The Unexpected Pedagogical Benefits of Making Higher Education Accessible

Abstract

Many ongoing efforts in online education aim to increase accessibility through affordability and flexibility, but some critics have noted that pedagogy often suffers during these efforts. In contrast, in the low-cost for-credit Georgia Tech Online Masters of Science in Computer Science (OMSCS) program, we have observed that the features that make the program accessible also lead to pedagogical benefits. In this paper, we discuss the pedagogical benefits, and draw a causal link between those benefits and the factors that increase the program's accessibility.

Full text

The Unexpected Pedagogical Benefits
of Making Higher Education Accessible
Abstract
Many ongoing efforts in online education aim to
increase accessibility through affordability and
flexibility, but some critics have noted that pedagogy
often suffers during these efforts. In contrast, in the
low-cost for-credit Georgia Tech Online Masters of
Science in Computer Science (OMSCS) program, we
have observed that the features that make the program
accessible also lead to pedagogical benefits. In this
paper, we discuss the pedagogical benefits, and draw a
causal link between those benefits and the factors that
increase the program's accessibility.
Author Keywords
Online education; accessibility; higher education.
ACM Classification Keywords
K.3.2. Computer and Information Science Education.
Introduction
Recently, there have been several efforts to make
higher education more accessible, with massive open
online courses—MOOCs—leading this charge [1];
however, MOOCs have proved to be divisive in higher
education, with some educators praising their potential
to make educational content accessible [7], while other
researchers have noted that MOOCs have not yet
delivered on many of their original promises [3, 8].
Permission to make digital or hard copies of part or all of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that
copies bear this notice and the full citation on the first page. Copyrights
for third-party components of this work must be honored. For all other
uses, contact the Owner/Author.
Copyright is held by the owner/author(s).
L@S 2016, April 25-26, 2016, Edinburgh, Scotland UK
ACM 978-1-4503-3726-7/16/04.
http://dx.doi.org/10.1145/2876034.2893383
David Joyner
Udacity, Inc.
2465 Latham Street NW
Mountain View, CA 94040
david.joyner@udacity.com
Georgia Institute of Technology
85 5th Street NW
Atlanta, GA 30332
david.joyner@gatech.edu
Ashok K. Goel
Georgia Institute of Technology
School of Interactive Computing
85 5th Street NW
Atlanta, GA 30332
ashok.goel@cc.gatech.edu
Charles Isbell
Georgia Institute of Technology
School of Computer Science
801 Atlantic Dr. NW
Atlanta, GA 30332
isbell@cc.gatech.edu
L@S 2016 · Work in Progress
April 25–26, 2016, Edinburgh, UK
117

