Research report: What are the biggest barriers to online learning?

It seems like higher education is beginning to enter the "Acceptance" phase of the Big Pivot. We're realizing that the shift to online instruction done on an emergency basis back in March was not a temporary event. More and more universities are moving to all-online instruction in Fall 2020, and some have even extended this shift into 2021. It's safe to say that regardless of developments with the pandemic, the sudden prominence of online learning in higher education is here to stay.

One of the most important things we can do right now to help make Fall 2020 a success for students, or at least keep it from being a complete disaster, is to try to understand what students are thinking about online learning and to ascertain what they need in order to be successful. We can learn from the recent past on this, because for as long as there has been online learning (the concept dates back to the 1960's, even predating ARPANET), there have been student perceptions of barriers to online learning. I read an article recently that studied this very thing:

Muilenburg, L. Y., & Berge, Z. L. (2005). Student barriers to online learning: A factor analytic study. Distance Education, 26(1), 29-48.

Official repository for the paper here, but it's behind a paywall; however a little Googling will turn up some unrestricted PDFs.

What is this study about?

The title says it all: It's a study that tries to determine what the perceived barriers are for student engaged in online learning.

Let me qualify that word "perceived". Barriers to learning that are perceived are, in fact, actual barriers, even if the thing that students perceive is not fully real. The paper points out that (as of 2005, and I think this is still true today) the preponderance of evidence indicates that although there are no significant differences in the effectiveness of online learning when comparing "well-designed" online learning experiences versus "well-designed" F2F experiences, perceived barriers can still degrade the student learning experience.

I don't like this fact any more than anybody else, but the fact remains. And it's on me to always take pains to gather intel on what my students are thinking and to address their concerns, and help them across those barriers.

What did the researchers do?

This is a questionnaire-driven study in which a sample of people in higher education (in the USA) were surveyed about their beliefs, experiences, and expectations with online learning. There was an initial pilot study done to determine an initial list of potential patterns and to shake out the bugs in the process. Then, eventually, the researchers looked at results from 1056 respondents. The sample was a mix of stakeholders — students who were currently taking online courses, students who had completed one or more such courses, students with no online course experience, and professionals gleaned from organization membership lists and email lists.

Using the pilot study, the questions on the main survey were grouped into six parts: technical, infrastructure/support services, social, prerequisite skills, motivation, and time/interruptions with a total of 47 different barriers on the survey. Subjects were asked to rate the barriers in terms of how much of a barrier they were. (Unfortunately there's not much detail on the survey itself, and the link to the pilot study given in the paper no longer works.)

What did the researchers find?

The full results of the paper include such items as demographic data and different instances of correlation. The findings below are the ones that really stood out to me.

735 of the respondents had studied or were studying online. (So 321 had not taken any online course yet.)
68% of the respondents said they were "comfortable and confident" with online learning. (It doesn't say how much overlap there is between this group and those who had studied or were studying online, which is 735/1056 = 69.6% of the sample.)
Of the respondents who had studied online, 33% said they cannot learn as well online as in a F2F setting; 44% saw no difference; and 23% said they learn better online.
The respondents who had not yet studied online couldn't answer the same questions from experience, so they were asked to predict how well they'd learn in the different modalities. 60% predicted they would learn poorly online; 32% predicted no difference; 7.5% predicted they would learn better online.
Of the respondents who had studied online, 31% said they enjoyed learning online less than they enjoyed learning F2F; 35% enjoyed online learning about the same as F2F; 34% said they enjoyed learning online more.
Of the respondents who had not studied online, 54% predicted they would enjoy online learning less, 35% predicted they would like online learning about the same, and 12% predicted they would like online learning more.

Since this was a factor analysis, the researchers also tried to isolate relationships between variables, in this case to examine how different barriers might correlate with each other and which ones were the most significant.

The most significant barrier to online learning was a perceived lack of social interaction. It's not clear from the study what "social interaction" consists of, but in practice this is a combination of student-student interaction, student-instructor interaction, and "instructor presence" which itself can take on three different forms (teaching presence, social presence, and cognitive presence).
Also significant, but less so than social interaction, were administrative/instructor issues, time and support for studies, and learner motivation.
Less important barriers were technical issues, a lack of technical skill, and a lack of academic skill.
The degree to which barriers to learning were perceived was inversely related to comfort and confidence levels with using online learning technologies. That is, the ones with the highest levels of comfort/confidence with using these technologies had the lowest perceived barriers in terms of number and degree; and as the comfort/confidence level dropped, the barriers were higher and more numerous.
The respondents who indicated they can't learn well online or who predicted a lack of success in learning online, had the highest barrier ratings. Those who said there's no difference in their learning, had moderate ratings on barriers Those who felt they learn better online had the lowest ratings on barriers.
Before you cry foul at the obviousness of the last couple of bullets, the authors make an important note: "It makes sense that perceived barriers decrease as experience with online learning increases. What is most interesting is the huge drop in barriers perceived after completing just one course, where fear of the unknown appears to be important." Students who had taken or were taking at least one online course had significantly lower barriers than those who had not taken any and were making predictions.
There was a strong association between social interaction and the degree to which respondents enjoyed, or predicted they would enjoy, an online course. In fact this was the strongest interaction among all variable pairings.

