THESIS

4‘

F .4.'-.~_:,k_ gut, n AM
5' 9““??? 53. WT“ I
1 exam; , a “at; 25%. ‘2?

' Eff-Es; 23333331: 35‘. ital-ate
l fluvmity J
\_—

 

 

This is to certify that the

thesis entitled

HIGHLAND RURAL MIGRATION IN NORTHERN CHILE

presented by

Osvaldo A. Mufiiz

has been accepted towards fulfillment
of the requirements for

. . . G r h
M A degree in eog ap y

 

 

A

MW

I
Mér professor

17 81
Date 9/ /

 

0-7639

 

 

RETURNING MATERIALS:
1V4f31.1 Rlace 1n 500E drop to
remove this checkout from
w your record. FINES wilI

be charged if book is

 

 

returned after the date
stamped beIow.

 

 

 

 

HIGHLAND RURAL MIGRATION IN NORTHERN CHILE

BY

Osvaldo A. Mufiiz

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Geography

1981

ABSTRACT

HIGHLAND RURAL MIGRATION IN NORTHERN CHILE

BY

Osvaldo A. Mufiiz

Despite of an important increase in rural to urban
migration research in Latin America, one finds a lack of
studies of rural to rural movements. This research of high-
lander migrants in northern Chile investigates migration
strictly within this rural area.

Assumptions focused on (1) rural to urban migration
and diffusion of information, (2) stage and stepwise migra-
tion and long-distance moves, (3) the attraction of inter-
vening opportunity centers, (4) the distortion of the
distance-decay function, and (5) differential migration by
sex, age, and marital status.

The results of both structural migration and spatial
mobility demonstrate that the rural migration process changes
from one of a stepwise function to one of long-distance
movements with time. Factor analysis and multiple regression
indicate that several variables are important. They are
(1) labor force, (2) work hours, and (3) movements within a

district.

ACKNOWLEDGMENTS

I would like to express my appreciation to all those
who collaborated in the preparation of this dissertation:
to Dr. Robert N. Thomas, my faculty adviser in the Geography
Department whom I first had the occasion and good fortune to
meet in Paipa, Colombia in 1977 and who was instrumental in
my decision to attend Michigan State University in 1979.

Dr. Thomas lent me long hours of assistance in the pain-
staking revision of my preliminary drafts for which I will
be ever grateful.

The guidance and insight into quantitative geographi-
cal methods provided by Dr. Bruce Wm. Pigozzi were an
invaluable contribution exceeded only by the patience he
showed me in the sometimes heated discussions in which we
found ourselves engaged over various points of disagreement
and for showing me the importance of detail and precision
in conducting an in-depth research project. I do not doubt
that I will be a better geographer for the valuable learning
experience I had with all those under whom I studied in the

Michigan State Geography Department.

ii

DEDICATION

To my entire family for their continual support
throughout these past two years in particular
my wife, Cecilia, a former student of mine, to
whom I am indebted for a large part of this
effort and a few helpful suggestions in the
manuscript and my dear sons who have patiently
learned to wait for the day when Dad will again

be available to play with them.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . .
LIST OF FIGURES. . . . . . . .
Chapter

I. INTRODUCTION . . . . . .

Scope and Content of Internal Migration

Research . . . . . .

Periodic and Rural-Urban Migration.

Migration, Diffusion, and Modernization

The Process of Acculturation.

Justification of the Study .

The Study Area . . . .
Statement of the Problem .
Major Purposes of the Study

Hypotheses and Operational Statements

Limitations of the Study .

II. REVIEW OF THE LITERATURE . .

Distance and Human Interaction.

Distance Factor . . . .
Intervening Opportunities.

Spatial Expression of Sociocultural

Transition. . . . . .

Stepwise Migration . . .
Long-Distance Movement. .

Replacement Movements and Selectivity

Intergenerational Movements as Chain

Migration. . . . . .

iv

Page
vi

vii

\IUIM N

11
12
16
16
17
19
19
20
23
24

25
27

29

31

Chapter Page

III. MIGRATION METHODOLOTY . . . . . . . . . 38
Demographic Change and Indirect Migration. . 39

Population Change and Annual Increase
Rate. 0 O O O O O O O O O O O 40

Average Annual Migration Rate . . . . . 41
Data PrOblem. O O O O O O O O O O 44

Structural Migration. . . . . . . . . 44
Volume and Spatial Effects . . . . . . 45
Differential Migration . . . . . . . 45
Data Problem. . . . . . . . . . . 46

Spatial Mobility . . . . . . . . . . 47
Factor Analysis and Multiple Regression. . 47

IV. ANALYSIS OF MIGRATION IN HIGHLAND COMMUNITIES . 49

Population Change and Indirect Migration . . 49
Structural Migration. . . . . . . . . 60

Migration Flows. . . . . . . . . . 63
Migration by Sex and Age. . 72
Migration by Marital Status. . . . . . 74

Spatial Mobility . . . . . . . . . . 78

V. CONCLUSIONS AND SUGGESTIONS FOR FUTURE
RESEARCH. O O O O O O I O O O O O 98

Hypotheses and Operational Statements . . . 98
Relationship of Labor Force to Migration . . 100
Application of Underlying Dimensions . . . 103
Data Quality to Predict Migration . . . . 104

BIBLIOGRAPHY . . . . . . . . . . . . . . 105
APPENDICES
A. Questionnaire, 1970-1978 . . . . . . . . 117

B. variables . . . . . . . . . . . . . 119

Table

1.

11.
12.
13.
14.

15.

LIST OF TABLES

Classification of Rural Settlements per Region,
Commune, and District . . . . . . . .

Average Annual Population Change, 1952-1960,
1960-19.70 o o o o o o o o o o o 0

Average Annual Population Change, 1952-1960,
1960—1970 0 O O O O O O O O O O 0

Average Annual Population Change and Average
Annual Net Migration Rate 1970-1978 . . .

Place-of-Birth and Place-of-Residence Matrix.
Migration Flows, 1970 . . . . . . . .

Intermediate Settlements 1970 . . . . . .
Average Number of Steps 1970 . . . . . .

Place-of—Birth and Place-of-Residence Matrix.
Migration Flows, 1978 . . . . . . . .

Average Number of Steps . . . . . . . .

Percentage Distribution of Age by Short-Distance
or Long-Distance and Sex, 1970, 1978 . . .

Migration by Marital Status 1960, 1970 . . .
Orthogonal Factors and Loadings . . . . .
Rural Settlements and Factor Scores . . . .

Stepwise Multiple Regression With Six
Orthogonal Factors . . . . . . . . .

Stepwise Multiple Regression With Twenty-One
Original Variables . . . . . . . . .

vi

Page

14

52

55

58

61
64
66

68

69

73
77
83
84

95

97

LI ST OF FIGURES

Figure Page
1. Northern Chile Reference Map . . . . . . 13
2. Highland Northern Chile Districts . . . . 15
3. Highland Northern Chile Settlements. . . . 43
4. Average Annual Population Change 1954-1960 . 51
5. Average Annual Population Change 1960-1970 . 53
6. VOlume and Direction of Flows 1970 . . . . 59
7. Volume and Direction of Flows 1978 . . . . 67
8. Migration by Age and Sex, Short- and Long-

Distance Flows 1978 . . . . . . . . 75

9. Migration by Marital Status 1970. . . . . 79
10. Labor Force Factor 1. . . . . . . . . 85
11. Local Movements and Distance Decay Factor 2 . 87
12. Primary Activities and Environment Factor 3 . 88
13. Mobility and Level of Information Factor 4 . 89
14. Intervening Opportunities Effect Factor 5. . 91
15. Movements to the City Factor 6 . . . . . 92

vii

CHAPTER I

INTRODUCTION

In recent decades, rural-urban internal migration
has become a phenomenon of great significance in all devel-
0ping regions, including Latin America. However, inter
regional migration between rural areas has common place
prior to the time when rural-to-urban migration became the
major internal migration force. The mobility of population
within its national boundaries has not received as early
attention as international migration, but it has been the
subject of increasing study in recent decades. Of all the
components of population change, internal migration has
been the most difficult to measure and predict.

Internal migrations are geographical pOpulation
movements while remaining within the same national boundaries.
Specifically, they involve a change of community (locality),
of usual residence or, more generally, they are movements
having a relatively permanent nature from one community to

another.1

 

1Juan C. Elizaga, "Internal Migration in Latin
America," International Social Science Journal, Vol. 17,

 

The geographer's interests have included (1) inter-
nal migration as a factor in population change and hence in
measures or estimates of movements; (2) internal migration
as the primary factor in pOpulation redistribution among
regional areas; and (3) differentials in short-distance
mobility and migration and the selectivity of migrants types

of human flows.

Scope and Content of Internal Migration Research

In the growing literature on the theoretical study
of internal migration, the perspective of rural-urban move-
ments changed from a push-and-pull or cause-effect movement,2
to a progressively complex system in which the effect of
changes in one part can be traced through the system.3 This
study investigates the migration histories of migrants in
highland northern Chile over a period of time. Therefore,
the research reveals the totality of places visited by the
migrant population for this given time period. Describing
and understanding the entire migration process through time
goes beyond the traditional generating center/receiving

center approach. To bring order from the large number of

individual migrant moves, the investigation collapses the

 

2R. Herberle, "The Causes of Rural-Urban Migration:
A Survey of German Theories," American Journal of Sociology,
Vol. 43 (1938), 932-950.

 

3Akin L. Mabogunje, "System Approach to a Theory of
Rural-Urban Migration,” Geographical Analysis, Vol. 2, No. l
(1970), 1-16.

numerous migrant movements into major migrant parts, streams,

or flows.

Periodic and Rural-Urban Migration

 

The main objective of periodic migration in which
migrants remain away from their original settlements for
short periods of two or three years and return continuously,
is to obtain better payment than in their permanent homes.
These periodic movements of the head of the household gen-
erally lead to a massive migration of families, once the
first migrants have received sufficient information about
job opportunities.

In Chile, the development of the mining industry in
some agricultural areas has been responsible for radical
changes in population distribution. The mining locations
have become a focus not only for those seeking employment,
but for family dependents and other subsequent migrants who
provide goods and services in the new mining settlements.

As a result of these movements the mining companies gain
access to a reliable, inexpensive source of labor.4

Another form of periodic migration that leads to a
definitive rural-urban movement is the farm-to-farm migra-

tion associated with the systems of share-cropping. Agri-

culturalists leave their original communities to farm.in

 

4Kenneth Swindell, "Sierra Leonean Mining Migrants,
Their Composition and Origins," Institute of British Geo-
graphers-Transactions, V01. 62 (March 1974), pp. 47-64.

 

other settlements for a period of years until the soils
become exhausted. This process of soil depletion generates
a rural-rural migration essentially different from those of
rural-urban migrations. It is known, for example, that in
many parts of the world the pattern of migration to towns
is one of movements from low-order central places to high-
order central places. In the case of rural-rural migrations
the direction of movement differs and is usually a movement
from areas of large population concentrations to sparsely
settled areas. The reason for this appears obvious; while
a large urban center with industries and a wide range of
services and employment Opportunities tends to attract more
and more peOple, the rural-rural migrant is attracted to
areas where farmland is abundant and where there is demand
for labor to develop primary production. Usually such rural
destinations are so sparsely settled that the available
population is inadequate to exploit fully the local re-
sources.5

Migrations from rural to urban areas are generally
the most important forms of internal movements, especially
in countries experiencing technological change. On the
whole, rural-urban migration involves the less fortunate.
Many rural migrants are changing their occupations and this
fundamental mutation in occupational structure is one of the

main results of this change in residence.

 

5Reuben K. Udo, Migrant Tenant Farmers of Nigeria
(Nigeria: African Universities Press, 1975), p. 2.

 

 

In rural areas the effect of migration on the gen-
erating center varies. It can lead to a better distribution
of the land for those who remain, or in some extreme cases,
may cause rural dep0pulation and land abandonment. Between
these two extremes lie various stages, such as the adverse
effects of unbalanced age and sex-structures and the vari-
ation of migration volume with distance and transport
facilities.6 The most seriously affected localities are
generally those with adverse physical conditions such as
mountain villages and specifically in northern Chile, small
towns in desert oases.

MigrationL,Diffusionyyand
Modernization

 

 

A substantial number of migratory farm workers enter
the migrant streams within the rural area each year in order
to find work. Most, but by no means all of them, move
through the central place hierarchy. These rural to urban
movements are facilitated by the diffusion of information
between an individual urban center and the surrounding rural
area.7

In the migrants' predominantly rural environment,

we expect communication to be limited among communities,

while within settlements the flow of information would be

 

6John I. Clarke, Population Geography (Oxford:
Pergamon Press, 1977), pp. 132-134.

7Peter E. Lloyd and Peter Dicken, Location in Space:
A Theoretical Approach to Economic Geography (Wheaton & Co.,
1977), p. 64.

more rapid and efficient. This availability of information
provides the potential migrant with the information on
which they can make a more rational choice of a possible

8 This ease of access to information about a

destination.
migrant receiving center, one's opportunity set, facilitates
the migration process.

Migration is an instrument of adjustment and cultural
diffusion. Specifically, the internal migrant who moves from
one settlement to another brings with him a cultural heritage
generated in his previous socio-economic environment.
Initially, he may act as a disruptive force in the community
he enters. Only by a process of assimilation, which some-
times requires a generation or more, the migrant group and
the receiving community finally adjust to each other. Where
internal migration process brings about a diffusion of
migrants over a wide.geographic area, the resulting cultural
diffusion may contribute to a reduction of intersectional
and interregional differences.9

The process of assimilation that affects rural

migrants is also a change from a traditional way of life

to a more complex and rapidly changing style of life. This

 

8P. Collier and J. M. Green, "Migration from Rural
Areas of Developing Countries: A Socio-Economic Approach,"
Oxfgrd Bulletin of Economics and Statistics, V01. 40, No. l
(1978), p. 26.

9Donald J. Bogue, "Internal Migration," The Study
of Population: An Inventory Appraisal, edited by Philip M.
Hauser and Otis D. Duncan (Chicago, 111.: University of
Chicago Press, 1959), pp. 486-509.

 

availability to initiate change is confined to a restricted
group of migrants who are capable of absorbing the cultural
shock of assimilation experienced in migrating from a more
traditional rural area into a more modern rural area close
to urban centers. Since a social adjustment implies suc-
cessive adoption of new behavioral patterns, the examination
of an economic profile of job types held before migrating
from traditional rural areas has shown a consistent shift
from farm to nonfarm occupations, though many still depend
on seasonal work.10
Some types of migrants become involved in a differ-
ent type of socioeconomic process as a result of their
migration. Involvement in the new setting requires adjust-
ment, participation, and eventual or gradual acculturation.

According to Germani,11

these three processes define the
degree of the migrant's assimilation into the urban environ-

ment.

The Process of Acculturation

Migration accelerates culture change as it trans-
forms the communities of origin through the exodus of the
younger members of the community and, more important,

through the influx of information and customs of the outside

 

loGhulalm Mohammad Sandami, Migration and Moderniza-
tion (Ithaca, N. Y.: Center for Urban Development Research,
Cornell University, 1972), p. 7.

11Gino Germani, "Migration and Acculturation," Hand-
book for Social Research in Urban Areas, edited by Philip M.
HauserTBelgium: UNESCO, 1964), pp. 159-168.

world brought back by the returning migrants. The migration
transition, the changing nature of the most important link
between urban and rural areas, causes and is effected by
transformations in the social structure of the migrant areas
of origin and destination.12
As migrants proceed through the rural-urban hierarchy

13

they undergo the process of acculturation. This process of

acquisition and learning of urban behavior takes place as
the migrant moves upward within the urban hierarchy.14
Generally the movement upward in the urban system brings

about greater economic mobility which can generate greater

social mobility.

Justification of the Study

 

Migration studies dominate population research in
Latin America. The dramatic impact of internal migration
upon the regions' urban areas has provided social scientists
an important area of investigation. Although geographers
have made important contributions to an understanding of

the process, Gonzalez points out that they continue to play

 

12Ronald Skeldon, "The Evolution of Migration
Patterns During Urbanization in Peru," The Geographical
Review, V01. 67, No. 4 (1977), 394-411.

13Dennis Conway, "Step-Wise Migration: Toward a
Clarification of the Mechanism," International Migration
Review, V01. 14, No. l (1969), 3-11.

 

14Arthur Conning, "Rural Community Differentiation
and the Rate of Rural-Urban Migration in Chile," Rural
Sociology, V01. 36, No. 3 (1971), 296-314.

15 According

a minority role in this vital area of research.
to Thomas' bibliography only eight major studies pertaining
to Latin America were included in a total of forty-eight
investigations on migration.16
Since 1971 one of the most important studies on
internal migration in Latin America that reflects a multi-
disciplinary contribution was that by Thomas and Hunter.17
The second and most detailed research on a specific region

18 with special

in Latin America has been done by Bahr,
emphasis on interregional migration. In spite of these con-
tributions, rural-rural migration remains a neglected aspect
of population movements in Latin America. Current research
and government policy are largely concerned with rural-urban
migrations, and have tended to give the impression that

internal migration in the continent is synonymous with

rural-urban migration. A survey of contemporary literature

 

15Alfonso Gonzalez, "Trends in the P0pulation Re-
search of Latin America," Population Dynamics of Latin
America. A Review and Bibliography, edited by Robert N.
Thomas (Boston, Mass.: The Conference of Latin Americanist
Geographers, 1971), pp. 44-62.

 

16Robert N. Thomas, "Population Bibliography of
Latin America," Population Dynamics of Latin America. A
Review and Bibliography (Boston, Mass.: The Conference of
Latin Americanist Geographers, 1971), pp. 122-200.

