo & Tocci, 1979; Wilson & Wilson, 1976).
Other studies have found that the male students rated female
instructors in a more harsh way than the male teachers. Bernard, Keefauver,
Elsworth and Naylor (1981) used an experimental design to test female and male
student responses to teachers. In that study, the researchers found that male
students were significantly less positive in their attitudes, expectations and
evaluations of female teachers than female students, and the male students
consistently rated male instructors more favorably than they did the female
instructors.
That finding was reinforced by Basow and Silberg (1987). In a study
of more than 1,000 college students of 16 male and female professors (matched
for course, teaching experience and tenure), male students gave female
professors significantly poorer ratings than they gave the male instructors for
each of six variables. Female students rated the female professor more
negatively than male teachers on only three variables. Findings, especially in
naturalistic settings, are complex. When variables such as experience and tenure
are not controlled for women tend to be ranked higher than men (Feldman, 1993).
According to this literature, when variables such as experience are
controlled for, women instructors are likely to be at a disadvantage. Certain
studies have emphasized that women instructors are expected to behave in a
certain manner consistent with social expectations. Other studies have pointed
out that the teaching function at the higher education level may run counter to
gender expectations. Rakow (1991) pointed out that women who teach on the
college level transgress their traditional gender role designations, assuming
instead a non-traditional role, a violation of the "natural" order or authority.
This hostile effect can be somewhat mediated by the students' continued
experiences with women in non-traditional roles such as university professor
(Fandt & Stevens, 1991).
Another body of research examines how societal expectations of
gender affect students. According to this body of research, students come to
expect certain behaviors from their female and male instructors. Findings
suggest that students expect certain gender-stereotypical behavior from their
instructors. Female teachers are expected to conform to culturally accepted
behaviors, and those who fail to conform to those stereotypes often find
negative student ratings. "Nurturing" and a close interpersonal relationship
with students fall within this cultural expectation.
Bennett's study (1982) of 253 students illustrates such research.
Bennett found that women instructors were perceived as warmer (a highly loaded
female characteristic [Bem, 1981]) within the classroom. Yet students demanded a
greater amount of interpersonal support and held women to a stricter standard of
accessibility. While Bennett did not detect any direct gender biases in the
formal student evaluations, she speculated that female faculty members were
subject to "cultural conditioned gender stereotypes" (pp. 170-179).
That expectation for a higher degree of interpersonal contact from
female instructors was also found by Cooper, Stewart and Gudykunst (1982) in
their study. Interpersonal contact as a variable had far more importance in the
students' judgment of female instructors than male instructors. "[W]omen are
caring and sensitive while men are competent," the authors concluded, pointing
to the stereotypes. "When evaluating instructors, students give greater
significance to the type of interpersonal responses they receive from female
instructors while giving greater significance to the accuracy of the grade they
receive from male instructors" (pp. 308-315).
These findings, however, remain controversial when placed in the
context of mixed research results. Other studies have not shown that culturally
conditioned gender stereotypes affect student evaluations. Bennett (1982) did
not detect direct biases, although she detected differing student expectations
of instructors, depending on gender. An early study by Elmore and LaPointe
(1974) detected no interaction between the faculty member and the student based
on sex and no differences between the mean ratings given male and female faculty
by students. Yet, these researchers emphasized that they did not control for
many variables in their study, such as class size and instructor's rank.
Recent research has indicated an interaction effect between
same-sex instructors and students. The researchers (1993) found an interaction
effect in journalism/mass communication courses at a midwestern university.
Basow (1995) also detected interaction effects -- across the curriculum -- in
her four-year study of a private liberal arts college. She found teacher gender
by student gender interaction describing same-gender preference. What she also
found was a teacher gender by course division interaction, with female teachers
in humanities rated similarly or slightly higher than male instructors, female
instructors in natural sciences rated slightly lower than male teachers, and
mixed results in the social sciences. She detected the three-way interaction of
teacher gender by student gender by division.
