Research report: Experiencing the hyflex model

Where, and how, will we hold classes this fall? That's the question of the moment in higher education. Back in March, when the Big Pivot was upon us, the choice to move classes online wasn't easy, but it was  simple. What other choice did we have? But now that we're moving into the second iteration of this pivot, we have the perspective to take a more intentional approach.

Some universities are committing to face-to-face instruction; some have announced they'll continue fully online; many others (like my university) haven't decided yet, but are considering a range of options and might deploy all of them. Each university will have to make its own decision, and we won't know whether those choices were right until years from now. Both the ambiguity and the stakes are high, so it's more important than ever to look to research to inform those choices.

Let's take a look at a research article on a particular, and peculiar, form of online teaching that has been around since at least 2006 and has had limited adoption in higher ed --- until now, and I predict you're going to be hearing a lot about it soon.

Malczyk, B. R. (2019). Introducing Social Work to HyFlex Blended Learning: A Student-centered Approach. Journal of Teaching in Social Work, 39(4-5), 414-428.

Direct access here, although there's a paywall.

What is this study about?

This study is about the "hybrid flexible" or hyflex model of instruction. No, not industrial strength work gloves or bespoke resume paper, but a specific form of hybrid instruction that focuses on maximizing student choice.

Discussions about online learning distinguish between face-to-face (F2F), online, and hybrid or blended formats, where the proportion of online instruction is 0%, 100%, and between 0% and 100%, respectively. Hybrid courses are often defined by that percentage. At my university, for instance, a class is "hybrid" if at least 15% (but less than 100%) of course meeting time has been moved online --- for a 3-credit course, that's about 6 clock hours. The percentage may vary by instructor; some hybrid instructors go way beyond 15% online, for example holding the entire class online except for three exam sessions.

Almost always, though, that ratio of F2F to online instruction is set by the instructor or institution, once and for all when the course is built. Students do not get a say in that percentage, and they have to follow the schedule once it's set. But what if we changed it so that the course was set up with equivalent F2F and online structures running simultaneously, so students could engage with the course either F2F or online, effectively letting each student determine their own F2F-online ratio and their own schedule, based on their needs and life situations in the moment?

If you did this, you would be using the hybrid flexible or hyflex model. Hyflex class sessions are available both F2F and online (through streaming video, or recordings, or both) and students can attend either mode of class, or both. Learning activities are available both for F2F and online students, and assignments can be selected from a range of options tailored to either the online or F2F environments.

When I first heard about hyflex several years ago through this EDUCAUSE whitepaper, I thought that an instructor would have to be either insane or underemployed to consider using it, because it sounds like it doubles your workload. But with the Covid-19 pandemic, hyflex now appears to be a compelling answer to the questions about what we're going to do in the fall. And as I read more about it, I'm starting to think it's not as unworkable as I thought.

That gets us to this study, which looks at the results of a hyflex experiment done in a social policy course at the University of Nebraska-Kearney. The researcher posed the following research questions:

• Given the opportunity to participate in various modalities, will student preferences for different modalities change?
• When given the opportunity to participate in hyflex, will students blend the modalities to suit their needs? That is, what will be the reasons students pick one mode over another?
• When given a choice of modality, which format do students actually select
• What benefits and challenges did students experience with each modality
• What are students' overall attitudes toward hyflex?
• Do students feel that rigor and quality of different modalities are comparable?

What did the researcher do?

The study involved one section (18 students) of a required undergraduate social policy course taken entirely by Social Work majors and minors. Class sessions were once per week from 4-7pm. A 5-week segment of the course was redesigned as a hyflex course. Before switching to hyflex, students filled out a questionnaire asking them which mode (F2F or online) they thought they would prefer. Then, during the five weeks, students were given the freedom to choose from week to week (that is, from class to class) whether they would attend class in person or online.

There were three ways students could participate:

• F2F: During the in-person meetings, students downloaded worksheets with questions and exercises from the LMS. The instructor and students worked through them together while students took notes and worked on the exercises. At the end of the meeting, students would upload their work to the LMS.
• Synchronous online: The F2F meetings were livestreamed over Zoom. Students who participated in this way would do the same activities as the F2F students.
• Asynchronous online: Students could also participate asynchronously, by doing the same exercises as the F2F and synchronous students, on their own time and agreeing to abide by established due dates.

Prior to the switch to hyflex, students were asked which modality (F2F, synchronous online, asynchronous online) they would prefer, the rank ordering (1-3 with 1 = most preferred) of their preferences, and which mode they predicted they would chose for each of the five sessions. Following the hyflex period, students were given the preference survey again and asked about the pros and cons of each modality, whether they would choose the hyflex format over fully F2F or online courses, and whether they felt the rigor and quality of the online coursework was equal to that of the F2F activities.

What did the researcher find?

Students tended to prefer the F2F modality both before and after the hyflex implementation; the average rank of the F2F modality was 1.3 before and 1.5 after hyflex. The student rankings for synchronous and asynchronous modalities were 2.6 and 2.1 before hyflex, and 2.9 and 1.6 afterwards. So there was a fairly significant increase in preference for the asynchronous modality after the switch to hyflex.

