June 26, 2008

Question 2: Why don't students retain what they've learned

Roger Sweeny asks:



During the year, students had "learned" various things. How do I know they learned them? Because they correctly answered questions on tests or used the appropriate formulas and ideas to solve problems (I teach physics and physical science).

But when they take the final, they show that they don't know lots of the things that they seemed to know earlier in the year.

So maybe it's incorrect to say that they actually "learned" those things. Maybe it's more accurate to say that they "memorized" various things and then forgot them.

Two questions: One, does anyone doing education research care about this? My impression is that it's a gigantic elephant in the room that no one wants to talk about. And yet teachers know it's there. Every teacher knows that if she gave the same test a month later, her students would do considerably worse.



The short answer is that learning something, even to mastery, is not sufficient. Merely mastering something does not guard it against the ravages of forgetfulness. In order to retain what you've learned you need to overlearn it past the point of mastery. Schools don't do this.

For the cognitive science behind all this read these three articles from Dan Willingham:

1. Students Remember...What They Think About

2. Why Students Think They Understand—When They Don’t

3. Practice Makes Perfect--But Only If You Practice Beyond the Point of Perfection

Roger continues:

So question two: Does anyone doing educational research use long-term learning as a measure of success? When researchers try to determine if various things "work," how many test for knowledge 24 hours later? a week later? a month later? more?


The DI people do, though the standardized testing instruments don't capture the long-term effects if learning.

Read Engelmann's Student-Program Alignment and Teaching to Mastery for one way how to truly teach to mastery with retention. Another way to do it is through Precision Teaching, as another commenter indicated.

The easiest way to learn how to teach to mastery is to take a look at one of the DI programs. I'd suggest getting a mid-level reading or math program from eBay (you must get the presentation book and teacher guide along with the student materials). Here's a series of posts I did analyzing the practice provided in a lesson from Reading Mastery rainbow Edition. Here's another series on how to teach a particular with distributed practice.

If anyone is interested I have a presentation paper that Don Crawford gives at the yearly DI conference on how to apply DI principles to content area texts. Here are a few points Don makes:


It’s not about you—it’s about the kids, and what they can articulate fluently.

Keep teaching and keep students practicing until knowledge is learned to the “walk around” level—known well enough that the kids can think about it and explain it when they are walking around without a book.

Why? Because the first prerequisite for knowledge to be used flexibly in higher order activities such as essays, comparisons, application, synthesis, evaluation, etc. is that it be learned to the “walk around” level.

Therefore you must test by production items (not multiple choice recognition level) to know if content knowledge is learned well enough. Also to capture relationships.


If you want a copy of the presentation, shoot me off an email.

2 comments:

Anonymous said...

The Precision Teaching folks would say mastery is not enough, rather the material must become "fluent".

The study of fluency was pioneered by Ogden Lindsay. Many people agree that the most effective learning method beyond compare is DI combined with PT.

This is the method the best LD school in the nation uses, Morningside Academy in Seattle.

RMD said...

4trogan brings up a great system: Precision Teaching.

Here's a great introduction to the benefits of fluency and some sample goals:

http://tinyurl.com/3ubdkf