Below the fold is a case study on the PBL setup I used for EC6012 last year. I wrote a paper about it, and presented it at a conference. Now it's a case study.

Some version of the stuff below the fold should be available on the economics network in a few days.

Update: The finished article is here. That was quick!

Teaching Structuralist Economics Using Problem

Based Learning and Weblogs

1 Problem

Teaching post-Keynesian monetary economics is not easy. The lecturer has to

overcome the natural apprehension masters students must feel when faced with

completely new material, which purports to rewrite much of what they already think

they know about macroeconomics. Students will ask of their orthodox courses,

”Was all that hard work in vain? Surely not”. The lecturer must justify the

use of seemingly familiar concepts in different ways to the students. There

is less textbook-style material to draw from, and structuralist models are

generally more complex, inviting, as they do, more reality into the average

model than more orthodox models. Time considerations usually preclude a

parallel treatment of economic issues using orthodox models all the way

through the module. Also, because the focus of the course was primarily

technical rather than a survey or history of economic thought-based module, the

content was likely to be harder than most students had been exposed to

previously.

The lecturer must have an answer for each of these issues. A large amount of

effort must be expended in comparing and contrasting the orthodox macroeconomics

learnt from textbooks on macroeconomics like Barro00A0;(2007) and Barro and

Sala-i Martin00A0;(2003) to justify one’s approach and contextualise unfamiliar

course material for the students. One spends significant amounts of class

time simply setting the basics up before the real benefits of post-Keynesian

macroeconomic modeling—dealing with uncertainty, business cycles, and

behavioural actors through their generation of stocks and flows of funds—become

apparent. Unless great care is taken, all but the best students are left behind to

muddle through as best they can. The course creates confusion, annoyance and

frustration and exposes the lecturer to the dreaded ”so what?” from their

students.

Godley and Lavoie00A0;(2006) have written a masters-level textbook that

reduces many of these costs. As they progress through the book, Godley and

Lavoie build models of increasing complexity from a basic methodological

premise: the decisions and actions of economic actors (banks, households,

governments, etc) create financial stocks and flows which must be accounted for in

the aggregate. This aggregate is the macroeconomy. Macroeconomics is a

series of explanations of how these accounting entities influence one another

through time, in the presence of endogenous money, fundamental uncertainty,

and differing expectations. Each macroeconomic model, then, is the set of

behavioural equations imposed by the modeler on the national accounts which

hold the information created by the economic actors as they go about their

businesses.

Each model is developed from a simplified balance sheet and transactions matrix

which records the stocks accruing to each actor and the flows between them. A set of

dynamic behavioural equations is posited effectively determining the direction of the

financial flows between the actors. Steady state conditions are derived, and the

simulated equilibrium system is shocked under various scenarios to test differing

actor and national response to changing policies or economic conditions. Each of

these simulations is available online to allow students to verify the authors’ claims for

themselves.

2 Problem Based Learning

I set out to teach a masters-level course using Godley and Lavoie’s textbook as a

guide. The details of the course (lecture notes, podcasts, student blogs, etc) are here.

I decided to use Problem Based Learning as the key to these new techniques and this

new modeling methodology.

Problem-Based Learning (PBL) methods have been used in medical

education for some time (Barrows,00A0;1986;00A0;Delisle,00A0;1997) and in economics,

PBL techniques were recently applied to US high-school students in

macroeconomics (Maxwell et00A0;al.,00A0;2004,00A0;2005;00A0;Sharp,00A0;2003). Given the potential

benefits1

of this style of teaching, a module was designed to try out this method

on a postgraduate economics class, but with an added twist—the PBL

methods would be used to introduce students to the concepts of structuralist

macroeconomics, and most of the group work would be conducted via

weblogs2 .

The class had 36 students. The average level of student preparation in economics

was intermediate undergraduate micro- and macroeconomics, taught from standard

textbooks. Motivation was not a problem as the students were preparing for a

professional qualification and represented an elite group already. The module was

designed to introduce structuralist macroeconomic modeling techniques, specifically

stock-flow consistent macroeconomic modeling via simulation, also called the

‘structuralist’ approach (Taylor,00A0;2004, pg. 19), to the students, and these new

concepts and techniques were to be delivered using PBL methods. The canonical

PBL method has students and teachers confront an ill-structured problem that, as a

result of investigation, research, and cooperative discussion, allows for more than one

possible correct solution. As students work on the problem, they discover that

understanding economics concepts are essential to framing and solving the dilemma.

Problems, although loosely structured to allow for student discovery and

independent learning, contain a fixed set of components. My treatment was

different from this because there was really only one answer to the problem

statement—the answer that lead students to the model developed in lectures later

on.

