THE INFLUENCE OF FAMILY IN THE
MIGRATION PROCESS

Thesis for the Degree of M. A.
MICHIGAN STATE UNIVERSITY
KAMUTONDO KAWALA SIMWANZA
1969

 

 

 

 

 

"1? V 3"
‘9 -

n“- .‘ao I .

 

 

 

  
  

g > _ I, ,
BINDING BY u ‘

IIDAB & SIIIIS'I i" I

800K BINDERY IIIC.
LIBRARY IINDEHS
”IMHO". IICIII‘II
t

   
  
     

IVIHUIC 2

M 51’: 134,198
a: I If

 

 

 

 

 

 

 

ABSTRACT

THE INFLUENCE OF FAMILY IN THE
MIGRATION PROCESS

BY

Kamutondo Kawala Simwanza

This thesis focuses on two inter-related prob-
lems: (l) the importance of kinship in the migration
process, and <E:>the level of attachment to family as a
motivating factor in the individual's migration deci-
sions and behavior.

Two sets of data are utilized in the present
study. Data were collected at two points in time by
means of questionnaires. The initial data were collec-
ted in the spring of 1957 from a sample of 269 high
school juniors and seniors in Ontonagon County, a remote
rural area in Michigan's Upper Peninsula. After a 10-
year lapse, this group was restudied, starting in the
spring of 1968. Data were collected by means of a mail-
back questionnaire.

Based on the 1957 project data, attachment to
family was measured by assigning scores to the subjects'
responses on statements expressing attitudes toward

their relatives. The objective was to examine the

Kamutondo Kawala Simwanza

possible relationship of migration behavior, as revealed
in the 1968 questionnaires, to the level of attachment to
family exhibited in 1957.

It was hypothesized that persons with strong at-
tachment to family would be more influenced by relatives
in their migration plans than would persons with weak
family ties. Results showed that:

l. Non-migrants had a slightly higher index score
on attachment to family than migrants.

2. The choice by migrants of destinations where
relatives were present or absent was related to the level
of family ties.

3. Migrants' perception of the effect of rela-
tives in their choice of destination was related to the
level of attachment to family.

4. The decision by migrants to return "home" was
related to the level of attachment to family.

5. Migrants' estimate of chances of returning
"home" was possibly related to the level of family ties.

These individual relationships between migrants'
behavior and level of family ties were not so strong as to
meet statistical tests of significance, but all were in
the direction predicted. As to the basic proposition that
family plays a significant role in migration, the data

strongly suggests that family was generally influential.

THE INFLUENCE OF FAMILY IN THE

MIGRATION PROCESS

BY

Kamutondo Kawala Simwanza

A THESIS

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

MASTER OF ARTS

Department of Sociology

1969

ACKNOWLEDGEMENT

This thesis would not have been completed without
the invaluable help of many individuals. I wish to ex-
press my sincere appreciation to the members of my Guid-
ance Committee for their interest and constructive criti-
cism: Dr. J. Allan Beegle, Chairman, Dr. Denton E.
Morrison and Dr. Ruth S. Hamilton. Special thanks to Dr.
Beegle who offered me patient and insightful guidance in
the development and completion of this work, and to Jon
H. Rieger for coding, encouragement, and his willingness
to follow through on various aspects. Emily Gibson also
deserves special thanks for typing the preparatory draft

of this thesis.

