Disclosure: While I have no business relationship with the California Community College system, I was involved in helping them set up the Online Education Initiative, which was renamed the California Virtual Campus. This post directly looks at the current state of online education in California, including information from the current and former executive directors of OEI / CVC based on a podcast interview I had with them as part of my research.
There's a microeconomics course at San Joaquin Delta College where the lectures were recorded over a decade ago, the quizzes are multiple choice, and the professor hands students the answers before they take the test. CalMatters found it and is right to be appalled. So am I. That is not assessment, and no department should have signed off on it.
But this is where the article goes wrong, and it's worth being precise about how. The CalMatters article takes that course — along with advising services that take weeks to respond, and students stuck in online sections they didn't choose — and extrapolates across a system where roughly 40% of California community college courses are now online, as if the modality were the diagnosis. The leap is the problem. Bad course design is bad in any modality, and under-resourced advising is under-resourced in any modality. That microeconomics course failed students because it was badly designed, not because it was delivered online — and conflating the two doesn't just misread one course, it points the fix in the wrong direction for the million-plus students the article claims to speak for.
What makes the leap harder to excuse is what the article left out. California’s Community College System (CCCS) runs one of the most developed online-quality operations in the country — a decade of course-review infrastructure built precisely to catch the kind of course CalMatters held up as typical. You wouldn't know it existed from the article. A student told the reporters she felt like she wasn't learning, and I believe her. But she deserves an honest account of why, and that account requires looking at what the system actually does and what actually happens to students inside it — not extrapolating from the worst example to the whole. If we want higher education to serve students well, we have to be willing to say where the failures actually are.
Awful course design is awful in any modality
Start with the microeconomics course’s assessments. Multiple-choice quizzes with the answers handed out in advance is not a weak assessment. It is not an assessment at all — a computer grading a test the student already holds the key to. Strip the word "online" off of it and you are left with a course that has abandoned the basic act of measuring whether anyone learned anything. That abandonment has nothing to do with the modality it was delivered in.
So the real question is not "what's wrong with online education." It is how a course like this gets offered at all, inside a system that has spent more than a decade building an apparatus designed to avoid these problems. Since 2014, CCCS, through the Online Education Initiative that was renamed as the California Virtual Campus (CVC), have run a voluntary course-design rubric and a Peer Online Course Review process (POCR) that not only provides guidance on design but also puts faculty courses in front of trained peer reviewers against a standard covering assessment, accessibility, communication, and engagement. A multiple-choice-with-the-answers course does not survive that review.
And this is not box-checking that faculty endure. In the podcast interview, Marina Aminy, the current executive director of the CVC and one of the first faculty members to put her own course through POCR, described the experience this way:
The feedback I received from my peer reviewers — this is 15 years ago — was more substantive than any feedback I had ever received for any evaluation in my career up to that point. It was so thorough, so detailed, and so helpful professionally.
That was one course, fifteen years ago. The reach today is systemic: four years ago roughly 50 colleges ran a local peer-review program, and today 92 of them do. So how does the CalMatters course exist inside a system like that? To their credit, the directors didn't dodge the question. Jory Hadsell — who helped launch the initiative in 2013 and now serves as its executive sponsor at the Chancellor's Office — gave the honest, unflattering answer:
We have this unified peer review process [snip] provided to many faculty, but we also have individual collective bargaining agreements at each campus. And in some places, this professional development can be optional. And so the anecdote that's described is completely counter to the type of instructional design and approach that we recommend.
The failure CalMatters found is real — but it is a failure of local commitment, a campus where the quality process is optional, and a department let an indefensible course run. It is not a failure of online education.
The irony that CalMatters missed entirely: online courses in this system carry more quality scrutiny than their face-to-face counterparts, not less. The article didn't mention the rubric, didn't mention POCR, and didn't mention that the people who are working to improve course design and delivery will tell you where the gaps are. None of that was hard to find. It just didn't fit the story.
Unavailable course choices are a real problem
Of the three anecdotes, this is the one where the extrapolation has some basis. When a student wants an in-person section and can't get one — because the campus didn't staff it, or couldn't afford to — and is left with an online seat as the only option, that is a genuine grievance, not a misreading, and it deserves to be taken seriously. But notice what kind of problem it actually is. It is a problem of scheduling, staffing, and campus finances, not a problem of online education's quality. The student forced into an online section they didn't want and the student in that gutted microeconomics course are describing two different failures, and only one of them has anything to do with the modality. Conceding the first does nothing to rescue the article's central claim about the second.
And here too the system is doing something the article didn't bother to find. The CVC Exchange exists to widen the menu — letting students enroll in courses across district lines when their home campus can't offer what they need, when they need it. It has gone from roughly 1,500 students using it to register four years ago to 60,000 this year, against a catalog of more than 100,000 online courses. That is not a complete answer; it doesn't conjure an in-person section where none exists, and as Jory Hadsell readily acknowledged, real gaps remain in specific disciplines — the hard sciences in particular — where a student can still struggle to find the right course at the right time. But it is the opposite of indifference. It is a statewide system treating limited choice as a problem worth solving, which is more than CalMatters managed to notice.
Data view of student outcomes
Given the issue of what I call faulty extrapolation, what do the data say?