The conflicting nature of the research results has led Cashin of
the Center for Faculty Evaluation and Development to conclude in 1988 that
gender of instructor did not appear to be a factor in student evaluations. Yet
two years later, while still noting that gender showed "little or no
correlation" in student evaluations, Cashin (1990) wrote, "[I]f the instructor
provides evidence of his or her self-report of these variables [including
gender], or if you or others have such evidence, that evidence should be taken
into consideration."
Much of the gender research has relied on the Bem Sex-Role
Inventory (BSRI) to measure socially desirable masculine and feminine traits.
The BSRI is based on extensive examination of culturally defined masculine and
feminine characteristics (Bem, 1981). For instance, based on two decades of
testing, Bem has found that an item, or behavioral characteristic, such as that
defined by the phrase "demonstrates leadership abilities" is highly loaded as a
culturally desirable masculine trait. Similarly, the phrase "eager to soothe
hurt feelings" was found through equally extensive testing to be a highly loaded
item designating a culturally desirable feminine characteristic. The BSRI has
found its use across the curriculum. Studies have verified the reliability and
the internal consistency of the BSRI. Yanico (1985) showed that the BSRI has at
least a moderate long-term reliability. Schmitt and Millard (1988) found that
while not all individuals can be shown to respond to sex-type coding, which the
BSRI endeavors to chart, people do tend to use culturally defined standards of
masculinity and femininity as yardsticks. Basow (1992) pointed out that the BSRI
reveals the fallacy of stereotypically masculine and stereotypically feminine
characteristics being considered polar opposites. She noted that there has not
necessarily been an inverse relationship between how one scores on the
stereotypically masculine and how one scores on the stereotypically feminine
characteristics. "How high someone scores on masculinity is unrelated to how
high he or she scores on femininity" (p. 7).
A study by Martin and Ramanaiah (1988) found that the more recent
Bem short form (the basis for the current study) is a better indicator of the
culturally defined standards of masculinity and femininity than the earlier
form. A thorough review by Spence (1991) showed that the BSRI measures broader
concepts, that it has construct validity and that the newer short form
eliminates the problems of the earlier longer form. Bem's inventory remains an
often utilized instrument in such research (Freeman, 1994; Stimpson, Neff,
Jensen & Newby, 1991). Even when researchers develop their own instruments, Bem
is often the basis for these new instruments (e.g., Street, Kromrey & Kimmel,
1995).
In their previous studies (Lueck, Caplan & Endres, 1994; Lueck,
Endres & Caplan, 1993) the researchers developed a 15-item instrument to chart
gender stereotypes in the journalism/mass communication classroom. To gauge
student perceptions of desirable masculine and desirable feminine traits, five
masculine statements and five feminine statements were used. An instrument was
formed from the pool of the highest loaded BSRI items for masculinity and
highest loaded BSRI items for femininity from the Bem short form. Statements
from a pre-existing departmental evaluation form were used as filler questions,
which are typically used in studies employing the BSRI. The items selected from
Bem's short form were not only from the pool of highest loaded masculine and
feminine characteristics but also those with potential for application to the
journalism/mass communication classroom. For example, one highly loaded feminine
characteristic selected was "eager to soothe hurt feelings," instead of another
highly loaded feminine characteristic, "affectionate."
In the journalism/mass communication field there are few studies
examining instructor sex and gender stereotyping as factors in student
evaluations. This is not to say that the subject of student evaluations has not
been studied in the journalism/mass communication field (see Hudson, 1989). Yet,
this field seems to be an ideal area for such a study for several reasons.
Current enrollment trends not withstanding, the journalism/mass communication
professional field has traditionally been dominated by males. The exception to
this is the employment situation in public relations (U.S. Department of Labor,
1989; AEJMC, 1989). Charting bachelor's degrees granted from 1966 to 1991,
Becker and Graf (1995) showed that women are responsible for "the dramatic
growth" of communication practitioners in general and journalists in particular
(pp. 4-6), but women instructors in journalism/mass communication on the college
level remain a minority (Schamber, 1989). Teaching journalism/mass communication
also calls for certain characteristics or talents that are not commonly
associated with "feminine behavior." Production classes are technology and
equipment intensive. Moreover, classes such as reporting require certain
aggressive questioning and other behaviors not associated with feminine
stereotypes.
