Running head: GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY

Gender Differences in the Workload of Faculty

Gender Differences in the

Workload of Faculty

The “ivory tower” of academia is often thought to be a place where people can meet and

exchange ideas freely without fear of being treated differently because of their gender. It is also

well known that there is a difference in the types of pay and tasks men and women are assigned

in the workplace. These differences are a source of friction within our society that has been

ongoing for many generation. Because this is something that is brought up in the politics of the

nation on a fairly common basis it made the research begin to question if this was also so of

faculty in higher education. The basis for this study is simply that. To determine if there is a

difference in the tasks assigned to faculty based on their gender. Another questions is raised in

this study, but is not the main focus, is there a difference in tasks assigned to faculty based on

their attainment of a doctoral degree or not.

Purpose

The purpose of this study is important because without knowing so, one would assume

that all things are equal in higher education faculty task assignments. That would seem to be the

fair thing to do. Theoretically that would also seem wise so that no one person is given special

treatment and their tasks are assignment based on merit. If we look at the concept of equity

theory Adams postulated that employees thrive in knowing that their inputs will receive a fair

reward from it (1965). Students who are working towards becoming faculty members very well

would like to know if they are entering a place of employment that is fair to their gender or not.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 3

Definitions

In this paper, when the topic of workload is discussed it is meant to mean the tasks to

which the faculty member is assigned to spend the most hours on each week. The tasks that will

be discussed include four things; research & development, teaching, management &

administration, and professional services to individuals. When it is discussed whether that

faculty member has a doctorate or not, this means that the faculty member has either a Ph.D.,

DSc. or Ed.D. or does not. The does not have a Ph.D., DSc. or Ed.D. category can include

faculty members with M.S., M.A., J.D., M.D., or other professional degrees that are considers of

masters level or higher in education.

Literature Review

At first glance of the literature of this data, it seems that it could be hard to tell if gender

does have a specific role on which tasks are assigned to faculty. One problem is that when you

look at the aggregate data of many studies they sometimes show that faculties where more

females and less males are less productive in the amount of research that they produce. But, then

when you break the data down to the individual level the numbers do not look the same and

females in the faculty might actually be more productive (Porter & Umbach, 2000).

In 2005 Pyke noted that in every faculty rank at the universities with the highest levels of

research activity women accounted for only 36% of the faculty. Women only help 7% of the full

professor positions of the whole faculty where as men help 28% of the full professor positions of

the whole faculty. This seems to show that women are not as prevalent in these types of

institutions and also are not promoted at the same levels as men to positions higher up in faculty.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 4

Another study made a point that is very important to this study. Bellas and Toutkoushian

noted in that based on their research women were more likely than men to have more dependents

(students to advise) and thus often had less time for research or completed research at a slower

pace (1999). With more students to advise it would be much more difficult to be able to

complete research and thus why these faculty members are relegated to teaching and other tasks

besides research. Further, women are more likely to have family responsibilities at home which

further inhibits their career growth and slow research (Ramirez, 2010), thus requiring them to be

assigned other tasks.

Finally, Gander and Pyke both noted that many female faculty members are given a

larger portion of the maintenance jobs, such as mentoring an institutional housekeeping (1999,

2011). They did not specifically say teaching and this leaves that question still left unanswered.

One further thought on the topic, a survey completed by Short in 2006 showed that 95% of

women contributed teaching to be one of the factors of job satisfaction. Whereas only 75% of

men said the same thing. So, it may not necessarily be negative for more women to be assigned

roles in teaching as their primary task. This could lead to greater job satisfaction.

Methodology
The first hypothesis for this study is to what extent the gender effects the tasks a faculty

member must spend most of their time. The second hypothesis is to what extent does the level of

degree a faculty member has attained effect the tasks they spend most of their time. The

dependent variable (DV) is the workload (task) at which the faculty spend most of their time per

week working on. The first independent variable (IV A) is the gender of the faculty member.

The second independent variable (IV B) is whether the faculty member has a conferred doctorate
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 5

degree (either a Ph.D., DSc., or Ed.D.). The data was gathered from Inter-University

Consortium for Political and Social Research’s survey entitled, A Data-Based Assessment of

Research-Doctorate Programs in the United States, 2005-2006. The target population for this

study was all doctorate granting universities in the United States of America. The survey was

sent out to all faculty members of these institutions. There were a 169,600 surveys that were

sent out. Of those surveys 115,550 had data that was usable for this study. To reach a power of

.80 the sample would have to have been 92,440. I calculated this by taking the power at

115,550, which was 1.0 and take out the difference to get that power down to .80.