What do these results mean?

Since this study was done in 2005, there are some questions about the applicability of the results to 2020 (see below). However, let's draw a few tentative conclusions from the data.

First: Experience with online learning seems to have a moderating effect on perceived student barriers. Obviously correlation is not causation (see below) but it's notable that the respondents with some online experience had fairly evenly-distributed beliefs about the effectiveness and enjoyability of online learning versus F2F, respondents with no experience were significant more pro-F2F in their responses. Experience tends to have a moderating effect on everything, so this isn't really surprising, but there are some potentially profound secondary effects, such as...

Second: Students' stated lack of belief that they can learn in an online setting might simply be the result of guesswork and fear of the unknown. When I was department chair back in March, as soon as we pivoted online I started getting emails from students telling me that they "can't learn math online" and wanted to know what they could do about it (for example, get a tuition refund). My response was to tell them to talk to their profs about what they needed, and give it a week and check back in. Most did, and all of them told me that after giving online learning a shot, it was actually OK. The lesson here is that when students say they "can't" do something, we should listen, but also put our filters on. Many times it's just a limiting belief, a fixed mindset at work, or a guess — all based on fear of the unknown. It could be more than that, but I'd bet that most of the time when students express their perceived inabilities, our best next move is to help them work through it by giving them a good experience, rather than helping them avoid the experience.

That leads me to the next conclusion, which is a little controversial:

Third: The notion that students – especially new students – overwhelmingly want F2F instruction in Fall 2020, might not be as real as we think. I've been hearing, ever since April, that if we (generally speaking) go all-online in the Fall, our enrollment numbers will crater and financial disaster will ensue. As a result, we (generally speaking) are trying to make F2F instruction happen, at great expense of time and money and putting faculty and students at considerable health risk. While I don't distrust the sources who tell me that F2F needs to happen to avoid a financial meltdown, I have also seen very little data supporting this point, and what I have seen is based on – you guessed it – perceptions of students, many of whom have no experience with online learning. Based on this study, I think it's entirely possible that we could be responding the wrong way.

The research suggests that even just a little experience with online learning can significantly change student perceptions. So instead of a fraught F2F experience, what if we instead committed to going fully online, and reinvest the time and money spent on making F2F happen into creating a well-designed, required 1-credit online course for all Fall students to take — maybe an intro to their discipline, or a "how to learn" course based on Make It Stick, or something — and offer it in the four weeks before Fall classes start, free of charge to all those who have deposited tuition? That would give every student the little bit of experience with the tools, workflows, and overall vibe of an online course that will help them succeed in an online environment later.

For most of us, the train's already left the station and it's too late to implement something like this. But perhaps this is something to try in the future, like Winter/Spring semester. At least, let's put our filters on institutionally just like we put them on individually.

Fourth and finally: The importance of social presence can't be overstated. The study points out that social interaction is the biggest correlate with enjoyment of online learning, and the lack of social interaction is the biggest barrier for student success. Simply put, if you want to make your online course a success, then in addition to having a solid structure based on clear and measurable learning objectives, you need to work hard to create a consistent and engaging social presence. I'll have more to say on this in another post.

Limitations and questions

First of all, 2005 was a very different time in higher ed than 2020. Online courses weren't as rare as you might think — according to a 2013 Babson Group study, over 3.1 million students in the US were enrolled in at least one online course in 2005, accounting for 18.2% of total enrollment at degree-granting post-secondary institutions. But still, back then, online courses were niche pursuits mainly for community colleges and for-profit schools. People who took them had different sets of needs and motivations as they do now. So be careful projecting this onto your freshmen in the fall, some of whom were still in diapers at the time.
Second, as mentioned, correlation is not causation. It's possible for instance that rather than experience having a moderating effect on perceptions, that perceptions drive experience — students who had a high comfort/confidence level with online learning went on to take more online courses and those with less comfort fewer courses, rather than experience determining the comfort level.

The only question I would have is whether this study still holds up. Somebody should replicate this study over the next academic year. It would certainly be a good thing to know what the perceived barriers are now, given our extraordinary circumstances.