Journalism/mass communication is a field that should be studied
discretely. But it is crucial at this time that it be studied in the context of
the humanities-based area of communication. Within a context of academic
mergers, the trend seems to be to merge schools of journalism and mass
communication under umbrella schools of communication. For instance, the Ohio
State University merger of their school of journalism with their communication
department was occasioned by this trend of resource consolidation. OSU President
E. Gordon Gee stated, "It is indeed time for us to revisit past organizational
decisions" (Magill, 1994, p. 20). More recently, in 1996, Bowling Green State
University, also in Ohio, has merged its School of Mass Communication with
Interpersonal Communication.
Methodology
With the prevailing climate of mergers in mind, researchers
examined a midwest university that already has incorporated journalism/mass
communication courses into its communication school. In addition to
journalism/mass communication courses, the school houses interpersonal,
organizational, public relations, rhetoric and speech courses. The undergraduate
program is divided into two major subject areas of study, with one area being
journalism/mass communication and the other area encompassing speech to
organizational communication. Because of its business orientation, public
relations in this school is grouped within this second communication subject
area. The school also offers a Master of Arts. Although on the graduate level
these two areas are less formally defined, students nonetheless choose to
concentrate in mass communication or communication. The structure of this school
represents one way such a consolidation may be modeled. In general ways it
adheres to the national definition of "communications," which has included
journalism since 1971 and mass communications since 1983, and also includes
public relations (Becker & Graf, 1995, p. 5). However, this school also houses
speech and rhetoric, which the Department of Education does not consider under
the umbrella of communications (p. 5). Working from within the larger
communication context, this study was designed to further the research on
student reactions to the sex of the instructor, student perceptions of the
instructor's gender and the effect these perceptions have on course evaluations
in journalism/mass communication.
To gauge the effect of gender on student evaluation ratings, one
of the most widely known and accepted instruments (Cook, 1985, pp. 40-41), the
Bem Sex-Role Inventory (BSRI), was used as the basis for this study's
instrument. This study used the 15-item instrument that the researchers
developed in their previous studies. For this study, the 15-item BSRI instrument
was imbedded in a 154-item course evaluation questionnaire. Piggy-backing their
instrument enabled researchers to reach a larger population and more variety of
classes throughout the school of communication than would have been otherwise
possible. Using this method, researchers were also able to compare the BSRI
results with student responses on a standardized evaluation. Nine full-time
instructors -- five journalism/mass communication and four communication; six
male and three female -- volunteered their classes for administration of the
questionnaire. Women taught four of the journalism/mass communication courses
and two of the communication courses.
The questionnaire was administered to students in 18 classes. Nine
of the courses were communication classes, including Communication Theories and
Organizational Communication. Seven of the courses were journalism/mass
communication courses, including Survey of Mass Communication, Editing and
Theories of Mass Communication. Two of the courses were quantitative research
methods courses that crossed into both areas. Data from the two research methods
courses were not included when the courses were broken out by area. Sixteen of
the courses were undergraduate and two of them graduate courses. Class
enrollment rosters showed a potential total student population of 458. On the
questionnaire, students provided demographic information, including sex and
major area of study. Instructors were indicated by sex, and their course by area
type. Researchers administered the questionnaire to students present in each
class on a voluntary and confidential basis within a two-week time span in the
middle of the Spring 1995 semester. The instructor was not present when the
students filled out the questionnaire.
Each student was asked to respond on the Likert scale as to his or
her degree of agreement with each of the statements, with "1" indicating the
highest degree of agreement. The majority of the 154 items, or 131 of them, had
been selected from the standardized Iowa evaluation, a widely used and well
tested evaluation instrument. The 15-item BSRI was included with five of the
items defined as feminine and five as masculine, according to Bem; the other
five were filler questions. Remaining items were for student demographics and
codes to identify the course. The students coded their responses on computerized
answer sheets.
The following hypotheses were tested:
1. Male students will rate male instructors more favorably than
female instructors on the standardized evaluation.
2. Female students will rate female instructors more favorably than
the male instructors on the standard evaluation.
