Open online homework systems

A showcase presentation for #OTESSA21

Day 2 – June 1

How Moodle encouraged me to build open online homework (without even realizing it)

All I’m going to write in this space today are words about my own work to provide free homework for my students, long before I was aware that there was any such thing as open education. So this is as much a personal reflection as it is anything at all, and it’s going to read like it – but stories like this form the foundation of scholarship that comes later. 

But first things first. If I’m going to tell the story of my personal engagement in the production of online homework for my students well, I have to start with how I first came into contact with Moodle. 

I was working at Shorter College in northwest Georgia (now Shorter University). I won’t name the for-profit tool that passed for the campus enrollment and courseware system (I don’t even know that it’s still made in its form them), but it was a tool that was made to be a jack of all trades, and master of none. It did the job of enrollment management passably, and course management less so. It had a quizzing tool, but unless you were interested in multiple choice or impossibly-precise short answer questions you were out of luck. Mostly it was just a tool that was reliably for sharing documents so that we could be “paperless”. (Remember 2007, and the era of going paperless?)

It wasn’t enough for what anybody was wanting or what we felt like we were capable of, and the school was investigating other options. Our technology roundtable got to the point where they felt like they had a viable option, and they gathered invested faculty on one of the satellite campuses to contribute to the discussions. 

The thing I most remember about the day was investigating everything about this new software they were talking about, Moodle, searching for this detail or that on my laptop while a sequence of presenters described details of this learning management system and how it worked – and how I completely stopped paying attention to the presentation when they reached the quizzing module, and discovering that Moodle was capable of supporting randomized calculated question authoring, in a fashion very similar to how WebAssign provided randomized calculated questions. 

Seeing a way to provide for free what somebody else was charging me to provide was a thunderbolt. 

I need to emphasize two things at this point. The first is that, as confining as the LMS seems to so many of us in 2021, and for all the ways that Moodle has corporatized and grown to feel like a for-profit provider in the intervening years, finding ways to open up web services in ways that undermined for-profit providers felt like a revolution to many of us in the mid-2000’s. Moodle did all the course management things that our for-profit provider did, and the software was (and is) free in every sense of the word. All we had to figure out was the hosting. And at the time, everybody concerned seemed very enthusiastic about the prospects. 

The second thing that needs to be emphasized is that I wasn’t alone in how I envisioned Moodle undermining a provider like WebAssign, and unlike me, some other scholars even published their efforts. I happened to be at an institution that was relatively isolated from academic discussions, with no pressure to participate in those discussions. My models for the work I ultimately did to provide resources to my students (in particular Ron DeLorenzo, who at Middle Georgia College innovated with a lot of computer tools from the 1980’s through the 2000’s in a rural, heavily isolated environment) didn’t construct their tools with sharing in mind, largely because the capacity to share those tools in any sort of critical way was so limited. I simply felt myself similarly free to construct my own tools in an isolated place.

At the time I was very grateful for that freedom, but increasingly in retrospect it feels like that freedom robbed me of a chance to report what I was doing at a time when there was a lot of opportunity (see Martín-Blas and Serrano-Fernández from 2009 as a work of people who were not merely doing the work, but publishing what they were doing). 

But I did spend the four years between 2007 and 2011 learning as much as I could learn about Moodle, experimenting with question-writing of my own, and throwing myself into providing as much free and randomized homework for my students as I could. I wasn’t terribly careful about copyright or sourcing on my early problems; I took whatever I could from any textbook I could and randomized the values in the problems to learn how to use the software. Again, I wasn’t working on this with any thought of how I might share what I was working on with anyone until much, much later; I simply was trying to figure out how to use the software to give the best experience of problem-solving to the students.

Through a different ed-tech exposure I was aware of the UC Irvine-developed mathematics learning tool ALEKS, and in particular the means by which ALEKS explored “knowledge spaces” within the topics covered in a course. (For a layman’s description of ALEKS, read this UCI press release from immediately after ALEKS Corporation was bought out by McGraw HillALEKS also provides a technical white paper describing knowledge space theory in more detail.) I used that theory within Moodle’s constraints to give homework sets I assigned to students in general physics at Shorter their ultimate shape – three problems, randomly selected from a “knowledge space” pool of anywhere between four and ten problems, with the individual numbers in a problem randomly selected, that students could submit as many times as were necessary to gain mastery. 

Example 1 of a problem set
Example 2 of the same problem set

Note that the two problem sets have largely different problems; when a problem is repeated, the numerical values in the question are changed.

The work I did in a vacuum at Shorter between 2007 and 2011 formed the foundation of how I still use Moodle in my classes today, primarily in the teaching of general physics. As I’ve moved from one institution to another I’ve reinvented the problem sets and made much more substantial efforts to make sure the problems I’ve written for those sets are original (or, if they’re not, that I’ve cited the source for the problem I’ve adapted). I’ve made halting attempts to adapt problems for general chemistry and physical chemistry when those teaching assignments have been added to my workload; nothing there is remotely ready for prime-time, however (although I post General Chemistry I homework for sharing anyway).

But the physics problem bases are quite complete, and I store them at – I’m willing to talk to anybody about sharing them at any time. 

I am still trying to settle myself as to how good an idea it is to share these problems widely, however. After all, one of the highest compliments I’ve ever received is the student who complained that “I can’t Google the answers to your homework!” – which I thought was hilarious at the time, but was only foreshadowing a lot of the discussion that would emerge about online homework later.