Despite student preferences, the actual choices students made for their modalities were a lot different than their predictions. Before hyflex, the numbers of students who predicted they would use the F2F mode in weeks 1-5 was 11, 12, 10, 8, and 12; but the actual numbers of students who attended the F2F sessions were 8, 6, 7, 6, and 6. For the asynchronous online mode, the predicted numbers were 5, 4, 6, 9, and 5 but the actual numbers were 10, 12, 11, 12, and 12. So students ended up using the asynchronous mode more than they expected.

If you do the math on those numbers, a surprising result surfaces: There were only a few students (2, 2, 2, 1, and 1) who predicted that they would use the synchronous online mode --- and none of the students ended up doing so. Not a single student chose to attend any class synchronously online.

Students blended their modalities in different ways. Eight of the 18 students (44%) participated in all five class sessions in asynchronous online mode, while only 3/18 (17%) went 100% F2F. The rest of the student choices were spread out in between, representing a mixed set of choices for each student about whether to use the F2F or asynchronous modes.

As to the benefits and challenges of each mode, students focused on the tradeoffs inherent in the two different modalities that they tried. The asynchronous mode allowed flexibility in scheduling and pacing, allowed students to stay at home (which seems important today), and gave the ability to control the flow of information by pausing and replaying video content. But conversely, the lack of structure that provides flexibility also made it difficult to stay on track and avoid procrastination. The synchronous mode, on the other hand, provided structure, guidance, interaction and accountability; but the structure made it less easy to fit the coursework into their schedules, and non-native English speakers found it hard to follow the flow of coursework. These are all fairly standard student perceptions with online, blended, and also flipped instruction.

When asked if they would choose hyflex again given the choice, 28% of students said they would choose hyflex over a F2F course and 39% said they would choose it over a fully online course. But also, 22% said they would pay an additional student fee (under $100) to have the option to choose hyflex. (39% said they would choose hyflex but not pay extra.)

Finally, 61% of the students felt the online and F2F learning modes were equivalent in terms of rigor and quality. The remaining 39% were split on their opinions of which mode was better or more rigorous.

What do these results mean for others?

There is a lot of counterintuition in the results of this study. I'm really surprised that none of the students tried the synchronous modality even once, and very few of them even intended to try it. This is a lot different than what faculty think students prefer. In my department, we pivoted online in March, we took it as given that keeping a synchronous format to mimic the in-class experience was the best approach. Now I am wondering if it's something students even want at all. So be careful what you assume about student preferences.

The results here also highlight the uncertainty and messiness of life that our students face when getting an education. Even before Covid-19 this was true, but the switching back and forth between modalities, and especially the greater-than-predicted use of asynchronous modes, highlights it even more. Students may predict they'll use F2F a lot and maybe even prefer it, but when life came at these students, a good number of them changed their minds.

Based on the nearly one-quarter of students who said they'd pay extra to have the hyflex option, I'll leave it as an open question for administrators whether charging differential tuition for hyflex courses is a good idea.

Limitations and questions

External validity is often an issue in educational research, and it's no different here. The study deals with a group that is quite small (18 students) and somewhat specialized (Social Work majors and minors in a required class in their program), so any attempt to make broad claims from these results has to be tempered.

Also note that no apparent attempt was made to determine if any of the quantitative results was statistically significant. The small group size (n = 18) mitigates against this, but I wish the author had done some kind of statistical analysis. As it is, it's fair to ask about the extent to which the numbers can really be attributed to the hyflex approach, or whether those are due to something else, or just to randomness. (There's enough data in the study that someone with free time could run that analysis after the fact.)

Finally, be aware of what this study doesn't examine – in particular, there's nothing here about whether students actually learn more, or better, under hyflex compared to pure F2F, online, or fixed-schedule hybrid courses. Only student perceptions and behaviors were studied, not student learning. This isn't "good" or "bad", but just a fact about what this study is and what it isn't.

I thought it was also curious that only one of the 18 students was a commuter student living more than 10 miles from campus. This seems non-representative, and it seems like commuter status would influence student choices and preferences. So I wonder what these results would look like with a population of students for whom physical co-location is more of an issue, like a commuter population --- or a campus under quarantine and social distancing rules.

I also wonder if the scheduling of the class affects student choices. A three-hour class makes it awfully hard to stay focused; many students (including myself when I was in college) would gladly deal with the issues of asynchronous learning just to get out of having to sit in class for that long. I suspect student preferences and choices would be different if the class were one hour, three times a week instead.

Generally speaking, I came away from this study thinking that hyflex is a very good response to our current situation because no matter what happens in the fall, things will continue to evolve and change rapidly and the one-size-fits-all approach of both pure F2F and pure online modalities seems like trouble. I've argued recently that what we really need heading into Fall is a means of instruction that is decoupled from physical location, so that we can pivot back and forth between F2F and online as often as needed in a given year without all the upheaval of March. Hyflex certainly seems like it offers this, and even better it gives students the driver's seat in making those choices.

As we evaluate our options heading into the Fall, hyflex is emerging as a very compelling one. Expect to see more about this here as the summer unfolds.