Learning by Analogy - Oracle Oracle Teaching Tips by Burleson Consulting

Many professors rely on analogy, metaphor and over-generalization to maximize complex learning, and it's a skill that is taught in graduate schools across the world.  I've taught over 80 classes in graduate school, and the only way to teach complex topics like statistics and database are by using analogies, building on the prerequisite conceptual framework classes such as "Algorithms" and "Data Structures" courses. 

Learning with Succedaneums

The term "succedaneum" mean something that can be used in-place of another, and teaching by succedaneum is a critical skill for science and engineering courses.  Poet William Wordsworth said it best:

Science appears as what in truth she is,
Not as our glory and our absolute boast,
But as a succedaneum, and a prop
To our infirmity.

William Wordsworth

The teacher starts with simple, over-general assertions,  building layer-upon-layer of complexity on top of each generalization until the whole concept becomes complete. 

Life is like a box of chocolates - The use of analogy in scientific instruction

Teaching by analogy and over-generalization is an integral part of the learning process, especially for complex concepts, and academic research confirms the centuries-old belief that children learn complex concepts best by analogy and over-simplification. 

For example, all junior High school kids like analogy jokes: "What does a oyster shucker with epilepsy have in common with a prostitute with diarrhea?"  (answer at bottom), and we all took the SAT analogy test and learned to express one thing in terms of anther:

MICROSOFT : DATABASE ::

(A) wild pack of rabid dogs : puppy
(B) Goo Goo Dolls : Replacements
(C) playa lake : oceanic trench
(D) I Can't Believe It's Not Butter : butter

Analogies are also helpful to understand complex science concepts and there are many examples:

• Biology - Ontology Recapitulates Phylogny is the analogy (untrue, BTW), that a developing zygote/fetus reproduces it's own evolutionary stages (e.g. a mammal fetus has gills at one point, allegedly mimicking it's evolution).
   
• Statistics - When we teach linear regression, we do it first in single terms first, showing sum of the least squares for examples on simple predictive models.  As the student learns more, we can then go into details, introducing exponential smoothing formulae and multivariate models. 
   
• Electrical Engineering - Introductory students are taught that electrical current is like a network of pressurized water pipes.

Other common analogies for leaning computer science include:

• Java - It's just like C, but without pointers. (Java is sometimes referred to as C++ ++ --).
   
• Polymorphism - A polymorphic interface is like a steering wheel on a car.  The steering may have many underlying types (power, rack and pinion), but all cars are unified by a standard interface.
   
• Class Definition - A class definition is like a stamping machine in a factory.   Just as the stamping machine stamps out products, a class definition stamps-out objects.
   
• Relational Databases - Relational databases are like naked people.  Stand them on their heads and they all look alike.

The ability to over-generalize and simplify complex concepts with analogies is a critical skill for anyone who teaches computer science.  Check-out the Google job interview questions where candidates must explain computer concepts to a hypothetical 8 year-old:

Q: "Explain a database in three sentences to your eight-year-old nephew."

Research on teaching by Analogy and Simplification

Teaching by analogy is especially useful for learning complex technologies, as noted by a wealth of academic research.  Professor John Robinson of the Memorial University of Newfoundland notes that the use of analogy and simplification is a critical skill for any engineer:

"Practicing engineers probably make use of analogy as often as practicing lawyers. Reference to previous jobs, identifying similarities and differences, making linkages between contexts, are all regular habits.

In many cases the analogies will be simple and direct, but, especially in systems engineering, the linkage can be between two very different domains.

The ability to see analogical situations, particularly in balancing the values of different criteria, is central to engineering judgment. The ability to explain these analogies, and argue their relevance, is engineering rhetoric."

This paper on cognitive science describes the use of analogies in leaning complex science concepts:

"Analogies are commonly used in science instruction -- for example, when an electrical circuit is compared to a water-flow system. Learners' models of a domain are shaped by such instructional analogies.

Analogy is used to apply arguments used in examples to new problems, and similarity is used to extract principles from repeated, overlapping experiences. "

Teaching Oracle Database concepts by analogy and simplification

I noticed that many aspiring Oracle teachers have trouble when "dumb-down" their communications to the level where they can be understood by novices, dilettantes and High School level students in Oracle Academy.  I've successfully presented Oracle technology to 9th-graders, and it you don't know how to communicate at the same level as your audience, you are wasting your time. 

For consultants, many top-level IT managers don't know (or care) much about the relative features of database products (SQL Server, Oracle, DB2), and it's necessary for a real Oracle guru to deliberately over-generalize and explain complex technology with simple analogies and illustrations. 

Many Oracle instructors use simple analogies, starting with overgeneralization, and filling-in the exceptions as the students learn:

• Oracle Index - An Oracle index is like a card catalog in a library.  If you visit the card catalog you can go right to the book without having to scan every shelf in the stacks.
   
• Freelists - To a beginner, being able to say "Oracle freelists are like one-way linked lists", gives the student an immediate frames of reference.  Of course, traditional Oracle freelists are not 100% like one-way linked lists, but these details can come later.

Analogy and the knowledge of the Audience

While the use of analogies and metaphors is the best way to "layer concepts" in beginner curricula, teaching practicing professionals presents different challenges.  Oracle professional Scott Swank (co-creator of the hilarious Oracle WTF blog) notes that can be a poor approach not to mention all of the subtle nuances when describing complex Oracle database features:

"I consider over-generalization to be a poor approach by definition. I consider analogies to be a poor replacement for concrete understanding."

True, for experienced, practicing Oracle professionals it's very important to completely describe all exceptions when explaining Oracle concepts.  On the other hand, a beginners eyes will glaze-over if their Oracle teacher starts listing dozens of exceptions, describing techniques that they do not comprehend.  As the kids say in college today "TMI", too much information.

Hence, the use of analogy is desirable for beginner courses, but you must make sure to follow-up with the details and exceptions when the student is prepared to understand and appreciate them.

Preventing conceptual errors with analogies

Professor Gertner explains the power of analogies and metaphors, noting that the "skill" of creating a good analogy is the key to preventing conceptual errors when teaching beginners with analogy:

"Even without explicit teaching, learners' naive mental models are often formed around their own spontaneous analogies and implicit similarity comparisons. Sometimes these comparisons lead to models that are wholly or partly misleading.

Thus, a better understanding of the psychology of analogy and similarity would not only improve our theories of learning and transfer but would also allow us to anticipate and perhaps prevent many conceptual errors."

References:

The Analogical Mind

 An Oyster shucker shucks between fits