3. Female journalism/mass communication instructors will be rated
higher on the BSRI masculine scale than female communication instructors.
4. Female communication instructors will be rated higher on the
BSRI femininity index than the female journalism/mass communication instructors.
5. Female instructors will be rated higher than male instructors on
the standardized evaluation.
Answer sheets were entered into the computer, and 328 cases were
analyzed using SPSS. Analyses of variance (ANOVAs) were used to test the
hypotheses for the relationships among the independent and dependent variables.
The independent variables were instructor sex, student sex and class type
(journalism/mass communication or communication). Numerical values for
masculinity and femininity traits were summed to create BSRI masculinity and
BSRI femininity indexes, which were created from the BSRI's desirable masculine
and feminine traits as the researchers had done in their previous studies.
The BSRI indexes were used as dependent variables. In other ANOVAs,
the dependent variable was the selected-item Iowa standardized evaluation. This
version of this standardized evaluation was highly reliable, with Cronbach's
Alpha = .9895.
A limitation of this study was that the instructor sample did not
allow for control of some instructor variables, such as experience and tenure,
or the further class breakdown, such as size and specific type, which have
proven meaningful in past studies.
Results
The description of the student population was taken from the
students' self described demographic data on the questionnaire (see Table 1).
Of the 319 students who reported their sex, 244 were male and 75
female. Of the 165 students who designated their majors within communication
areas, 133 of them stated that they were in the area of journalism/mass
communication and 32 indicated that they were majoring in communication.
Graduate students were totaled with "other."
Of the 165 undergraduates who reported both their sex and major,
107 of the males and 26 of the females stated their major area as
journalism/mass communication; 28 of the males and four of the females
designated communication as their major.
Table 1
Student Population Described by Sex and Major
MAJOR
Communication Mass Communication Row
SEX
Male
Count: 28 107 135
Row %: 20.7 79.3 81.8
Col. %: 87.5 80.5
Total%: 17.0 64.8
______________________________________________________
Female
Count: 4 26 30
Row %: 13.3 86.7 18.2
Col. %: 12.5 19.5
Total%: 2.4 15.8
______________________________________________________
Column: 32 133 = 165
Total 19.4 80.6 = 100.0%
Number of missing observations: 163
______________________________________________________
In order to test H1. "Male students will rate male instructors more
favorably than female instructors on the standardized evaluation," and H2.
"Female students will rate female instructors more favorably than the male
instructors on the standard evaluation," the interaction effect for instructor
sex and student sex was evaluated. H1 and H2 were not supported.
An ANOVA performed on the data using the standardized evaluation as
the dependent variable with instructor sex by student sex by type of course
resulted in a 2x2x2 factorial design that showed that the two-way interaction
effect for instructor sex and student sex was
not significant (see Table 2). Therefore, H1 and H2 were not
supported. This indicated that there was no interaction between same-sex
instructors and students.
Table 2
Analysis of Variance: Standardized Evaluation
Source of Variation
Sum of
DF
Mean
F
Sig of F
Squares
Squares
Main Effects
196505.917
3
65501.972
10.138
.000
Instructor sex
183749.671
1
183749.671
28.439
.000
Student sex
9819.721
1
9819.721
1.520
.219
Type of course
2936.524
1
2936.524
.454
.501
Two-Way Interactions
98548.321
3
32849.440
5.084
.002
Instructor sex and
Student sex
6206.195
1
6206.195
.961
.328
Instructor sex and
Type of course
73718.287
1
73718.287
11.409
.001
Student sex and
Type of course
14654.217
1
14654.217
2.268
.133
3-Way Interactions
4.740
1
4.740
.001
.978
Instructor sex and
Student sex and
Type of course
4.740
1
4.740.
.001
.978
Explained
295058.978
7
42151.283
6.524
.000
Residual
1770352.657
274
6461.141
Total
2065411.635
281
7350.219
Cases processed: 328
Cases missing: 46 (14.0%)
______________________________________________________
With journalism/mass communication a traditionally male-identified
field demanding masculine behaviors, the researchers proposed H3. "Female
journalism/mass communication instructors will be rated higher on the BSRI
masculine scale than female communication instructors." This hypothesis was not
supported.