 

17Robert N. Thomas and John M. Hunter (Eds.), Inter-
nal Migration Systems in the Developing_W0rld (Massachusetts:
Schenkam Publishing Co., 1980), PP. 1-176.

18Jiirgen Bahr, "Migration im Groben Norden Chiles,"
Bonner Geographische Abhandlungen, Heft 50, 1975, pp. 1-286.

10

on migration confirms that even in the more developed
countries of Latin America the research emphasis has focused
on rural-urban movements. As a result, the process of rural-

19 Indeed, to obtain

rural migration is poorly documented.
a true understanding of rural-urban migration it is necessary
to comprehend this segment of the migration process as it
comprises the total migration system of a given country.

It is also an important part of this research problem to
verify the stepwise and/or stage migration process as it
occurs within the context of the total migration system.

This human spatial movement has been described in numerous
ways,20 however, the literature basically ignores the first
stage in the process, that of the role played by the rural
component.

An objective of this research is to combine the
demographic and geographic approaches to migration. While
progress has been made by several disciplines (i.e., geog-
raphy, sociology, anthrOpology, and demography) toward a
general understanding of the social and spatial dimensions
of migration, there is still much to be done to integrate

these two divergent approaches.21

 

19Udo, op. cit., pp. 1-4.

20Conway, op. cit., pp. 3-11.

21Edward A. Ackerman, "Geography and Demography,"
The Study of Population: An Inventory and Appraisal, edited
by Philip M. Hauser and Otis D. Duncan (Chicago, 111.: Uni-
versity of Chicago Press, 1959), pp. 717-727.

11

The methodological purpose of this research is to
analyze the spatial variability in migration to the nearest
city. The advantage of a multivariate approach is that it
extracts from a volume of data some major dimensions or
factors. For the geographer an additional advantage of this
model is the opportunity to map and analyze the factor

22 The

scores for these independently-derived dimensions.
factors extracted from this model were regressed on a

migration variable.

The Study Area

 

The Great North of Chile includes the desert regions
of Tarapaca and Antofagasta. Much of the population of the
Atacama Desert lives in coastal cities and many p0pulation
clusters in the northern interior have grown up around mines.
The other significant far northern pOpulation settlements
are scattered highland oases which support small farming
communities.23

The study area selected for this investigation con-

stitutes two sub-areas from the Regions of Antofagasta and

Tarapaca. The first sub-area, known as the administrative

 

22Stanley D. Brun and Robert N. Thomas, "The Migra-
tion System of TeguCigalpa, Honduras," Population Dynamics
of Latin America. A Review and Bibliography, edited by
Robert N. Thomas (Boston, Mass.: The Conference of Latin
Americanist Geographers, 1971), pp. 63-82.

 

23The concept of "community is accepted as an ecol-
ogical or social organization in which people share a limited
territorial space for residence and for sustenance and
functions to meet common needs.

12

unit of Calama, covers more than 60 percent of a vast high-
land surface and includes the eastern half of the province
of Antofagasta. The second subarea is the southeastern
highland sector of Tarapaca which comprises three adminis-
trative units, Huara, Pozo Almonte, and Pica. Both surfaces
cover more than 65,000 square kilometers, approximately

36.4 percent of the total area of these two regions (Figure
1).

For this study, only villages and mining sites with
more than fifty persons have been considered, with the
exception of those settlements with less than fifty persons
that might be intermediate centers for those migrants that
try to reach larger villages, towns, or cities (Table 1).
Any intermediate center near Tarapaca's border, but in
another administrative unit, may draw people from the south-
eastern part of this region. Since this research deals with
the cited administrative units of the highland areas near
two urban centers namely: Calama and Chuquicamata, it is

assumed that this area is also a geographic unit (Figure 2).

Statement of the Problem

It is the purpose of this study to examine the
spatial pattern of rural mobility among highland rural
settlements in northern Chile in such a way that a stepwise
and/or stage migration thesis will be tested, and to account
for the local differences of out-migration volume with

respect to migration differentials.

l3

 

NORTHERN CHILE
REFERENCE MAP

  

 

 

 

 

 

 

0 CITY

Till M. VILLA“
—-- "mum ”I
M flmclvu. um!”

 

 

 

 

DO
50 IOD Km.

 

IOO Ml.

 

Figure 1

 

14

Table 1.--Classification of Rural Settlements Per Region,

Commune,

and District.

 

 

 

 

 

 

 

 

 

 

 

 

 

Region Commune District Settlement No.
Antofagasta Calama Calama quuina 1
Caspana 2
Cupo 3
Chiu-Chiu 4
Lasana 5
San Pedro 6
Toconce 7
Turi 8
Collaguasi Puquios 9
Ujina 10
Ollague Amincha ll
Buenaventura 12
Conchi 13
Ollague 14
S.P. de Rio Grande 15
Atacama Machuca l6
S.P. de Atacama 17
Toconao Talabre 18
Toconao 19
Socaire Peine 20
Socaire 21
Tarapacé Huara Chusmisa Chusmisa 22
Cariquima Villablanca 23
Tarapaca Huavina 24
Tarapacé 25
P020 Mamina Mamina 26
Almonte
Pica Pica Pica 27
Guatacondo Guatacondo 28

 

Source: Instituto Nacional de Estadisticas, Chile.

15

 

 

HIGHLAND NORTHERN CHILE
COMMUNES AND DISTRICTS

HUARA

RALMONTE { H4

PICA S"~..-""°.L,‘.é

1"”! \I
8 i.
.. 7‘.
.1
9 .I.
....... \

1. Cariquima
2. Chusmizo 11
3. Turopocd
4. Momino

5, Pica 12 '-..,

6. Guatacondo /./
7. Collaguasi r"
8. Ollague i
9. Column
10. San Pedro de Atacama
11.Toconoo
12. Socaire

 

 

 

 

 

 

 

 

 

 

 

 

) CALAMA

 

 

”Figure 2

 

16

Major Purposes of the Study

It is assumed that a migrant will move to a more
accessible rural settlement or small mining location in
successive and short distance moves. These same migrants
will construct a pattern of spatial mobility according to
the number of economic opportunities at those intermediate
villages and mining centers. To prove this assumption, a
first approach examines the step-wise system at an early
stage of the rural migration process before migrants initiate
the more extensive movements toward the cities. A second
approach examines migrant selectivity such as age, sex, and
marital status. Moreover, the investigation reveals the
influence of the distance variable upon the selection pro-
cess.

Hypotheses and Operational
Statements .

 

The research focuses upon two main areas of inquiry
in migration theory, the mobility process, and the charac-
teristics of migrants. The hypotheses and their specific
operational statements follow:

1. Through time the step-wise or stage migration
process breaks down and becomes a more direct
rural to urban movement.

a. The distortion in the distance-decay function
might be attributed to the rapid development
of a country's internal infrastructure.

b. Direct rural to urban migration may be
facilitated by information flows established
by previous migrants residing in the major
urban centers.

17

c. The role of intervening opportunities decreases
through time as a migrant's awareness space
increases.

2. Selectivity of migrants is for the most part affected
by the distance obstacle.

a. Males move longer distances than female migrants.

b. Young married migrants move shorter distances
than older married migrants.

c. Migrants who move to intermediate centers are
found in the 18-29 cohort while migrants found
in small villages near major urban centers tend
to be older (30-35 years cohort).

Limitations of the Study

 

This study will not attempt to examine the specific
causes of migration in highland northern Chile or the impact
of in-migration upon the urban centers. However, it will
examine the spatial movements that rural population experi-
ence. Although causes and consequences clearly exPlain
migration, they are beyond the scope of this research.

In the following analyses, twenty-eight communities
will be considered as settlements affected by rural-rural
migration. Selected communities were examined as centers
of out-migration that provide migrants to the secondary
urban centers Calama-Chuquicamata. .There are other highland
villages in northern Chile that belong to other communes in
which the migration phenomenon may behave in a similar way.
But, these communities will not be included in the study
area since they are considered as settlements of possible
out-migration to other urban centers such as Iquique and

Arica.

18

A third limitation of this research is related to
movement within the rural-urban system. Although a complete
view of rural-urban migration includes the rural environment
as well as the urban destination centers, this study refers

only to the rural area.

CHAPTER II
REVIEW OF THE LITERATURE

Distance and Human Interaction
Perhaps one of the most important contributions in
geography can be summarized by the concept of "spatial inter-

action."24

Most explanations of this concept have contained
elements included within a given spatial distribution of
places. Gunnar Olsson points out that most explanations of
spatial distributions have involved elements included within
the broad framework of spatial interaction,25 for example
the diffusion of people among various points. Olsson also
considers the implication of the concept of functional
distance for the understanding of human behavior in the
geographical site.

In part, movement of people is a result of the

degree of connectivity within a given system. This con-

nection and subsequent action among centers implies specific

 

24Edward L. Ullman, "Human Geography and Area Re-
search," Annals, Association of American Geographers, V01.
43’ 1053’ Pp. 54-66.

25Gunnar Olsson, "Distance and Human Interaction:
A Review and Bibliography," Bibliography, Series No. 2
(Philadelphia: Regional Science Research Institute, 1965).

19

20

complementarity. However, the interchange of migrants
between two centers will not totally develop if intervening
opportunities are created between them.

A final factor found in a system is distance which
generally creates a barrier to human interaction. If the
distance between two settlements were too great and too
costly to overcome, interaction would not take place in
spite of perfect complementarity and absence of intervening

Opportunities.

Distance Factor

 

The question of how far migrants travel has been
the focus of migration studies in recent decades and has
given rise to the formation of many concepts and theories.
These studies confirm the earlier generalization presented
by Ravenstein in which he recognized the relevance of dis-
tance as a factor of migration. He stressed the relation-
ship between distance and the propensity to move by stages
in his first "law of migration" where he states that most
migration occurs over a short distance and that the number
of migrants enumerated in a given center of absortion will
decrease as the distance from the center increases.26

One of the basic works on migration and distance

investigates population movements from one city to another.

George Zipf tried to explain urban-urban migration by the

 

26Edward G. Ravenstein, "The Laws of Migration,"
Journal of the Royal Statistical Society, V01. XLVIII,
Part II (June 1855), 183.

21

principle of least effort. According to him, the number of
migrants from one city to another is a function of the
distance separating the cities, since the effort required
to cover greater distances would increase with the dis-

27 Although it appears that Zipf's work is outside

tance.
the scope of the present study, it seems certain that both
rural-rural migration and rural-urban migration has little
new to offer but merely confirm these conclusions.

There is also a temporal dimension to a migrant's
perception of distance in that today people tend to travel
greater distances than in the recent past. This change has
come about as a result of several factors such as improved
roads and networks of transport, and the greater amount of
information which potential migrants have about various
destinations. Distance perception by migrants is one of
the issues which has been discussed in recent migration

studies. Olsson,28 and Claeson,"9 among others, have
pointed out that purely physical distance does not adequately

explain interaction intensity between places and should

 

27George Zipf, Human Behavior and the Principle of
Least Effort (Cambridge, Mass.: Addison-wesley Press, Inc.,
1949).

 

28Gunnar Olsson, "Distance and Human Interaction:
A Migratory Study," Gepgrafiska Annaler, Vol. 47, Series B,
No. l (1965), 8-9.

29C. F. Claeson, "Distance and Human Interaction:
Review and Discussion of a Series of Essays on Geographic
Model Building," Geografiska Annaler, Series B, No. 2
(1968), 142-161.

22

therefore be replaced by a new concept of "functional" or
"real distance." Claeson suggests that distance tolerance
is a more realistic concept than distance perception, but
the problem of quantifying these concepts is yet to be
resolved. Olsson refers to this discussion as the informa-
tion concept of distance and is largely a product of the
feed-back effect from earlier migrants.

Migration is selective of individuals with given
characteristics. The intensity and pattern of selectivity
depend, in some way, on the distance factor. Everett Lee
wrote:

It is also commonly noted that as distance of
migration increases, the migrants become an increasingly
superior group. At the other extreme, we have the
milling-around in restricted areas of persons who, by
any definition, are less capable; for example, unedu-
cates slum dwellers often move round and round within
a few block radius.30

It is proven-fact that the friction of distance
acts on human movements and that the frequency of these
moves decreases with the increasing distance. This distance-
movement relationship, known as "distance decay" or inverse
distance relationship has been well documented since

Ravenstein emitted his "laws of migration."31

 

30Everett Lee, "A Theory of Migration," Demography,

Vol. 3, No. 1 (1966), S7.

 

31Ravenstein, op. cit., pp. 167-235.

23

InterveningyOpportunities

Forty years ago Stouffer established the fact that
the number Of migrants at a given destination would be
affected by the number Of opportunities available between
their places of origin and destinations. The theory assumed
that there is little relationship between mobility and
distance. Instead, he introduced the concept Of intervening
Opportunities which states that the number Of migrants going
a given distance is directly prOportional to the number Of
Opportunities at that place and inversely proportional to

32

the number Of intervening Opportunities. Stouffer argued

that the volume Of migration from one place to another is
associated not only with the number Of people in the two

33 but with the number of Oppor-

places, as Zipf assumed,
tunities in each place.
By stressing Opportunities, Stouffer Opened to
migration analysis a wide array of sociological character-
istics that make a city differentially attractive in addition

34

to its economic characteristics. Although Stouffer did

 

32Samuel A. Stouffer, "Intervening Opportunities:
A Theory Relating Mobility and Distance," American Sociol-
pgical Review, V01. 5 (1940), 845-867.

 

33George K. Zipf, "The P1P2/D Hypothesis: On the
Intensity Movement Of Persons," American Sociological Review,
Vol. 11 (1946), 677-687.

34James J. Zuiches, "Migration Methods and Models: A
Demographic Perspective," Internal Migration Systems in the
Developing W0rld, edited by Robert N. Thomas and John M.
Hunter (Cambridge, Mass.: 1980), pp. 3-16.

 

24

not attempt to define "Opportunities," and the term has been
defined differently among social scientists, it seems clear
that for most purposes (particularly where interest is in
migration Of working-age people) some variable representing
employment Opportunities would be appropriate.35

Under certain conditions, intervening Opportunities
might help to create interaction between two distant comple-
mentary areas. They might act as a bridge for a distant
migrant located far from urban centers. A town or settle-
ment closer to a major city may be either an intervening
Opportunity site or a stOp over to a migrant.

Spatial Expression Of Sociocultural
Transition

 

If we examine contemporary migration literature, it
becomes clear that the stepwise migration process has been
broadened to include the spatial manifestation Of a social

process Of adjustment.36

With rural-urban migration, there
is considerable evidence to support the thesis that move-
ments are usually from small villages to medium size towns,
and from the latter to larger cities. Almost all the avail-

able evidence Of the step-by-step process in rural-urban

migration has included the cultural adaptation Of migrants

 

35Roger L. Burford, "An Index Of Distance as Related
to Internal Migration," Southern Economic Journal, Vol. 29
(1962), 77-81.

 

36Dennis Conway, "Step-Wise Migration: Toward a
Clarification of the Mechanism," International Migration
Review, Vol. 14, NO. l (1979), 3-14.

25

and the consequent transformation of their way Of life.

The acceptance Of the stepwise process is well understood.
This change Of residence views stepwise migration as the
spatial expression Of a sociocultural transition Of an indi-
vidual moving progressively from a traditional rural area

to the modern urban centers.37

Step-Wise Migration

 

It has been established in the general literature
on migration that rural-urban movements Often take place by
steps. Within the step process two particular spatial types
Of movement must be defined: "stepwise" and "stage" migra-
tions. Thomas and Mulvihill have defined stepwise migration
as movement by an individual from a rural area to a major
city through a central-place hierarchy. On the other hand,
they distinguished stage migration as the spatial movement
that emphasizes replacement. Thus, an individual moves from
a secondary center to a major city and is replaced in the
secondary center by someone from a smaller city, and so on
down the hierarchy.38

The first theoretical description Of the step process

was made by Ravenstein who wrote:

 

37Conway, Op. cit., p. 6.

38Robert N. Thomas and James L. Mulvihill, "Temporal
Attributes Of Stage Migration in Guatemala," Internal Migra-
tion Systems inlthe Developing World, edited by Robert N.
Thomas and John M. Hunter (Cambridge, Mass.: 1980), pp.
51-61.

 

26

The inhabitants Of the country immediately sur-
rounding a town Of rapid growth, flock into it; the
gaps thus left in the rural population are filled up
by migrants from.more remote districts, until the
attractive force Of one of our rapidly growing cities
makes its influence felt step by step, to the most
remote corner Of the kingdom.3
According to this thesis, the instigator Of the step-by-step
process is the city but, in the wake Of the consequent pull
Of the city, rural-rural migrations follow. Here, it is
necessary to consider the rural-urban hierarchy in the step
migration process. Conway's elaboration Of the notion Of
a series Of spatial steps has been fused with the idea Of a
stepwise progression upward in the urban hierarchy.40
After the first discussion by Ravenstein concerning
the step process through the rural-urban hierarchy, a few
attempts have been made to verify it. Two Of the most
recent studies in developing countries have been done by

4L and Thomas and Catau.42

Riddell and Harvey, Both results
concluded that the rural migrant living near a primary city
by-passes the local secondary center and moves directly to

the primary city, while a rural migrant living a great

 

39Ravenstein, Op. cit., p. 199.

40Conway, Op. cit., pp. 3-14.

41J. Barry Riddell and Milton E. Harvey, "The Urban
System in the Migration Process: An Evaluation Of Step-Wise
Mibgration in Sierra Leone," Economic Geography, Vol. 48,
NO. 3 (1972), 270-283.

 

42Robert N. Thomas and John C. Catau, "Distance and
the Incidence Of Step-Wise Migration in Guatemala," Pro-
ceedings Of the Association Of American Geographers, V01. 6
(1974), pp. 113-116.