ii

TABLE OF CONTENTS

Chapter Page
I. INTRODUCTION . . . . . . . . . . . . . . . 1
Background of the Problem . . . . . . . 1
The Relevance of Reference Group
Theory to Migration . . . . . . . . . 3
Literature Review . . . . . . . . . . . 12
II. THE PROBLEM, METHODS AND HYPOTHESES . . . 22
The Problem Area . . . . . . . . . . . . 22
The Significance of the Problem . . . . 27
Conceptual Statement of the
Hypotheses . . . . . . . . . . . . . . 28
Methodology . . . . . . . . . . . . . . 29
The sample 0 O O O O O O O O O O O O 29
Variables . . . . . . . . . . . . . 30
Uncontrolled Variables . . . . . . . 33
Operational Statement of the
Hypotheses . . . . . . . . . . . . . . 34
Method of Analysis . . . . . . . . . 35
III. HYPOTHESIS-TESTING AND RESULTS . . . . . . 36
Level of Attachment to Family and
the Migration Process . . . . . . . . 36
summary 0 O O O O O I I O O O O O O I O 43
IV. DISCUSSION AND CONCLUSION . . . . . . . . 45
Introduction . . . . . . . . . . . . . . 45
Evaluation of the Results . . . . . . . 45
Conclusion and Recommendations for
Further Research . . . . . . . . . . . 51
BIBLIOGRAPHY O O O O O O O O O O O O O . Q 0 O O O O 52
APPENDIX 0 O O O O O O O I O ,0 O O O I O O O O O O 55

iii

Table

1.

LIST OF TABLES

Page

Migration by Level of Attachment to
Family 0 O O O O O O O O O O O O O O O O O 37

Reasons for the Initial Move Out of
Ontonagon County by sex, in Rank
order I O O O O I O O O O O O O O O 0 O O 3 8

Destinations with Relatives Present or
Absent by Level of Attachment to
Family, for Persons who Migrated for
Reasons Other Than Military Service
or Education . . . . . . . . . . . . . . . 38

Effect of Relatives on Choice of Initial
Destination by Level of Attachment to
Family; Migrants with Relatives at
Destination who Migrated for Reasons
Other Than Military Service or
Education . . . . . . . . . . . . . . . . 40

Effect of Relatives on Choice of Initial
Destination by Level of Attachment to
Family; Migrants with Relatives at
Destination who Migrated for Reasons
Other Than Military Service or Educa-
tion, and who Later Returned to
Ontonagon County . . . . . . . . . . . . . 41

Return/non-Return to Ontonagon County
by Level of Attachment to Family . . . . . 42

Migrants' Estimates of the Chances of

Returning to Ontonagon County by
Level of Attachment to Family . . . . . . 43

iv

CHAPTER I
INTRODUCTION
Background of the Problem

Many studies of migration focus on the economic
factors associated with migration behaviors and tend to
play down the role of non—economic factors such as kinship
and friendship networks.1 This tendency may have been
fostered by some of the views of human behavior which were
once common among social scientists. The validity of some
of these views has either already been challenged or is
now being questioned.

According to Bott, for instance, it is still a
common belief among some sociologists and anthropologists
that in industrial societies kinship does not play a very

2

important role except in certain rural areas. Comple-

mentary to this view of industrial societies is another

 

1For a summary of various economic theories of
Inigration, see: Brinley Thomas, Migration and Economic
Growth, A Study of Great Britain andithe Atlantic Economy,
LEhdbn: Cambridge University Press, 1954.

2Elizabeth Bott, Familygand Social Networks,
London: Tavistock Publications Ltd., 1957, p. 115.

idea, once common among sociologists, that America has be-
come a "mass society" in which human behavior is largely
influenced by mass media. The validity of this idea,
according to Rogers, was first challenged by Lazerfeld's
1940 Presidential study which showed that personal influ-
ence was still very important in major decision-making.
Implicit in these views is the idea that the con-

cept of the primary group, so popular in sociological

 

textbooks, does not have fruitful application to the anal-
ysis of many types of behavior in industrial societies.
The view that the family does not play an important role
in industrial societies is now being rehevaluated. Evi-
dence from recent sociological research on the family in
industrial societies has shown that, despite lack of close
geographical proximity, the extended family persists in
industrial settings as exemplified by continued interaction
of kinfolk of different generations, the existence of
strong, affective ties among family members and the per-
formance of various services for one another. A study by
.Axelrod has revealed intense interaction among relatives

in Detroit, irrespective of social class.4 Wilmont and

 

3E. M. Rogers, Diffusion of Innovations, New York:
The Free Press, 1962, p. 211.