Course completion
The primary reference to a broader view of online education in the California Community Colleges was a 2025 report from two University of California and one University of Pennsylvania researchers that showed a performance gap between online and face-to-face student performance, but that the gap was shrinking over time.
That study was based on the primary metric shared in CCC public data - course completion rates with grades of A, B, or C. This, of course, is not an ideal metric, which can be seen in the aforementioned microeconomics course example. Passing that course would be easy, whether you learned anything or not. But there is some value in this metric.
The report looked at students taking classes in face-to-face, online asynchronous, and online synchronous versions from 2015 - 2021. The primary conclusion shows a performance gap, but one that is shrinking over time.
These highly-saturated estimates show that asynchronous/face-to-face performance gaps were shrinking even prior to the pandemic, decreasing from an 8.3 percentage point gap in course passing rates in 2015 (Column 1) to a 7.1 percentage point gap in 2018 (Column 2). These gaps fell further to 5.8 percentage points in 2021.
The report also looked at synchronous online but noted that the results “limit generalizability.”
There are broader data from the same source. If you look at all years for that performance gap, from 2016-17 to 2024-25, what is remarkable is that the performance gap has almost disappeared. Note that the gaps will be off by a few percentage points from the research report’s results due to statistical handling.
Furthermore, the pandemic presented an outlier, where most students were forced into online courses in “emergency remote learning” situations. If you look at the pre- and post-pandemic years, what jumps out to me is the significant increase in performance for online course taking with much smaller increases for traditional, or face-to-face course taking. In fact, the performance gap has all but disappeared in the system.
The gold standard?
But even here there seems to be an assumption that face-to-face education is the gold standard, which neither Hadsell nor Aminy buys.
Marina: The gold standard is the modality and the strategy that best suits the student and what the student's needs are at that point in time in their learning. And if online can do better — and in some colleges, in some fields, in some content areas, they are doing better. This data is the average, but if we look at the pockets, I think it's doing much better depending on what we're looking at.
Jory: I agree with Marina. We also have to remember the success numbers aren't telling us how well students are doing, they're telling us how many students did well. [snip] Putting certain types of disciplines into a distance or an online modality can be more challenging. So I think that line crosses as we continue to figure out the ways to — content creation gets easier, simulations become easier. There are a lot of elements that can go into those courses where I think we could see the number of students who are successful exceeding in the online space. It's just different. If you put your same lectures into an online class shell, that's not a great design. So there's an intentionality to how we engage students online that can help with that.
Program completion
Looking more broadly, the state system commissions an independent review of online education in the system every two years, and in the most recent RP Group report from 2025 we see that online education is associated with program completion. In fact, the data show a relatively straight line correlating the percentage of online course taking per student and their program completion rates.
If the CalMatters’ headline of “Community college students struggle with online education” refers to students having to work hard, then that would make sense. But what I believe they actually mean is that students are not succeeding because of online education.
But even a cursory review of available data actually mentioned in the article does not support that argument.
The system already named a real variable
Perhaps the most useful document in this whole debate isn't the CalMatters article — it's the CVC's own 2024 POCR Landscape Report, which lands on two findings that, set side by side, tell the story the article didn't. First: where the process is applied, it works. Colleges that compared the same instructors' courses before and after POCR alignment found success rates rose — 12% overall at Chaffey, 5% at LA Pierce with larger gains for Black, Hispanic, and Asian students. That pre/post design matters, because it rules out the easy objection that strong teachers simply self-select into the process. And the gains carry past the badged course into the same faculty member's other classes, online and face-to-face.
Second, less comfortably: the process isn't everywhere, because it isn't prioritized. The report describes a patchwork of local funding and incentives in which more than a quarter of faculty who align a course are paid nothing for the work, and only 12% of colleges so much as flag a quality-reviewed course to students in the schedule. Put the two findings together and a very credible diagnosis writes itself. The variable that predicts course quality is not online versus in-person — it's whether a campus chose to invest in quality. Associated with that second investment point is whether the college implements standards in the quality process. We often find that courses of the microeconomics type are often ad hoc, by an individual faculty member, and not tied to an online program. The San Joaquin Delta course is what the second finding looks like on the ground; the closing performance gap is what the first looks like at scale. CalMatters found the former and blamed the modality, when the system's own report had already named the real variable.
Same old, same old
We need to improve online education quality and outcomes, just as we do for other modalities. And online does have unique challenges - the need for time management skills, difficulty with student engagement (at least without intentional design), and more difficulty maintaining academic integrity, in particular. But this is a statewide system investing time and money on exactly this type of improvement. And the results are improving. If the article showed that there were no results from these initiatives, then that would be a problem. But that is not the case.
We've seen this argument before — online is cheaper, online is easier, online is worse — and we'll see it again. That's the same old, same old, but repetition isn't evidence. A statewide system has spent more than a decade building an apparatus to find and fix bad courses, the data show the gap closing, and the people who run it will tell you where the real problems are if you ask. CalMatters didn't ask. It found one indefensible course and extrapolated it into a verdict on a million students' education. That's the failure here — not the modality, the extrapolation — and the students the article claims to speak for are the ones who pay when we aim the fix at the wrong target.
The main On EdTech newsletter is free to share in part or in whole. All we ask is attribution.
Thanks for being a subscriber.