27

distance from the primary city is more likely to move to
the local secondary center. As a consequence, the effect
Of distance demonstrates an evident positive association
with stage-migration process. However, as the relative
accessibility to the large urban center improves, the stage
migration process progressively decreases in importance as
people move directly to the major urban center from all
parts Of the study area.

Although the scales Of the former studies are quite
different from the study area in northern Chile, the effect
Of distance may reflect similar patterns if the stage migra-
tion process is tested. A combination Of village case
studies and census data analysis would seem to provide the
best possibility to better understand the significance Of
short- versus long-distance movements of rural-urban

migrants.43

Long-Distance M0vement

 

It was also Ravenstein who stressed the condition
in which long-distance movements take place.' He indicated
that migrants who proceed long-distances generally move to
one Of the great centers Of commerce and industry.44 From

a basic short-distance move, migration evolves through

 

43Robert V. Kemper, "Rural-Urban Migration in Latin
America: A Framework for the Comparative Analysis of Geograph-
ical and Temporal Patterns," International Migration Review,
Vol. 5, NO. 1 (1971), 36-47.

 

44Ravenstein, Op. cit., p. 199.

28

complexity tO a long-distance movement. As a complement to
the former statement, Skeldon added that the ideal end point
of the process Of urbanization in a region occurs when
migrants from rural settlements at all levels Of the hier-
archy are engaged in permanent direct long-distance migra-
tion.45
Before long-distance movement to the primary city
occurs a period Of migration adjustment is required. During
this period, when the migration field Of each rural settle-
ment is at its greatest areal extent, both periodic and
permanent migration streams prevail.46
Long-distance migration does not diffuse uniformity
throughout the whole system. Furthermore, its influence can
involve an extensive area within the rural-urban hierarchy.
Indeed, the real contributor to the decline Of rural isola-
tion has not been a gradual spreading out Of local contacts
but a rapid expansion Of long-distance links in which
distance itself becomes less relevant. In relation to this
phenomenon, Morel has shown that beyond the local region
knowledge is generally restricted to towns and cities whose
real distance from home is frequently distorted by incomplete

information and differently perceived attractiveness.47

 

45Ronald Skeldon, "Evolution Of Migration Patterns
During Urbanization in Peru," The Geographical Review, Vol.
67, No. 4 (1977), 394-411.

46Ibid., p. 407.

47A. Morel, "L'espace social d'un village picard,"
Etudes Rurales, Vol. 45 (1972), pp. 62-80.

29

The drastic reduction Of the rural population is
due tO the existence of a greater variety Of occupations
found in distant urban centers. By analyzing the occupa-
tional structure Of migrants Odgen suggests that the
general effect Of this factor is to reinforce spatial
patterns Of long-distance movements.48

The relationship between formal education and long-
distance migration is rather complex. When education has
been introduced and accepted for a long period Of time
within the rural environment, students who receive an ele-
mentary education are forced to leave their homes if they
wish to continue their education. This necessitates greater
contact with the distant cities and acts tO increase popula-
tion mObility. Even when it is to be expected that the
importance Of long-distance movements would Often be exag-

d,49

gerate the effect Of these flows will be felt not only

locally but also in the entire rural-urban system.

Replacement Movements and Selectivity
Migration is essentially selective. Despite some
exceptions, for example forced migrations or movements to

colonization projects, the vast majority Of migrations

 

48F. E. Ogden, "Migration, Marriage and the Collapse
of Traditional Peasant Society in France," The Geographical
Impact of Migration, edited by Paul White and Robert WOOds
(London: Longman Group Limited, 1980), pp. 154-179.

 

49Dorothy S. Thomas, "Research Memorandum on Migra-
tion Differentials," Bulletin 43 (New York: Social Science
Research Council, 1938).

30

contain an element Of migrant selectivity or differentiation.
In this context migrants are not randomly selected from the
population of the place Of origin. Specific groups Of
people are likely tO react in distinct ways to the differ-

50 they are likely to value different

entiation Of places;
attributes Of places, to have different information avail-
able tO them, and therefore to react in different manners.

It has been suggested that movement between stages
in the life-cycle Of a migrant is an important cause Of
Spatial movement, and it can be argued that within each
stage there may be less migration. The initial migrant's
aspirations should be completed once he has reached the last
stages Of his life-cycle. In general however, it may be
said that selectivity occurs in out-migration from one
place to another because there generally are distinct differ-
ences between the interests Of the individuals comprising
various groups within the population. Consequently, such
attitudinal differences are manifested in behavioral differ-
ences with respect to staying in or leaving the community.51

Migration is Often selective among potential migrants

due to a wide range Of demographic attributes Of the

 

50Differentiation Of places is related here tO the
concept Of "areal differentiation" introduced by Richard
Hartshorne in his book "On the Nature Of Geography," published
by The Association of American Geographers, V01. XXIX, 1939.

 

51Paul White and Robert WOOds, The Geographical
Impact of Migration (London: Longman Groups Limited, 1980),
pp. 1-245.

 

 

31

population, but those attributes which appear to engender
over-representation Of certain population groups differ
from place to place and from migration flow to migration
flow.

Migration trends consistently reflect the social
and economic changes taking place within the system. Since
these changes vary from place to place and from time to time,
socio-demographic characteristics of the migrants can hardly
be expected to remain static.

Intergenerational Movements as
Chain Migration

 

 

It has been suggested that stage migration may take
place across generations with each successive generation
moving to a more urbanized place. However, whether this
kind Of migration is intra-generational and/or intergenera-
tional may not, at first sight, appear to be a crucial

distinction.52

Wendel's migration study Of a large sample
Of individuals in Sweden, was one Of the first to define

the intergenerational movements Of migrants as chain migra-
tion. However, Wendel did not point to the conclusion that
migration by stages in terms Of intergenerational flows was

common . 5 3

 

52Conway, Op. cit., p. 6.

53Bertil Wendel, "A Migration Schema: Theories and
Observations," Lund Studies in Geography, Series B, Human
Geography, NO. 9, 1953, pp. 1-38.

 

32

By analyzing intergenerational stage movements,
Hagerstrand did infer that there is a complex intergenera-
tional flow through the settlement hierarchy that he defined
as chain migration. From his survey Of a sample of Swedish
migration fields, he concluded that:

The migration pattern Observed is not necessarily

to be regarded as a kind of "migration by stages" in
the meaning that the rural inhabitants are moving indi-
vidually toward the bigger towns gradually via smaller
towns. Just as important may be the alternative, which
we could term "chain migration," meaning that young
people, born and bred in small urban centres, are
turning to bigger centres, to be compensated--sometimes
insufficiently, sometimes more than sufficiently--by
in-migration from still smaller centres and from rural
districts.54
Hagerstrand was convinced that the majority Of movements
are replacement movements. In other words, they arise to
fill vacancies, which in many instances are due to deaths
or superannuation and dependent on the demands Of training
and age imposed upon those who will fill the vacancy.55

The concept Of "chain migration" cited by Hagerstrand
is Of considerable importance to this research in that inter-
generational movements occur through the settlement hierarchy.
Nevertheless, the duplicity in the literature when using

"chain migration" tO refer to the migrational process Of

 

54Torsten Hagerstrand, "Migration and Area. Survey
Of a Sample Of Swedish Fields and Hypothetical Considera-
tions On Their Genesis," Migration in Sweden: A Symposium.
Lund StudiesipGeography, Series B, Human Geography, NO. 13,
1953, PP. 27-158.

55Torsten Hagerstrand, "On the Definition Of Migra-
tion," Population Research Institute, V01. XI (1969), p. 65.

33

families and relatives is also useful. In the context Of
this study the spatial mobility Of highlanders in northern
Chile might be conceived as MacDonald and MacDonald indicated
as the movements Of families from their place Of birth to
intermediate towns or final destination where immigrants
instigate a "chain migration" by helping their relatives

to move . 56

Differential Characteristics
Of Streams

 

Several studies have suggested the importance Of
migrant selectivity on the basis of age and sex as well as
other socio-economic characteristics that migrants hold at
various stages in their life-cycle. One Of the most common
conclusions to emerge from studies of migrant selectivity
is that age is Of particular importance in explaining the
likelihood Of migration. There is a close correlation, but
not a direct one, between migration and age; it is suggested
that the direct association is between migration and life-
cycle stages. Each stage in one's life-cycle is associated
with a particular age group and is the link tO his or her

57

age. Speare compared the annual probability Of migrants

by their ages and life-cycle and concluded that both age

 

56J. MacDonald and L. D. MacDonald, "Chain Migration,
Ethnic Neighborhood Formation and Social Networks," Social

Research, Vol. 29 (1962), pp. 433-448.

57White and WOOds, op. cir., p. 14.

34

and life-cycle have some independent importance so that both
should be considered as significant factors Of migrant
selectivity.58

When investigating migration streams we might dis-
cover that the predisposition Of young people to migrate
could be consistently higher than other age groups when the
area Of origin is rural. In rural areas the predisposition
of this age group to migrate is accentuated by lack Of

59 Bogue has found the same

occupational Opportunities.
generalization valid in many places and for a long period
Of time: persons in their late teens, twenties, and even
early thirties are much more mobil than younger or Older
persons. Thus, migration is closely associated with the
first commitments and acts Of adjustment to adulthood that
are made by adolescents as they mature.60
It is a proven fact that sex distribution is as a
rule, unequal, favoring either men or women according to

whether flows come from a "short" or "long" distance.61

 

58Alden Speare, "Home Ownership, Life Cycle Stage,
and Residential Mobility," Demography, V01. 7 (1970), pp.
449-450.

59James M. Beshers and Eleanor N. Nishiura, "A
Theory Of Internal Migration Differentials," Social Forces,
V01. 39, NO. 3 (1961), pp. 214-218.

60Donald J Bogue, "Internal Migration," The Study
Of Population: An Inventory Appraisal, edited by Philip M.
Hauser and Otis D. Duncan (Chicago, 111.: University of
Chicago Press, 1959), pp. 486-509.

 

61Juan C. Elizaga, "Internal Migration: An Overview,"
International Migration Review, V01. 6, NO. 2 (1972), pp.
121-126.

35

The predominance Of females among short-journies was initially

62 Although he did conclude that men

stressed by Ravenstein.
were more migratory over long-distances, this conclusion has
been reviewed by Grigg who discovered a drastic reduction Of

63 Some scholars have further

sex differences in migration.
refined the concept and to state that females predominate

in well-established, secure migratory movements while, males
prevail in those streams directed toward pioneering or

frontier regions.64

We have to accept that sex differences
in migration is evidently complex and related to differ-
ential employment Opportunities and a number of other con-
ditions as well.

The impact that migrants have on their final desti-
nation and area Of origin is associated with their education
and occupational skills. While education may be readily
measured in terms Of years of schooling, occupational skills
are more difficult to determine. Simmons and Cardona

classified urban workers to determine whether they differed

from non-migrants in social standing or their own

 

62Ravenstein, Op. cit., p. 288.

63D. B. Grigg, "E. G. Ravenstein and the 'Laws Of
Migration,'" Journal Of Historical Geography, Vol. 3 (1977),
47-54 0

64Donald J. Bogue, Principlgsof Demography (New
York: John Wiley & Sons, 1969), p. 765.

 

36

65 However, it is also important to

occupational skills.
determine the occupational skills a migrant had before he
left his original site. There, a complete classification

Of occupational skills is not as important as the number Of
hours a rural person utilizes tO perform a job.

Much less information is available on migration
differentials by marital status than by sex and age, and
most discussions Of the former are less systematic in their
coverage. A considerable difficulty in this connection is
the problem Of inferring marital status at the time Of
migration. One Of the most recent studies along these lines
was conducted by Gupta who measured the marriage-distance

relationship.66

Marital migration normally is a short-

distance movement and involves the flow Of mostly women.

On the contrary, long distance moves associated with marriage

are restricted by many factors, such as cultural barriers

and cost Of movements.67
In general, migrants leaving home for the first time

do not normally take their wives with them. Rather they go

alone to investigate the new location and to arrange

 

65Alan B. Simmons and Ramiro Cardona 6., "Rural-
Urban Migration: Who Comes, Who Stays, Who Returns? The
Case Of Bogota, Colombia, 1929-1968," International Migra-
tion Review, V01. 6, NO. 2 (1972), 166-181.

 

66H. S. Gupta, "Marital Migration in the Rural Chha-
ttisgarh," The Geographical Review, V01. 39 (1977), 126-133.

67Ihid., p. 126.

37

accommodations before sending for their families. However

Udo's study demonstrated that the use Of family labor on the
farms is an important reason why rural-rural migrants prefer
to take along their families since those children who go to

school can provide farm labor when needed.68

 

68Reuben K. Udo, Migrant Tenant Farmers Of Nigeria
(Nigeria: African Universities Press, 1975), p. 41.

CHAPTER III

MIGRATION METHODOLOGY

Although migration philosophies can usually be
inferred from the works of migration researchers, unfor-
tunately very few researchers provide us with their philo-
sophical perspective from which we may better understand and

69 As Harvey pointed out70

judge their research methods.
it is also assumed that a philOSOphical approach is the
basic condition tO orient a particular methodology in order
to build an organized body Of thinking.

If the fundamental question in this research is how
do migrants decide where to move, perhaps one Of the ways
to determine the philOSOphy of migration analysis will be
to Observe both the needs' structure of the migrants and the

scale Of the study. On the one hand, highlanders in northern

Chile are conditioned by an environment that makes a

 

69Stephen E. White, "A PhilOSOphical Dichotomy in
Migration Research," The Professional Geographer, Vol. 32,
NO. 1 (1980), pp. 6-7.

70David W. Harvey, Explanation in Geography (London:
Edward Arnold, 1969).

38

39

permanent life based upon a symbiotic farming-herding system

71 On the other hand,

and an unstable mining exploitation.
there is also a general decision to choose a place Of
destination given the increasing information within the
rural area. Then, it may be acceptable to develop a pro-
cedure in which the Objective/cognitive philOSOphical

synthesis appears to be the cornerstone.72

Demographic Change and Indirect Migration

It is frequently desirable to describe the demo-
graphic change that occurs during more than one time period.
This approach, utilizing the time perspective enables the
researchers to see more clearly what has happened within a
given area.

An indirect estimate of net migration may be computed
from population figures by using previous censuses to esta-
blish an average Of annual rate Of increase. Employing this
rate over the desired number Of years provides an expected
population. The difference between this expected population
and the actual census count is Often attributable to migra-
tion alone. It should be pointed out that many rural peOple

in Chile have escaped census registration. Therefore, this

 

71Khairul Bashar Sajjadur Rasheed, "Man and the
Desert in Northern Chile" (unpublished Ph.D. dissertation,
Columbia University, 1970).

72For further explanation Of these two philosophical
approaches and the ways that each position influences the
selection of the independents variables, see Stephen E.
White, Op. cit., pp. 6-13.

40

population group to some degree, is always under-registered.
we should keep this discrepancy in mind when estimating
populations and drawing inferences from these figures. If
vital statistics are available for a given area and the
population is known, and indirect measurement of net migra-
tion is relatively straightforward. The degree of inaccuracy
is merely the result Of imprecision due to the fact that,

for a large number of subjects, it is impossible to deter-
mine the exact places Of birth and death.

Population Change and Annual
Increase Rate

 

 

The first analysis undertaken by this research
describes the population change by district. Employing
three census figures 1952, 1960, and 1970, the average
annual increase rates were computed using the following

formula:

P1 - Po

" jaE(1>l + Po)

 

1
r ° 3 k

where: P

o the initial population within the time period,

the population at the end Of the time period,
the time period in years, and
a constant, to convert to percentages.

P1
11
k

Two computations were made on the basis Of the
figures for the three years. With unequal time periods as

it was in our case, the average Of the populations at the

41

beginning and the end of the period (8(P1+P°)) were calcu-

lated thereby eliminated a possible bias.73

In computing the average change within a time period,
an assumption was made as to the growth over time. Highland
rural population experiences a linear growth, or follows an
arithmetic progression, i.e., there is a constant increase
per unit Of time. It is difficult, however, tO hypothesize
a set of demographic conditions under which population would
increase or decrease by arithmetic progression. Nonetheless,
a straight line frequently has been used not only to describe

population but also to project population.

Average Annual Migration Rate

The averages Of annual migration rates were computed
in order tO determine whether any appreciable net migration
could be detected in each rural settlement, and if so,
whether it might be in-migration or net out-migration. The
average Of annual migration rate, m, for the time period

1970-1978 was computed according tO the following formula:74

 

73A general formula to compute the annual increase
rate has been frequently applied by using the population at
the beginning Of the intercensal period. These rates are
comparable, however, only when the periods are Of equal
length. For further elaboration on the meaning and the use
Of this formula, see, Henry S. Shryock and Jacob S. Siegel,
The Methods and Materials Of Demography, V01. 2 (U.S. Bureau
Of the Census, 1975), p. 387.

74This same formula has been applied by Ball in his
study about migration in Mexico. John M. Ball, Migration
and the Rural Municipio in Mexico (Atlanta: Bureau Of
Business and Economic Research, Georgia State University,
1971).

42

(P + N.I. - P

70-78) se78

Ps70
n

370

where: = the population Of a settlement in 1970,

P
370
N.I. = the total natural increase Of a settle-
70-78 ment
I
Pse78 = the population estimate Of a settlement,
n = the number of years recorded, and
k = a constant, to convert to percentages.

