Online education

Online education needs data analytics to find out what works

The editorial board operates independently of the UT newsroom but is held to similar ethical standards. We base our editorials and recommendations on rigorous reporting, interviews and debates, and strive to be accurate, fair and courteous in our section. To disagree? Let us know.

American professional sports, after initially despising data analytics as the province of ignorant nerds, has almost entirely returned. The value of walks is now universally understood in baseball. Whether cornerbacks can make or break a team in a way that can be right behind quarterbacks now determines football decisions. The inefficiency of taking long, contested two-point shots is now understood by just about everyone in basketball without the name of Russell Westbrook.

So when will this transformation hit K-12 education as it relates to remote learning? The key to data analysis is being able to evaluate large amounts of statistics to uncover patterns that show what is working or not working. With tens of millions of students forced into a year or more of online learning in March 2020, there should be plenty of evidence to sift through to establish best practices. It’s folly to think that every school district should try to do this on an individual basis. As students across the country begin to be forced back into distance learning due to Omicron’s highly contagious version of the coronavirus, what America needs is the online educational equivalent. of the Manhattan Project, perhaps led by a coalition of top universities. That need was underscored by Friday’s release of statistics for the 2020-21 school year by the state Department of Education. They showed a substantial increase in the number of students not meeting basic standards in English and math, especially in lower grades, and an increase in D and F grades.

A first best practice couldn’t be more obvious: ensuring K-12 students are actually engaging in virtual learning. There are many anecdotes from teachers about students disappearing for weeks during the pandemic. Even before the virus hit, it was a huge problem in classroom teaching. In 2018-19, about 720,000 of California’s more than 6 million K-12 students were labeled “chronic absentees,” missing at least 10% of their classes. In 2020-21, this number has increased to around 840,000.

Because many districts that receive state funds based on average daily occupancy have struggled with their budgets, State Sen. Anthony Portantino, D-La Cañada Flintridge, introduced a bill this week. which would instead link the financing of education to annual registration. Portantino says the measure includes provisions to push districts to reduce absenteeism. But unless the bill is crafted with exquisite care, it could cause districts not to care whether students show up. It is telling that one of the strongest supporters of the proposal – the Los Angeles Unified School District – has such a wrong record on school attendance.

Beyond the overriding importance of student participation, however, recommended best practices for remote learning are unconvincing. They represent common sense and platitudes, not specific evidence-based guidelines. “Teaching Explicitly and Systematically How to Use New Instructional Materials,” a website Remarks. But what is the most efficient way to do it? Another site disdains the idea that experimentation is crucial to improve virtual learning: “Don’t think you have to reinvent your teaching persona to be effective. You can use approaches that bring you joy as an educator.

Even if the pandemic subsides, online learning seems certain to be a lot more mainstream than it used to be. It’s time for data scientists to get involved in determining the best way to help children learn from home. California’s plummeting pandemic test results show just how high the stakes are.


Source link