Table 3
Analysis of Variance: Masculinity Index
Source of Variation
Sum of
DF
Mean
F
Sig of F
Squares
Squares
Main Effects
127.138
3
42.379
3.667.013
Student sex
3.324
1
3.324
.288.592
Type of course
6.927
1
6.927
.599
.439
Instructor sex
116.887
1
116.887
10.114
.002
Two-Way Interactions
32.402
3
10.801
.935
.424
Student sex and
Type of course
8.718
1
8.718
.754
.386
Student sex and
Instructor sex
1.616
1
1.616
.140
.709
Type of course and
Instructor sex
19.136
1
19.136
1.656
.199
3-Way Interactions
9.504
1
9.504
.822
.365
Student sex and
Type of course and
Instructor sex
9.504
1
9.504
.822
.365
Explained
169.044
7
24.149
2.090
.045
Residual
3155.134
273
11.557
Total
3324.178
280
11.872
Cases processed: 328
Cases missing: 47 (14.3%)
______________________________________________________
The three-way ANOVA with the BSRI masculinity index as dependent
variable and student sex, type of class and instructor sex as independent
variables showed a significant difference for instructor sex; however, a
significant difference was not found for type of class (see Table 3).
With the cell mean of 11.53, the female instructors as a whole
ranked significantly higher on the BSRI masculinity index than did the male
instructors, whose mean was 12.96 (lower score = higher rating). Females ranked
higher than males on the masculinity scale, but when further differentiated by
type of course, communication ranked slightly higher than mass communication
instructors (see Table 4).
Table 4
Cell Means on BSRI Masculinity Index: Instructor Sex by Course Type
INSTRUCTOR SEX
Male Female
COURSE TYPE
Mass Comm 12.84 11.87
Communication 12.99 11.00
_____________________________________________________
Although there is not necessarily an inverse relationship between
masculine and feminine indexes on the BSRI, because of the hypothesized tendency
for female journalism/mass communication to favor masculine behaviors,
researchers proposed H4. In other words, it was expected that female
journalism/mass communication instructors would embody some of the male-defined
attributes of their traditionally male profession, and that this would not be
the case with female communication instructors. H4: "Female communication
instructors will be rated higher on the BSRI femininity index than the female
journalism/mass communication instructors." Although the cell means support this
hypothesis (see Table 6), there was not a significant difference between type of
class on the BSRI femininity index (see Table 5). This hypothesis was not
supported.
Using the BSRI femininity index as the dependent variable, the main
effect was only significant for instructor sex. The female instructors ranked
"higher" on this BSRI index as well. But the main effect for instructor sex and
type of class was not significant (see Table 5). With a cell mean of 10.36 for
the female instructors and a mean of 12.92 for the male instructors, female
instructors as a whole ranked higher on the femininity index.
Table 5
Analysis of Variance: Femininity Index
Source of Variation
Sum of
DF
Mean
F
Sig of F
Squares
Squares
Main Effects
409.005
3
136.335
9.914.000
Student sex
20.150
1
20.150
1.465.227
Type of course
19.370
1
19.370
1.409
.236
Instructor sex
369.484
1
369.484
26.869
.000
Two-Way Interactions
180.392
3
60.131
4.373
.005
Student sex and
Type of course
49.165
1
49.165
3.575
.060
Student sex and
Instructor sex
.813
1
.813
.059
.808
Type of course and
Instructor sex
136.550
1
136.550
9.930
.002
3-Way Interactions
8.389
1
8.389
.610
.435
Student sex and
Type of course and
Instructor sex
8.389
1
8.389
.610
.435
Explained
597.786
7
85.398
6.210
.000
Residual
3754.100
273
13.751
Total
4351.886
280
15.542
Cases processed: 328
Cases missing: 47 (14.3%)
______________________________________________________
The female instructors as a whole were rated more feminine than the
male instructors, with the female journalism/mass communication instructors
second in the ranking, but not significantly differentiated from the higher
rating (see Tables 5 & 6).