Within each Of the twelve districts considered in
this research a number Of settlements were chosen. Their
average annual migration rates were compared with those
annual increase rates already computed for each settlement
(Table 1, Figure 3). The Objective was to find the best
correlations between the annual increase rates and the
average annual migration rates. In a somewhat similar
research conducted in rural Mexico, Stoltman and Ball found
a greater correlation between migration rate and population
change than between natural increase and population change.75

Using the average annual migration rate as a depen-
dent variable in a multiple regression analysis study, I
attempted to ascertain the nature Of the migration and
grouped the subjects involved according to differential

characteristics. On the basis Of these data I arrived at

the most probable explanation Of the phenomenon Observed.

 

75Joseph P. Stoltman and John M. Ball, "Migration
and the Local Economic Factor in Rural Mexico," Human Organi-
zation, Vol. 30, NO. 1 (1971), pp. 47-56.

43

 

 

omqmmbwmw

h:hahahah:h>h:h9h>hahahahlhoh:h-hahahd
G)\JO\Uiébu1h)h‘C>K>G)\IO\UIhnu)kJF'O

HIGHLAND NORTHERN CHILE

SETTLEMENTS

quuina
Caspana
(hub
Chiu-Chiu
Lasana

San Pedro
Toconce
Turi
Puquios
Ujina
Amincha
Buenaventura
Conchi
Ollague

Rio Grande
Machuca
S.P. Atacama
Talabre
Toconao
Peine
Socaire
Chusmiza
Villablanca
Huavina
Tarapaca
Mamina

Pica
Guatacondo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3

 

44

Data Problem

 

The basic land units used by the Chilean Institute
of Statistics changed among the census periods using in
this study. For this reason the original classification Of
entities or localities has been characterized by misspecifi-
cations and erroneous categorizations. Therefore, in order
tO Obtain uniformity among the census categories, some
villages were reclassified according tO size and region.
This assisted in updating census data for the intercensus
periods.

A possible source Of inaccuracy arises from a dis-
crepancy between the time periods covered to register vital
statistics and the time Of their occurrence in each village.
The author updated vital statistic data from direct field

Observation.

Structural Migration
In a rural-urban migration system in which the
analysis is focused in one subsystem or dimension, the rural
component, the study Of those migrants who successfully make
a movement must be approached from a perspective Of struc-

76

tural growth. The lack Of a second dimension, the urban

 

76Here, I refer to the growth process as Boulding
has pointed out. The growth process of an aggregate with a
complex structure Of interrelated parts. Problems Of struc-
tural growth seem to merge almost imperceptibly into the
problems Of structural change or development, so that fre-
quently "what grows" is not the over-all size Of the struc-
ture but the complexity or systematic nature of its parts.

For further details about the principles that

control this kind Of growth see: K. E. Boulding, "Toward

45

system, has to be compensated by a thorough analysis of
movements and migrant characteristics as well as direction
and spatial extension Of flows. The meaning Of this central
analysis is given by the question: What spatial changes does

a migrant undergo in the migration process?

Volume and Spatial Effects

The study Of direct migration within the rural area
has been resolved through the structural analysis Of flows.
Two matrices of place-Of-birth and place-Of-residence Of
migrants that successfully made one or more moves were con-
structed for 1970 and 1978. A complementary matrix measuring
the average number of steps required in moving from place-
Of-birth to place-of—residence was built to verify the step-
wise process.

This structural analysis not only covers volume,
direction of movements, and step process but also the rele-
vant characteristics Of migrants. Maps Of flows and age-
sex differentials were plotted to visualize both structure

of flows and intrinsic characteristics Of migrants.

Differential Migration

 

Since marriage is one Of the most important variables

affecting migration, marital status was included in the

 

a General Theory Of Growth," General Systems, Yearbook of
the Societyjfor the Advancement of General Systems Theory,
Vol. 1 (1956), pp. 66-75.

46

77 TO measure the relationship between it and

analysis.
distance from the urban center, migration of young peOple
who recently married was compared with the migration his-
tories Of Older married couples. The location of these two
migrant groups was mapped and analyzed. For the purpose Of

this study the variable was divided into two time periods,

1960 and 1970.

Data Problem

 

The migration data were extracted from question-
naires administered for two time periods: 1970 and 1978.
The Objectives Of the survey were twofold: (a) to gather
factual data among a rural population in each settlement
not available from census records, and (b) tO update the
data from the 1970 census (Appendix A).

A classic example Of inaccuracy appeared when
questionnaires demonstrated typical interviewer biases
(behavioral reactions and level of education). In order to
preclude this possible source of bias, additional interviews
were arranged which used "leave it to the respondents"

78

approach. This method separates the question into a set

Of separate components which can be answered free from the

 

77H. S. Gupta, "Marital Migration in the Rural
Chhattisgarh," The Geographical Review, V01. 39, NO. 2
(1977), PP. 126-133.

 

78Russell L. Ackoff, Scientific Method: Optimizing
Applied Research Decision (New York: John Wiley & Sons,
1962), p. 209.

 

47

Observer's prejudices. This approach (informal verbal

testimony) complemented the previously cited direct approach.

Spatial Mobility

In order to study a migration process affected by
physical constraints, economic conditions, and social factors
one should employ physical, economic, and social variables.
Once the settlements affected by out-migration were detected,
the variety Of flows, direction of movements and selectivity
Of participating migrants could be Observed. This approach,
structural analysis, gave us the spatial dimension Of
migration based on population characteristics. At this
point, further analysis was conducted by using independent
variables representing the fundamental philosophical
approach: the Objective-cognitive dichotomy. The final
question can be summarized as follows: What are the variable
associations related to the spatial mobility Observed?

Factor Analysis and
Multiple Regression

 

The use Of regression techniques within the frame-
work Of migration analysis has become increasingly important.
They provide a method to test variables generated

79

from the literature. However, when migration is evaluated

principally on the basis Of average figures between time

 

79Kenneth G. Willis, "Regression Models Of Migra-
tion," Geografiska Annaler, Vol. 57, Series B (1975),
pp. 45-54.

 

48

periods, the use Of these techniques to predict migration
is difficult.

This research has utilized average data classified
into twenty-three variables. In order to lessen the com-
plexity, among these, factor analysis was employed to yield
a set of orthogonal factors, new independent variables.80
At this point, a step-wise regression was conducted in order
to Obtain significant factors. In the statistical process,
it proved expedient to reinterpret the results found among

the six factors by using a second step-wise regression

analysis uSing twenty-one Of the original variables.

 

80Factor analysis is concerned with defining the
patterns Of common variation among a set of variables in
which the unique variation is ignored.

CHAPTER IV
ANALYSIS OF MIGRATION IN HIGHLAND COMMUNITIES

Population Change and Indirect Migration

Frequently, it is desirable to describe the pOpula-
tion change occurring within a region over more than one
time period. The absolute change is Obtained by subtracting
the population at the earlier date from that at the later
date. The percent Of change is Obtained by dividing the
absolute change by the population at the earlier date.
When the average amount Of change is required, a simple
division of the total change by the number of years should
be sufficient; however, an initial approximation to popula-
tion data in Chile indicated that the time periods were not
Of equal duration. With unequal periods and decreasing
population the longer period tends to have the highest
approximate rate. The elimination Of this bias yielded an
accurate comparison between the two time periods.

There was a clear tendency for population change in
highland northern Chile tO decrease between the first inter-
census period 1952-1960, and the second intercensus period

1960-1970. The average annual population change between

49

50

1952 and 1960 was positive in those districts most remote
from Calama and Chuquicamata such as Tarapaca, Mamina, Pica,
and Collaguasi; while those districts in near proximity to
the urban centers experienced a negative pOpulation change
(Figure 4). There was one exception to this general pattern,
that Of Socaire district, which showed an annual negative
change Of -4.12 percent. In terms Of absolute values,
population change in Socaire was greater than in the Calama
district, San Pedro de Atacama district, and Toconao district,
areas located closer to the urban centers than Socaire.
While the average annual population change between
the intercensus period 1952-1960 indicated some positive
values, the figure for 1960-1970 show a clear decrease.
The twelve districts Of this highland region also have
experienced a downward trend that ranged between -0.59 and
-7.58 percent (Table 2). If we Observe the location Of
these districts according to the average annual pOpulation
change values, it is possible to note that absolute higher
negative values correspond tO districts located closer to
the urban centers than those with lower negative values
(Figure 5). Here again there is an exception. Between the
time period 1952-1960 and 1960 and 1970, the district Of
Collaguasi changed from one Of the highest absolute positive
population growth rates tO one Of the highest absolute

negative growth rates. These demographic fluctuations are

51

 

 

AVERAGE AN NUAL

POPULATION CHANGE

1952-1960

e
4.)
Yo
‘0
7

2.99 OOO

 

- 0.01 - 3.00
-301 -6.00

 

 

 

 

 

 

 

 

 

 

 
 
 

 

 

 

 

 

 

 

 

 

L—hgi

um.

 

 

Figure 4

52

Table 2.--Average Annual Population Change 1952-1960,

 

 

 

 

 

1960-1970.
Population POpulation Change
Districts
1952 1960 1970 1952-1960 1960-1970
Chusmiza 376 308 170 -2.48 -5.77
Cariquima 528 555 543 +0.62 -0.22
Tarapaca 858 1161 744 +3.75 -4.37
Mamina 732 558 474 -3.37 -l.62
Pica 1213 1947 1302 +5.80 -3.97
Guatacondo 211 I 203 170 -0.48 -l.76
Ca1ama* 2052 1784 1251 -1.74 -3.51
Collaguasi 95 140 63 +4.78 -7.58
Ollague 1250 1293 1055 +0.42 -2.02
S.P. Atacama 1208 1131 755 -0.82 -3.98
Toconao 738 693 533 -0.78 -2.61
Socaire 798 572 539 -4.12 -0.59
*The district Of Calama includes both Calama and
Chuquicamata. Since this study considers rural population,

the figures for both cities were not added.

Source: Instituto Nacional de Estadfsticas, Chile.

53

 

 

 

AVERAGE ANNUAL
POPULATION CHANGE

1960-1970

 

- 4.00 - 599
- 6.00 - 799

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 5

54

primarily a result of market changes in the sulphur mining
Operations.81
The average annual population change estimates for
each settlement demonstrates a realistic and precise defini-
tion Of intercensus variations (Table 3). Between 1952 and
1960 two groups Of settlements share the highest positive
values of annual change. The first group is comprised Of
Chusmiza, Tarapaca, and Pica, with 1.17, 4.68, and 6.62
annual percent change respectively. The second group com-
posed Of Puquios, Conchi, Lasana, and Chiu-Chiu, indicate
average annual population fluctuated between 13.99 and 2.91
percent. The main difference between these two groups is
based upon distance among the cities and the influence Of
one or two Of the towns within the group. The group of
towns of the Tarapaca region is more removed from Calama-
Chuquicamata than the group Of settlements associated with
the Antofagasta region. Within the cluster of towns in the
first group, Pica has the largest population and is the
major intervening Opportunity. On the contrary, the second
group is divided into two subgroups. The first, which is
closer to the cities, maintains values which suggest a
positive influence Of in-migration. The second subgroup
represents mining Operations and services. Their high
average annual population changes confirm the existence Of

a migratory influence.

 

81William E. Rudolph, "Sulphur in Chile," The Geo-
graphical Review, V01. 42 (1952), 562-590.

55

Table 3.--Average Annual POpulation Change 1952-1960,

 

 

1960-1970.
Entities 1952-1960 1960-1970
1 quuina +0.24 -6.08
2 Caspana -0.05 +2.00
3 Cupo -4.41 -7.54
4 Chiu-Chiu +2.91 -0.33
5 Lasana +5.56 +0.12
6 San Pedro +0.67 +4.59
7 Toconce +1.39 -0.80
8 Turi -2.01 +0.84
9 Puquios +13.99 -7.85
10 Ujina -l.78 +1.42
ll Amincha +9.08 +3.45
12 Buenaventura -0.70 -7.63
13 Conchi +9.05 +3.36
l4 Ollague -4.64 -3.45
15 Rio Grande -1.64 -3.94
16 Machuca -2.67 -7.39
17 S.P. Atacama -2.42 -l.00
18 Talabre -3.24 -3.79
19 Toconao +0.33 -0.61
20 Peine -3.45 -0.57
21 Socaire -2.11 -2.24
22 Chusmiza +1.17 -4.93
23 Villablanca -0.91 -0.79
24 Huavifia -l.04 -l.l4
25 Tarapaca +4.68 -5.38
26 Mamina -2.05 -2.28
27 Pica +6.62 -4.20
28 Guatacondo -3.02 +0.96

 

Source: Instituto Nacional de Estadisticas, Chile.

56

During the 1960-1970 intercensus period the general
trend depicts a more radical effect Of negative average
values Of annual population change. By comparing these
intercensus figures, it should be noted that only seven
settlements show positive values and several are close to
1 percent. Conchi, Amincha, and San Pedro present the
highest values (3.36, 3.45, and 4.59 respectively). These
three settlements are mining and/or service centers in which
agriculture does not play an important role. The influence
of the railway that connects Bolivia to Chile might be a
contributing factor to the mobility Of peOple within these
areas. Furthermore, the construction of a dam in Conchi
has attracted workers to this area. However, while some
settlements such as Conchi and San Pedro still maintain a
high positive annual population change, others within the
same area such as Puquios have shifted from high positive
values tO the absolute lowest. This is attributable to the
unstable mining Operations seen in these areas. The negative
effects have been more generalized for the most remote
settlements. As an example, Villablanca and Peine contri-
buted values equal to -0.79 and -0.57 respectively, while
an inverse relationship is seen in those settlements at
intermediate distances to the city (quuina, Cupo, Machuca,
and Rid Grande) whose annual population change figures for
1960-1970 fluctuated between -7.54 and -4.00 percent.

Only by analyzing the average annual population

change for the two time periods it is possible to postulate

57

a definitive trend due to migration. Although fluctuations
might be, in part, due to a natural increase, the other two
components Of change in the total population, immigration
and emigration, most likely exerted a pronounced effect on
the population change Of the Chilean highland settlements.
TO verify the association between population change
and migration in each settlement the average annual pOpula-

82 was compared with the average

tion change due to migration
annual pOpulation change between 1970 and 1978 (Table 4 and
Figure 6). A simple linear correlation analysis was used

to show the extent Of association. A correlation of 0.93
(0.1% significance level) was Obtained between migration
rate and population change inferring that population change
statistics do indeed indicate migration among highland rural
populations in northern Chile. The coefficient Of deter-
mination (the amount Of variation in one variable that is
explained by the variation in another variable) is 0.87.

In other words, 87 percent Of the variation in population
change is explained by migration. Likewise, the positive
correlation between the variables allow us to view the

migration rate as a crucial factor in explaining population

change.

 

82Stoltman and Ball point out that a more realistic
label for the average annual net migration rate would pro-
bably be "average annual pOpulation change due to migration."
Joseph P. Stoltman and John M. Ball, "Migration and the
Local Economic Factor in Rural Mexico," Human Organization,
Journal of the Sogiety for Applied Anthropology, V01. 30,
NO. 1 (1971), 47-56.

58

Table 4.--Average Annual POpulation Change and Average

Annual Net Migration Rate 1970-1978.

 

 

 

Entities Population Change Migration Rate
1 quuina -4.47 -5.35
2 Caspana +1.66 +1.42
3 CupO -6.66 -6.57
4 Chiu-Chiu +2.37 +2.11
5 Lasana +2.09 +1.75
6 San Pedro -0.85 -l.37
7 Toconce +0.88 -0.65
8 Turi -2.43 -3.22
9 Puquios -9.90 -7.43
10 Ujina -0.86 -0.83
11 Amincha +0.74 +0.51
12 Buenaventura -0.16 -0.48
13 Conchi +6.03 +6.62
14 Ollague -0.21 -0.31
15 Rio Grande -9.33 -7.89
16 Machuca -10.93 -8.69
17 S.P. Atacama +3.10 +2.26
18 Talabre -6.37 -6.25
19 Toconao +1.36 +0.52
20 Peine -l.38 -2.04
21 Socaire -2.92 -3.62
22 Chusmiza -5.90 -5.90
23 Villablanca -O.34 -2.06
24 Huavina -0.30 -l.82
25 Tarapaca -5.57 -5.57
26 Mamifia +0.97 +0.23
27 Pica +0.90 +0.44
28 Guatacondo -0.53 -2.18
Source: Instituto Nacional de Estadisticas, Chile. Field

work 1970-1978.

59

 

 

VOLUME AND DIRECTION OF FLOWS
1970

 

 

., ................. :2
"I

7“

 

 

 

 

  

NUMBER OF MIGRANTS
5- 9 __.

10-14=¢
15- 19 ‘
20-24 ‘

25-29 m -......,_.
30-34 - x. * .

 

 

 

60

Structural Migration

In order to understand a migration system for a
specific place and time period it is pre-supposed that not
only is the volume Of flow important but the direction Of
the movements as well. In this context, the process Of data
disaggregation related to migration has tO be based upon the
detailed measurement Of the steps within a migratory flow,
thereby furthering our understanding Of the structural
changes Observed in the context of any migration system
through time. Moreover, the analysis, Of the step-migration
process is understood through the interpretation Of the
differential characteristics Of the migrants.

The analysis Of structural migration among highland
communities in northern Chile has been divided in two time
periods: those which occurred immediately prior tO, or
during 1970, and those which were Observed directly pre-
ceeding and including the year 1978. Both cases are part
Of a unique process characterized by a rural to rural move-
ment.

The migration flows for the time period 1970 are
characterized by direct movement toward intermediate towns
(Table 5). From the total net in-migration Of 772 subjects,
49.30 percent is directed toward intermediate centers Of
attraction. Six towns maintain a stable pull over the rest
Of the settlements. Thus, Momina and Pica serve as sources
of attraction within the rural highland areas of Tarapaca.