4Morris Axelrod, "Urban Structure and Social Par-
ticipation", American Sociological Review (XXI), 1966, pp.
13-18.

Young have found elaborate and intense interaction among
kin in English urban areas.5 Gans observed the same
phenomenon among Italians in a Boston slum'area.6 And
Sussman demonstrates that middle class parents continue
helping their children even after they are married.7 A1—
together, the evidence indicates that primary groups do
play important roles in the lives of people in industrial
societies. This being the case, studies of migration be-
havior should take into consideration the possible role
that non-economic factors such as kinship and friendship
networks play in migration decisions.

The Relevance of Reference Group Theory
to Migration

 

 

In this study of the migration process, we shall
use the level of attachment to family as the explanatory
variable in the individual's migratory behavior. Attach-
ment here refers to the feelings one has towards his rela-
tives including his perception of the attitudes of rela-

tives toward himself. Attachment in this sense may be

 

5M. Young and Peter Wilmont, Family and Kinship
in East London, London: Routledge and Paul Kegan, 1957,
pp. 57-68.

6H. Gans, The Urban Villagers, New York: Free
Press of Glencoe, 1962, pp. 45-46.

 

 

 

7Marvin Sussman, "The Help Patterns in the Middle
Class Family." American Sociological Review, (18), 1953,
pp. 22-28.

viewed as a social psychological phenomenon experienced by
an individual as a result of association with relatives.
It is hypothesized that such feelings of attachment have
an influence on the individual's behavior in respect to
his relatives.

The theoretical frame of reference for using levels
of attachment to family as predictors of migration behav-
ior is that of "reference group theory." The concept

reference group, first introduced by Hyman, has been used

 

by Merton, Newcomb, and other social psychologists to de-
note a group to which individuals refer as they make deci-
sions and take action. This view suggests that individuals
act partly in relation to role expectations of those groups
that are important to them.

The term reference group, as Kelley8 points out,
refers to more than one group and can also be used to des-
cribe two kinds of relationships between a person and a
group. In one sense, the term denotes a group in respect
to which one is motivated to gain or maintain acceptance.
Such a group sets and reinforces certain group norms or

standards. An example of this application is discussed by

 

8Harold Kelley, "Two Functions of Reference
Groups," in G. E. Swanson et al, (ed), Readings in Social
Psychology, 1952.

 

 

Merton.9 The setting and reinforcement of standards is
referred to by Kelley as the "normative function" of
reference groups. The other usage of the term is in the
situation in which an indiviuaal refers to a group in eval-
uating himself and others. This was Hyman's original ap-
plication of the term. This evaluative use of the group,
Kelley termed the "comparison function." Kelley has
pointed out that both these functions of reference groups
can be performed by a single group.

In any community the family and close friends are
perhaps the most important reference groups as well as
agents of socialization,10 and can be seen to perform both
functions discussed above. These primary groups not only
shape one's attitudes, which in turn affect one's behavior,
but they also give an individual psychological and material
support. Through the process of socialization and beyond,
an individual comes to rely on his primary groups not only
for his interpretation of reality but also for certain

material needs. The dependency of individuals on their

 

9R. Merton and A. Kitt, "Contribution to the
Theory of Reference Group Behavior" in R. Merton and Paul
Lazerfeld (eds.), Continuities in Social Research Studies
in the Sco e and Methods ofiflThe AmericanISroldierflr
Glencoe: T e FreeFress, 1950, pp. 40-105.

loJ. Riley and Matilda White Riley, "Mass Com-
munication and the Social System," in R. Merton, et a1.
Sociology Today, New York: Basic Books, Inc., 1962, pp.
537-578.

 

primary groups should be expected to be greater during
periods of stress, adversity, or major decision making.

The reference groups do tend to maintain them-
selves by creating some internal cohesion. Such social
cohesion is created by a combination of various factors
operating within and outside the group. Making members
conform to certain group norms is one of the ways by which
group cohesion is achieved. This conformity is maintained
through a combination of various factors such as coercion,
ridicule, and the development of friendly feelings among
group members. It is mainly through such "normative
functions" of the reference group that individuals develop
group attitudes.

