Content uploaded by Tawanna Dillahunt
Author content
All content in this area was uploaded by Tawanna Dillahunt on Aug 29, 2016
Content may be subject to copyright.
Content uploaded by Tawanna Dillahunt
Author content
All content in this area was uploaded by Tawanna Dillahunt on Oct 16, 2014
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
! !
!
!
Democratizing-Higher-Education:-Exploring-
MOOC-Use-Among-Those-Who-Cannot-Afford-a-
Formal-Education-
Tawanna Dillahunt1, Zengguang Wang1,2, Stephanie Teasley1
1University of Michigan, Ann Arbor, MI, 2Peking University, Beijing, China
Abstract
Massive Open Online Courses (“MOOCs”) provide free access to higher education for
anyone with Internet access. MOOCs are considered a means for democratizing
education. These courses will hopefully provide an opportunity for individuals to learn
from the best educators in the world, as well as help expand their personal networks,
and facilitate their career development. However, research thus far shows that the
majority of people taking advantage of these courses are already employed, have post-
secondary degrees, and have encountered few barriers related to the affordability of
higher education. Little is known about MOOC learners with financial constraints and
who do not fit the typical profile of MOOC learners. This paper presents the results of
the analysis of data from six Coursera courses offered by the University of Michigan
from fall 2012 through winter 2013. In this analysis learners who self-identified as being
unable to afford to pursue a formal education (the target group) were contrasted to
other learners (the comparison group) in terms of demographics, motivations, course
enrollment, engagement and performance. Learners in the target group were primarily
male and over 25 years old. A statistically significant portion of the target group held
less than a 4-year college degree than the comparison group. Target learners were also
significantly underrepresented in the enrollment of the courses examined here.
Although the comparison group had a significantly higher completion rate overall than
the target group, the target group had a statistically significant higher rate of completing
courses with certificates of distinction. This article provides a discussion of these results
and suggests how MOOCs could be adapted to better address the needs of learners who
feel financially unable to pursue a more traditional path to a post-secondary education.
Keywords: Massive Open Online Courses; education; online learning; affordability! !
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
2
Introduction
Massive Open Online Courses (MOOCs) are seen as an opportunity to gain access to
education and professional development, and to develop new skills to prepare for high-
paying jobs (Pappano, 2012). Recent articles on MOOCs in both the scholarly literature
and the popular press emphasize the fact that hundreds of thousands of people around
the world now have access to courses offered by elite universities (Lewin, 2012).
Information and communication technologies have increased opportunities for higher
education, though the key beneficiaries are individuals from affluent families from the
Western Province (Liyanagunawardena, 2012). In addition, research thus far
consistently shows that the people taking advantage of MOOCs are already employed,
young, well educated, predominantly male, from developed countries, have higher levels
of formal education, and are unlikely to encounter barriers related to the affordability of
higher education (Christensen, Steinmetz, Alcorn, Bennett, Woods, Emanuel, 2013). In
short, the individuals expected to benefit most from MOOCs are inadequately
represented among the early adopters of this new form of education (ibid.). Although
MOOCs are seen as one possible path toward upward mobility, few studies have
examined whether and how the populations with the most to gain leverage these
resources. Therefore the goal of this study was to address the following question to
complement prior research: How do the demographics, enrollment, personal
motivations, performance and engagement of learners unable to afford a formal
education compare or contrast to learners who do not report being motivated by
financial constraints?
This paper provides the results of a comparison between MOOC learners who self-
identified as being unable to afford to pursue a formal education (the target group) with
other learners (the comparison group), looking specifically at demographic data and
motivations across 11 Coursera offerings from fall 2012 to winter 2013. The results
detailed here contribute a better understanding of an understudied and
underrepresented group. The aim is to determine how MOOCs might better serve those
who feel financially unable to pursue a more traditional path to a post-secondary
education studies.
Related Work
Massive Open Online Courses are considered a means for democratizing education
(Lewin, 2012; Wulf, Brenner, Leimeister, 2014). MOOCs address an unlimited number
of participants (“massive”); are offered free of charge or impose only low participation
fees (“open”); are not dependent on location as they are available via the Internet
(“online”); and the content consists of instructional lectures and assessment (“courses”)
(Wulf, Brenner, Leimeister, 2014; Clow, 2013; McAuley et al., 2010; Vardi, 2012).
However, research shows that MOOCs are reaching a fairly homogeneous population
and that those thought to benefit most from these courses are underrepresented in
course enrollments (e.g., Christensen, Steinmetz, Alcorn, Bennett, Woods & Emanuel,
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
3
2014). Therefore, it is unclear how people who are financially constrained, who may be
unemployed, and who have less formal education are taking advantage of these courses.
The question of whether they can benefit from participating in this new educational
context also remains open.