Limitations

There were several limitations to this study and its data. First, the surveys were only sent

out to universities that grant doctoral degrees in the United States. This was part of the purpose

of this original study from which this data collected. However, it limits this study to only those

institutions and does not include data from every other university and community college in the

United States. This data does also not include information from the doctoral granting institutions

of the for-profit industry since they are generally not considered research institutions.

Research Design

The design that was chosen for this study is a two way factorial analysis of

variance (ANCOVA). This was chosen because the dependent variable of this study is has four

levels and each of the two independent variable has two levels. The statistical methods that will

be need to be used to complete the analysis of this data will first be Levene’s Test of the Equality

of Error Variances. This test will be used because there is more than one variable and in a two

way ANOVA it is assumed that the variances that of the populations that the samples are taken
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 6

from are equal. Next, a test of between-subjects effects will be performed. This test will tell us if

there is a statistically significant interaction between each of the independent variables and the

dependent variable. It will also tell us if the interaction between the two independent variables

and the dependent variable was statistically significant and the error in the model. Finally, there

will be two different charts showing the interactions between the independent variables. The

first, a profile plot, will show if there was an interaction between the independent variables and

also if they had a significant effect. The second chart, an error bar, will show if there was an

interaction between the two independent variables and which one of the size of the independent

variables. The hypothesis for the first independent variable is:

𝐻0 = 𝜇1𝑗 = 𝜇2𝑗 = 𝜇3𝑗 = 𝜇4𝑗

𝐻1 = 𝐴𝑡 𝑙𝑒𝑎𝑠𝑡 2 𝑚𝑒𝑎𝑛𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑎𝑐𝑡𝑜𝑟 𝐴 ≠ 0

The hypothesis for the second independent variable is as follows;

𝐻0 = 𝜇𝑘1 = 𝜇𝑘2 = 𝜇𝑘3 = 𝜇𝑘4

𝐻1 = 𝐴𝑡 𝑙𝑒𝑎𝑠𝑡 2 𝑚𝑒𝑎𝑛𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑎𝑐𝑡𝑜𝑟 𝐵 ≠ 0

The hypothesis for the interaction is as follows;

𝐻0 = 𝜇𝐽𝐾 = 𝜇𝑗𝑘 … = 𝜇𝑗𝑘 = 0

𝐻1 = 𝐴𝑡 𝑙𝑒𝑎𝑠𝑡 𝑜𝑛𝑒 𝜇𝑗𝑘 ≠ 0

The alpha level that will be used for this study is .05.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 7

Statistical Data Analysis

Descriptive Statistics

When looking at the descriptive statistics for this study there are a few conclusions to be

drawn. First, there are 113,430 faculty members that have an earned doctorate degree while only

2,124 in the study did not. This shows a very large difference. Further, there are 87,461 total

men in the study and 28,093. This means that women only make up slightly less than a quarter

of the sample of this study.

Levene’s Test

The Levene’s Test of Equality of Error Variances for this test showed that the assumption

for equality was violated with a 𝑝 = .001. Thus, we did not fail to reject the null hypothesis that

population variances would be equal. Thus, population variances are not equal. Although

looking at the descriptive statistics it is fairly clear why Levene’s test was not meant.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 8

It was decided to run a Kolmogorov-Smirnov test. Based on this test we see that our data is not

normally distributed.

Test of Between-Subject Effects

When looking at the results for the test of between-subjects effects the results were as

expected based off of the literature that was reviewed. The first thing looked at is the interaction

between the two independent variables, gender and if the faculty have an earned doctorate, and

the dependent variable of the tasks that the faculty member spent the most time on. This showed

there is not a statistically significant interaction for these variables with a 𝑝 = .080, 𝑓 =

3.067 𝑎𝑛𝑑 𝑎 𝑝𝑜𝑤𝑒𝑟 𝑜𝑓 .417 (highlighted in red on the table). However, there was a statistically

significant main effect for gender with 𝑓 = 38.104, 𝑝 = .001 𝑎𝑛𝑑 𝑝𝑜𝑤𝑒𝑟 = 1.0 (highlighted in

green on the table). There was also a statistically significant main effect if the faculty had an

earned doctorate or not with 𝑓 = 38.095, 𝑝 = .001 𝑎𝑛𝑑 𝑝𝑜𝑤𝑒𝑟 = 1.0 (highlighted in blue in the

table). Gender did have a lower overall mean between the two main effects.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 9

Box Plot
The box plot for this two way ANOVA shows a couple of different items that are

important to look at. First, it did show that there was no interaction between the two independent

variables, neither ordinal nor disordinal. It also showed that the independent variables of gender

and level of education both had a strong effect on the dependent variable.