Since an individual's attitudes are partly
anchored in his normative reference groups, and because
of his dependency for material and psychological needs on
such groups, especially those of the primary nature, an
individual who becomes involved in major decision-making
usually evaluates the meaning and possible consequences of
alternatives in terms of what he thinks is in line with
his reference group. The decision to migrate can be con-
sidered a "major decision" and individuals should be ex-
pected to refer strongly to their primary groups in making
such decisions.

Comparative reference groups may also influence

the migration process. This may happen when individuals

who are evaluating themselves against others in their own
primary groups or against those in groups in which they
are not members, feel that they are relatively worse off
than their desired units of reference. Such comparisons
may stimulate aspirations which can not be fulfilled with-
in the aspirants' community for certain reasons, thus
leading individuals to migrate to other places where they
hope to acquire the necessary things for the satisfaction
of their aspirations. .
In most cases, the most aspired for thing may be
social recognition among one's reference groups. Many
people will, for instance, work hard to get a certain job,
not so much because its monetary rewards are higher than
those of other jobs, nor because they would enjoy the job
as an end itself, but because such a job carries with it
high social status. Even in cases where people want a job
because of its high monetary rewards, such people may not
be happy with such rewards unless they can be translated
into symbols of prestige through conspicuous consumption.
The realization of these aspirations is viewed by
the individuals as enhancing their standing in the eyes
of their reference groups. The View that reference groups
do play a significant role in influencing people's ambi-
tions is perhaps succinctly expressed by Cooley, who says:

"the individual will be ambitious, but the chief object

of his ambition will be some desired place in the thought
of others."11
As indicated earlier, family and friends are per-
haps the most important references when individuals make
major decisions such as that to migrate or not migrate.
The idea that reference groups are important in the migra-
tion process has received some support from empirical
studies. Beegle and Goldsmith, on the basis of their
findings in a study on the "initial phase of voluntary
migration," suggested that individuals with high attach-
ment to their families would be more likely to migrate if
they receive support from them than would individuals with
high attachments but without support.12 Crawford tested
this hypothesis on a sample of rural youth from New York
and found that although high attachment to family can in-
hibit migration, such inhibition can be overcome if the
individual is given support by his family in his decisions

to migrate.13

 

llCharles Cooley, "The Primacy of Primary Groups,"
in Rosenberg, (ed.) Mass Society in Crisis, Social Prob-
lems and Social Pathology, New York: Macmillan, 1964,
p. 70.

 

 

I
12J. A. Beegle and H. Goldsmith, "The Initial
Phase of Voluntary Migration,” East Lansing: Michigan
Agricultural Experiment Station, Rural Sociology Studies
No. l, 1962, p. 11.

13C. Crawford, "Family Attachment, Family Support
for Migration and Migration Plans of Young People," Rural
SociolQQY: (31), 1966, pp. 293-300.

 

Friendly feelings among members of a group, such
as a family, are some of the means by which it maintains
itself as a unit. The stronger such feelings are, the
more cohesive it is likely to be and the more difficult
for the members to leave. In the rural environment, with

its presumably stronger Gemeinschaft, the influence of

 

family and friends in the migration decisions of the in-
dividual should be expected to be high, because in such
communities, neighborhood and kinship relationships tend
to be the most meaningful interaction patterns and also
familistic norms tend to be the most important mechanisms
of social control.14

In sum, it would appear that reference group
theory is relevant to the analysis of migration behavior
because the decision to migrate is a major decision in
which one's friends and relatives play an important role,
and because the carrying out of the actual migration act
may very well have direct impact on one's relatives and
friends.

One of the complexities involved in this analysis
is the multiplicity of reference groups to which an in-

dividual refers at any single point in time. These groups,

as indicated earlier, may consist of those in which one

 

14Ferdinand Tonnies, Community and Society:
Gemeinschaft und Gesellschaft, translated by C. P. Loomis,
New York: Harper and Row, 1963.