In a systematic review of 45 peer-reviewed papers in the MOOC-related literature
published between 2008-2012, Liyanagunawardena, Adams and Williams (2013) found
that the majority of articles discussed MOOC challenges and trends. McAuley, Stewart,
Siemens and Cormier (2010) advocated for a clear research agenda to help evaluate
both the feasibility and the potential of the MOOC model for opening up access to
higher education and the circumstances in which MOOCs might achieve this potential.
They identified several open questions and challenges, such as the role for MOOC
accreditation, understanding depth versus breadth in MOOC participation,
understanding the conditions in which MOOC participation can expand beyond those
with broadband access and advanced social networking skills, and the viability of
MOOCs from an economic perspective. Understanding how underrepresented learners
compare to the majority MOOC learners in terms of demographics, motivations,
engagement and performance could help to evaluate the feasibility of the conditions by
which MOOCs might achieve their potential for democratizing education.
Researchers from the University of Pennsylvania analyzed more than 400,000 surveys
from individuals enrolled in 32 Coursera courses (Christensen, Steinmetz, Alcorn,
Bennett, Woods, Emanuel, 2013). In these courses, 83% of the registered learners had
two-or four-year degrees, and of those, 44% had some graduate education. According to
an analysis of 17 online courses offered on the edX platform, Ho et al. (2014) found that
of those reporting, the most typical edX MOOC learners were males with bachelor’s
degrees who are 26 and older (31% of learners). Learners reporting their gender as
female represented 29%. Learners enrolling in these MOOCs appeared to be diverse in
terms of highest education achieved (33% reported high school and lower), age (6.3%
reported being 50 or older) and 2.7% of the students had mailing or IP address from the
least developed countries as listed on the United Nations (Ho, Reich, Nesterko, Seaton,
Mullaney, Waldo & Chang, 2014). The authors reported that despite the low percentages
of learners from typically underserved populations, these courses were still reaching a
large number of these learners and that the edX MOOCs were attracting diverse
audiences.
MOOCs are still relatively new (Clow, 2013) and unexplored in the literature on distance
education and online distance learning. Many research questions are still open in regard
to the learning analytics on MOOCs and understanding trends such as the high drop
rates (Clow, 2013). In particular, future research should further investigate the types of
learners taking advantage of MOOCs and their motivations. The research discussed in
this article provides an understanding of a population that has not yet been studied—
learners who report being unable to afford a formal education. This data for this study
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
4
comes from learners who registered in Coursera courses offered by the University of
Michigan, a large midwestern university. The goal of the research was to address the
following research question: How do the demographics, enrollment, personal
motivations, performance and participation of learners compare or contrast to learners
who do not report being motivated by financial constraints? This article provides the
results of this analysis, and suggests how MOOCs could be adapted to better meet the
needs of this population. The results help to further develop hypotheses regarding the
performance and demographics of these populations across multiple MOOC courses,
platforms, and universities.
METHODOLOGY
As this study was exploratory in nature, statistical methods consisted of a series of
comparative analyses between the target and the comparison group. These data were
collected from the demographic surveys jointly administered by Coursera and the
University of Michigan at the beginning of six courses (see Table 1). The surveys were
voluntary and could be answered at any time during the course session. These surveys
were designed to provide learner demographic information and their motivation for
taking the MOOC (see Table 1, Question #3 for a complete list of motivations, and note
that learners could select more than one answer). Any learner who included the answer
indicating that they were unable to afford a formal education were classified as the
“target group” for analyses. The “comparison group” comprised of those learners who
selected any of the reasons other than affordability for enrolling in the course. Although
determining the target group on the basis of a single survey question does not reveal
possible variability in what affordability means to respondents, affordability was the
major factor identifying underrepresented learners in previous MOOC research.
Course enrollment, engagement, and performance data was available via the data
provided by the MOOC platform. Course engagement data included whether learners
accessed course material, watched videos and engaged in discussion forums. In terms of
forum engagement, learners could engage in four distinct activities: view a forum, view
a thread, up vote a thread or down vote threads. Forum posting data was not readily
available and does not appear in this analysis. Course performance data is available
through grades achieved in the course. There were two types of course completion
certificates—a basic “certificate of completion” and a “certificate of completion with
distinction.” In general, earning a certificate required completion of the course with a
minimum grade, or meeting a set of requirements set by each instructor; earning a
certificate of distinction required passing with a higher-grade threshold (which varied
from course to course). Before discussing the results of the analysis, details of the course
survey are presented next, as well as an overview of the courses analyzed.
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
5
Table 1
!
Survey'Questions'used'in'Analysis'
!
1.!What!is!your!gender?!(open!ended)!
2.!What!is!your!age?!
☐ Under!18!
☐ 18924!
☐ 25934!
!
☐ 35944!
☐ 45954!
☐ 55+!