Table 6
Cell Means on BSRI Femininity Index: Instructor Sex by Course Type
INSTRUCTOR SEX
Male Female
COURSE TYPE
Mass Comm 12.39 11.20
Communication 13.05 9.03
_____________________________________________________
Since researchers would not be able to match instructors for
variables such as experience, they proposed H5. "Female instructors will be
rated higher than male instructors on the standardized evaluation." This
hypothesis was supported.
The three-way ANOVA with the standardized evaluation as dependent
variable and instructor sex by student sex by type of class showed a significant
main effect for instructor sex (see Table 2). On the standardized evaluation,
female communication instructors received best ratings and male communication
instructors the worst evaluations, with journalism/mass communication
instructors falling between, females ranking higher than the males (see Table
7).
Table 7
Cell Means on Standardized Evaluation: Instructor Sex by Course Type
TYPE OF COURSE
Mass Communication Communication
INSTRUCTOR
SEX
Male 294.62 313.29
Female 271.63 227.15
______________________________________________________
An interaction effect was revealed, but it was not the anticipated
same-sex interaction. The interaction effect was a two-way interaction effect
observed between instructor sex and type of class on both the three-way ANOVA
using standardized evaluation as dependent variable (see Table 2). A two-way
interaction effect was also observed in the ANOVA using the BSRI femininity
index as the dependent variable (see Table 5); in other words, there was a
significant difference by type of course and instructor sex when designating
degree of adherence to desirable feminine characteristics.
Discussion
Attempts to chart an interaction effect that previous studies
suggested between instructor sex and student sex were foiled. Researchers found
no significant interaction effect between same-sex instructors and students in
the aggregate. However, since the student population became overwhelmingly male
in this broader context, female instructors did not suffer adverse effects by an
observed lack of interaction effect, or same-sex favorable ratings between
student and instructor. With 244 of the 319 students who reporting their sex
being male, males dominated the sample in this study. When variables are
controlled for, an interaction effect could be revealed, with the female
instructors being at a disadvantage in the evaluation process.
Traditional ranking was discovered. Females who rated high on the
BSRI femininity index as well as the masculinity index also ranked high on the
standardized evaluation. In fact, all three indexes exhibited the same
high-to-low ranking of instructors: female communication, female journalism/mass
communication, male journalism/mass communication, male communication. In its
support for the last hypothesis, this study bore out what the literature
predicted, in that when instructor variables such as experience are not
controlled for, females are often ranked higher than males on standardized
course evaluations.
Two-way interaction effects were observed between instructor sex
and course type on both the ANOVA using the standardized evaluation as dependent
variable and the ANOVA using the BSRI femininity index as dependent variable. In
the journalism/mass communication courses, female instructors were ranked higher
than the male instructors. In communication courses, the same ranking was
observed. The females seem to have been rewarded for their high adherence to
desirable feminine characteristics (when paired with high ratings on the
masculinity index and when instructor factors remain uncontrolled).
With regard to major designation, this study's reported student
population differs from the school's reported student body. With 133 of the 165
students who reported their major area of concentration being journalism/mass
communication, journalism/mass communication majors dominated the population in
this study. However, these major designations are not apparent in the formalized
processing and reporting in the school of communication. A Spring 1996 count of
those students enrolled in the school of communication who had actually signed
contracts designating tracks, revealed 345 in communication areas and 215 in
journalism/mass communication areas.
Conclusions and Further Research
The dust has settled at this midwestern university, and some of the
results are rather unsettling. Unanticipated results caught the attention of the
researchers and beg further study. With 244 of the 319 reporting their sex as
male, males dominated the sample in this study. Can this overwhelming majority
of males in the classroom be attributed simply to the broadening of the field,
or does it portend a reversal in enrollment trends? In either case, it is an
unusual situation for journalism/mass communication educators, particularly
women, to find themselves in. When factors such as instructor experience are
controlled for, literature indicates that an interaction effect would be
anticipated. In such cases, women instructors would be at a disadvantage in the
evaluation process. Therefore, building gender-based alliances across the
communication disciplines seems to be a healthy strategy in this consolidation
climate, and a step that AEJMC President Shoemaker is already taking.