 

 

10

holds membership and those of which he is not a member.
This multiplicity of groups increases the difficulty of
identifying and measuring the effect of reference groups
in the analysis of the migration process. In this study,
we shall attempt to evaluate the role of the individual's
family in influencing his migration.

In view of what has been suggested in this theory
about the influence of reference groups in migration,
plus the facts brought out in the literature review below
on the role of family in the migration process, one may
raise a basic question: if an individual aspires to
achieve a certain social position and in the process mi-
grates to get education or work which he hopes will help
him get the desired goal, on which of these factors should
we put emphasis as the explanation of the migration? Is
it the aspiration to the ultimate goal (social position)
or the immediate purpose (education or employment)? The
position taken here is that while the immediate purpose
should be taken into consideration, emphasis should be
placed on the long range goal. Placing emphasis on the
former would seem to have neither logical nor emperical
validity. It would explain the cause of an action by em-
phasizing what may be merely the means used to carry out
such action to the exclusion of the ends sought. If
human behavior is goal-motivated, as much evidence makes

it appear, then it is the "ends" which should be treated

11

as the principal factor. Analyses of migration which em-
phasize only immediate economic factors and ignore or play
down complex social and psychological factors like long
term motivations and aspirations, are vulnerable to cri-
ticism on such grounds. An example of this is Gugler's
recent study, (1968) of the migration process in Africa.15
Gugler maintains that African rural-urban migration is
mainly affected by economic conditions and very little by
non-economic factors.

This error of misplaced emphasis is easily com—
mitted in migration studies probably because of the obvi-
ousness of economic factors in the migration process and
perhaps because of the type of questions researchers ask
migrants. If a migrant is asked why he came to a certain
place, he will most likely give a very short response
such as, "to look for work," "to go to school," etc. Very
seldom does the migrant say "to look for work so that I
may build myself a better house than the one I have now."
A broader set of questions asking a person why he decided
to seek more education or the better job should give us a
better picture of the complexity of the migration process

which some researchers have simplified by viewing it as a

simple response to economic conditions.

 

15Josef Gugler, "The Impact of Labor Migration on
Society and Economy in Subsahara Africa. Empirical Find-
ings and Theoretical Considerations." University of Zam-
bia Institute of Social Research. Nos. 6; Dec. 1968.

 

 

12

Literature Review

 

In the sociological literature on the relationship
of industrialization to group membership, especially as
exemplified in urban areas, there is a traditional View
which emphasizes impersonality of relationships in the in-
dustrial communities, greater importance of formal and
secondary group associations and the decline of the kin-
ship group. This view of urban life owes much to the work

of the "Chicago School."16

As indicated earlier, Bott
maintains that there are many sociologists and anthropolo-
gists who think that in industrial societies the importance
of kinship has declined.17
This tendency to play down the importance of pri—
mary group associations in industrial settings is also
observable in migration literature. Contemporary research
on migration tends to ignore the importance of primary
groups in the migration process. This tendency, exhibited
by researchers in developed nations, is also observable in
some research on migration in the developing nations.18

Where primary groups appear in migration literature, they

tend to be treated as the dependent variables. Migration

 

16Morris Axelrod, gp. cit., p. 13.

17Elizabeth Bott, gp. cit., p. 115.

18Josef Gugler, 9p. cit.

13

is generally viewed that way in developing nations today
and is pictured as having devastating effects on family
and friends.19 Burgess' theory, for instance, suggests
that migration detaches and disorganizes individuals and
societies, and, according to Brown and Tilley, this belief
has been accepted without question.20
Some of the traditional conceptions of the rela-
tionship of migration and primary groups are being ques-
tioned. Recent research is also beginning to treat family
and friends not as passive reactors to the effects of
migration, but rather as active participants or explana-
tory variables in the migration process. Findings from
these studies indicate that family and friends do play a
much more important role in facilitating or inhibiting
migration than is commonly realized.21
The newer view that emerges from recent litera-
ture on the relationship of migration to family is that,
although the immediate effects of migration on the family

seem to indicate family disorganization, the long range

effects imply a transformation of the family roles to suit

 

19W. J. Goode, World Revolution and Family Pat-
terns, New York: Free Press, 1963.