!
☐ I!prefer!not!to!answer!
!
!
3.!Which!options!best!describe!your!motivations!for!taking!this!class?!(please!check!all!
that!apply)!
☐ Cannot!afford!to!
pursue!a!formal!
education!
☐ Supplement!other!
college/university!courses!
☐ Geographically!isolated!
from!educational!institutions!
☐ Extending!
current!knowledge!
of!the!topic!
☐ General!interest!in!the!topic!
☐ Decide!if!I!want!to!take!
college/university!classes!on!
the!topic!
☐ Professional!
development!
☐ Interest!in!how!these!courses!
are!taught!
!
4.!What!is!your!highest!level!of!education?!
☐ Some!high!school!
☐ High!school!
☐ Some!college!
☐ Associate’s!
degree!(2!year’s!of!
college)!
☐ Bachelor’s!degree!(BA/BS!49
year’s!of!college)!
☐ Master’s!degree!
5.!What!is!your!current!occupation?!Select!all!that!apply!
☐ Student!
☐ Faculty!
☐ Teacher!
☐ Other!
!
!
Survey and Course Overview
Survey responses were gathered from multiple offerings of six distinct courses offered in
the Fall of 2012 through the Winter of 2013, for a total of 11 course offerings. These
courses were 5 to 15 weeks long and taught by university professors at the University of
Michigan. The advertised workload for the courses ranged from 4-12 hours per week.
The courses were classified into three categories: 1) Humanities, 2) Economics and
Finance, and 3) Technology. Specific course names are included in Table 2.
Table 2
Summary of Courses Offered
!
Course Type
Course Names
Course #
Humanities
Fantasy and Science Fiction
1
Economics and
Finance
Model Thinking
Intro to Finance
2
3
Technology
Internet, History, Technology and
Security
4
Social Network Analysis
5
Securing Digital Democracy
6
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
6
RESULTS
In total, 666,407 learners registered for the six courses and approximately 6.3%
(N=42,097) took the demographic surveys. Note that not all of the questions were
answered by every student thus the n varies by item (see survey response rates in Table
3). Only 9.08% (N=3,812) of those completing surveys represented the target
population (i.e., stated that they were not able to afford a formal education).
Table!3!
!
Number of survey participants and (response rates) for each course by term offered
!
Course Type
Course #
#Participants
Fall 2012
Winter 2013
Humanities
1
37,118 (2.90%)
23,318 (.77%)
Economics and Finance
2
102,802 (2.47%)
38,429 (17.50%)
3
125,332 (5.42%)
89,362 (9.50%)
Technology
4
41,683 (10.94%)
34,218 (18.67%)
5
61,754 (1.97%)
35,363 (10.29%)
6
19,582 (2.43%)
*NA
Note. *Survey responses were unavailable.
Table 3 shows that the courses with the highest survey response rates occurred in the
Winter of 2013. While the causes of these variations in response rates are unclear,
factors such as the popularity of MOOCs in general or the commitment of learners after
evaluating the courses the first time around may reflect response rates.
The next section provides the results of the demographic survey, motivations for
enrollment (e.g., other than an inability to afford a formal education), and course
enrollment details. The section concludes with details about how the target population
performed relative to the comparison group. Where applicable, statistically significant
differences between the target and comparison groups are specified.
Demographics
Overall, 41,636 learners that responded to the question of gender and 68.65%
(N=28,585) were male. Of the 41,734 learners that responded to the question of age, the
largest age group taking courses was 25-34 (39.78%, N=16,603), and the second largest
age group was 18-24 (22.67%, N=9,461). For the total number of learners responding to
the survey, 99.68% (N=41,961) answered the question regarding their motivations for
taking the course. Of these, approximately 9.08% (N=3,812) were in the target
population. The remaining 90.92% (N=38,149), those in the comparison group, did not
select affordability as their motivation for taking the courses.
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
7
Tables 4-1 and 4-2 below provide a breakdown of the gender and age of the target and
comparison groups.
!
Table!491!!
Gender Breakdown for Learners in the Target vs. Comparison Groups
Target
(i.e., Cannot Afford, N=3,762)
Comparison
(i.e., Other, N=37,788)
Gender
Count
Percentage
Gender
Count
Percentage
Male
2,467
65.58%
Male
26,053
68.95%
Female
1,295
34.42%
Female
11,735
31.05%
Table!492!!
Age Breakdown for Learners in the Target vs. Comparison Groups
Target
(i.e., Cannot Afford, N=3,798)
Comparison
(i.e., Other, N=37,855)
Age
Count
Percentage
Gender
Count
Percentage
18-24
764
20.12%
Male
8,678
22.92%
25-34
1,690
44.50%
Female
14,883
39.32%
The target and comparison groups are relatively the same in terms of gender and age.