In the two years since the Georgia Tech online Masters
of Science in Computer Science debuted, we have
noted that in some ways, this program has delivered a
superior pedagogical experience to the traditional
residential program, specifically due to the factors that
made the program accessible. Here, we describe and
connect that accessibility, to the pedagogical benefits
we have observed. We conclude that with the right
planning, the relationship between accessibility and
pedagogy can be mutualistic, not parasitic.
OMSCS Program
The Georgia Tech OMSCS program was developed
through a partnership with AT&T and Udacity, and has
since grown to over 3000 students in two years. The
experience is similar to residential students: students
apply for admission, pay tuition, earn grades, and
complete the same assignments, projects, and exams.
The chief difference is that class material is delivered
via Udacity.com in lieu of classroom lectures.
Each OMSCS class is built around a MOOC developed by
Udacity. However, the for-credit offering of the class
includes individual instructor involvement, expert
human grading, complex projects, and other factors
that are uncommon in MOOCs. The courses themselves
are similar to distance learning classes, with the MOOC
as a textbook. The OMSCS, however, is built from the
ground-up to take advantage of the internet; it is
remarkably affordable compared to residential and
distance learning degrees; it is capable of admitting a
greater number of students; and it is an equivalent
degree to the residential program. Thus, the OMSCS is
large, flexible, affordable, and partially pre-produced
like MOOCs, but is also rigorous, accredited, and
subject to admissions and tuition like distance learning.
Increased Accessibility
The OMSCS program addresses four traditional barriers
to access: scheduling, geography, cost, and exclusivity.
 Schedule Flexibility. OMSCS classes require no
synchronous activities and present most class
materials on the first day of the semester, allowing
students to succeed without taking time off work.
 Geographic Flexibility. The OMSCS has no
residency requirement. Students may succeed
without uprooting their family or leaving their job
for two years. One student wrote, "I wrote and
submitted Assignment 1 from Liberia. Project 1 was
coded in Israel, and submitted in Palestine.
Assignment 4 was written and submitted from
Mongolia. And Project 2, from home in Portland."
 Economic Affordability. The program costs
approximately $8000 including books and fees. The
low tuition and flexibility noted above dramatically
reduce the true costs of the program. As a result, a
plurality of students enroll in the program not for
career benefits, but for personal enjoyment.
 Enhanced Inclusivity. The program has no
physical barrier to create an artificial student
capacity. As a result, while the on-campus program
fills up with only 10% of applicants, the OMSCS
accepts all qualified students. This creates a much
more diverse student body, with backgrounds and
experiences from outside computing.
Student Demographics
Table 1 compares the demographic information of the
online and residential sections of CS7637 in Fall 2014
[2]. Notably, online students tend to be older, more
educated, and more experienced. They are also more
likely to be male, domestic, and employed, although
Tools of the Program
 Piazza: The forum and
de facto "classroom" for
OMSCS classes, for class
discussions, questions,
announcements, and
other virtual analogues of
the in-class experience.
Every OMSCS course
heavily leverages Piazza.
 Peer Feedback: A tool
developed at Georgia
Tech for pairing students
for peer reviews. Peer
review is used as a
learning activity, not to
generate grades,
although experiments
have been performed on
using peer review to
support expert grading
[4]. Roughly one third of
the classes in the OMSCS
use Peer Feedback.
 T-Square: The Sakai-
based learning
management system
used for assignment
submission and expert
evaluation.
 Udacity: The host of the
program's prepared
lecture material, as well
as interactive exercises
and some class projects.
L@S 2016 · Work in Progress
April 25–26, 2016, Edinburgh, UK
118

the different gender ratios are due to the drastically
different international and domestic ratios between the
programs. Within each location, the gender ratios are
the same between the online and residential programs.
Pedagogical Benefits
The accessibility of the program has led to a student
population that is more intrinsically motivated to learn,
more experienced, and more professionally diverse. We
argue this unique student body and the structure of the
OMSCS combine to create unique pedagogical benefits.
Student-Initiated Discussions
In traditional residential classes, time is a finite
resource that must be budgeted carefully. Online, there
does not exist a finite amount of synchronous class
time to budget, and thus student-initiated discussions
do not compete with planned activities. In CS6460 in
Fall 2015, for example, the class concluded with over
500 discussions initiated by students.
Parallel Discussions
In residential classes, students must take turns and the
class (or groups within the class) may only discuss one
topic at a time. This makes individualization difficult: a
particular discussion may confuse the low-achieving
students or bore the high-achieving students. Online,
discussions may proceed in parallel, allowing students
to self-select which discussions are appropriate for their
ability. This also allows the volume of discussion to
expand; in CS6460 in Fall 2015, ~100 students and
instructors generated 11,000 posts in 17 weeks.
Extended and Self-Documenting Discussions
Because discussions in residential classrooms are time-
boxed within the assigned class time and conducted
verbally, the discussion must end with little left behind
except students' notes and memories. Online,
discussions have no need to come to an artificial end,
and all discussions inherently self-document. In
CS6460, for example, class projects are captured in
single long-running threads that document the project
team's progress across the entire semester.
Help-Seeking and Student-to-Student Interactions
In the OMSCS, students ask any questions on Piazza.
While the question may be implicitly targeted to the
instructor, students may enter and answer or follow-up.
Thus, student-to-student interaction dominates the
class experience. As a result, students receive rapid
answers and feedback from that diverse and
experienced student body instead of just instructors.
The average time to receive an answer is 30 minutes.
De Facto Teaching Assistants
With most classes enrolling anywhere from 200 to 500
students, there are reliably certain students who are
excessively helpful. To an outside observer, these
students would appear to be teaching assistants with
the extent to which they help their classmates.
Anecdotally, it is not unusual for students to note that
certain classmates dramatically improved their class
experience. This has also allowed instructors to identify
candidates for future teaching assistant positions.
Expert-Level Peer Feedback
As noted previously, many OMSCS use peer review [4]
to allow students to review classmates' assignments.
Although MOOCs often use peer feedback to generate
assignment grades [6], the OMSCS program instead
leverages the known pedagogical benefits of both
giving and receiving peer review [5]. The unique
Student Profiles
Online
Residential
Age
<24 15% 82%
25-34 47% 17%
>35 39% 1%
Gender
Female 10% 24%
Male 90% 76%
Location
Domestic 87% 32%
International 13% 68%
Level of Prior Education
Bachelor's 87% 94%
Master's 11% 6%
Ph.D. 2% 0%
Employment Status
Full-Time 90% 5%
Part-Time 5% 15%
None 5% 80%
Prior Coding Experience
<8 years 45% 90%
9-18 years 41% 10%
>18 years 14% 0%
Table 1: Comparison of online
and residential students in the
KBAI class in Fall 2014.
L@S 2016 · Work in Progress
April 25–26, 2016, Edinburgh, UK
119