20Charles Tilley and H. Brown, "On Uprooting,-Kin-
ship,and the Auspices of Migration?"; International Journal
of Comparative Sociology, (8), Sept. 1967, p. 139.

21H. Schwarzweller and J. Brown, "Social Class
Origin, Rural Urban Migration and Life Chances," Rural
Sociology, (32), 1967, pp. 5-19.

 

 

 

14

the contemporary industrial-urban settings. 'The idea that
migration transforms, rather than destroys, the family and
kin groups has been strongly advocated by Litwak who main-
tains that a new type of extended family called the RESET
fied extended family does exist in contemporary industrial-
urban societies.22
Family and friends are very influential in the
migrant's choice of destination. Research has indicated
that although individuals may decide to move for economic
reasons, their initial destination tends to be to a place
where relatives or friends or both are already established
rather than merely to a place where economic opportunities

happen to be objectively good.23

Perhaps one of the
oldest studies on the influence of social relations in the
migrants choice of destination was published by Park in
1925. Park observed that migrants from EurOpe and Asia,
on arrival in the United States, tended to cluster in cer-
tain places and "reproduce spontaneously the home com-

munity." Commenting on the Italians, Park wrote:

Among the more important immigrant groups, the
Italians show perhaps the strongest wish to remain in

 

22Eugen Litwak, "Geographic Mobility and the Ex-
tended Family COhesion," American Sociological Review,
(25), 1960, pp. 385-394.

23H. K. Schwarzweller, "Socio-Cultural Origins
and Migration Patterns of Young Men from Eastern Ken-
tucky," University of Kentucky Agricultural Experiment
Station Bulletin 685, Dec. 1963, p. 23.

15

solitary communities. They settle here by villages
and even by streegi, neighbors in Italy tending to
become neighbors.

Park suggested that this behavior by Italian mi-
grants could be attributed to the Italians' "intense
family pride" in Italy. The family includes those re-
lated by blood and ritual bonds. Because of "very strong
family loyalty," migration tended to be under the aus-
pices of kin with the subsequent tendency by migrants to
cluster in their areas of destination.

A recent study by the McDonalds on the role of
social networks in chain migration and ethnic neighborhood
formation reported results similar to those of Park.25
McDonald found that most Italians who came to the United
States between 1880 and 1914 tended to form "Little
Italies," clusterings of fellow Italian townsmen in the
U.S. cities. This formation of "Little Italies" and
"chain employment of Italians" was attributed to the fact
that most migrants were involved in chain migration which
he defines as a "movement in which prospective migrants

learn of opportunities, are provided with transportation

and have initial accommodation and employment arranged

 

24Robert'Park, Old World_Traits Transplanted,
Chicago: University of Chicago Press, 1925, p. I46.

25John and Leatrice McDonald, "Chain Migration,
Ethnic Neighborhood Formation and Social Network,"
Milbank Memorial Fund Quarterly, 42 (1964), pp. 82-97.

 

 

16

by means of primary social relationships with previous
migrants."26
The role of social networks in facilitating inter-
national migration has been observed among other national—
ities as well. Kiser reported that friends and relatives
were a positive influence for Negroes migrating from St.
Helena Islands to Harlem and other U. S. urban centers.27
Litwak also notes that the existence of the extended
family encouraged migration from Ireland during the potato
famine in the 1840's.28
Recent research by sociologists on migration to
cities has indicated that, in the U. 8., family and friends
play a powerful part among the "auspices" of migration.
Tilley and Brown's study indicates that for most migrants
in Wilmington, Delaware, the initial move was under the
auspices of family first, friends second and work third.29
Evidence from Choldin's study in Chicago indicates that
kinsmen were involved in the migration in three ways:

(a) most migrants decided to move into Chicago after kin

had done so; (b) migrants travelled in the company of

 

261bid., p. 82.