Learners in both groups were primarily male (~70%) and between 25-34 years old. This
finding is consistent with the age and gender demographics reported in prior research
(Christensen, Steinmetz, Alcorn, Bennett, Woods, Emanuel, 2013).
Overall, 37,148 learners responded to the question of motivation and country of origin.
Table 5 below provides a summary of the country of origin of both groups.
Table 5
!
*Country of Origin, Count and Percentage of Enrollment for Learners in the Target vs.
Comparison Groups
!
Target
(i.e., Cannot Afford, N=3,191)
Comparison
(i.e., Other, N=33,957)
Country
Count
Percentage
Country
Count
Percentage
US
1,065
33.38%
US
9,615
28.32%
IN
236
7.40%
IN
2659
7.83%
GB
154
4.83%
BR
1,502
4.42%
*Note that the survey asks for country of origin rather than the current country of residence
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
8
The majority of learners in the target and comparison groups were from the United
States, followed by India. Great Britain was third among the target group while Brazil
was third among the comparison group. Consistent with prior research, the single
largest group of learners is from the U.S., but there were also learners taking courses
from developing regions (Ho, Reich, Nesterko, Seaton, Mullaney, Waldo, & Chuang,
2014).
Educational Achievement
Overall, 41,709 participants responded to the survey question regarding their
motivations for taking the course and their highest educational degrees achieved (Figure
1 questions 3 and 4). See Table 6 below for details.
Table 6
Degree Achievement for Learners in the Target and Comparison Groups
!
Group
Target
(i.e., Cannot Afford,
N=3,790)
Comparison
(i.e., Other,
N=37,919)
Some high school
72
12.33%
512
87.67%
High school
355
13.40%
2,294
86.60%
Some college
593
17.78%
2,742
82.22%
Associate’s degree
(2 years of college)
251
14.14%
1,524
85.86%
Bachelor’s degree
(BA/BS, 4 years of
college)
1,519
9.83%
13,931
90.17%
Master’s degree
834
5.95%
13,194
94.05%
Professional degree
(MD, JD)
85
5.91%
1,354
94.09%
Doctoral degree
81
3.31%
2,368
96.69%
As shown in Table 6, approximately one third (33.63%, N=14,028) of all individuals
responding to the survey reported that their highest degree achieved was a master’s
degree, and 37.04% (N=15,450) had a bachelor’s degree. These results also show that a
statistically significantly higher percentage of the target population reported having a
bachelor’s degree than those in the comparison group (40.08% vs. 36.74%, z=4.06, p<.
01).
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
9
In addition, a larger majority of target learners had bachelor’s degrees (40.08%
N=1,519) than master’s degrees (22.01% N=834), which is significantly different from
the comparison group (36.74% N=13,931 vs. 34.80% N=13,194 respectively) (z=52.18
p<0.01). In fact, learners in the comparison group had a statistically significant higher
proportion of advanced degrees (e.g., master’s degree and higher) than the target group
(44.61% vs. 26.39%, z=21.61, p<0.01), while a statistically significant portion of the
target group had less than a four-year college degree in contrast to the comparison
group (33.54% vs. 18.65%, z=21.84, p<0.01).
Motivations
Figure 1 shows a comparison of the reported motivations (excluding the ability to afford
a formal education) for taking MOOCs between the target and comparison populations
(N=42,097).
Given the large number of learners, all differences were statistically significant at the
p<0.01 level (except to supplement other college/university classes/courses). The target
learners, however, were five times more likely to indicate being motivated to take
courses due to issues of geographic isolation than the comparison learners. The target
learners were twice as likely to indicate being motivated to decide if they wanted to take
college/university classes on the topic.
Course Enrollment
Course enrollment data (Figure 2) was analyzed based on education level in addition to
affordability to compensate for any barriers to entry in terms of course difficulty.
!
Figure 1
Additional motivations of learners based on affordability
81.90%'
67.47%'
58.34%'
31.77%'
17.44%'
16.19%'
17.39%'
79.53%'
57.40%'
54.01%'
26.54%'
16.14%'
8.20%'
3.46%'
0.00%' 20.00%' 40.00%' 60.00%' 80.00%' 100.00%'
'General'interest'in'the'topic'
'Professional'development'
'Extending'current'knowledge'of'the'topic'
'Interest'in'how'these'courses'are'taught'
'Supplement'other'college/university'classes/courses'
'Decide'if'I'want'to'take'college/university'classes'on'
the'topic'
Geographically'isolated'from'educaKonal'insKtuKons'
Mo#va#on'for'MOOC'Enrollment'by'Affordability'
Comparison'(i.e.,'Other)'
Target'(i.e.,'Cannot'afford)'
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
10
All learners:
!
