Monday, June 10, 2013

Online Remedial Courses Considered Harmful

Online remedial courses are inherently absurd. Even though this is UDacity's (for example) great-white-hope moment, as it has begun offering elementary algebra courses for college acceptance in California (link), it practically beggars the mind that this will help the crushing wave of need that students in such courses evidence.

The reason why is that everything that online courses do well is precisely the opposite of what remedial students need. We know that online courses require a higher level of discipline, dedication, and self-starter initiative than in-person courses do. Online courses are inherently tougher to follow than live courses. You also need a certain technical proficiency just to interface with the platform (and occasionally troubleshoot problems). This is all well and good for high-functioning academic-types who fundamentally love to learn on their own.

But our remedial students have none of that. One of the first overwhelming problems is that they don't have self-discipline in schedule or study habits -- frequently helping with this is itself part of the remedial math course. And they don't like the subject; surveys routinely show an overwhelming and long-seated hatred for the discipline; often a large proportion of a remedial class doesn't show up for the very first class. Minor technical problems will routinely frustrate them and throw them off completely.

Frankly, what the remedial student needs is clear -- if we were serious about getting these students educated, then they would need more individual, one-on-one interaction to address their deep level of need (not less). They need a personal touch to get them over their often pathological resistance for the technical subject matter. They need personal tutors -- but the cultural structure is not one that is interested in paying for that.

Here is a quote, based on the reform experience at a community college near Philadelphia, as they focused effort on the classes with the highest failure rates throughout their college, and in many cases improved their statistics by as much as half (based on interventions such monitoring no-shows on the first day, requiring early evidence of class participation, academic probation communication procedures, etc.):
In some cases, Hayden said, the college's analysis has led officials to believe that some courses were being offered in inappropriate formats. For instance, several of the highest failure rates were in online developmental courses (around 60 percent) -- and various reforms didn't budge those numbers. So the college has ended online remedial education. "The failure rates were so high that it seemed almost unethical to offer the option," Hayden said. (link)
This quote inspired me to write this post tonight. "Unethical to offer the option" (for online remedial education) seems about right. We'll see how quickly MOOCs such as UDacity, and those partnering, paying, and linking their reputation with them, re-learn this lesson.


  1. Thanks for writing this. I share your concern.

  2. As you may have heard now, Udacity and SJSU have stopped offering their online-only for credit classes while they re-evaluate based on poor results. I took their remedial math class not for credit, just for curiosity and review and like their other classes I had a positive experience. However, I wouldn't disagree with you that remedial classes seem like they might be the worst application of online-only courses. There's a lot of reasons a student might need remediation and I don't see an online course addressing many of them. Their strong points are being able to work at your convenience and being able to practice over and over. They don't provide encouragement or discipline. They apparently are having much better success with an edX hybrid class that includes classroom time. I'd be interested in the success rate of people like myself who took the course voluntarily and probably self-selected to have the minimal self- discipline needed to go online-only.

    1. Yes, I did just see that a few hours ago and I agree with your observations. I would expect that people coming to the courses with a pre-existing desire to learn would of course do better.

    2. By the way, Gerald: You made a nice post on the "Score Mangling" thread that came to my email, didn't show up on Blogger. Do you mind if I re-post it for you?

    3. Please do.

      I would like to know if there is a typical remedial student, or frequent characteristics, and how you go about helping them. For instance, are they typically convinced they 'don't get' math' and hate it? Do they want to learn or just get the course behind them? Can you instill any self discipline or just settle for providing external discipline? Is it usually these kind of attitude roadblocks or just a matter of reviewing the math itself and they are good to go? How far back in their schooling do you have to go? A lot of questions for maybe a whole blog post some day rather than here in the comments.

    4. Really good questions; I'd say the problems have a pretty wide variation. Of course the brick wall is algebra, but I think about 1/3 of our students need a basic arithmetic course first (basic operations on integers, fractions, decimals, percent). All of the students I see know how to add & subtract, but a very common problem is not knowing times tables or a long divide procedure -- which holds a student back in terms of fractions, factoring, solving equations, etc. Also the single most common mistake from start to finish is tripping up over integer operations. Usually in the quick algebra review someone expresses shock at realizing that the rules for add/multiply are not the same. As much as I harangue students to fix those problems, I'm not sure students in that situation ever do.

      So there's a lot of problems with the most basic symbolic manipulations (times tables, signs), and likewise students haven't been taught grammar (so have even greater difficulty parsing word problems in English), logic, etc. The algebra class is the first time they've been required or assessed on structural details, and often they're scrambling up symbols. Possibly a large part of the whole cohort is dyslexic or something, I don't know?

      On the other hand, there are always 1-3 students in a class who have a good chunk of the skills, consider the class beneath them, and need to make a show of being insanely combative with me in anything I ask the class to do (esp. in terms of disciplined writing). This only happens in an algebra class, and it's totally bewildering and disruptive; these students usually have a 50/50 pass rate. So I tend to be contrarian on the "attitude" problem -- the most confident students tend to do really poor work.

      I also give study-habit surveys to all my classes in the past year. The one and only thing that's strongly correlated with success is simply attendance (noting that I almost uniquely *don't* have an attendance requirement for the class). Other stuff shows no effect (study time, learning definitions, doing a
      1st-day review, studying in different locations, attending the math workshop, utilizing tutorial services, and watching online
      videos). Surprisingly: in the basic arithmetic class, more hours spent in an outside job is *positively* correlated with results (and this synchs well with my night classes, people who worked all day, generally being more serious and doing better).

      One last thing is that lots of people come into the algebra class being able to simple solve 1st-degree equations -- the 1st test is uncorrelated with final scores, maybe this early review lulls them into complacency? But it's the later subjects and tests on equations with exponents, radicals, factoring, and graphing that clearly predict success on the final exam (and on which the wheels fall off for many students).

      I'm glad you asked!

    5. Gerald wrote:
      Interesting. I hadn't considered some students being so resentful of being there that they self-destruct. The success of working folks in the basics does make sense; maybe more mature and able to relate that math to their jobs. Is there any correlation between age and success?

      I was surprised in taking that Udacity class to see basic arithmetic covered. I would have considered it unnecessary for me but my policy is to not skip anything and when I found myself making some mistakes it was humbling. There was a lot of emphasis and practice later on factoring and solving polynomials; more thorough than I remember from high school. The graphing and visualizing/predicting the graph by looking at the equation took some thought too. The class was no pushover but doable. I can see people getting stuck sometimes despite being able to review it over and over. A video isn't going try paraphrasing the explanation. There was help available in the forums. I wonder if there was any correlation between use of the forums and success? I think the for credit students also had access to TAs although no actual classroom.

      I hung around in the forums for while answering questions (I became a bit of a karma points whore). It is quite gratifying to help somebody get that aha! moment. It must be a lot of what keeps teachers going despite the frustrations. Kind of like golf; the occasional successes are sweet and keep you coming back.

    6. I'm not sure about the age -- sometimes I get a student in their 40's or 50's and they have a very hard time with the remedial material. I would guess "not correlated", without any hard data in front of me.

      Sounds like the UDacity class was a reasonable effort (I like the predict-the-graph exercise). I'm pretty sure part of their philosophy is to depend on the forums as a kind of large-scale study group. You're totally right that getting someone to the "aha" moment is the major joy of teaching, glad you got a taste of it. :-)