27Clyde Kiser, Sea Island to City, A Study of St.
Helena Islanders in Harlem and other Urban Centers, New
York: Columbia University Press, 1952.

28

 

Eugen Litwak, gp. cit., p. 387.

29Charles Tilley and H. Brown, 92. cit.

17

other relatives; and (c) they were received by kinfolk who
had preceded them in the migration.30
The influence of family in directing migrants to
certain locations seems to be more important than it is
commonly assumed. Bell and Blumberg reported that about
two thirds of the Negroes in their sample said that they
came to Philadelphia because relatives were already
there.31 Omari also found that nearly all Negroes studied
in Bolt had relatives there before migrating.32 And
Killian reported a "clustering" of Southern white migrants
in Chicago's ethnically heterogeneous neighborhood.33
Clustering of migrants in place of destination has
also been found by Brown, 25 31., in a study of the migra-
tion patterns out of three Kentucky neighborhoods with
strong family ties. This study showed that family ties
generally guide migrants to certain areas and also aid in

the migrants' adjustment in these new areas.34

 

30Harvey M. Choldin, "Kinship Networks in the Migra-
tion Process," unpublished article, Sociology Department,
Michigan State University, East Lansing, 1968.

31L. Blumberg and Robert Bell, "Urban Migration and
Kinship Ties," Social Problems, XI, (1959), p. 330.

32

 

Ibid., p. 331.

33Lewis Killian, "The Adjustment of Southern White
Migrants to Northern Urban Norms," Social Forces, XXXII,

34James Brown, H. K. Schwarzweller, and J. J.
Mangalam, "Kentucky Mountain Migration and the Stem Fam-
ily: An American Variation of a Theme by Le Play," Rural
Sociology, (28), March 1963, pp. 48-69.

 

 

18

Schwarzweller and Brown, in their study of Kentucky youth,
reported that instead of the youth migrating into Ohio's
northern industrial cities which have better job oppor-
tunities, migrants tended to cluster around the less in-
dustrialized areas of Dayton, Middletown, and Cincinnati.
The two researchers suggested that this clustering could
be an indication of the importance of non-economic factors
such as kinship and friendship ties in determining the
destination of the migrants.35
In the literature reviewed above, it has been sug-
gested that relatives and friends are influential in the
migrants' choice of destination. Evidence of this influ-
ence is found in what some migrants have said and also in
the clusters of migrants in the places of destination.
Once kinship networks have directed the migrants
to the destinations where other relatives are already es—
tablished, these networks continue to operate in various
ways, helping the migrants in their re-settlement phase.
The findings by Choldin indicate that some migrants to
Chicago were met at the bus, train stations, and airport
by relatives and friends. Relatives and friends also pro-
vided the migrants with the initial accommodations, infor-

mation on jobs and housing, and generally helped them make

 

35H. Schwarzweller, gp. cit., p. 23.

19

36 McDonald reported a similar

new social connections.
phenomenon among Italian migrants into the U. S. For some
of these migrants, even their transportation was financed
by friends and relatives who had migrated earlier or by
the Padroni, or "God-father."37
The already established migrant families and
friends indirectly aid the newcomer in his adjustment to
the community by providing him with a social environment
which combines the customs of his home and those of the
host community. This is especially true in cases where
migrants form their own colonies such as in the "Little
Italies." In such colonies migrants tend to develop group
consciousness. Killian notes for instance that Southern
white migrants in Chicago had a "defensive group conscious-
ness." They tended to keep to themselves, owned their own
"hillbilly tavern--a place to get together with the boys
. . . [and] . . . to meet the girls" and also showed evi-
dence of the persistence of Southern norms as exmplified
by their praise of racial discrimination-against blacks.38
This tendency for migrants to interact with people

who are socially similar to them has also been observed

by Jitodai in his study of the participation patterns of

 

36Harvey M. Choldin, gp. cit., pp. 12-13.

