Economic Geography

What approaches and methods will we use?
Just as the questions posed by economic geographers have broadened in the past
few decades, so have the approaches and methods that economic geographers use.
Thinking back, thinking ahead
27

---

I’ll briefly review some aspects of this broadening and provide an example using
a study that examined access to employment.
The theoretical frameworks, approaches, and methods adopted by the leaders
of the quantitative revolution in geography were largely borrowed from neoclas-
sical economics. As a result, an approach that was widely used by proponents of
spatial analysis in the 1960s entailed the development of deductive models and
the subsequent testing of these models with available (secondary) data (Dacey
1966). At the core of these deductive models was economic man, that completely
rational, profit-maximizing, all-knowing, a-social individual whose decisions
drove the models. Economic man was the main target of behavioral geography,
which emerged in the late 1960s as a reaction against the rather rigid, mechanistic
form of agency implied in this neoclassical actor.
I see behavioral geography very much as an outgrowth of the quantitative
revolution. For one thing, students who were inspired by this alternative to the
neoclassical view of the world had been deeply immersed in mathematics, philos-
ophy of science, mathematical modeling, and quantitative methods. The goal of
behavioral geographers was not so much to reject these accoutrements of the
quantitative revolution as it was to humanize economic man – to recognize that
people do not have complete information, are not always distance minimizing,
are embedded in networks of social relations, and therefore may base decisions on
factors other than sheer economic rationality. If people do not follow the precisely
prescribed decision-making calculus of economic man, how do they actually make
decisions?
Although it is currently fashionable to ridicule behavioral geography as 
positivistic and somewhat simplistic, I see it as a highly significant phase of our
disciplinary history insofar as it was the very beginning of the idea that people
are different and that these differences matter immensely to decision making 
and to behavior. Behavioral geography represents the beginning, in geography,
of looking at agency in a meaningful way by trying to grapple with some of 
the complexities that agency poses. From these origins have grown distinctly 
un-positivistic interests in, for example, difference, identity construction, meaning,
positionality, and the social construction of space and scale. These more recent
developments also incorporate approaches that emerged in reaction to (i.e. in
order to correct some of the deficiencies in) behavioral geography, namely those
approaches that pay attention to social structures, institutions, and cultures.
One example of studies of access that in some ways trace their origins to
behavioral geography yet integrate other approaches is the work that Gerry Pratt
and I carried out on gender and urban labor markets in Worcester, Massachusetts
in the 1980s and 1990s, summarized in Hanson and Pratt (1995). This series of
studies grew out of one simple finding from analyzing travel data from Baltimore,
Maryland, namely that women and men who work in female-dominated occupa-
tions (such as clerical work or elementary teaching) work closer to home than 
do those who work in other occupations (Hanson and Johnston 1985), a finding
that raised the big question that became the focus of our study: What is the rela-
tionship between geography and gender-based occupational segregation?
28
Susan Hanson

---

Through in-depth interviews with household members and employers, and
through analyses that ranged in scale from intra-household to the workplace, the
community, the nation state, and global movements of workers, we examined
how spatial divisions of labor, and therefore (often lack of) access to opportuni-
ties, are created from the decisions of people within households and employers 
as well as from social networks, community structures (such as different norms
and forms of child care), and local cultures (such as rootedness to place or the
practice of the intergenerational transfer of housing within families).
My hope is that in their approaches and methods economic geographers of 
the future will combine theoretical and empirical work, will pose genuinely 
open-ended research questions and be open to surprises, and will productively
mix qualitative and quantitative approaches. I believe it is now apparent that the
scale of analysis is not related to the level of sophistication of the theoretical
analysis (that is, small scale does not equate to being a-theoretical), empirical
work does not equate to a lack of abstraction, and attention to complexity does
not equate to disregard for theory (Massey 1994). Theory and data are closely
linked, and this link can be especially productively explored via fieldwork and
comparative studies. In line with my view that difference encourages creativity, 
I would also encourage economic geographers to engage in collaborative fieldwork
with international colleagues in field sites outside one’s country of origin. 
I am not enthralled by studies in which the investigator knows (or at least
appears to know) the answer to the research question before launching the study,
and I urge students to pose research questions that are genuinely open-ended,
rather than questions that are aimed at proving something the investigator is
already quite certain about. Why spend several years of your life researching some-
thing you are not genuinely curious about? A related point is the challenge to be
continually open to surprises throughout the research process; every open-ended
research question leads to surprises, and sometimes probing these surprises is
enlightening indeed. It’s often interesting at the end of a project to reflect on
what you learned that was truly unexpected. In this regard, a piece of advice that
Professor Shalom Reichmann
2
gave me years ago was, ‘Don’t ignore the outliers;
they can tell you a lot about what’s going on’.
I am encouraged by the increasing number of economic geographers who 
are combining quantitative and qualitative approaches in their work, a trend that
bodes well for the future. As Eric Sheppard has pointed out, there is no need to
create a dualism between the quantitative and the qualitative as some are wont
to do, nor is it sensible to equate one (the quantitative) with positivism and the
other (the qualitative) with critical approaches in geography (Sheppard 2001).
The two approaches are complementary, so that when employed together they
provide a much fuller analysis than can either in isolation. Moreover, they often
enable the investigator to communicate effectively with different audiences;
government officials may be partial to quantitative information, for example
Hanson and Pratt (2003). In using each of these approaches, however, I think 
we need to be as transparent as possible in describing the methods and especially
the categories we use; for example, how much heterogeneity is hiding within any
Thinking back, thinking ahead
29

---

one category? I would also like to see us think carefully about – and raise the bar
for – the standards of evidence that we collectively agree on as acceptable in
economic geography scholarship, whether that scholarship uses quantitative or
qualitative methods.