Error Bar

The error bar showed a couple of pieces of useful data as well. First and most important

it showed that there was not an interaction between the independent variables. The second and

also interest point is that it shows that the more of faculty in this data set had an earned doctoral

degree than faculty that did not. This could be part of the reason why the Levene’s Test of

Equality of Error Variances was not met.

Conclusions

Because the “ivory tower” of academia holds a certain respect in the world. Many view

it as a safe place to pursue their academic dreams and exchange knowledge with some of the

greatest minds of the day. This study was created to shed light on this world and help get a

glimpse of the inner workings of higher education faculty. Thus, it is of interest to many to

know the workload (tasks) assigned to faculty members of higher education institutions. This is

an especially important topic for anyone in academia or thinking about entering this field. Being

able to see if there is a difference in tasks that genders are assigned or if it matters what level of

degree you have earned can be an important influence on the decisions that students and future

academics must consider to determine their next move. With this information they can have a

clearer picture of what to expect.

When looking at the data collected from this study there are several conclusions that can

be drawn. The first big conclusion is that there does seem to be a strong statistically significant

effect between the variables of tasks assigned to faculty members and gender. This tells us that
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 11

there is a difference in gender roles in the higher education faculty workplace. This may be

something for students and other academics that are considering entering into this field to

consider and whether it fits their desire. Second, if the faculty member has a doctorate or not

also does seems to have a strong statistically significant effect on the variable of tasks assigned

to faculty members. This shows that it may be important to earn the doctorate degree to be able

to perform that tasks in the higher education faculty workplace that you want to do and this

should be considered. However, the effects of gender on tasks may not be as significant as it

seems thanks to the point made by Short as stated earlier in the literature review that female

faculty members place a higher value on teaching as part of the job satisfaction than men in their

role as higher education faculty (2006). So, this may not be a large problem for women wishing

to enter the role as faculty.

The data in this study could be used to further develop the knowledge of the roles that

gender plays in higher education. It could also help to shed some light on the importance of the

level of degree in higher education. Future research on this topic could look more closely at

what specific degrees are most useful to obtain what kinds of tasks as a faculty member’s

primary task. One could also break down the faculty into different fields of study to see if there

is a difference in what tasks are assigned to faculty by gender based on the field in which they

are working. For instance, if a Biology department was more likely to have women doing

research and development and the men were more likely assigned to the teaching/administrative

roles. While a chemistry department did the exact opposite. This would provide a better look at

the differences in gender within faculty working in higher education. It would also be interesting

to add other universities and colleges into a similar study to see if that changed how the results

ended up.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 12

Limitations

The first important thing to take note of is that because Levene’s Test of Equality of Error

Variances was not met and this means that the type 1 error rate was probably inflated. This is

due to the data samples not being normally distributed. There were far more faculty members in

the data that had an earned doctorate degree than not and far more men than women. This

caused the data to be skewed.

This study is also limited because the data would not allow delineation between different

types of advanced degrees so that it could be determined if there was a difference in tasks

assigned say in someone that has a M.S. degree vs. someone that has an M.D.

The data did not include information from many other universities and colleges in the

United States. These include the non-research doctoral granting universities, the non-doctoral

granting universities, colleges and community colleges. This leaves out a large number of

faculty in the country that could vastly change how the data appears.
GENDER DIFFERENCES IN THE WORKLOAD OF FACULTY 13

References

Adams, J. S. (1965) Inequity in social exchange. In L. Berkowitz (Ed.), Advances in

experimental social psychology (Vol. 2, pp. 267-299). New York: Academic Press.

Porter, S. (Director) (2001, May 22). Analyzing Faculty Workload Data Using Multilevel =

Modeling. Annual Forum of the Association for Institutional Research. Presentation

conducted from Association of Institutional Research, Cincinnati, OH.

Pyke, K. (2011). Service and Gender Inequity among Faculty. PS: Political Science & Politics,

44(1), 85-87.

Ramirez, S. (2010). Gender Inequities in Academe and Faculty Perceptions of Family-Friendly

Policies (Doctoral Dissertation). Retrieved from Proquest.

Short, H.N. (2006). Gender Disparities in Tenure-Track Faculty Job Satisfaction (Doctoral

Dissertation). Retrieved from Proquest.

Toutkoushian, R., & Bellas, M. (1999). Faculty Time Allocations and Research Productivity:

Gender, Race and Family Effects. The Review of Higher Education, 22(4), 367-390.