On the other hand, situated within the influence Of the

£5].

 

 

 

 

 

 

vva a v0 Nm m N v MN N O 0' N NNN v o nN NO ON ov a n mN on ON mm NO ON ON NN
NNN NN v N N o o v N N .coonudso
Nn QN a o OONA
mN n mN n N N N ch552
an v MN NN N N n N N n nummuuuh
NN N o o N N N N N N N ch>usz
NN N o N N m o N Andi;
Nv aN N n m N N N n uuNsaagu
0v mN N 0N N afiwuoow
mm N N N NN N 05.2—
ON NN N N oacoooa
NN mN o oundea
NN N o NN N .'03¢.m.w
ON N N N n c9501:
on Na v v coccuo.m
NN NN N N n v N onuoNNo
«N N NN Nzocoo
mN n MN N N n N .cosacosa
mN N n o N v n N N N «sochc
ON N N N N n N N N chnD
on N m o o. n N N N N 3:68
an n vN N v n m n Nuns
vv 0 n v m N N a N 00:0008
ON a n N n n n N n N ouoom.m
a a «sauna
a o 5N:UI5N£U
on N n N N N v N n n v v n onno
o N N N N N 2330
an n N v NN N v ON N N IcNuva¢
d I A a ea c. H 0. Nu c. I
mammmuemmNWLZW mm :mem as.
V

“.mmwmmomim .m 1:“ .8...

o - o n a

. no

oocNm

 

.ONON .IIDNh GONHGHUN!

.xauua: ooeuoNuoz.NouooaNN on. euuauuuocooadeug.m oNhaa

62

Antofagasta to Bolivia railway, Amincha and Conchi draw
population from the neighboring mining settlements and the
smaller villages. Finally, Toconao and San Pedro de Atacama
Of the southern Calama area, generate a continuous influx
Of migrants. In contrast, the total net out-migration
within the highland communities was Of a similar magnitude.
On the basis Of this figure, Tarapaca, Chusmiza, Puquios,
Rio Grande, Peine, and Socaire, among others, demonstrated
the highest out-migration. If we note the location Of these
settlements, (Figure 3) it is apparent that the attraction
is influenced by short distance moves. The local influence
is significant when it is possible to measure this attrac-
tion in terms Of the percentage Of movements within local
areas in relation to long distance moves.83 A 45.30 percent
Of the total movements are local flows.

With respect to direction, by selecting the greatest
flows among settlements, we see a clear tendency toward the
intermediate centers. Furthermore, the other villages (Chiu-
Chiu and Lasana) indicate an area Of attraction that is
self-explanatory since they represent the last step needed
to reach the city. Being the closest villages to Calama
and Chuquicamata and sharing similar functions as the first
six intermediate centers mentioned above, the differences

among them are based on the intensity of flow or volume of

 

83Movements within local areas shall be understood
to mean migratory flow between the two settlements in the
closest proximity to one another.

63

migration. It is a foregone conclusion that this volume Of
migrants is not directly prOportional to the population Of
the centers of destination but to their number Of functions
and distanced to the city (Table 6). Thus, San Pedro de
Atacama and Chiu-Chiu demonstrate a greater attraction than
their counterparts. However, it should be pointed out that
the flow map for 1970 reflects only movements greater than
five. The resulting analysis Of the 1970 matrix indicates
the fundamental importance Of small volumes. Here again,
the short-distance move dominates. As we will Observe
around 1978, time period long-distance flows do not appear
in a generalized form. Moreover, as the migrants initiated
short-distance moves their awareness Of other Opportunities
were progressively broadened. Two variables can support
this contention namely, illiteracy and transportation mode.
Since 1970, the percentage Of illiteracy for those settle-
ments affected by out-migration toward intermediate centers
has decreased from 43.6 to 21.7 percent. Accordingly, the
importance of the transportation variable has increased in
importance thereby augmenting the outflow Of information

flow about the better economic Opportunities.

Migration Flows

At the outset, it might be advisable to ascertain
how migrants move among settlements, or what is the relevant
itinerary Of successive movements. By comparing the average
number Of steps, up to and including 1970, with those in

1970, the general trend has been marked by a large number

64

Table 6.--Intermediate Settlements* 1970.

 

 

Distance

Population Net In- TO Calama Number of
Settlements 1970 Migration Km. Functions
Mamina 271 51 402 5
Pica 1074 64 393 10
Amincha 343 40 223 5
Conchi 126 69 82 3
Toconao 425 48 144 5
S.P. Atacama 546 117 103 10
Chiu-Chiu 148 92 38 5
Lasana 164 55 46 4

 

*"Intermediate settlements" refers to those settle-
ments located between the most rural communities and the
cities. They are synonymous to intervening Opportunity
centers.

Source: Instituto Nacional de Estadisticas, Chile Field
W0rk, 1970.

65

Of steps (Table 7). Furthermore, the absence Of long-
distance moves, in 1978, was becoming apparent around 1970.
These movements were characterized by periodic flows among
settlements located near the cities. If we compare the
number Of steps from one village tO another between 1970
and 1978, we see that the migration flows characterized by
two, three, and four steps were more prevalent in 1970 than
in 1978. In contrast, one step moves were more numerous
in 1978 than in 1970.

The analysis of migration flows in 1978 has revealed
a small but ever-decreasing influence of the intermediate
centers (Figure 7). FrOm the total net in-migration Of 677
subjects, 45 percent were attracted to intermediate centers.
On the other hand, the short distance movements were less
pronounced compared to the migration flow Of 1970; however,
these types of moves reveal a similar pattern within the
study area (Table 8). The real change in the evolution Of
migration flow between 1970 and 1978 is related to long-
distance movements. The first migrants from Villablanca
and Huavina (the most remote points) that reached Chiu-Chiu
and Lasana in 1970 followed a course of four steps. An
increasing number of migrants were later to travel the same
distance in fewer steps, two. This group is comprised
largely Of family members Of the initial group Of migrants.
If we compare the number Of steps, it is clear that a
greater number Of flows in 1970 required more steps than

was the case in 1978 (Table 9). Almost the same effect was

.aNmN guNt OOINuqnloo :N nucolu>OI ucoond I x

 

N
n
K
C

.coooucso
oONm
«CNaa:
audacuua
2:23:
.NncNNN>
cuNIlng
ouNuoom
ocNom
oacoooa

H

MN v

H
MNKH

”MN
CNN

an n

NMnNHN

HHHH

cm N

MHRNH
MNNH
NXM'MNM

N0 N

NMMV'VK"

 

 

NHNN
M

CHAINIB
.g¢.m.w

IroNh NO .2 unoum N

 

H
H
H

 

66

82.68.
8:30...
2638
26:8
.cosucdam
26:3
2: no
oONsusm
«NE.
095008
Boom . m
05.3
QNBUnsNgu

NM
H
.4

MNN

N
H
N
N

N fio-IHH
HNHMMNN
H
MMMN
NMNV'FIMF'M
HHHM N
NHHNNNN

HMNNIfl'"N'
NNN

HNNH
HF‘N
HH

N N ocNauac

N

 

'HT“'“H
'TQ‘IITA
oaruaos
outed
oeuooom
cannon“
THOUOD

onbuxto

poodbxo;
qu'rvr
fiPuvzs'u
'ROHTI!
mm
00m
tan;
'8
outing

-uoooavn9

'OTd

'UT'UH
-uoanuona

“Tub-“TED

odna

W":
IuTnhK

"Ofl" J's

 

 

.22 .98 «6 Neal... 8322..-.N .38.

67

 

 

VOLUME AND DIRECTION OF FLOWS
1978

 

 

 

 

 

 

 

NUMBER OF MIGRANTS
5‘ 9 _’ s ’l
10 '14 = :0 3. I
15 ‘ 19 fl "°-..,,,/

20-24 ‘ :7" J
25-29 - ....... ., .//

30-34 - “Tum/f.

 

 

68

 

 

 

 

 

 

vaN N no NN N o N NN N o oN N voN v N NN no NN NN N N NN vN NN No No vN Nv NN
New vN N N o o v N N .eooouuso
NN N o N uoNN
NN N NN N N N N N ucNag:
on NNNo NNo N N «N moan—qua...
NN N N o N N N N N N aeNsudz
NN v N N N v N N N o .NhuNNN>
Nv NN N N e N N v NN uuNuusno
NN NN N NN N ouNuoom
NN N o N NN N ocNoN
NN NN N NN oaeoooa
NN NN a ouhuNoa
NN NN N .Iofiadd
oN NN N N o cannon:
NN NN. v o ooeouo.m
NN o N N N N .soaNNo
oN N N Neoeoo
oN N o o N N N N .eoeneosn
NN N N N N N N N N N «noeNn:
oN N N N N N N N N acNN=
NN N N N v v N N N N .oNnuse
NN N oN N N N v N Nada
v. N v N o o N NN N ooeoooa
NN N v N v N N N N N oaooe.m
v v neon-N
N N :NguaaNhu
«N N N N N N N N N v N N v capo
oN N N N N N neauaau
No v N N NN N v o N N
d I A 58 80 n.. S
mom awn mmmnrmoum m: mm. mmmu
m I 10M? on Mr. 4
mm em .1 mm . w m
m 'o "I. 0m. mm. I O m
. n . n

 

 

.oNNN .usoNN eoNuauoN: .xNuuq: ooeuoNuoNINoiouaNe can :uuNNuNouooaNNnn.N «Nana

.ONON suNt cOONNOQIOO cN cued-Ibo! accuse I x

 

. 900336
no.3
i
$03938
05155:

. NANNN 35
63.328
0.3108
050m
03.0006

H
N
K
V
C

O

In

H
”Ni-CH

Fl
HH
Hod—4H
HHNHHH

NN n

NNNNN
MNNH
NNN

NN
NNKNNNN
NNNNN

Nm N N N
N

HNNNNNNN

MON N

 

 

NHHH

ounaNuu.
0.6.: no oz .93 N .Uoout.m.m

H

 

 

69

N
N N N 50:503.
N ope—cue..—
SEINNO
20:00
€022.95
£23!
ucNNa
no?
Nash.
006002.
9.60m . m
333
53:53
095
Sim-co
N N SNE‘

HN
H

H
N
H
N

HHF‘NMHN

HHH

H

HH

NNNN
HHHNMNMF‘H
HHHMMMNF‘H

HHHHN N

NHH
HHHNNNN

N
H N
H N
M
N
H H
,4
Flo-OH
HHHH
HH

 

“tomato
Md
m-a
pardon;
mans
'tm‘rtm
'
”woos
med
mooox.
03W].
'nmv'rs
'mum
W19 '8
“mo
mama
mam
mm
coming
1m
'3
mm
tum-um
cam
mm
mm!

 

 

.22 Room an anal: noon->716 oNn-N.

70

felt in the most remote southern villages of this study
area. Peine and Socaire experienced increasing out-
migration that was not significant in terms of volume but
was important when the distance traveled is considered.
Apparently, these outflows were less pronounced in terms of
distance than in the case Of those from the most remote
northern area; nevertheless, they were not local movements
as was clearly seen in 1970. In fact, migration from these
distant towns, as was the case in the far northern areas,
indicates the Obvious influence Of both better roads and
improved privately owned transportation. These conditions,
initially generated by both the media and "word Of mouth,"
increased reports Of Opportunities within the city areas
and surrounding towns. However, it is important to point
out that the migration pattern Of the southern Calama area
(San Pedro de Atacama, Toconao, Socaire, and Peine) was
more stable in terms of short distance migration than the
northern study area (Chusmiza, Villablanca, Huavina,
Tarapaca, Mamifia, and Pica). Both San Pedro de Atacama and
Toconao played increasingly stronger roles as intervening
Opportunities than did the towns Of Pica and Mamina. In
light Of this fact, the southern study area presents a more
compact migration field than the northern area. As seen in
the latter analysis, the attraction Of the secondary towns
as intervening Opportunities increased in direct prOportion

to the distance from Calama and Chuquicamata.

71

Of the total 1544 in-migrants registered in 1970 and
1978, approximately 288 migrant family heads were inter-
viewed for their migration histories. For this research,
it was crucial to determine the migrants' places Of resi-
dence prior to their most recent moves. The data were
divided into two categories according to the year Of migra-
tion: (a) those who moved before 1970, and (b) those who
moved between 1970 and 1978. For each Of highland's twenty-
eight communities, the number Of migrant family heads were
recorded and their migration histories classified according
to the number Of steps completed before reaching the settle-
ments. The interviews of 1970 demonstrated that distance
has had a positive association with stage-migration process.
As the distance from the city increased, the migrants
reported a greater number Of steps before reaching villages
close to the cities than did the migrants whose places Of
birth fell within a shorter distance (Table 7 and 9). More-
over, migrants from the intervening Opportunity villages
were replaced by new migrants from the smaller remote
settlements and communities located at higher altitudes
(Table 5 and 8).

Between 1970 and 1978, as the relative accessibility
to the intervening Opportunity centers and towns close to
Calama and Chuquicamata improved, the stage migration process
progressively decreased. Long-distance migrants reduced
both the number of steps and time required to reach villages

near those cities. The replacement effect, which was an

 

72

important characteristic in the intervening Opportunity

villages in 1970, had decreased by 1978.

Migration by Sex and Age

There is a general agreement in the Latin American
migration literature that, just as in other parts Of the
world, migration is highly selective with regard to age and
sex. This research analyses data within a time period Of
sufficient length (1970-1978) to Obtain a meaningful com-
parison between what is occurring at present with that Of
ten years ago. With the exception Of those males between
25-29 years who in 1978 conducted long-distance moves,
middle-aged females (30-34 years Of age) exceeded the other
age groups as principal migrants (Table 11). This table
also shows the movements by sex with respect to short- or
long-distance migration. In this regard, in 1970, short-
distance migration was sufficiently more important for
females than for males (56.1%); whereas long-distance migra-
tion proved more predominant in 1978 with 58.2 percent Of
male migrants. It is predominantly the people Of median
age group (30-34) who are more prone to move than the Older
people, hence the process shifts toward younger ages (25-29
and 18-24). In 1978, males between the ages 25-29 began to
undertake more long-distance moves. However, there are some
figures corresponding to the ages 45-54 that might suggest
return migration or short-distance movements. These changes

in residence may occur when migrants seek out health care

73

Table lO.--Percentage Distribution Of Age by Short-Distance or Long-
Distance and Sex, 1970, 1978.

 

Migrant by Distance and Age

 

 

 

 

 

 

1970 1978
Age Group
Short-Dist. Long-Dist. Short-Dist. Long-Dist.

Male Female Male Female Male Female Male Female
25 - 29 8.0 11.1 10.6 9.0 11.5 12.5 17.8 8.1
30 - 34 13.5 19.6 15.5 14.6 12.1 16.5 16.1 14.3
35 - 44 7.0 8.0 11.0 8.5 6.0 7.5 7.0 5.5
Total Aged
18-44 Years 40.8 54.7 51.3 45.4 43.1 51.6 55.4 40.4
45 - 54 2.1 1.1 1.0 009 1.7 103 2.0 100
55 - 64 1.0 003 008 0.6 1.0 1.0 008 0.4
65 & over 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0
Total Aged
45 and Over 3.1 1.4 1.8 1.5 2.7 2.6 2.8 1.4
Total 43.9 56.1 53.1 46.9 45.8 54.2 58.2 41.8
Grand Total 100.0 100.0 100.0 100.0
Number of
Cases 394 378 281 395

 

 

Source: Field work, 1970, 1978.

74

facilities or attempt to strengthen family ties within those
areas dominated by intervening Opportunity centers (Figure

8).

Migration by Marital Status

Since young married couples generally do not wish
tO separate, they usually migrate together. And, tO in-
crease their possibility to Obtain employment, they usually
migrate to nearby areas for which they have reliable reports
Of economics Opportunities. On the other hand, for those
Older married workers (those married around 1960), the
migration pattern is rather complex. One group has moved
from the most remote northern communities toward towns such
as Chiu-Chiu and Lasana. The second group initiated short-
distance moves tO both Toconao and San Pedro de Atacama from
the most remote southern communities such as Socaire and
Peine.

Of the entire group Of Older married (1960) in-
migrants living in the intervening Opportunity towns, more
than 40 percent were encountered in Chiu-Chiu and Lasana;
whereas the percentage decreased considerably as the distance
from these two villages increased in the direction Of the
most remote centers. In the same context, the difference
in number Of Older married males and females migrants is
evidenced between agricultural towns and mining towns (Mamina,
Pica, and Amincha-Conchi, respectively). Younger married

(1970) inmigrants showed the most positive correlation to

75

 

 

MIGRATION BY AGE AND SEX

SHORT AND LONG-DISTANCE FLOWS
1978

 

 

 

 

 

 

 

Egg

AGE I
A : MIGRANTS (18-24)
B : MIGRANTS (25- 29)

C :MIGRANTS (30-34)

W“.

 

 

Figure 8

 

76

distance among the intervening Opportunity centers. The
relative attraction of the two most important towns, Pica
and San Pedro de Atacama, is demonstrated in that more than
50 percent of the number is constituted by young married
persons Of both sexes (Table 12). This phenomenon must be
interpreted, then, as a dual process: migration toward inter-
vening Opportunity centers and migration to towns closer to
the cities. Within the intervening Opportunity locations
(Pica and Mamina, Toconao and San Pedro de Atacama) the
highest number Of male migrants were those married around
1970. Female migrants were characterized by marriage Of
longer duration (before or in 1960). The latter group, in
which short-distance movements predominated, is associated
with repeatedly short-distance moves. To Obtain better
Opportunities as miners in sulphur mines, as share-croppers
in villages close to Calama, or as unskilled workers in
railway stations, these married people Of the same age had
left their wives behind in their home towns or in inter-
vening Opportunity centers.