Target Learners:
Comparison Learners:
Figure 2 – Percentage of course enrollment by degree achievement
Course 1: Fantasy and Science Fiction
Course 2: Model Thinking
Course 3: Introduction to Finance
Course 4: Internet, History, Technology and Security
Course 5: Social Network Analysis
Course 6: Securing Digital Democracy
Figure 2 shows that the two courses with the highest enrollment percentage for those
with less than a 4-year degree across both target and comparison learners were Courses
4, which is a basic, technology-related course and 2 (Economics). By contrast, the
highest percentage of enrollment for those with 4-year degrees and higher was Course 5,
a more advanced technology course, followed by Courses 2 (Economics) and 3
(Finance).
Table 7 provides details regarding the actual number and percentage of the enrollment
of each course per term offered. Note that none of the learners enrolled in the
Humanities and Technology courses offered in the fall 2012 term reported an inability
to afford a formal education (Table 7).
24.16%'
15.42%' 16.68%'
31.72%'
11.91%'
19.37%'
0.00%'
20.00%'
40.00%'
60.00%'
80.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5' Course'6'
Percentage)Enrolled)
Less)than)a)four)year)college)degree)
75.84%'
84.58%' 83.32%'
68.28%'
88.09%'
80.63%'
0.00%'
20.00%'
40.00%'
60.00%'
80.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5' Course'6'
Percentage)Enrolled)
Four)year)college)degree)and)higher)
60.53%'
28.28%' 25.83%'
55.78%'
22.74%'
0.00%'
10.00%'
20.00%'
30.00%'
40.00%'
50.00%'
60.00%'
70.00%'
80.00%'
90.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5'
Percentage)Enrolled)
Less)than)a)four)year)college)degree)
39.47%'
71.72%' 74.17%'
44.22%'
77.26%'
0.00%'
10.00%'
20.00%'
30.00%'
40.00%'
50.00%'
60.00%'
70.00%'
80.00%'
90.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5'
Percentage)Enrolled)
Four)year)college)degree)and)higher)
23.08%'
14.42%' 15.37%'
29.47%'
11.19%'
19.53%'
0.00%'
10.00%'
20.00%'
30.00%'
40.00%'
50.00%'
60.00%'
70.00%'
80.00%'
90.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5' Course'6'
Percentage)Enrolled)
Less)than)a)four)year)college)degree)
76.92%'
85.58%' 84.63%'
70.53%'
88.81%'
80.47%'
0.00%'
10.00%'
20.00%'
30.00%'
40.00%'
50.00%'
60.00%'
70.00%'
80.00%'
90.00%'
100.00%'
Course'1' Course'2' Course'3' Course'4' Course'5' Course'6'
Percentage)Enrolled)
Four)year)college)degree)and)higher)
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
11
Table 7
!
Representation'of'target'and'comparison'groups'by'course:'
!
Course type
Term
Target
Learners
Percentage
of Target
enrolled
Comparison
Learners
Percentage
of
Comparison
enrolled
Humanities
Fall 2012
-
0.00%
3,607
100.00%
Winter
2013
688
9.99%
6,199
90.01%
Economics
and Finance
Fall 2012
847
12.52%
5,917
87.48%
Winter
2013
1,062
12.54%
7,407
87.46%
Technology
Fall 2012
-
0.00%
6,239
100.00%
Winter
2013
1,215
12.16%
8,780
87.84%
Grand Total
3,812
9.08%
38,149
90.92%
Despite the low percentage of the target group in the overall sample, results showed a
significant increase in the population over each course term offered (Table 7).
To better understand how issues of affordability may interact with educational
attainment, Table 8 provides details regarding which courses may have attracted target
learners who held less than a 4-year degree. Course 6 was removed from the table as
survey responses were unavailable for Winter 2013.
Table 8
!
Percentage of Learners in the Target Group who have less than a 4-year Degree by
Course.
Course 1
Course
2
Course
3
Course 4
Course
5
Fall 2012
22.41%
12.60%
14.42%
26.99%
9.50%
Winter
2013
34.64%
16.47%
18.48%
35.12%
12.72%
z-statistic
z=-3.54,
z=-4.57,
z=-6.67,
z=-8.98,
z=-2.99,
p<0.01
p<0.01
p<0.01
p<0.01
p<0.01
As shown in Table 8, target learners with less than 4-year degrees had the highest
enrollments in Courses 1 (Fantasy and Science Fiction) and 4 (Internet, History,
Technology and Security). In addition to Humanities, these students are enrolling
heavily into technology courses, which may suggest areas of future research. In fact, the
increase in enrollment from Fall 2012 to Winter 2013 are statistically significant. Details
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
12
of enrollment suggest that helping these populations may require access to courses that
provide marketable skills.