Initially, students found the new teaching method difficult to cope with, and to a

certain extent the novelty in lecture structure retarded progress in content delivery

for the first few weeks. This ‘learner resistance’ is well documented in the PBl

literature (Schmidt and Moust,00A0;2000), though it represents a significant cost in

lecturing time lost. Once the ‘rules of the game’ had been learned though, in general

students found the module useful and stimulating. In a questionnaire distributed

during the last class, students rated the module highly relative to other modules

taken, though in the qualitative section of the questionnaire their comments showed

there was definite room for improvement in the module design, especially with regard

to the formation of groups, group size in particular, and feedback on problem

sets.

3 Issues in PBL setup and delivery

As in all PBL-based modules, each lecture began with a problem statement, the

lecturer gave the students access to the resources to try and solve the problem, with

the lecturer acting as a guide to these resources, solving simple conceptual blocks for

the students as the lecture progressed. The students worked through various

problem-based scenarios, using the internet and available informational resources as

guides, and reported the results of their problem-solving methods via weblogs

set up by each group. The lecture theatre had wireless access, so students

could work on their laptops in groups during the lecture. Each model was

simulated for the students using Eviews, and students were taken through the

estimation and simulation of the different models using different parameters,

situations, etc, in computer labs lead by a teaching assistant later on in the

week.

Teaching a complicated, technique-laden subject like structuralist macroeconomics

from a PBL standpoint also requires some modification of the textbook PBL premise.

In essence, there are no open ended answers that groups might come useful p

with—the correct answer is always the model to be taught in lectures. PBL

gave the students a clearer motivation for the development of these models,

and students could definitely see the rationale for developing these models

in a stock-flow consistent way, but in the end, the models still had to be

developed, discussed, solved for its steady state(s), shocked, and simulated.

This is the methodology of structuralist macroeconomics. The PBL method

allowed students to see why these methods were being chosen above other

neoclassical or neo-Keynesian alternatives, but the models still had to be

solved. So a blend of PBl and traditional lecturing and problem-sets was

required.

4 Module Design and Assessment

The module was designed around twelve two-hour lectures with the lecturer, followed

by one-hour tutorials and computer labs with a teaching assistant to embed the

concepts learned in the module for the students. The module began with an overview

of the differences between the stock-flow consistent (or structuralist) approach and

the traditional orthodox Neo-Keynesian approaches the students had previously been

exposed to. The differing philosophies of the two schools were presented, as well as

their respective policy prescriptions for dealing with the various macroeconomic

pathologies of inflation, unemployment, etc. Each subsequent lecture introduced

students to a new and successively more complex model of the macroeconomy using a

problem statement and in class exercises to guide student inquiry. The primary

learning objective of the module was to give students the tools to actually

create such models themselves using computer programs such as Eviews or

Mathematica.

To get a grade in the module, students had to complete all their problems and

submit them as a group online for 20%, produce and deliver a 30 minute presentation

and summary of a seminal paper in the area worth 30% of their grade, with the

intended audience being their peers in the class. Students were graded on the quality

of this presentation by the other groups with the lecturer having the final say in

grading. They were also required to complete a final written exam solving standard

models for 50%.

5 Conclusion: Student Feedback, and Lessons Learned

As a result of the continual assessment and large group size, as well as the high

quality of initial student preparation, no student failed the module in 2006. The most

interesting remarks are found in the qualitative comments. Some students feel

the PBL module is working well, writing “Lecturer is knowledgeable and

very enthusiastic about all aspects of economics. The module objectives

could be a little clearer.” However, others find the method confusing and

annoying “The lectures are not structured, there are no clear objectives and

it is very badly taught. The groups for projects are too bifg and the [sic]

self-based learning is not working.” The comments that appear most often is the

seeming lack of a clear objective in each lecture followed by group size. It is

reasonable from the student’s point of view that each lecture would look

unstructured: one does start out with a sentence and a set of web links, after all.

The issue of group size, 2-3 versus 4-5 students, came up over and over.

Because students were randomly allocated to rather large groups at the

start of the module there was some infighting and shirking, though this

was not reported until after the module had finished, and even then only

conversationally. The area of most contention was the presentation, where

students were asked to summarize a technical paper in such a way that

they could teach the paper to their peers. Those groups who by chance had

a more technically minded member or two seemed to shine, while other

groups did not fare so well, and this caused some friction in the class. The

next class will thus have students ‘self-selecting’ onto groups of between

2 and 3 at a maximum. While this increases the amount of work for the

lecturer, it also removes many of the issues that came up during this semester’s

module.

References