A new trend begins to take shape for those migrants
who returned to their places Of origin. They took their
families with them to initiate a new movement toward towns
closer to the city than their original villages. Chiu-Chiu
and Lasana have received new migrants from Huavina and
Chusmiza in 1978. This group is comprised of Older couples
and their children ranging from 10 to 14 years Of age. The

wives remained in Mamifia or Pica while their husbands worked

 

 

77

Table ll.--Migration by Marital Status 1960, 1970.

 

 

 

 

 

Intervening In-Migrants Married In-Migrants Married
Opportunity in 1960 in 1970

Centers

Male Female Male Female

Mamina 4.0 13.0 3.0 6.5
Pica 5.5 14.5 18.0 26.0
Amincha 6.6 2.0 1.5 0.0
Conchi 19.5 0.5 12.0 0.0
Toconao 12.0 11.3 18.5 18.0
S.P. Atacama 9.9 14.5 25.0 25.5
Chiu-Chiu 24.0 23.2 12.0 14.0
Lasana 18.5 21.0 10.0 10.0

TOTAL 100.0 100.0 100.0 100.0
Source: Field work, 1970.

 

78

in the mines. At the later date, the entire family then
moved to agriculturally prosperous areas.

In conclusion, migration Of couples or families con-
sisted cf a multi-staged process. First, young couples moved
to intervening Opportunity centers in short-distance flows.
Thereafter, males left their wives and children behind for
short periods (4 or 5 months) while working in the mines.
Finally, their families joined them in the new setting to

complete the family migration (Figure 9).

 

Spatial Mobility
I have pointed out that the phiOlosophical approach
to northern Chilean migration phenomenon reflects the

Objective/cognitive dichotomy.84

This position has been
determined by the structure.of needs among the migrants and
the spatial range Of the study. 'On the one hand, the
mobility decisions are associated first with needs for
survival and safety, and secondly, to advance oneself socio-
economically. At the same time, periodic migration char-
acteristic Of short-distance moves, have begun to shift
toward longer distance movements. Furthermore, internal

migration in northern Chile is bringing about a change in

migrant occupations. Former agriculturalists leave their

 

84The Objective/cognitive dichotomy reflects the
rational-man/satisficer dichotomy. According to the assump-
tion Of economic rationality a migrant selects the most
beneficial potential destination available. The satisficer
assumption is one Of bounded rationality, which suggests
finding a course of action that is good enough for the
situation as the migrant comprehends it.

79

 

 

 

 

MIGRATION BY MARITAL STATUS

 

 

 

 

 

 

 

o MIGRANTS MARRIED IN 1960
o MIGRANTS MARRIED IN 1970

 

 

80

their farms to take up work in one Of the mining communi-
ties Of the region.

Based on the Objective/cognitive dichotomy, twenty-
two variables were selected which represent both socio-
economic and physical characteristics in the study area
(Appendix B). The data were Obtained for two dates, 1970
and 1978. In order to Obtain an easier manipulation Of the
variables, data from seventeen variables were averaged for
the 1970-1978 time period. The original labels Of these
variables were as follows:

1. Labor force rate

2. Index Of market volume

3. Mobility coefficient

4. Illiteracy rate

5. Uninhabited houses rate

6. Average hours worked per day

7. Percentage Of return migrants

8. Percentage Of active pOpulation in agriculture
9. Percentage Of active population in mining
10. Percentage Of active pOpulation in small

factories

11. Percentage Of active population in services
12. Percentage Of local movements

13. Percentage Of districtal movements

14. Percentage Of communal movements

15. Percentage Of movements to the city
16. Travel time

17. Number Of trucks (main transportation mode).

In addition, five other variables were considered, and from
these only two were recorded in 1970: non-owner Operator
rate, and mean farm size, and one for 1978: number Of ser-
vices. The last two are absolute values: travel distance
(Km), and altitude (meters above sea level). The dependent
variable was computed from the average annual migration

rate between 1970 and 1978.

 

 

81

It has been established under the first hypothesis
that, in time, the stepwise or stage migration process
breaks down and becomes a more direct rural to urban move-
ment. Operationally, this assumption has been expanded by
postulating that there is a distortion in the distance-decay
function, that long-distance moves may be facilitated by F
information flow accompanying the return migrants, and that
the role Of intervening Opportunities might tend to decrease

in time as a migrant's awareness space increases. Under

 

this general assumption and associated Operational state- a
ments, the structural analysis has revealed nothing new and
has tended to confirm these migration trends. In addition,
this analysis has identified the associations between long-
distance migration, the level of information, changes in the
structure Of flows (from migration with replacement to long-
distance moves), and improvements in both roads and transpor-
tation modes. Furthermore, the structural analysis found
an association between intervening Opportunity centers and
attraction of labor force via short-distance moves. It has
to be kept in mind that the volume of migrants is directly
proportional tO the number of functions (services) within
the intermediate centers and the distance to the cities,
Calama and Chuquicamata, which also act as centers of
attraction for the migrant labor force.

TO verify the associations cited, the set Of twenty-
two variables was subjected to factor analysis in order to

Observe the independent variation within the set. As a

82

result, six factors with eigenvalues greater than 1.00

account for 82.0 percent Of the total variance. However,

the communality Of ACTIVAGR exceeded 1.0 after six itera-

tions.

Six orthogonal factors were extracted and rotated

using varimax rotation.
these factors (Table 12), it was possible to group the
underlying elements that, in order to facilitate the analy-

sis, were descriptively labeled as follows:

Factor
Factor
Factor

Factor
Factor
Factor

O\U'lub MIDI-d

85 By Observing the loadings in F'

 

Labor Force 1
Communal movements and distance

Primary activities and environmental

constraint

Mobility and level of information

Intervening Opportunity effects

Movements to the city.

The spatial distribution Of the factor scores (Table

13) proves useful in the analysis (Figure 10 through Figure

15).

Factor scores for the first factor showed that those

settlements which had high scores gave indications Of a high

labor force (LABORF), high number Of work hours (WORKHOUR),

and a very low percentages Of uninhabited houses (HOUSES).

Migrants from these same towns showed very little movement

within the desert (DISMOV). Conchi, Amincha, Chiu-Chiu, and

Lasana had the highest scores. This factor is labelled

Labor Force (Figure 10).

 

85

Varimax rotation tries to minimize the number Of

variables that go highly on a given factor.

 

83

 

 

NMNmN.I NmmmN.I oNomN.I mmNNN. vmvmo.t mmmwN.l mmBNNNH
OGONN. vawm. mONmm. mommm. mvmmo.l mmmmo. BflMfiflZ
Nmmmo. ommmo. GNQNo. «mmON. ONva. momNN.l Hn>dm9
mmmom.l NNNoo. vommm. mNmmm.l mmvNo. Nmmmv. mMUDmBZ
mNmMN.I mmNNm. NNNmm. mmvmo. wamo. mmomm. DZUH>mmm
mmwmo.l mwemo. omMNm. MNmmN. vaNo. mNNmm. mmbommmoz
mmNNm. mmmmN. vmmmm. NN¢¢N.I mmNmN. mmNNo.l NBHNHmOS
NvNNo. mmmmo. momNo.l ONmmN. mNmmm. mewN.l HB>4MB
Nwmmo.l NNNmm.l mmmmo.t NNme. MNme.I vaNN.I mDDBHBN4
mNmmo. NNmNN. mohmo. Nva¢.I NmNON.I mNMNO. OUflMOfl
GMQMN. NonN. mvaN. omwmm.l mmvho. omew. >QSMBHU
mmNmN.l ommwo. mmmNo. mwaN.I mvam. mommm. >QZEOU
NmmmN.l mvaN. mNNNN.I mmNmN. mmmoo. mmva.l >OZBmHQ
NNmNN.I «NNNM.I vamo.I ommNN. NNmNN.I NNNmN.I >OZUON
movmo. vmvmo. NommN. mmNmm. ONONo.I mN¢NN. mmm>HEU<
vaNN. ommNN. Novmc. wNomo.I NGNGN. Mommo. Umm>HBU¢
Nvac.I mNmmN.I NNmMN.I MMOON. NmNoo.l omGVN. 2H2>HBU¢
mmmmo.l moomc. mNNmm.l vmvNN.I Nmmoo.l mwomN.I m0d>H804
vmhwv. mmmmm. NoomN. omNNN.I mammm. Nmomo. szmsamm
NNNGN. NNmNN.I mMNmN.I NNmmN.I OGMNo.I MNNmo. 2302202
N¢¢No.I mmoNN.I waoo.l mmmNN. NGQON.I mmNNm.I mambo:
MGNON. mmomo. wmmwo.l ommmo. NmmNN.l mmmom. mmom¢fi
@ Houomm m Houomm v HOHOMh m HONOmm N HOUOMh N Houomh

 

.mchomON pom muouomm chomonuNOII.NN mNnma

84

 

 

 

 

ONMN. ONOO.I NmNN.NI ONom.I Nmmm.N NOON. ON
vao.NI m¢m¢.N mOmm. mmmo.NI OmNN.N mmNO.N NN
«Nmm.I NOON.I NMON.N NNNN.NI NOON.N mmmw.l ON
mono. NOON.I NONo.I mOmm.NI Nomm.N NovN.NI ON
mmNN. mNmN.I mNOm.I mmmO.I NNNm.N Omom.I ON
mwNm. ONOv.I NOmN.I OOON. «ONO.N NmmN.I mN
mNNN.m NOom.I NNNO. NNvm.N NOON.N NNNN.N NN
omNO.N Nmmm.I wONN.I mNmm.N vam. Ommo.N NN
mNNm.NI mmmm.N NNmN.NI NOON.mI MOmm.I momm.NI ON
m¢o¢.N owwm.N NOON. Ovmm. mNmm.I Nomv.I mN
mNmo.NI mNmm. vOmN.NI mva.NI NNom.I NNNO.NI mN
momN.I OmmN.N ammo. NNNO.NI NOON.I «New. NN
NONO.I mNNN.I NNNm.NI mva.I Ova.I OMON.I ON
NONm.I MNNO. Nvmm.l NNON.I vaO.I mNNo.NI mN
«NON.N mme.I MNmo.N mmNN.O omm¢.I Nmmm. NN
mmoO. omNN.NI mOON.N mNmO. mNmN. Nm¢m.N MN
NNNN.N OOmN.I omNN.I omOm.m «Nmm.I Nome. NN
mNom.NI mmmm.l OONO.NI OmON.I NNNv. mmmN.N NN
Nmmm.l mNmm.I ONMN.N «mom.m OOON.I NNmN.I ON
vam.NI Nmmm. Nmmo.N OOON.¢I Ova.I mONO.NI m
ONON.I NNNm.I mNmo.I OONN.N mmmm.| mmOm.I m
«Nvo. mmmN.NI mNNw. mmmN.m NmNo.I Nmmm.N N
mmNN. ONNN.NI NmNo.N ONmN.NI NNO¢.I OOOO.I O
OMNN.N mmmN.I NmNo.I Nomm.MI vam.I OmOv.N m
ONON.N Oomm.I NmmN. mNom.NI NON¢.NI ONmN.N v
MOMN.NI Nmmm. NmmN. ONOO.I «NN¢.NI Nmmm.NI m
mmON.NI mmmm. OONO.NI 0Omm.I ONMO.I Nva. N
vao. omNo.NI OMNO.I OONm.I mOmm.I «Own. N
O m o m N N
.mONOOm Houoom 0cm mucoEwNuumm NmuomlI.mN ONQOB

85

 

 

FACTOR 1

LABOR FORCE
FACTOR SCORES

 

 

 

 

 

 

 

 

 

 

Figure 10

 

86

The second factor showed the highest factor loadings
in communal movements, travel time, and travel distance.
Factor scores indicate that those settlements located at the
greater distances from the cities were highly correlated
with high communal movements. These most remote northern
settlements yielded the highest positive scores (Figure 11). r

The highest factor loadings on factor 3 were seen
in the non-owner operator rate (NONOWN), the active popula-

tion engaged in agriculture (ACTIVGR), the active population

 

engaged in mining (ACTIVMIN), and altitude (ALTITUDE). High E,
positive scores are found in the most remote areas as well

as at the highest altitudes of the mountain range. Since

this factor consists of variables suggesting the primary

activities linked to certain physical conditions (sulphur

mines around volcano areas, agriculture with large amount of

water) and, since the scores are most significant at higher

altitudes, it was labeled "the primary activities and environ-

mental constraint" (Figure 12).

The fourth dimension, labeled "mobility and level of
information," is comprised of the active population in ser-
vices (ACTIVSER), a mobility coefficient (MOBILITY), and
illiteracy rate (ILLITER). High positive values were found
along the railway routes leading to the mining sites and
Conchi. 0n the other hand, negative scores indicate no clear
or defined pattern (Figure 13).

Loading highest on factor five were the return

migrants (RETURMI), the active pOpulation in small factories

87

 

 

FACTOR 2

LOCAL MOVEMENTS
FACTOR SCORES

 

 

 

 

 

 

 

 

 

 

 

 

5. a so too DIED.
t:::::---t:::::---l

 

 

Figure 11

 

 

88

 

 

FACTOR 3

PRIMARY ACTIVITIES
FACTOR SCORES

 

 

 

 

 

 

 

 

 

 

Figure 12

 

89

 

 

FACTOR 4

MOBILITY AND INFORMATION
FACTOR SCORES

”O

a c. ' . .O..'. A.
441.0 I F'?'\‘
\ LN .\

 

 

 

 

 

 

 

 

 

 

Figure 13

 

90

(ACTIVFAC), and the number of trucks (NTRUCKS). The highest
scores were found in those settlements considered to be the
intervening opportunity centers most distant from the city
with the exception of Lasana and Chiu-Chiu, whose loading
level where negative and low. For these reasons, this factor
has been labeled "the intervening Opportunities effects" E‘
(Figure 14)-

The highest loading on factor 6 was movement to the

cities (CITYMOV). The highest positive scores were assigned

 

to Chiu-Chiu, Lasana, Toconao, Socaire, and Ollague. The E
lowest scores were located around the intervening opportunity
centers. This factor was labeled according to the highest
loading: "movements to the city" (Figure 15).

One of the most revealing results obtained from this
partial analysis is related to the role of intervening oppor-
tunity centers in the migration process. These centers
attract migrants seeking employment from nearby rural areas.
The influence of these intermediate rural centers is expressed
in three variables: return migrants, active population in
small factories, and number of trucks (the main tranSporta-
tion mode within the rural area). The potential importance
of this finding lies in the attraction of people from urban
areas back to rural areas and the indirect control of rural
migration toward urban areas. It has been stated that the
influence of these intermediate settlements is expressed in
terms of both the highest return migration and the highest

percentage of active population in small factories. Therefore,

91

 

 

FACTOR 5

INTERVENING OPPORTUNITIES
FACTOR SCORES

 

 

 

 

 

 

 

 

h.

 

 

' Figure 14

 

92

 

 

FACTOR 6

MOVEMENTS TO THE CITY
FACTOR SCORES

 

 

 

 

 

 

 

Figure 15

 

 

93

one of the possible solutions for curtailing rural to urban
migration might be the reinforcement of these small factories
within the intermediate settlements. Since the nature of
activities in these small factories is primarily agricul-

tural, their expansion and perhaps their extension toward

 

other settlements, will reduce an already increasing migra- 5-
tion toward urban areas. On the other hand, an increase in E
truck transportation will not serve to increase rural-urban i
migration but rather to create contacts between complemen-

tary areas (markets in urban areas, and small industrial L

sector based on agricultural production in rural areas).
After obtaining six orthogonal factors a multiple
regression analysis was performed with the average annual
migration rate (MIGRAT) as the dependent variable having
six factor score vectors as the independent variables. By
using stepwise process the six factors entered as follows:
(1) Labor Force (PAC 1), (2) Communal Movements and Distance
(FAG 2), (3) Intervening Opportunity Effects (PAC 5),
(4) Mobility and Level of Information (FAC 4), (5) Primary
Activities and Environmental Constraint (PAC 3), and (6)
Movements to the City (PAC 6). Multiple regression analysis
demonstrated that only the first of the factors (PAC 1) was
significant. If we accept a 5 percent margin of error (95%
confidence level), this factor presents an alpha (a) value
smaller than 0.005. However, if we carry the operation
further, the second factor entered in the stepwise process

showed a F significance equal to 0.233 which indicates an

94

insignificant factor similar to the remaining factors. The
b value for the significant factor was 2.30 and the constant
(a) -2.08 (Table 14). From this result it is evident that
the most important composite variable is Labor Force (FAC 1).
It might be logical to infer a pronounced association between
migration and labor force characteristics and that, in fact,
migration is related to the economic conditions within the
rural area, specifically, within the intervening opportunity
centers as intermediate settlements between the most remote
rural communities and the cities. It must be kept in mind
that Labor Force (PAC 1) is comprised not only of the labor
force as a variable with highest loading (+0.90) but also

of two other variables with high loadings (average hours
worked per day, and market volume). It should be acceptable
to extend the association of migration to these two vari-
ables.