Results seem to indicate that female communication and female
journalism/mass communication educators share commonalties, which could provide
the foundation for important gender-based alliances. Mergers with communication
provide opportunities for journalism/mass communication instructors to build
such alliances. For women just now finding satisfaction within journalism/mass
communication, particularly in the recent opportunities to mentor women
students, the consequences of this study could be unsettling. It still appears
that women educators must fulfill gender androgyny -- perceptions of traditional
femininity in order to be accepted and perceptions of traditional masculinity in
order to be considered competent in their highly technical field.
With the important implications for journalism and mass
communication education that this research suggests, this study calls for more
studies to be done in the area of gender-based evaluation research, controlling
for instructor rank and experience as well as student demographics such as age
and GPA and the courses by size and type (i.e., production or lecture). The
researchers would also like to encourage more studies on student expectations of
women instructors in the humanities-based field of communication, to see if they
contrast with those expectations held in journalism/mass communication.
In addition to quantitative studies to help chart these phenomena,
qualitative methods are also called for in order to explore the questions this
study raises.
30
References
Arreola, R.A. (1984). Defining and evaluating the elements of
teaching. Proceedings of Academic Chairpersons: Evaluating Faculty, Students,
and Programs, 1 - 14.
Association for Education in Journalism and Mass Communication.
(1989). Women, Men and Media: Next Steps for Journalism and Mass Communication
Educators, special report.
Basow, S.A. (1992). Gender Stereotypes and Roles, Third Edition,
Pacific Grove, CA: Brooks/Cole Publishing Company.
Basow, S.A. (1995). Student evaluations of college professors: When
gender matters. Journal of Educational Psychology, 87, (4), 656-665.
Basow, S.A. & Silberg. N.T. (1987). Student evaluations of college
professors: Are female and male professors rated differently? Journal of
Educational Psychology, 79, 308-314.
Becker, L.B. & Graf, J.D. (1995) Myths & trends. (special report)
Arlington, VA: The Freedom Forum.
Bem, S.L. (1981). Bem Sex-Role Inventory: Professional Manual. Palo
Alto, CA: Consulting Psychologists Press Inc. This manual explains the newer,
shorter form and elaborates on the original Bem Sex-Role Inventory, initially
published in the Journal of Consulting and Clinical Psychology.
Bernard, M.E., Keefauver, L.W., Elsworth, G. & Naylor, F.D.
(1981). Sex-Role behavior and gender in the teacher-student evaluations. Journal
of Educational Psychology, 73, 681-695.
Bennett, S. (1982.) Student perceptions and expectations for male
and female instructors: Evidence relating to the question of gender bias in
teaching evaluation. Journal of Educational Psychology, 74, 170-179.
Cashin, W.E. (1988). Idea paper no. 20: Student ratings of
teaching: A summary of the research. Center for Faculty Evaluation and
Development, Manhattan, KS: KS State U.
Cashin, W.E. (1989). Idea paper no. 21: Defining and evaluating
college teaching. Center for Faculty Evaluation and Development, Manhattan, KS:
KS State U.
Cashin, W.E. (1990). Idea paper no. 22: Student ratings of
teaching: Recommendations for use. Center for Faculty Evaluation and
Development, Manhattan, KS: KS State U.
Centra, J.A. (1977). How universities evaluate performance: A
survey of department heads. Princeton, NJ: Educational Testing Service.
Cook, E.P. (1985). Psychological Androgyny. New York: Pergamon
Press.
Cooper, P.J., Stewart, L.P. & Gudykunst, W.B. (1982.) Relationship
with instructor and other variables influencing student evaluations of
instruction. Communication Quarterly, 30, 308-315.
Elmore, P.B. and La Pointe, K.A. (1974). Effects of teaching sex
and the evaluation of college instructors. Journal of Educational Psychology,
66, 386-389.
Fandt, P.M. & Stevens, G.E. (1991) Evaluation bias in the business
classroom: Evidence relating to the effects of previous experiences. The Journal
of Psychology, 125(4), 469-477.
