Posted in Teaching mathematics

Use of exercises and problem solving in math teaching

Mathematical tasks can be classified broadly in two general types: exercises and problem solving tasks. Exercises are tasks used for practice and mastery of skills. Here, students already know how to complete the tasks. Problem solving on the other hand are tasks in which the solution or answer are not readily apparent. Students need to strategize – to understand the situation, to plan and think of mathematical model, and to carry-out and evaluate their method and answer.

Exercises and problem solving in teaching

Problem solving is at the heart of mathematics yet in many mathematics classes ( and textbooks) problem solving activities are relegated at the end of the unit and therefore are usually not taught and given emphasis because the teacher needs to finish the syllabus. The graph below represents the distribution of the two types of tasks in many of our mathematics classes in my part of the globe. It is not based on any formal empirical surveys but almost all the teachers attending our teacher-training seminars describe their use of problem solving and exercises like the one shown in the graph. We have also observed this  distribution in many of the math classes we visit.

The graph shows that most of the time students are doing practice exercises. So, one should not be surprised that students think of mathematics as a a bunch of rules and procedures. Very little time is devoted to problem solving activities in school mathematics and they are usually at the end of the lesson. The little time devoted to problem solving communicates to students that problem solving is not an important part of mathematical activity.

Exercises are important. One need to acquire a certain degree of fluency in basic mathematical procedures. But far more important to learn in mathematics is for students to learn to think mathematically and to have conceptual understanding of mathematical concepts. Conceptual understanding involves knowing what, knowing how, knowing why, and knowing when (to apply). What could be a better context for learning this than in the context of solving problems. In the words of S. L. Rubinshtein (1989, 369) “thinking usually starts from a problem or question, from surprise or bewilderment, from a contradiction”.

My ideal distribution of exercises and problem solving activities in mathematics classes is shown in the the following graph.

What is teaching for and teaching through problem solving?

Problems in mathematics need not always have to be an application problem. These types of problems are the ones we usually give at the end of the unit. When we do this we are teaching for problem solving. But there are problem solving tasks that are best given at the start of the unit. These are the ones that can be solved by previously learned concepts and would involve solutions that teachers can use to introduce a new mathematical concept. This strategy of structuring a lesson is called Teaching through Problem Solving. In this kind of lesson, the structure of the task is king. I described the characteristics of this task in Features of Good Problem Solving Tasks. Most, if not all of the lessons contained in this blog are of this type. Some examples:

  1. Teaching triangle congruence through problem solving
  2. Teaching the properties of equality through problem solving
Click the links for more readings about Problem Solving:
Posted in Curriculum Reform, Mathematics education

Why math education is failing

A backlink to my post What kind of mathematical knowledge should teachers have?   brought me to the essay by Matthew Brenner titled The Four Pillars Upon Which the Failure of Math Education Rests (and what to do about them). Here’s the quote from the essay posted in Wild about Math.

Kids are taught math as pets are taught tricks. A dog has no idea why its master wants it to perform. With careful training many dogs can be taught to perform complex sequences of actions in response to various commands and cues. When a dog is taught to perform a trick it has no need or use for any “understanding” beyond which sequence of movements its trainer desires. The dog is taught a sequence of simple physical movements in a specific order to create an overall effect. In the same way, we teach children to perform a sequence of simple computations in a specific order to achieve an overall effect. The dog uses its feet to move about a space and manipulate objects; the student uses a pencil to move about a page and manipulate numbers. In most cases, the student doesn’t know any more than the dog about the effect he creates. Neither has any intrinsic motivation to perform nor any idea why the performance is demanded. Practice, practice, practice, and eventually the dog can perform reliably on command. This is exactly how kids are trained to perform math: do a hundred meaningless practice problems, and then try to do the same trick on the test.

Mr.Brenner’s observation is as true in America as it is here in the Philippines. This is a painful truth but something that we all must take seriously. I strongly encourage our teachers, those writing our new curriculum framework (I think this is our third within the decade), textbook publishers and our DepEd officials to read the entire essay. The author outlined the reasons why math education is failing but he also offers solutions which I believe are doable even if our average class size here is 60! Let me list the 10 point solutions:

  1. Understanding Must be Central in Math Education
  2. Textbooks Must Not be Allowed to Undermine Math Education
  3. Teachers Must Stop Teaching Math as They Learned It
  4. Curricula Must be Coherent and Cumulative
  5. Worked Examples Must be Emphasized for New Material
  6. Curricula Must Include Examples of Excellent Performance
  7. Assignments Must Draw on the Old and the New
  8. Content Must be Meaningful and Contexts Must be Rich
  9. Metacognitive Activity Must Pervade Mathematical Activity
  10. Language Must be Taught, Used and Evaluated Fairly
I do not agree with #5 proposal because I believe that mathematics should be taught in the context of solving problems but I think this is a very good list. Find time to read it. Mr. Brenner also offers very good sample lessons. You may also want to read 10 signs there’s something not right in school maths and let us know your thoughts.
Posted in Mathematics education

What kind of mathematical knowledge should teachers have?

As a result of her research, Liping Ma developed the notion of profound understanding of fundamental mathematics (PUFM) as the kind of mathematical knowledge teachers should possess. She discusses this kind of knowledge in her book Knowing and Teaching Elementary Mathematics: Teachers’ Understanding of Fundamental Mathematics in China and the United States (Studies in Mathematical Thinking and Learning Series). This book is now considered a classic by many mathematics educators. The ‘elementary’ in the title does not mean the book will be valuable to elementary teachers only or those engage in the training of prospective elementary teachers. The book is for all mathematics teachers, trainers, and educators. This book is a must-read to all that has to do with the teaching of mathematics.

Here’s what Liping Ma says in the introduction:

Chinese students typically outperform U.S. students on international comparisons of mathematics competency. Paradoxically, Chinese teachers seem far less mathematically educated than U.S. teachers. Most Chinese teachers have had 11 to 12 years of schooling – they complete ninth grade and attend normal school for two or three years. In contrast, most U.S. teachers have received between 16 and 18 years of formal schooling-a bachelor’s degree in college and often one or two years of further study.

In this book I suggest an explanation for the paradox, at least at the elementary school level. My data suggest that Chinese teachers begin their teaching careers with a better understanding of elementary mathematics than that of most U.S. elementary teachers. Their understanding of the mathematics they teach and -equally important – of the ways elementary mathematics can be presented to students continues to grow throughout their professional lives. Indeed about 10% of those Chinese teachers, despite their lack of forma education, display a depth of understanding which is extraordinarily rare in the United States….

Why the word ‘profound’? Profound has three related meanings – deep, vast and thorough – and profound understanding reflects all three. From the paper delivered by Liping Ma and Cathy Kessel in the Proceedings of the Workshop on Knowing and Learning Mathematics for Teaching conference, Liping and Cathy offered the following explanation:

  • A deep understanding of fundamental mathematics is defined to be one that connects topics with ideas of greater conceptual power.
  • A vast or broad understanding connects topic of similar conceptual power.
  • Thoroughness is the capacity to weave all parts of the subject into a coherent whole.

A teacher should see a ‘knowledge package’ when they are teaching a piece of knowledge. They should know the role of the current concept they are teaching in that package and how that concept is supported by which ideas or procedures.

To further explain the kind of mathematics knowledge a teachers should possess, Liping and Cathy used the analogy of a taxi driver  who knows the road system well. The teachers should know many connections so that they are able to guide students from their current understandings to further learning.

I think this is how designers of curriculum, writers of curriculum materials, and teachers should interpret the standard “Making connections”.  It is not simply linking.