Monday, July 29, 2019

2019 #VWEDU #DTA: Suppose the real objective in school is to pass multiple-choice tests. Can a digital TA handle test-based learning?

Suppose the real objective in school is
 to pass multiple-choice tests. 
Can a digital TA handle test-based learning?

If passing a multiple choice test on the content in a reading assignment is the objective, most of the learning can be automated.   It may not need any educators except to keep the system current.  The rest could be handled by managers and proctors.
The Empty Classroom

Test-based learning

  • It is a legitimate concept:
  • Test-based learning (search)
  • Scholarly articles for Test-based learning
  • Are they still doing that?  
  • I don't know.  Probably not in corporate training
  • Corporations want results, not test scores.
  • But maybe in education.  
  • Test-based learning is easy to provide.
  • And achievement is easy to measure.
  • You just pick some books and provide some lectures
  • Then you give tests on what the students remember.
  • You could automate all of that.
  • Give links to online books and video-captured lectures.
  • Use multiple-choice tests drawn from an existing database.
  • Maybe give online study sheets that have the answers to likely test questions.
  • You don't need teachers.  Just a test manager.  And a test proctor.
  • You don't need educators for superintendents and principals.  
  • You could use retired military officers instead.  
  • Students take tests online with school-supplied computers so they can't cheat.
  • You need a proctor to make sure they aren't cheating.  
  • There are limitations.

Advantages of test-based learning

  • Most content and test items remain reusable for a number of years
  • Most content and test items could be used across a whole nation.
  • Almost all content could be made available on the browser.
  • Students who have their own laptop would only need to come to school on test days.
  • The money saved on buildings could probably buy every student a laptop.
  • Assuming tests once a week, a school building might handle far more students.
  • Students who could stay on task would require no live teacher. 
  • Or at most an online teacher who could answer questions. 
  • In time a digital TA could be developed to handle common questions.
  • This method could dramatically lower costs.  
  • And it might produce students with great skill at reading text and answering tests.
  • If there ever is a market for those skills.

Limitations of test-based learning

  • You can't use it in the lower grades.
  • In grade school, you have to teach useful skills: Reading, writing, arithmetic.
  • People learn skills by practice
  • To get little children to practice you have to give them tasks
  • You can still use tests to assess those skills.
  • But you have to wait till they are older to use tests to force them to "study."
  • Especially when what they are studying is boring and of no obvious value.
  • "Studying" does give practice in reading boring texts and listening to boring lectures.
  • Why high schools teach tolerance for boredom and what they can do about it.
  • Maybe that is still good college prep (as it was when I went to college).
  • But I doubt that it is part of any job description.
  • So if you want the students to have marketable skills, you need to think of other learning objectives.

Would they really set that objective?

  • They would not admit to it. They would insist they are preparing students for life.
  • But it is hard to find "reading text and answering tests" in any job description.
  • The text and tests may deal with content relevant to some jobs.  
  • The students may come away able to talk about what people do in those jobs.
  • But employers know there is a difference between talking about and doing a job,


Visit me on the web



  • Original text in this blog is CC By: unless specified public domain
  • Use as you please with attribution: link to the original.
  • All images without attribution in this blog are CC0: public domain.
  • Second LifeLindenSLurl, and SL are trademarks of Linden Research Inc.
  • Annotated screenshots made with Jing
  • This blog is not affiliated with anything.   Ads are from Google.
  • Selby Evans in Kitely and hypergrid, Thinkerer Melville in Second Life

No comments:

Post a Comment