On The Utility and Definition of Regions in Comparative Political Sociology

• Author: Kevin R. Cox
• DOI: 10.1177/001041406900200104
• Published date: 01 April 1969
• Date: 01 April 1969
• Subject Matter: Articles

ON

THE

UTILITY

AND

DEFINITION

OF

REGIONS

IN

COMPARATIVE

POLITICAL

SOCIOLOGY

KEVIN

R.

COX

KEVIN

R.

COX

is an Associate PIofessor

of

Geography at Ohio Statc

University.

His

mjor

interests lie in the

arms

of

the quontitative study

of

the spatial structure

of

political beltavior and in the application

of

behav-

ioral models

to

spatial problems. He

is

the author

of

tw

papers to

appear

in

a

volume

edited

by Mattei Dogan and Stein

Rakkan,

Quantitative

Ecological Analysis in the Social Sciences,

and co-editor of the

forth-

coming volume

of

essays,

Behavioral Models in Geography:

A

Symposium.

HE

REGION

IS

a concept which finds wide applicability and wide-

spread use in the work of comparative political sociologists.

This

is

reflected not only in the policies

of

such data-gathering agencies as Gallup

Poll but also in the analytic designs employed by researchers. For example

the seminal study

of

American voting,

The American Voter

(Campbell et

al.,

1960),

is at continual pains

to

differentiate between correlation pat-

terns distinctive

of

the South as opposed

to

the nonSouth. A more recent

study (Alford,

1963)

of survey data across four nations has elevated region

to

the status of a major explanatory factor and other work has demon-

strated the utility

of

the regional concept

in

ecological analysis

of

voting

statistics.'

AUTHOR'S

NOTE:

The research upon which this paper

is

based

nus

supported by a

grant

from

the National Science Foundation; this support

is

gratefully acknowledged

An

earlier draft was presented

at

the Loch Lomond Voting Group Conference, Alex-

andria, Scotland, July,

1968.

It received extensive review

from

Dr.

Lpwrence

A.

Brown

of

the Department

of

Geography at thdohio State University; some

of

his

suggestions, along with some

from

Dr.

Richard Rose

of

the University

of

Strathclyde,

have been incorporated into this version. While the author remains entireIy respon-

sible for the content

of

the paper, he would like

to

offer his thanks

to

the gentlemen

concerned.

UTILITY

AND

DEFINITION OF

REGIONS

1691

Despite the widespread use of the region in comparative political sociol-

ogy

research, one notes some problems

in

its application. Thus, the

Understanding

of

the regional concept appears to lack the sophistication

with which more sociological concepts such as social class are handled;

many studies in comparative political sociology focus upon the social bases

of the vote but few concentrate upon the regional or spatial bases. Con-

comitantly there seems to have been little critical examination of how

regions should be defined and measured. Thus, political sociologists attend

to the difference between subjective and objective social class but do not

pay a similar attention

to

different types of region. Given that rigorous

measurement and careful definition depend upon the careful conceptual-

hation of the role of a variable, this is probably as one might expect.

The viewpoint taken here is that the region is a geographic concept and

that geographical theory and geographical techniques-in combination

with a knowledge regarding the generation of voting differences across

individuals-can provide the basis for an improvement

in

the methodology

of comparative political sociology. The specific aims therefore are three-

fold: (a) to explain the utility of region as a concept in comparative

political sociology; (b) to critically examine methods of defining regions

for comparative political sociology; (c) to apply some of these ideas and

techniques to an empirical problem in comparative political sociology.

To

the first end, the region will be defined in terms analogous to the

definition

of

other concepts used in comparative political sociology: just

as social class is a classification of a variety of indicators,

so

is region. The

predictive power of region in past research will also be pointed out and the

theoretical utility of the region in comparative political sociology will be

clarified.

Second,

I

will demonstrate how there are a variety of ways by which

one can arrive at the variables to be employed in one's regional definition.

I

will then review the recent work of geographers concerned with the

problem of r'egional definition and measurement, and conclude with a

demonstration of the predictive power of regions defined

in

a rigorous

manner with full attention to theory and to taxomonic principles.

The

Utility

of

Regions

in amparative Political

Sociology

REGION AS A CLASSIFICATION OF PLACES

It is commonplace in the methodology of comparative political sociol-

ogy

to

measure a number of variables such as income, occupation, and age