It has been postulated that the volume of migrants
is directly prOportional to the number of functions (ser-
vices) within the intermediate centers including those rural
centers close to Calama and Chuquicamata. These centers
have the highest averages of working hours and the highest
indices of market volume (regional market). Therefore, it
is possible to assume an association between labor force
and migration by considering the importance of the latter
two variables. All of the variables are associated to the

migration flows which first develOped toward the intervening

 

 

 

95

 

Hoe. msA.OH mam. Omo.~- unmumcoo
Ahooo.u mmsmo.- Amm. HAN. was. «A~.I m can
smaoo.- assoa.I Nam. Hmm. mam. mmN.- m o<m
Asaoo. osmom. EON. mmO.A omm. cam. s cam
mmooo.u ommma. vmm. ham. NNm. mam. m can
mmooo.- oamma.u mmm. mom.A mam. mmm.- N cam
Amsoo. mamas. coo. mso.m~ Has. aom.~ A van

suflofiummam mumm .mflm a . m .m.m m mannaum>

 

pomm.NmH suflafinmflum> mo unmoo moms. mumsvm m mud

moo. .mflm mmm.s mmN.mmN .HN Amsuwmmm sv¢5.~ >mn cum
vsm.q -s.~m Nmm.OmN .O coflmmmummm mmmm. mumsvm m
O .OO cams mmumswm saw no m>oq¢ «was. m mNmNust

 

.muouomm Nmoomoauuo me suNz :ONmmmHmmm mNmNuNoz mmNsmmvmll.vN ONQMB

96

centers and consequently led to the initiation of long-
distance moves.

To verify the degree of influence exercised by the
variables included in the first factor as well as the influ-
ence of other variables, a second multiple regression
analysis was performed by incorporating the twenty-one T
original variables. One variable (ACTIVAGR) was removed
from the original set in order to preclude extreme multi-

collinearity. The stepwise process indicated that three

 

variables to be significant: (1) average hours worker per

llI

day (WORKHOUR), (2) percentage of uninhabited houses
(HOUSES), and (3) movement within the district (DISTMOV).
The significance level for each of these variables were
0.000, 0.001, and 0.010, respectively, at 95 percent con-
fidence level. The contant was -2.01 and the b values were
1.05 (0.209)*, -0.05-(0.013)*, and -0.06 (0.021)*, respec-
tively. Labor Force was not included within the significant
variables (Table 15). However, its influence is closely
related to the first significant variable: average hours
worked per day (WORKHOUR). Therefore, it is possible to
assume multicollinearity between these two economic vari-
ables. Furthermore, the inverse association between migra-
tion and movement within a district (short-distance moves)
indicates that long-distance migration predominates within

the highland rural area in northern Chile.

 

*Standard Errors of the b values.

97

 

 

 

 

OvN.
mmOmmom.N| vavam.N NNOtho.NI Aazmamzouv

OOOmm. ONO.
OOONOON.I Nwmhomh.NI NOIMONOOONNN. NoumNthOmOm.l >OZBmNo

mNOOh. Noo.
mmmvonm.| vmmmomm.mn NoumOhmOONmN. NoummmmONva.| mmmDOm

bthO.NI coo.
Omhmwom. hOO¢¢mo.m ommumeN. NOmmvmo.N moomxmoz

muNoNummNm mocmoNMNomNm
m Honum new m mNnmNum>
mumm B

 

.mmNQMNHm> NmonNHO woolmuomza saw: :ONmanmmm mNmNust mmNzomumll.mN ONQMB

98

CHAPTER V

CONCLUSIONS AND SUGGESTIONS FOR F‘

FUTURE RESEARCH

This research has examined two aspects of the

 

migrant population in the rural highlands of northern Chile.

"5

It has identified the changes within the stepwise and stage
migration processes that occur in the area and points out
migrant selectivity within the process. However, it should
be kept in mind that the main objectives of this research
were not to determine the causation of migration within the
rural highlands, but rather to observe the spatial movements
of the migrants as well as to test several hypothesis

relating to migrant selectivity.

Hypotheses and Operational Statements
The operational statement of the first hypothesis
indicates that short-distance, multiple stage migration
shifts to one of long-distance direct migration. Several
reasons account for this occurrence. The decrease is due
primarily to improved information flows throughout the study

area. These flows are enhanced by linkages that former

98

99

migrants maintain with their home areas. Furthermore,
periodic and return migrants may serve as important carriers
of positive information about the potential receiving centers.
These contentions are supported by the following statistical
association. There is a positive correlation (0.57) between
periodic and return migrants and migration to the city.

This contention appears to negate the distance variable
which is implicit in the stage migration process, since

most periodic and return migration involves those cities at
great distance from receiving centers. Moreover, this
belief is supported by the relationship between low illiter-
acy rate and a number of functions performed in a given

town (40.54). This assumes that those towns which have a
greater number of functions and a highly literate popula-
tion have peOple whose awareness space is greater than their
Opposite numbers. Therefore, this relationship assumes an
improved communication system within the migration network,
and as mentioned before, assists in breaking down the stage
migration system.

The operational statements of the second hypothesis
were based on migrant selectivity in which males move longer
distances than female migrants. Males make the long-distance
move from their home area to a town before their spouses and
children. After the male family head becomes settled, their
families generally follow. This reinforces the contention
that males indeed move via a stepwise process whereas women

and children by-pass the intervening centers and move

 

100

directly to those settlements close to the cities. However,
both sexes provide about an equal number of migrants to
nearby towns, movements that for the most part are of a
short distance. Once these short-distance moves were com-
pleted however, males migrate longer distances than females.
Nevertheless, these appear to be governed by time of occur- E”
rence. It is quite possible that this action infers that
stage migration is operative where replacement via short

distance moves is occurring.

 

For the entire time period 1970-1978, age also has
proved to influence the migration act. Generally, migrants
within the 30-34 cohort move longer distances than younger
migrants (18-29). However, today younger migrants appear
to be more mobile than young peOple of former years.

With reference to migrant selectivity by marital
status, older married migrants move longer distances than
young married migrants. More than 40 percent of the older
population moved from the most remote areas to Chiu-Chiu
and Lasana, towns close to Calama and Chuquicamata. On the
contrary, young married persons who comprise approximately
50 percent of the migrants, moved shorter distances usually
to one of the nearby intervening opportunity centers.

Relationship of Labor Force
to Migration

Multiple regression analysis was employed to test
the first hypothesis of this study. One factor, labor force

proved significant in develOping an association with the

101

migration rate. Within this factor, three variables (labor
force, average hours worked per day, and percentage of un-
inhabited houses) yielded results contradictory to those
expected. Normally, an area with out-migration should have
low rates of population in the labor force, low average
hours worked per day, and a high percentage of uninhabited
houses. The reverse pattern observed in the study area has
a possible explanation at its roots, the traditional way of
life. Nomadic pastoralism keeps periodic migrants out of
town for days and sometimes weeks. This ancient trans-
humance activity makes the rural population highly mobile
and renders even the best population register (interview)

inadequate. Furthermore, the concept of community cohesion

within these settlements is extremely strong. Rural dwellers

moved back and forth among communities to help relatives and
friends in various activities such as agricultural labor,
house construction, or other manual tasks. This periodic
phenomenon may mask the high rates of unemployment. It also
could inflate employment statistics as well as the number of
hours worked.

Another possibility exists in that the surplus labor
that formerly existed in these communities of high out-
migration, has moved to other places. This outflow of labor
creates a better relationship between jobs and workers for
those who remain. Furthermore, this process also may bring

about an increase in the number of hours worked per laborer.

 

l-
x

102

In addition, the unreliability of the housing vari-
able could affect its positive association with the migra-
tion rate. ‘During the time of religious festivals both
rural and urban dwellers spend several weeks within the
settlements and occupy the houses that otherwise remain
empty. In retrospect, the author realizes that in both
instances, the data were collected during the times of
religious activities in the study area.

It is important to keep in mind that periodic migra-
tion gives rise to increased information about both economic
and social advantages. This form of migration occurs pri-
marily among those settlements which exhibit high rates of
population in the labor force and whose population generally
works long hours. Therefore, the concept of high periodic
migration must be considered when attempting to understand
the migration process. On the case of this study, in-
creasing movement toward the cities does not occur in those
settlements with fewer job opportunities and lower rate of
persons in the labor force but in those rural centers with
high percentage of persons in labor force where workers
spend long hours at their jobs.

When a multiple regression analysis was conducted
with the original twenty-one variables, it was ascertained
that it is not labor force but work hours which is the most
important variable. The result of this analysis compels me
to conclude that the two variables act in conjunction with

each other to account for the association of these two

 

103

significant variables to the migration rate. The presence
of one of these variables readily suggests the existence of
the other variable since both of them are indicative of a

common economic effect: employment.

Application of Underlying Dimensions

Future research might stress the importance of
findings that deal with the orthogonal factors and some of
their underlying dimensions. For example, rate of labor
force and average hours worked per day necessarily consti-
tute important variables in observing the effect of migra-
tion between rural and urban areas. On the basis of these
two economic variables, future migration studies might
investigate these effects with greater precision.

Furthermore, the study area is composed of two
economic activities, agriculture and mining. Both are
related to the envirOnmental constraints of altitude and
distance. In order to obtain greater precision in future
studies it may be useful to refine these phySical variables,
since they may well influence the migration process.

In addition, the Antofagasta to Bolivia railway has
been shown to be directly associated with movements to
Calama. Some mining sites and railway stations located
along this transportation route were affected by this
linkage in that supply a greater number of migrants than
originally expected. Further investigations might wish to
determine the relationship between accessibility to the

rail line and migrant flows.

 

104

Data Quality to Predict Migration
In conclusion, future research might also consider
the importance of data quality. Data averages could induce
error. Also, if one wishes to predict migration, the data
should allow for a time component. More reliable data
would enable the researcher to better understand the changes

within the migration system among several time periods.

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Books

Abler, R.; Adams, J.: and Gould, P. Spatial Organization.
Prentice-Hall, 1971.

Ackoff, Russell L. Scientific Method: Optimizing Applied
Research Decisions. New York: John Wiley & Sons,
1962.

 

Amiran, Solon. E1 Desarrollo del Norte Arido de Chile.
CorporaciOn de la ProducciOH, (CORFO), Santiago,
Chile, 1959.

 

Barraclough, Solon. Notas sobre la Tenencia de la Tierra.
Instituto de CapacitaciOn e InvestigaciOn en Reforma
Agraria, (ICIRA), Santiago, 1971.

Billinghurst, G. Estudio sobre la Geografia de Tarapaca.
Santiago, Chile, 1886.

. La Irrigacién de Tarapaca. Santiago, Chile, 1893.

Bogue, Donald J. Principles of Demography. New York: John
Wiley & Sons, 1969.

 

Bruggen, Juan. Fundamentos de la Geologia de Chile.
Santiago: Instituto Geografico Militar, Chile, 1950.

Butland, G. J. Chile. An Outline of Its Geography. London,
1969.

Clarke, John I. Population Geography. New York: Pergamon
International Library, 1972.

Corporacion de Fomento de la Produccidn (CORFO). Geografia
Economica de Chile. Primer Apéhdice, Santiago, 1966.

Dollot, Louis. Les Migrations Humaines. Paris: Presses Uni-
versitaires de France, 1965.

 

105

106

Elizaga, Juan C. Migraciones a las Areas Metropolitanas de
America Latina. Centro Latinoamericano de Demogra-
fia, Santiago, 1969.

. Formas de Asentamiento de la Poblacién en America
Latina. Centro Latinoamericano de Demograffa,
Santiago, 1969.

 

Harvey, David W. Explanation in Geography. London: Edward
Arnold, 1969.

Hauser, Philip, and Duncan, Otis (Eds.). The Study of
POpulation. Chicago: The University of Chicago
Press, 1959.

 

Isard, Walter. Methods of Regional Analysis. An Intro-
duction to Regional Science. New York, 1960.

 

Lloyd, Peter E., and Dicken, Peter. Location in Space: A
Theoretical Approach to Economic Geography.
Wheaton, & Co., 1977.

 

 

Mattelart, Armand. Atlas Social de las Comunas de Chile.
Santiago, 1966.

Morales E., and Portilla, Belfor. Demanda de Alimentos Pere-
ceferos para Tarapaca'and Antofagasta'en Relacidn a
ggs Recursos Productivos Agropecuarios. Santiago,
Corporacidh de Fomento de la Produccidh. Depto.
de Desarrollo Agricola, Facultad de Agronomia, U.
de Chile, 1968.

Navarro, Isolda. Seminario de Problemas y Desarrollo Regio-
nales de Antofagasta. Santiago: Edicién del Depto.
de ExtensiOn Cultural de la Universidad de Chile,
1957.

Nie, Norma H., 25 a1. Statistical Package for the Social
Sciences. McGraw-Hill Book Company, 1970.

 

Plandes. El Hombre en la Zona Arida del Norte de Chile.
La Serena: Primeras Jornadas de Estudios de las
zonas Aridas del Norte Chileno. Santiago, 1970.

Pressat, Roland. Statistical Demography. Methieu & Co.
Ltd., 1978.

Sadie, John L. Poblacién y Mano de Obra de Chile: 1930-1975.
Santiago: Centro Latinoamericano de Demografia, 1969.

Saez, R. Chile_y e1 Cobre. Santiago: Departamento del
Cobre, 1965.

107

Sandami, Ghulam Mohammad. Migration and Modernization.
New York: Center for Urban Development, Cornell
University, 1972.

Sauvy, Alfred. Teoriade la PoblaciOn. Madrid: Editorial
Aguilar, 1967.

Shryock, Henry S., and Siegel, Jacob S. The Methods and
Materials of Demography. U.S. Bureau of the Census,
vol. 1, 1975.

 

. The Methods and Materials of Demography. U.S.
Bureau of the Census, Vol. 2, 1975.

Spiegelman, Mortimer. Introduction to Demography. Cambridge,
Mass.: Harvard University Press, 1968.

Taylor, Peter. Quantipative Methods in Geography. Houghton-
Miffin Co., 1977.

 

Thomas, Robert N., and Hunter, John M. (Eds.). Internal
Migration Systems in the Developing WOrld. Mass.:
Schenkam Publishing Co., 1980.

Trewartha, Glenn. A Geography of Population: World Patterns.
New York: John Wiley & Sons, Inc., 1969.

Udo, Reuben K. Migrant Tenant Farmers of Nigeria. Nigeria:
African Universities Press, 1975.

White, Paul, and WOods, Robert (Eds.). The Geographical

Impact of Migration. London: Longman Group Limited,
1980.

 

Yeates, Maurice. An Introduction to Quantitative Analysis
in Geograp_y. McGraw-Hill series in geography, 1974.

Zelinsky, Wilbur. A Prologue to Population Geography.
Englewood Cliffs, N.J.: Prentice-Hall Publishers,
Inc., 1966.

Zemelman, Hugo. E1 Migrante Rural. Santiago: Instituto de
Investigacién y Capacitaci6n en Reforma Agraria
(ICIRA), 1971.

Zipf, George. Human Behavior and the Principle of Least
Effort. Cambridge, Mass.: Addison-Wesley Press
Inc., 1949.

108

Articles

Ackerman, Edward A. "Geography and Demography." In Philip
Hauser and Otis D. Duncan (Eds.), The Study of Pop-
ulation: An Inventory and Appraisal. Chicago:
University of Chicago Press, 1959, pp. 717-727.

 

Adams, Dale W. "Rural Migration and Agricultural Develop-
ment in Colombia." Economic Develppment and Cultural
Change, V01. 17 (1969), 527-557.

Peru: Un Inventario de Proposiciones." Estudios de _
Poblacion y Desarrollo, V01. 1, No. 4, Serie Original I
No. 2, 1968. ’

 

Alers, J. 0., and Appelbaum, R. P. "La Migracion en el E

Allen, James P. "Changes in the American Propensity to
Migrate." Annalspgf the Association ofgphe American
Geographers, Vol. 67, No. 4 (December 1977), 577-
587.

 

m

Aranda, Ximena. "San Pedro de Atacama: Elementos Diagnosti-
cos para un Plan de Desarrollo Local." Informa-
ciones Geograficas, U. de Chile, V01. 11, No. 14,
Santiago, 1961-1964, pp. 19-61.

Bachi, Roberto. "Geostatical Analysis of Internal Migra-
tion." Journal of Regional Science, V01. 16, No. l
(1976).

Bahr, Jfirgen. "Migration im Groben Norden Chiles." Bonner
Geographische Abhandlungen, Heft 50, 1975.

Beals, Ralph E.; Levi, Mildred B.; and MOses, Leon N.
"Rationality and Migration in Ghana." Review of
Economics and Statistics, VOl. 49, No. 4 (1967),
480-486.

Bedford, R. D. "New Hebridean Mobility: A Study of Circular
Migration." Research School of Pacific Studies.
Australian National University, Camberra, Dept. of
Human Geography, 1973.

Beshers, J. M., and Nishiura, E. "A Theory of Internal
Migration Differentials." Social Forces, Vol. 39
(1961).

 

Bock, Wilbur E., and Iutaka, Sugiyama. "Rural-Urban Migra-
tion and Social Mobility: The Controversy on Latin
America." Rural Sociology, VOl. 34, No. 3 (1969),
343-355.

109

Bogue, D. J., and Thompson, W. S. "Migration and Distance."
American Sociological Review, Vol. 14 (1949), 236-244.

Boulding, K. "Toward a General Theory of Growth." General
Systems Yearbook, V01. 1 (1956): pp. 66-75.

Bowman, J. "Desert Trails of Atacama." American Geographi-
cal Sociepy, Spec. Pub., No. 5, 1924.

 

 

Brown, Lawrence A., and Holmes, John. "Intra-Urban Migrant
Lifelines: A Spatial View." In Demography, VOl. 8,
No. 1 (February 1971), 103-122.

Browning, Harley L., and Feindt, waultraut. "Patterns of
Migration to MOnterrey, Mexico." International
Migration Review, VOl. 50, No. 3 (1971), 309-324.

 

. "Selectividad de Migrantes a una Metrépoli en un
Pais en Desarrollo: Estudio de un Caso Mexicano.“
Demografia y Economia, Vol. 3 (1969), 186-200.

