This is a writing sample by “nycghostwriter,” AKA Barbara Finkelstein. It is “The Learning Analyst,” published in 2014 for Teachers College/Columbia University. You can get professional ghostwriting services from a published non-fiction writer. Email me or fill out the short form on my contact page.
Ryan Baker believes that education data mining can answer the question: What makes students tune in?
By Barbara Finkelstein
It should come as no surprise that Ryan Shaun Baker has turned his own work habits into a data set. A look at his computer log shows an itemized record of the 7,000-plus minutes he has invested in constructing TC’s first Massive Open Online Course (MOOC) – data whose granularity will help Baker and his colleagues understand the strengths and weaknesses of MOOC education.
An Associate Professor of Cognitive Studies in TC’s Department of Human Development, Baker designed the MOOC, as well as a new master’s degree program in Educational Data Mining (EDM), to bring Teachers College into the exploding field of big data analytics. Now a staple of medical research, financial services and retail, where big data aims to enhance profitability and product development, big data analytics is also revealing trends and patterns in online education. Using advanced computing technologies to mine ”petaflops” of information gathered from intelligent tutoring systems such as MOOCs and other Web-based programs, EDM has the potential to produce actionable insights into teaching and learning behavior. Among the critical questions Baker and other EDM researchers want to consider: Why does a student keep getting the wrong answer to a math problem? How can a teacher pinpoint which homework problems stump his algebra class? How can teachers make better use of classroom time? How can a school superintendent know which science curriculum is more effective with students across the district?
All of these educational issues fall squarely in Baker’s wheelhouse. At 35, Baker has published more than 150 peer-reviewed papers on the use of EDM to assess boredom and cheating among students who use online tutoring systems, and on the mining of educational data to better understand metacognition, motivation and self-regulated learning. His ultimate goal: Use big data to create computer-based environments that engage students in difficult subjects, especially in mathematics and computer science.
Baker’s interest in student engagement dates back to the mid-2000s when he was a graduate student at Carnegie Mellon University. Armed with pen and clipboard, Baker spent hours in a suburban Pennsylvania school trying to identify key moments when students went “off task,” that is, when they lost interest in the program, expressed confusion or gamed the Web-based system itself.
“After six weeks of slow and laborious note-taking, I said, ‘Agh!’” says the Texas-born data scientist. “There has to be a better way.”
Sensors and webcams, typical modes of capturing moment-to-moment student behavior, proved to be intrusive and financially unsustainable. Baker and his fellow grad students instead chose to design a set of computer-based math lessons that recorded clicks and keystrokes from which they could infer how long it took students to solve a problem before moving on to the next one. By analyzing data housed in the computer log, Baker and his group could even determine when students resorted to guesswork. The test was the first-ever automated detector of disengaged learning behavior based on moment-to-moment data.
Before arriving at Teachers College in fall 2013, Baker taught at Worcester Polytechnic Institute, where he worked with fellow faculty member Neil Heffernan on an intelligent tutoring system called ASSISTments. This free Web-based program could generate sets of increasingly more complex math problems that gave students instantaneous cues whenever they answered incorrectly. Heffernan and Baker are now conducting an eight-year study, funded by the National Science Foundation, to determine if ASSISTments has changed the way students fundamentally approach new and ever more complex material.
Baker‘s work with intelligent tutoring systems gained additional traction with the appearance in 2012 of the MOOC. Adopted mostly by large universities, such as Harvard, Stanford and MIT, the MOOC has the potential to educate millions of people around the world at little or no financial cost to the institution or the student. Indeed, measured by the number of course enrollees – Baker’s MOOC clocked in at an impressive but relatively modest 44,000-plus students – the MOOC is becoming a core educational offering at universities in the U.S., France, India, Lebanon, Russia, Australia, Vietnam, Israel and China.
While the Web is awash in anecdotal evidence that alternately aggrandizes and condemns MOOC education, serious research is sparse in two important areas: first, the pedagogical methods and technologies most responsible for successful MOOC learning; and second, the problem of keeping great numbers of registrants committed from start to finish. A universally acknowledged statistic: Course completion rates hover between just 5 and 13 percent.
Baker intends to riddle out critical questions about MOOC usage and utility by mining the data from his own MOOC. Among the questions he will focus on are the ones MOOC instructors everywhere are asking themselves: What is the goal of MOOC education? What draws large numbers of potential students to a MOOC – an online course consisting of educational modules, discussion forums, benchmarks and homework assignments? Why do so few course registrants persevere throughout the MOOC semester to receive a “statement of accomplishment?” How effective are MOOC feedback systems and student forums? Is the MOOC business model sustainable? Most important, are students better served by MOOC or classroom instruction?
Although he has only just begun to scrub the data in his computer log, Baker is ready to share some anecdotal evidence about the efficacy of his MOOC instruction. Among his observations:
- MOOCs are an excellent vehicle for introductory material. Many intro courses are given by teaching assistants relatively new to teaching. To attract a massive number of targeted users, MOOC lecturers have to be subject matter experts who carefully hone their lectures and continuously work to improve them.
- MOOCs fill an educational void when subjects are locally unavailable. Programs in educational data mining, for example, existed in about five cities worldwide before Baker made his course available to anybody with a computer and Internet connection.
- MOOCs are advertisements for more formal education programs, such as TC’s officially named “Masters in Cognitive Studies in Education, Focus in Learning Analytics” in the Department of Human Development. Indeed, a January 2014 Sloan Consortium report cites “increased institution visibility” and “drive student recruitment” as the top reasons academic leaders give for offering MOOCs.
The big question, of course, concerns how MOOC instruction compares with in-classroom learning. Baker takes a measured approach. He thinks MOOCs in their current state will not dominate education partly because of enrollment fall-off rates, but also because MOOCs have been made into an educational hybrid. The lectures, Baker says, are “mostly pretty good,” while homework assignments are “mostly awful.” Baker cautions against making the MOOC all things to all people.
Baker recognizes that a certain amount of frustration is part and parcel of any challenging learning experience – and that wrestling with the content in an intelligent tutoring system can strengthen one’s resolve to learn.
“To date, I have spent one hundred and seventy-three hours to create my MOOC,” Baker says. “I sacrificed writing four research papers in the process. But my MOOC will have been worth all the effort if it can achieve two goals: One, teach us how to consistently engage student attention; and two, bring more data scientists into an area of educational research where so much is at stake for so many.”