Engagement and Performance
Log data of student activity in the course was used to analyze the participation, or
engagement and performance between the two groups. These data included the number
of times learners watched videos and completed assessments, forum engagement as well
as the outcome earned in each course (no certificate, certificate, certificate of
distinction). Overall, 48.88% of those that registered, including those not completing
the surveys (N=325,743 Ntotal = 666,407), performed some activity within the course
(e.g., actually watched a video, up or downvoted a thread, viewed a thread or a forum,
looked at course materials and/or conducted an assessment). Consistent with prior
research on MOOC completion rates (Christensen, Steinmetz, Alcorn, Bennett, Woods,
Emanuel, 2013; Ho, Reich, Nesterko, Seaton, Mullaney, Waldo & Chuang, 2014), only
4.40% of all learners registered for these courses completed them and earned a
certificate. Table 9 below details the course completion results based on affordability.
Table 9
Level of completion based on affordability
Achievement Level
Target
(i.e., Cannot Afford, N=3,812)
Comparison
(i.e., Other, N=38,149)
Count
Percentage
Count
Percentage
Certificate with
Distinction
339
9.11%
2,274
6.09%
Certificate only
716
19.24%
13,645
36.58%
None (e.g., did not
complete)
2,757
71.65%
22,230
57.33%
There were no significant differences between the two groups’ engagement in terms of
watching videos, accessing course materials and/or conducting assessments. However,
as measured by the total count of forum activities (up vote, down vote, view thread and
view forum), participation among the target population (94.65%) was significantly less
than the percentage of the comparison population (96.68%, z=-6.5, p<0.01). In
addition, those in the comparison group had a higher percentage of course completion
(36.58% vs. 19.24%, z=21.07, p<0.01). Despite these findings, a higher percentage of the
target group completed a course with a certificate of distinction than the comparison
group (9.11% vs. 6.10%, z=7.18, p<0.01).
Summary of Results
In summary, the demographics of learners from both groups were similar in coarse
grain terms of gender (i.e., majority male), age (25-34 years old), and country of origin
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
13
(i.e., majority U.S.), and are consisted with demographics reported in prior research
(Christensen, Steinmetz, Alcorn, Bennett, Woods, Emanuel, 2013; Ho, Reich, Nesterko,
Seaton, Mullaney, Waldo & Chuang, 2014). The demographic results also showed that
the second highest percentage of target and comparison learners were from India, which
provides evidence of learners from developing regions (Ho, Reich, Nesterko, Seaton,
Mullaney, Waldo & Chuang, 2014).
In terms of educational achievement, results showed that a statistically significant
portion of the target group (33.54%, N=1,271) had less than a four-year college degree in
contrast to the comparison group (18.65%, N=7,072). Similar findings have been
reported in a more focused study (Dillahunt, Chen & Teasley 2014). Target learners
were also significantly more motivated to enroll in MOOCs than the comparison
learners for all reasons except to supplement other courses.
While learners in the target group primarily enrolled in Economics and Finance, those
with less than a four-year degree enrolled at higher rates in the courses with content
focused on basic technology. Consistent with the fact that they had higher levels of
education, the comparison learners had higher enrollment in the more advanced
technology course. Nevertheless, there was an indication that the percentage of learners
indicating an inability to afford a formal education had increased with each offering of
the course (e.g., 12% increase in Course 1, 4% increase in Courses 2 and 3, 8% in Course
4 and 3% in Course 5 per Table 8).
Finally, and perhaps the most interesting result, although comparison learners had a
higher completion rate overall, target learners had a signicantly higher rate of
completing courses with certificates of distinction (36.58% vs. 19.24%, z=21.07,
p<0.01). This is despite the finding that participation among the target population was
significantly less than the percentage of the comparison population. These results
contribute insight into an unexplored MOOC population and additional insight into
these learners’ demographics, motivations, enrollment and performance; however,
these findings raise additional questions and directions for future research.
DISCUSSION AND LIMITATIONS
The motivation behind this work was to understand the differences in demographics,
motivations, course enrollment, and engagement and performance between learners
who enrolled in a MOOC for reasons related to the affordability of traditional higher
education as compared with learners who enrolled for reasons other than affordability.
While the target learners could potentially reap the most economic benefit from taking
these courses, the study findings show that this group only represents 9.08% of the
surveyed population. A promising finding is that when these learners do complete a
MOOC, they are more likely to earn a certificate with distinction than those who
enrolled in the MOOC for reasons other than educational affordability. Understanding
more detail about the motivations of these individuals is worth further investigation.
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
14
For example, are these individuals primarily motivated for professional development? If
so, are they specifically motivated to transition to new jobs, or to refresh their current
skillset? Are the key differences in motivations between the target learners from the U.S.
versus other regions related to geographic locations? These questions were beyond the
scope of this initial exploration and fully understanding these findings will be aided by
qualitative data focused on the nuances behind affordability.