Brun, Stanley D., and Thomas, Robert N. "The Migration
System of Tegucigalpa, Honduras." Robert N. Thomas
(Ed.), Population Dynamics of Latin America. A
Review and Bibliography. Boston, Mass.: The Con-
ference of Latin Americanist Geographers, 1971,
pp. 63-82.

Burford, R. L "An Index of Distance as Related to Internal

Migration." Southern Economic Journal, Vol. 29
(1962), 77-81.

Claeson, C. F. "Distance and Human Interaction: Review and
Discussion of a Series of Geographic Model Building."
Geografiska Annaler, Series B, No. 2, 1968.

Collier, P., and Green, J. M. "Migration from Rural Areas
of Developing Countries: A Socioeconomic Approach."
Oxford Bulletin of Economics and Statistic, V01. 40,
No. 1 (1978). PP. 23-35.

Concha, M. "Establecimientos Humanos en el Altiplano
Chileno." Estudios Geogréficos. Santiago, Chile,
1966, pp. 55-82.

Connell, John, £3.2l- "Migration from Rural Areas: The
Evidence from Village Studies." Delhi, Oxford
University Press, 1976.

Conning, A. M. "Estimacién de la Migracién Interna Neta,
Clasificada por Edad y por Sexo en las Provincias y
Regiones de Chile durante los afios: 1930-1940, 1940-
1952 y 1952-1960." Celade, Serie D, No. 36, Santiago,
1966.

110

. "Rural Urban Destinations of Migrants and Com-
munity Differentiation in a Rural Region of Chile.
Interpational Migration Review, Vol. 6, No. 2 (1972),
148-157.

Conway, Dennis. "Step-wise Migration: Toward a Clarification
of the Mechanism." International Migration Review,
Vol. 14 (1969).

Cummings, F. M. "Internal Migration and Regional Develop-
ment Planning: Thailand, The Philippines, Indonesia."
Journal of Tropical Geography, Vol. 41 (1975).

Ducoff, L. F. "The Role of Migration in the Demographic
Development of Latin America." Milbank Memprial
Fund Quarteriy, VOl. XLIII, No. 4 (1965), 197-216.

Elizaga, Juan C. "Internal Migrations in Latin America."
Milbank Memorial Fund Quarterly, VOl. 43, Part 2
(1965), 144-161.

 

"Internal Migration: An Overview." International
Migration Review, Vol. 6, No. 2 (1972), 121-146.

 

Flinn, William L. "Rural to Urban Migration: A Colombian
Case." R.P. Land Tenure Center. University of
Wisconsin, No. 19, 1966.

Gauthier, Howard L. "Migration Theory and the Brazilian
Experience." Revista Geogréfica, VOl. 82 (1975),
pp. 51-61.

Germani, Gino. "The Process of Urbanization in Argentina."
Paper presented to the Seminar on Urbanization

Problems in Latin America, UNESCO, Santiago, Chile,
1958.

. "Migration and Acculturation." Philip M. Hauser
(Ed.). Handbook for Social Research in Urban Areas.
Belgium: Unesco, 1964.

Goddard, A. D.; Gould, W. T. S.; and Masser, F. J. "Census
Data and Migration Analysis in Tropical Africa."
Geografiska Annaler, Series B, Human Geography,
Vol. 57, No. 1 (1975). PP. 26-41.

Grigg, D. B. "E. G. Ravenstein and the 'Laws of Migration.'"
Journal of Historical Geography, Vol. 3 (1977).

Gupta, H. S. "Marital Migration in the Rural Chhattisgarh."
Geographical Review of India, VOl. 39, No. 2 (1977),
pp. 126-133.

111

Hagerstrand, Torsten. "Migration and Area. Survey of a
Sample of Swedish Fields and Hypothetical Consider-
ations on Their Genesis.” Lund Studies in Geography,
Series B, Human Geography, No. 13 (1953).

Hart, R. A. ”The Economic Influences on Internal Labor
Force Migration." Scottish_Journa1 of Political
Economy, V01. 5, No. 19 (1972).

Hartshorne, Richard. "On the Nature of Geography." The
Association of the American Geographers, VOl. XXIX
(1939).

Herberle, R. "The Causes of Rural-Urban Migration: A Survey
of German Theories." American Journal of Sociology,
VOl. 43 (1938).

Herold, Joan M. "Female Migration in Chile: Types of MOves
and Socioeconomic Characteristic." Demography,
VOl. 16, No. 2 (1979), pp. 257-277.

Hobbs, Alfred H. "Differentials in Internal Migration."
The University of Pennsylvania Press, 1942.

Iyoba, M. A. "A Theory of Rural-Urban Migration in Developing
Countries." Nigerian Journal of Economic and Social
Studies, Vol. 17, No. 2 (1975), 29-39.

Jones, Richard T. "E1 Potencial Latente Migratorio entre
una RegiOn Deprimida y Destinos Alternativos: Un
Ejemplo Venezolano." Revista Geografica. Mérida,
.Venezuela, V01. 14, No. 1-2 (1973), 127-138.

Kemper, Robert V. "Rural-Urban Migration in Latin America:
A Framework for the Comparative Analysis of Geogra-
phical and Temporal Patterns." International Migra-
tion Review, VOl. 5, No. 1 (1971), 36-47.

 

Lee, Everett. "A Theory of Migration." Demography, Vol. 3,
No. l (1966), 47-57.

Li, wen L. "Matrix Analysis of Migration Streams." Inter-
national Migration, VOl. 8, No. 4 (1970). pP. 174-
181.

Light, Mary, and Richard. "Atacama Revisited 'Desert
Trails.'" The Geographical Review, V01. 36, No. 4
(1946), 525-545.

 

Mabogunje, Akin L. "Systems Approach to a Theory of Rural-
Urban Migration." 'GeographiCal Analysis, Vol. 2,
No. 1 (1970), 1-18.

 

112

MacDonald John, and MacDonald Leatrice. "Chain Migration,
Ethnic Neighborhood Formation and Social Networks."
SociallResearch, Vol. 29 (1962).

Mangalam, J. J., and Schwarzweller, Harry K. "General
Theory in the Study of Migration: Current Needs and
Difficulties." Internatignal Migration Review,
VOl. 3, No. 1 (1968), 3-17.

Miller, Edward. "Return and Nonreturn In-migration."
Growth and Change, V01. 4, No. 1 (1973): PP- 3-9.

Morel, A. "L'espace social d'un village Picard." .Estudes
Rurales, VOl. 45 (1972).

Norris, Robert E. "Migration as Spatial Interaction."
Journal of Geography, VOl. LXXI, No. 5 (1972),
294-301.

 

Olsson, Gunnar. "Distance and Human Interaction: A Review
and Bibliography." Bibliography Series No. 2.
Philadelphia: Regional Science Research Institute,
1965.

Parr, John. "Outmigration and the Depressed Area Problem."
Land Economics, V01. 5, No. 42 (1966), pp. 149-159.

Petersen, W. "A General Typology of Migration." American
Sociology Review, Vol. 23 (1958), 256-266.

 

Poole, M. A., and O'Farrell, P. N. "The Assumptions of the
Linear Regression Model." Transactions of I.B.G.,
Vol. 52 (1971), pp. 145-158.

 

Ravenstein, E. G. "The Laws of Migration." Journal 9f the
Royal Statistical Society, V01. 48 (1885), 167-227.

 

Rees, Teresa L., and Rees, Gareth. "Alternatives to the
Census: The Example of Sources of Internal Migration
Data." Town Planning Review, Vol. 48, No. 2 (1977),
123-140.

Riddell, J. Barry. "On Structuring a Migration Model."
Geographical Analysis, VOl. 2 (1970), 403-409.

. "Regression Analysis of Aggregate Migration Data
in the Third World. " International Rggional Science
Review, VOl. 12, No. 2 (1972), pp. 195- 212.

, and Harvey, Milton E. "The Urban System in the
Migration Process: An Evaluation of Step-Wise
Migration in Sierra Leone." Economic Geography,
VOl. 48, No. 3 (1972).

 

 

113

Rudolph, W. E. "Antofagasta to Salta Railroad." Geographi-
cal Review, V01. 17 (1927), 553-585.

 

. "Chuquicamata: Twenty Years Later." Geographical
Review, V01. 41 (1951), 88-113.

. "Sulphur in Chile." Geographical Review, Vol. 42
(1952), 565-590.

. "Vanishing Trails of Atacama." American Geogra-
phical Sociepy, Res. Ser. No. 24 (1963).

 

. "Water Supply Problems of a Desert Region."
American Society of Civil Engineers, V01. 94 (1930),
600-625.

Rummel, R. J. "Understanding Factor Analysis." Journal of
Conflict Resolution, V01. 42, No. 4 (1967), 444-480.

Rasheed, Sajjadur K. B. "Depopulation of the Cases in
Northern Chile." Revista Geografica, IPGH, V01. 74
(1971), pp. 101-113.

Schnell, G. A. "Geographic Association between Migration
and Selected Aspects of Age Structure." Proceeding
of the Pennsylvania Academy of Science, Vol. 43
(1969). PP. 153-156.

Schwartz, Aba. "Interpreting the Effect of Distance on
Migration." Journal of Political Economy, Vol. 81,
No. 5 (1973), 1153-1169.

"On Efficiency of Migration." Journal of Human
Resources, VOl. 6, No. 2 (1971), 193-205.

 

Shaw, Paul R. "A Note on Cost-return Calculations and
Decisions to Migrate." Population Studies, Vol. 28,
No. 1 (1974). PP. 167-169.

Simmons, Alan B., and Cardona, Ramiro G. "Rural-Urban
Migration: Who Comes, Who Stays, Who Returns? The
Case of Bogota? Colombia, 1929-1968." International
Migration Review, VOl. 6, No. 2 (1972).

Singh, Jasbir. "A Reconnaissance Analysis of Net-Migration
and the Tendencies of Migration of Rural Population
in Hissar District, 1951 to 1961 (Haryana-India)."
The National Geographical Journal of India, Vol. 15,
No. 3-4 (1969), 155-166.

 

Sjaastas, Larry. "The Costs and Returns of Human Migration."
Journal of Political Economy, V01. 70, No. 5 (1962),
80-93.

 

A I-_‘I

114

Skeldon, Ronald. "The Evolution of Migration Patterns during
Urbanization in Peru." The Geographical Review,
V01. 67, No. 4 (1977), 394-411.

 

Speare, Alden. "Home Ownership, Life Cycle Stage, and
Residential Mobility." Demography, Vol. 7 (1970).

Sterstein, Larry. "Migration to and from Bangkok." Annals
of the Association of American Geographers, Vol. 64,
No. 1 (1974). pp. 138-147.

Stevens, R. P. "Spatial Aspects of Internal Migration in
Mexico, 1950-1960." Revista Geografica, V01. 69
(1968). pp. 75—90.

 

Stewart, Norman R.; Belote, Jim; and Belote, Linda. "Trans-
humance in the Central Andes." Annals of the Associ-
ationgf American Geographers, V01. 66, No. 3 (1976),
377-397. '

 

Stoltman, Joseph P., and Ball, John. "Migration and the
Local Factor in Rural Mexico." Journal of thg
Society for Applied Antropology, Vol. 30 (1971).

 

Stouffer, S. A. "Intervening Opportunities: A Theory Re-
lating Mobility and Distance." American Sociology
Review, V01. 5 (1940), 845-867.

 

. "Intervening Opportunities and Competing Migrants."
Journal Regional Sociology, V01. 2 (1960), 1-26.

 

Swindell, Kenneth. "Sierra Leonean Mining Migrants, Their
Composition and Origins." Institute_9f British
Geographers-Transactions, VOl. 62 (1974).

 

Thomas, Dorothy. "Research Memorandum on Migration Differ-
entials." Bulletin 43. New York: Social Science
Research Council, 1938.

Thomas, Robert N. "Population Bibliography of Latin America."
Robert N. Thomas (Ed.), Population Dynamics of Latin
America. A Review and Bibliography. Boston, Mass.:
The Conference of Latin Americanist Geographers,

1971, pp. 122-200.

 

, and Catau, John G. "Distance and the Incidence
of Step-Wise Migration in Guatemala." Proceedings
of the Assoc. of American Geographers. VOl. 6 (1974).

 

 

, and Mulvihill, James L. "Temporal Attributes of
Stage Migration in Guatemala." Internal Migration
System. Robert Thomas and John Hunter (Eds.), 1980.

 

115

Ullman, Edward. "Human Geography and Area Research." Annals
of the Association of American Geographers, VOl. 43
(1953).

Uzzeil, Douglas. "The Myth of Bipolar Migration." Ekistic.
Athenes, Greece, V01. 44, No. 263 (1977), pp.
216-219.

Vanderkamp, J. "Return Migration: Its Significance and
Behavior." Western Economic Journal, VOl. X, No. 4
(1972), 460-465.

. "Migration Flows, Their Determinants and the
Effects of Return Migration." Journal of Political
Economy, V01. 79, No. 5 (1971), 1012-1031.

Wendel B. "A Migration Schema: Theories and Observations."
Lund Studies in Geography, Serie B, Human Geography,
V01. 9 (1953)

Whetten, N. L., and Burnight. "Internal Migration in
Mexico." Rural Sociology, Vol. 21 (1956).

 

White, Stephen. "A PhiloSOphical Dichotomy in Migration
Research." The Professional Geographer, V01. 32,
No. 1 (1980).

Willis, K. "Regression Models of Migration." Geografiska
Annaler, Series B, No. 57 (1975), pp. 42- 54.

Wolpert, Julian. "Behavioral Aspects of the Decision to
Migrate." Papers of the Regional Science Associa-
tion, VOl.15,No. 15 (1965), PP. 159- 169.

Zelinsky, Wilbur. "The Hypothesis of Mobility Transition."
Geographical Review, Vol. 61, No. 2 (1971), 219-249.

 

Government Repprts and Documents

Espildora, C. Basilio. "Informe sobre un Reconocimiento
Preliminar del Rio Loa del Rio Zapaleri." Univer-
sidad de Chile, Mayo, 1968.

 

Oficina de Planificacidn Nacional (ODEPLAN) y Carabineros
de Chile, "Antecedentes sobre Pueblos del Interior,
Provincia de Antofagasta. MacroZona Norte. Infor-
maciones entregadas por Carabineros de Chile a1 30
de Septiembre de 1972.

 

116

Theses

Croner, Charles M. "Spatial Characteristics of Internal
Migration to San Pedro Sula, Honduras." Unpublished
Ph.D. dissertation, Michigan State University, 1972.

Khairul Bashar, Sajjadur Rasheed. "Man and the Desert in
Northern Chile." Unpublished Ph.D. dissertation,
Columbia University, 1970.

Stephens, John D. "Migration and Distance: A Case Study of

Urban In-migration in Guatemala." Master's thesis
Indiana University of Pennsylvania, 1970.

Other Sources

 

Personal interview with the Presidents of villages, 1970,
1978.

Personal interview with Mr. Mario Farandato, Director Ser-
vicio Agricola Ganadero (SAG) Antofagasta, 1976.

APPENDICES

APPENDIX A

QUESTIONNAIRE, 1970-1978

 

APPENDIX A

QUESTIONNAIRE, 1970-1978

 

 

 

Lugar Encuestado: Fecha:
Encuestado: (Posicién en el grupo fa-
miliar).

Composicién del Grupo Familiar y Edades ()

 

Esposo Esposa h i j o s O t r o 5 Total
Hombres Mujeres

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lugar de Origen: Lugares Intermediog:
1 2 3 4 5 6
Esposo: /7( ) ( T (T ) ( 7) ( ‘T ( )
Esposa: . //( ) ( ) ( ) ( ) ( ) ( )
Hijos : //( ) ( ) ( ) ( ) 1, ) ( )
/7( ) ( ) ( ) ( ) ( ) ( )
77( ) ( ) ( ) ( ) ( ) ( 1)
/7( ) ( ) ( ) ( ) ( ) ( )
Otros : /7( ) ( ) ( ) ( ) ( ) ( )

 

 

 

//( ) ( ) ( ) ( ) ( ) j )

Actividades gue desarrollan: ponga () e1 lugar y no. de horas

 

 

 

 

 

Esposo: / ( ) no.
IEsposa: .- ,, __g ] ( . g ) no.
Hijos : >/ ( ) no.

 

117

'118

 

 

 

 

 

 

 

/ ( ) no.
7 ( ) no.
Otros : / ( ) no.
if ( ) no.

 

Emigrados Temporales:

 

 

No. Sexo Salidas a1 mes Duracidn Lugar(es) Causas

 

 

 

 

 

 

Emigrados Definitivos:

 

 

No. Sexo Afio de salida Edad Lugar Causas

 

 

 

 

 

 

APPENDIX B

VARIABLES

MIGRAT
LABORF
HOUSES
NONOWN
RETURNMI
ACTIVAGR

ACTIVMIN

 

ACTIVFAC

ACTIVSER
LOCMOV
DISTMOV
COMMOV
CITYMOV
AGRECO

ALTITUDE

 

TRAVTI

MOBILITY

 

WORKHOUR

 

SERVICNU

APPENDIX B

VARIABLES

Annual migration rate

Labor force rate

Uninhabited houses rate

Non-owner operator rate

Percentage of return migrants

Percentage of active population in agriculture
Percentage of active population in mines
Percentage of active population in small
tories

Percentage of active population in services
Percentage of local movements

Percentage of districtal movements
Percentage of communal movements

Percentage of movements to the city

Mean farm size

Altitude in meters (a.s.1.)

Travel time

MObility coefficient

Average of daily working hours

Number of services in each settlement

119

NTRUCKS

TRAVDI
MARKET

ILLITER

120

Number of trucks (main transport mode)
Travel distance (Km)
Index of market volume

Illiteracy rate