student body present in the OMSCS means that experts
are participating as peers in these exercises as well.
Organic Student-Instructor Interaction
As instructors, we have been struck with how the online
forum structure creates a 24-hour classroom where we
may come, view discussions in their entirety, and
participate in every conversation, with everyone able to
listen. This has led to the most common positive
feedback we have received: students have commented
that the student-instructor interaction in these classes
exceeds any residential class they have taken.
Conclusion
In this paper, we have worked as participant-observers,
noting pedagogical benefits to students in the OMSCS
program not previously seen in residential classes.
OMSCS students agree: in Summer 2015, 90% of
online CS7637 students rated the course as better than
residential courses. CS6460, a course that relies on the
benefits outlined above, drew a similar assessment;
80% rated it as better than other courses. Students
frequently praise the teaching teams' involvement, the
chance to review classmates' work, and the active
student communities, all of which come from efforts to
increase accessibility. With the right circumstances and
planning, the relationship between accessibility and
pedagogy can be mutualistic, not parasitic.
Acknowledgements
The authors are instructors who have taught both
OMSCS and residential classes. We thank Joe Gonzales,
the creator of Peer Feedback and the staffs of the
Georgia Tech College of Computing, Georgia Tech
Professional Education, and Udacity for their support.
References
1. Daniel, J. (2012). Making sense of MOOCs: Musings
in a maze of myth, paradox and possibility. Journal
of Interactive Media in Education, 2012(3), Art-18.
2. Goel, A., & Joyner, D. (2016). An Experiment in
Teaching Artificial Intelligence Online. In D Haynes
(Ed.) International Journal for the Scholarship of
Technology-Enhanced Learning(1) 1.
3. Guzdial, M. (2014). Ubiquity symposium: MOOCs
and technology to advance learning and learning
research: limitations of MOOCs for computing
education: addressing our needs. Ubiquity,
2014(July), 1.
4. Joyner, D., Ashby, W., Irish, L., Lam, Y., Langston,
J., Lupiani, I., … Goel, A. (2016). Graders as Meta-
Reviewers: Simultaneously Scaling and Improving
Expert Evaluation for Large Online Classrooms. In
Proceedings from the Third ACM Conference on
Learning @ Scale.
5. Li, L., Liu, X., & Steckelberg, A. L. (2010). Assessor
or assessee: How student learning improves by
giving and receiving peer feedback. British Journal
of Educational Technology, 41(3), 525-536.
6. Lu, J., & Law, N. (2012). Online peer assessment:
effects of cognitive and affective feedback.
Instructional Science, 40(2), 257-275.
7. Milheim, W. D. (2013). Massive open online
courses (MOOCs): Current applications and future
potential. Educational Technology: The Magazine
for Managers of Change in Education, 53(3), 38-42.
8. Popenici, S. (2015). Deceptive Promises: The
Meaning of MOOCs. In McKay, E. (Ed.) Macro-Level
Learning Through Massive Open Online Courses
(MOOCs): Strategies and Predictions for the Future,
158-167.
L@S 2016 · Work in Progress
April 25–26, 2016, Edinburgh, UK
120