STUDY LIMITATIONS
Perhaps the most significant limitation of this study is the potential for sampling bias
inherent in opt-in surveys. Specifically, the survey method lends itself to a self-selection
bias where learners choosing to respond to the pre-course surveys are usually more
likely to be active course participants. In addition, those with the ability to respond to
these surveys were more likely to respond—it is possible that those underrepresented
populations in which this study was designed to explore were the least likely to complete
the surveys due to issues of affordability, accessibility and time. For example, certain
regions may have intermittent Internet access or impose fees based on the amount of
time spent online. Secondly, the reliability and accuracy of survey responses are always
uncertain, and the issue of “affordability” is relative. For example, indicating, “I cannot
afford to pursue a formal education” could mean that someone cannot afford to pursue
a formal education financially, but it could also be interpreted as “I cannot afford to take
time out of my schedule to pursue a formal education”.It is also possible that some
learners from the comparison group were also not able to afford a formal education but
they chose not to select this answer in the survey. Finally, the study data is limited to
data from courses offered by a single U.S. university, though with a worldwide audience.
Despite these limitations, the results of the analysis do offer an initial insight into an
underrepresented and unexplored population of learners. These limitations alone
provide implications for reaching underrepresented learners in the future.
Future MOOC Research
MOOCs are considered a means for democratizing education. An open question and
challenge is to understand the feasibility of the MOOC model for opening up access to
higher education and the potential to do so (McAuley, Stewart, Siemens & Courmier,
2010). The demographic results of this study are consistent with prior research showing
that MOOCs are primarily taken by well-educated males, 26 years and older, from
developed regions and who are unlikely to encounter financial constraints for pursuing
their education. Learners who have less formal education, are women, older adults,
individuals from developing regions, and those with financial constraints are
underrepresented in MOOCs.
As mentioned in the study limitations, leveraging the survey to understand
demographics and motivations of MOOC learners presents sampling bias and
difficulties in reaching targeted populations. To better understand the factors related to
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
15
issues of affordability, future research should explore whether and how MOOC
platforms can capture more detailed information about learners during their activity
and engagement in the courses. For example, is there a way to determine if learners are
accessing courses via broadband, mobile, dial-up or from public facilities such as
libraries, Internet cafes (which may be more common in developing regions), or
universities? How can statistical models be used to detect enrollment of learners from
these populations? What features can be used to identify these learners and barriers
they may face (e.g., IP address to identify location, engagement trends, the type of
technology being used to access the MOOCs)? What interventions could reduce these
barriers?
Christensen, Steinmetz, and Alcorn (2013) describe a lack of technological access as the
key reason poor people have not taken the opportunity to study online. Indeed research
reveals that information and communication technologies have increased opportunities
for higher education, although primarily for those individuals from affluent families
from the Western Province (Liyanagunawardena, 2012). In an overview of the
educational developments in open, distance, and technology-facilitated learning to
reach world-wide populations deprived of education, Gulati found that new technologies
have done little to help deprived groups gain access to educational opportunities
(2008). Gulati’s research has shown that these groups continue to be marginalized due
to their lack of access to basic education and adequate learning resources. However, the
rapid growth of mobile devices in developing countries may enhance the development of
mobile learning to educate the masses (Gulati, 2008).
With worldwide penetration of the mobile-broadband subscriptions—almost 3 billion
Internet users, two-thirds from the developing world and mobile-broadband uptake
growing at double-digit rates by the end of 2014 (ITU, 2014)—access constraints may be
declining. It is unclear, however, whether learners are leveraging mobile phones to
access MOOC content. The results from this research suggest that financially
constrained learners are finding ways to access these courses though these possibly
represent the most motivated and the most affluent learners in certain regions. A better
understanding of the methods in which learners access these courses could help to
further understand these issues.
Although access is a concern, another issue could be a lack of awareness of the potential
benefits MOOCs could offer. It is unclear how learners find out about MOOC courses
and interesting to know whether sources differ from learners from the target and
comparison groups. This could help to understand how information about the courses is
currently being disseminated within these learner communities. Advertising MOOCs via
billboards, radio and television, job placement offices, Internet cafes and libraries could
help to raise awareness to the people who might benefit most from MOOCs. Although
not discussed in the context of this study, it is also unclear whether and the extent in
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
16
which English as the primary language of instruction in MOOCs presents access barriers
to learners from developing and non-English speaking regions.
Finally, many unknowns still exist; including new pre-course survey questions could
shed additional light on those learners that choose to complete the surveys. For
example, requesting specific occupation information, current salary, place of degree
attainment, and job type (e.g., full-time, part-time) could be beneficial. Understanding
these factors could help to tease apart information about each cohort of learners and
how these cohorts change over time. As mentioned earlier, exploring MOOC features to
detect details such as methods of access and creating new models to predict when these
learners engage could offer additional insight to ways to better meet the needs of these
populations.