Feldman, K.A. (1986). The perceived instructional effectiveness of
college teachers as related to their personality and attitudinal
characteristics: A review and synthesis. Research in Higher Education, 24,
129-213.
Feldman, K.A.. (1983). Seniority and experience of college teachers
as related to evaluations they receive from students. Research in Higher
Education, 18, 3-124.
Feldman, K. (1993). College students' views of male and female
college teachers: Part II -- evidence from students' evaluations of their
classroom teachers. Research in Higher Education, 34, 151-211.
Freeman, H.R. (1994). Student evaluations of college instructors:
Effects of type of course taught, instructor gender and gender role, and student
gender. Journal of Educational Psychology, 86, (4), 627-630.
Hudson, J.C. (1989). Expected grades correlate with evaluation of
teaching. Journalism Educator, 44, 38-44.
Kierstead, D, D'Agnostino, P. & Dill, H. (1988). Sex role
stereotyping of college professors: Bias in student ratings of instructors.
Journal of Educational Psychology, 80, 342-344.
Lombardo, J. & Tocci, M.E. (1979). Attribution of positive and
negative characteristics of instructors. Perceptual and Motor Skills, 48.
Lueck, T.L., Endres, K.L. & Caplan, R.E. (1994). The effect of
gender on course evaluations in mass communications: A pilot study. Ecquid Novi,
15 (2), 2-19.
Lueck, T.L., Endres, K.L. & Caplan, R.E. (1993) The interaction
effects of gender on teaching evaluations. Journalism Educator, 48 (3), 46-54.
Magill, S.H. (1994, March). Ohio's new land grant university. Ohio
State Alumni Magazine, 19-21.
Marsh, H.W. (1984). Students' evaluations of university teaching:
Dimensionality, reliability, validity, potential biases and Uuility. Journal of
Educational Psychology, 76, 707-754.
Marsh, H.W. & Ware, J.E. (1982). Effects of expressiveness, content
coverage and incentive on multidimensional student rating scales: New
interpretations of the Dr. Fox effect. Journal of Educational Psychology, 74,
1126-1134.
Martin, H.J. & Ramanaiah, N.V. (1988). Confirmatory factor analysis
of the Bem Sex-Role Inventory. Psychology Reports, 62, 343-350.
Naftulin, D.H., Ware, J.E.& Donolly, F.A. (1973). The Doctor Fox
lecture: A paradigm of educational seduction. Journal of Medical Education, 48,
630-654.
Schamber, L. (1989). Women in mass communication education: Who is
teaching tommorrow's communicators? In P.J. Creedon (Ed.), Women in Mass
Communication: Challenging Gender Values. Newbury Park, CA: Sage: 148.
Schmitt, B.H. & Millard, R.T. (1988). Construct validity of the
"Bem sex role inventory" (BSRI): Does the BSRI distinguish between
gender-schematic and gender-aschematic individuals? Sex-Roles, 19, 581-588.
Shoemaker, P.J. (1996, January). From the president. AEJMC News.
Association for Education in Journalism & Mass Communication, 2.
Spence, J.T. (1991). Do the BSRI and PAQ measure the same or
different concepts? Psychology of Women Quarterly, 15, 141-165.
Stimpson, D., Neff, W., Jensen, L.C. & Newby, T. (1991). The caring
morality and gender differences. Psychological Reports, 69, 407-414.
Street, S. Kromrey, J.D. & Kimmel, E. (1995). University faculty
gender roles perceptions. Sex Roles, 32, (5/6), 407-422.
Swim, J., Borgida, E., Maruyama, G. & Myers, D.G. (1989). Joan
McKay versus John McKay: Do gender stereotypes bias evaluations? Psychological
Bulletin, 105 (5), 409-429.
U.S. Department of Labor. (1988). Labor force statistics derived
from the current population survey, 1948-87, Bulletin 2307. Washington, D.C.:
U.S. Government Printing Office.
Wilson, D. & Wilson, K.G., Jr. (1976). Student ratings of
instruction. Journal of Higher Education, 47, 465-470.
Yanico, B. (1985). BSRI scores: Stability over four years for
college women. Psychology of Women Quarterly, 9, 277-283.
|
