Justin Reich blogs about the CREATE Equity Research Report at Ed Tech Researcher in Education Week this week. The research report, released earlier this year, shares findings from a summer 2013 study on the in‐person roles that humans play in online learning. Read his blog post here or full text below.
The Role of Humans in Blended Learning
By Justin Reich on May 26, 2014
Last month a colleague at U.C. San Diego sent along a lovely piece of research that begins to address one of the most important gaps in online learning research: what roles do people play in blended learning environments? When people take an online course for credit, or a MOOC, or some other self-contained online learning experience, they don’t just learn from the software or the online content. Sometimes they learn online but off-platform, by looking up terms on Wikipedia, posting questions on sites like StackOverflow, or looking up answers in search engines. And sometimes they learn from the people around them, either by deliberately seeking expert help, deliberately creating social venues for working together like study groups, or they find people in their network who can support their learning. These activities are invisible to people studying data from online learning platforms, so at present we know very little about them; they are not in our Big Data big data sets. The best entry point into these unstructured, informal experiences is to conduct qualitative research that asks people to describe their learning behaviors or better yet, to observe them as they go about the business of learning.
Enter the work of Mica Pollock at UCSD’s CREATE lab. She and her students spent the summer of 2013 conducting observations in classrooms from UCSD’s Early Academic Outreach Program (EAOP), a program that lets underserved high schools students take online courses, called UC Scout Courses, that earn them credit towards the admission requirements for the University of California system. The online Scout courses are developed and administered by the University of Santa Cruz, but students in the EAOP program take the courses in local high school classrooms where each student has their own individual workstation as well as access to teachers and undergraduate teaching assistants for support. Thus, it is a blended model where local teachers provide supplementary support for an online curriculum.
The Scout courses are designed to be taken entirely individually, so this motivates the question: what are the teachers in the EAOP doing to help these ambitious but underserved students take the Scout courses they need to get into the UC system? To answer this question, Pollock and her colleagues observed a summer’s worth of classes, and they created a taxonomy of seven kinds of work that humans took on in these blended learning environments:
Humans as fixers and explainers of technology
Humans as digesters of content
Humans as explainers of content
Humans as extenders of content, toward application
Humans as providers of feedback and assessment
Humans as regulators of student behavior
Humans as peer supporters
One way to think of all of these human supports is that they identify places where the online software falls short, where humans have to step in and do what computers cannot yet. For software designers, these might be useful inspirations to identify ways that teaching software can be improved. But these findings also may be a reminder that, especially when trying to serve the students who most need our support, there are plenty of crucial dimensions of learning for which humans remain irreplaceable. Pollock’s conclusion:
Some students, including low‐income students from under‐resourced schools, can use Scout largely independently and successfully to move quickly and sequentially through core course material and assessments. But, the gold standard of Scout implementation and indeed, of online learning ‐‐ where students dive deeply into course concepts, synthesize, extend, and apply their learning, become curious about a field, or discover and fill key holes in what they know – will likely require more flexible integration between humans and online course curriculum. Since supporting high‐need students particularly is Scout’s main goal, designers and users will have to navigate a classic core tension between balancing essential access to and completion of college‐preparatory A‐G courses and credit (basic equity) with ensuring the highest quality learning experience that will set students up well for future coursework (deep equity).
The full report is worth reading and a great piece of anthropological research to inspire deeper thinking about the balance between people and machines in blended learning environments.