CONCLUSION
The aim of this study was to address the question: How do the demographics,
enrollment, personal motivations, performance and engagement of learners unable to
afford a formal education compare or contrast to other learners? Results from six
Coursera courses offered by a large research university from fall 2012 through winter
2013 show that while learners who self-reported an inability to afford a formal
education were majority males, primarily over 25, they also:
1. Had a significant portion of learners with less than a four-year college degree
than learners in the comparison group (33.54% vs. 18.65%, z=21.84, p<0.01);
2. Were generally more motivated to enroll in MOOCs than those in the
comparison group due to issues of geographic isolation (five times more likely
to select this motivation than comparison) and deciding if they wanted to take
college/university classes on the topic (twice as likely to select this motivation
than comparison);
3. Were significantly more likely to be awarded a certificate of achievement (9.11%
vs. 6.10%, z=7.18, p<0.01) than those in the comparison group.
The goal of this research was to explore underrepresented MOOC populations as a
starting point to better understand how to open up access to higher education to
economically constrained populations. Future work includes obtaining more qualitative
data about targeted learners via interviews to better understand their MOOC
experiences, whether their goals are to obtain certificates with distinction and why, and
investigating models that help to predict targeted learners. Future work also includes
updating surveys to obtain details about targeted learners such as income, place of
degree attainment, and employment status.!!
!
!
!
!
!
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
17
ACKNOWLEDGMENTS
The! authors! would! like! to! thank! the! Bill! and! Melinda! Gates! Foundation! for! funding!this!
work.!Moreover,!many!thanks!go!to!the!invaluable! feedback!from!our! reviewers!and!from!
those!in!the!!USE!Lab!at!the!University!of!Michigan’s!School!of!Information.!
!
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
18
REFERENCES
Christensen, G., Steinmetz, A., Alcorn, B., Bennett, A., Woods, D., Emanuel, E., The
MOOC Phenomenon: Who Takes Massive Open Online Courses and Why?
(November 6, 2013). Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2350964.
Clow, D. (2013). MOOCs and the funnel of participation. In: Proc 3rd international
conference on learning analytics and knowledge (LAK ’13), New York, pp 185-
189.
Dillahunt, T., Chen, B. Teasley, S. (2014). Model Thinking: Demographics and
performance of MOOC students unable to afford a formal education. Learning
at Scale, Work in Progress, 2014.
Gulati, S. (2008). Technology-Enhanced Learning in Developing Nations: A
review. The International Review of Research in Open and Distance Learning.
9(1), 1-16. Retrieved from
http://www.irrodl.org/index.php/irrodl/article/view/477.
Ho, A. D., Reich, J., Nesterko, S., Seaton, D. T., Mullaney, T., Waldo, J., Chuang, I.
(2014). HarvardX and MITx: The first year of open online courses (HarvardX
and MITx Working Paper No. 1).
International Telecommunications Union (ITU). (April 2014). The World in 2014: ICT
Facts and Figures. Retrieved from http://www.itu.int/en/ITU-
D/Statistics/Documents/facts/ICTFactsFigures2014-e.pdf
Lewin, T. (2012, March 24). Instruction for Masses Knocks Down Campus Walls. The
New York Times. Retrieved from http://goo.gl/ilPEc.
Liyanagunawardena, T., Adams, A., Williams, S. (2013). MOOCs: A systematic study of
the published literature 2008-2012. The International Review of Research in
Open and Distance Learning. 14(3), 202-227. Retrieved from
http://goo.gl/VMOqt.
Liyanagunawardena, T. (2012). Information and Communication Technologies and
distance education in Sri Lanka: a case study of two universities. PhD thesis,
University of Reading. Retrieved from http://centaur.reading.ac.uk/32337/
McAuley, A., Stewart B., Siemens G., Cormier D. (2010). The MOOC model for digital
practice. Retrieved from
http://www.elearnspace.org/Articles/MOOC_Final.pdf.
Democratizing Higher Education: Exploring MOOC Use Among Those Who Cannot Afford a
Formal Education
Dillahunt, Wang and Teasley
Vol x | No x Month/yr.
!
19
Pappano, L. (2012, November 2). “The Year of the MOOC.” The New York Times, ED26.
Retrieved from
http://www.nytimes.com/2012/11/04/education/edlife/massive-open-online-
courses-are-multiplying-at-a-rapid-pace.html.
Vardi, M.Y., (2012). Will MOOCs destroy academia? Communications of the ACM
55(11):5.
Wulf, J., Blohm, I., Brenner, W. & Leimeister, J.M. (2014): Massive Open Online
Courses. In: Business Information System and Engineering (BISE),
Ersheinugsjahr/Year: 2014. Seiten/Pages: Online First.
