Sherman’s style note: An algorithm is not personal

Style note to education beat reporters: an algorithm is not “personalization” of education, no matter how many people make the claim. As computerized algorithms currently exist, here are the things that an algorithm cannot identify in an educational context:

  • An algorithm does not know when to pull a student aside at a quiet moment to ask about her mother. When a teacher knows and reaches out, that’s personalization.
  • An algorithm does not know when a joke will disarm a tense moment for a particular class. When a teacher knows and launches the joke, that’s personalization.
  • An algorithm does not know how long to wait before the student in the second row will not be able to remain silent for one minute longer and just has to speak up. When a teacher knows and waits, that’s personalization.
  • An algorithm does not know when the uncomfortable student leaning out of his chair needs to go use the bathroom. When a teacher knows and gives permission, that’s personalization.
  • An algorithm does not know who needs to be persuaded against starting a fight at lunch. When a teacher knows and talks the student down, that’s personalization.
  • An algorithm does not know when the blank look means “you lost me a while ago and I hope my friend understood what you just said,” and when it means “I’m not buying what you’re selling today.” When a teacher knows and responds to that student, that’s personalization.

This is not a slam against all algorithms — they have their uses. But the claim that an algorithm can personalize education? Not currently possible.

If you enjoyed this post, please consider subscribing to the RSS feed to have future articles delivered to your feed reader, and sign up for my irregular newsletter below!

2 responses to “Sherman’s style note: An algorithm is not personal”

  1. William J McKibbin

    I tend to agree with the premise that “algorithm” are not personal. But, even I have had to update my own thinking (and feelings) about computerized education. Consider the effectiveness of Rosetta Stone language education programs, which are delivered “algorithmically.” Likewise, consider the effectiveness of YouTube videos to deliver software training. Finally, consider the effectiveness of the Khan Academy in the 21st century. Depersonalization of education may even have some benefits, like leaving those learners who refuse to commit to learning behind. The future of education is going to be effected by the algorithm whether we as educators want to realize that or not.

    David Berlinski (2000) wrote his book entitled, “The Advent of the Algorithm: The 300-Year Journey from an Idea to the
    Computer” (2000) that two great ideas have most affected technological progress of the Western world: “The first is the calculus, the second the algorithm. The calculus and the rich body of mathematical analysis to which it gave rise made modern science possible; but it has been the algorithm that has made possible the modern world.”

  2. CCPhysicist

    As they currently exist, on some present generation of computer hardware built on some present data base. Algorithms can learn how long to wait for a particular student to respond. They can, in principle, even learn from watching different teachers. That is, IMHO, the main purpose of on-line homework systems: to collect data to inform (potentially automated) algorithm development.