37John and Leatrice McDonald, gp, cit., p. 87.

38Lewis Killian, 22, cit., pp. 66-69.

20

Detroit migrants.39 This study indicated that most mi-
grants, irrespective of their social class, when compared
with non-migrants, tended to participate more in informal
organizations involving relatives and friends than in
formal organizations.

The literature reviewed above indicates some of
the different ways in which family and friends influence
the migration process. Although it has been argued that
friends and relatives do perform important services in the
adjustment of new migrants, this should not be taken as
evidence that migrants who are received by either friends
or relatives or both, are necessarily happier or do better
than those without such receptions.

Some of the available evidence indicates that mi-
grants who are received by someone in the city are much
slower than those without any such receptions in making
new acquaintances and also in finding jobs. Evidence from
Choldin's Chicago study also indicates that migrants with
friends or relatives and those who received a lot of help
from such groups tended to show more signs of ”homesickness"
than the group without such privileges.40

Choldin's findings can be interpreted in such a

way as to suggest that persons with strong attachments to

 

39Ted Jitodai, "Migration and Kinship Contacts,"
Pacific Sociological Review, Fall, 1963, pp. 49-54.

40Harvey M. Choldin, QB! cit., p. 18.

21

family and friends will tend to migrate under the auspices
of such groups and will also tend to receive more help
from such groups than persons with weak attachments. Once
they are in a new environment, whether with or without any
members of their primary groups, persons with strong at-
tachments to primary groups left at home will tend to ex-
perience more "homesickness" than persons with weak
attachments.

This review of literature is, of course, not
exhaustive, but, for the purpose of this study the works
reviewed above indicate the type of available evidence
pointing to the role played by the family in the migration

process.

CHAPTER II

THE PROBLEM, METHODS AND HYPOTHESES

The Problem Area

 

This study focuses on the influence of kinship in
the migration process. The term "migration" has numerous
meanings. We think of migration as involving at least
some spatial mobility, and immediately we are confronted
with the question of how much distance people must move in
order to be classified as migrants. When migration is
viewed as a movement of people from one community to
another, this raises the question of the meaning of the
term "community," which, according to Hillery, has had at
least ninety-four different definitions.1 Sociologically
speaking, spatial movement of people has been called mi-
gration if it involves the severance of some meaningful
interaction between the people who have moved and those
who are left behind.

Since the way one conceptualizes and defines mi-

gration affects the way he studies it, and thereby his

 

1George A. Hillery, Jr., "Definitions of Commun-
ity," Rural Sociology, XX (June, 1955), p. 111.

 

22

23

findings, it is important that a common definition be
sought so that the results of different investigators are
comparable. The definition of the term migration used in
this study is that given by Beegle and Goldsmith, who
have conceptualized migration as
. . . any relatively permanent spatial change of
residence an actor (or a set of actors) makes that
necessitates the severance of day-to-day contacts with
members of concrete interaction systems. Thus migra-
tion is to be regarded as the movement of individuals
beyond and outside their interaction systems of a
given community of residence.2
In this study we attempt to test propositions
bearing on the issue that kin are influential in (a) the
individual's decision whether or not to migrate; (b) the
migrant's choice of destination; and (c) the migrant's
subsequent decision whether or not to return "home." It
is recognized, of course, that factors such as kingroup
size, location, and cohesiveness may contribute to varia-
tions in the pattern of influence. In this study we in-
tend to examine only the effects of professed familial at-
tachments on the part of the individual in the migration
process.

The attempt will be made to find out whether or

not high attachment to family inhibits or facilitites

 

2J. A. Beegle and H. Goldsmith, "The Initial
Phase of Voluntary Migration," East Lansing: Michigan
Agricultural Experiment Station, Rural Sociology Studies
No. l, 1962, p. 11.

24

migration. High attachment to family may inhibit migra-
tion when the individual with such feelings towards his
significant relatives finds it hard to leave them or if
his intentions to migrate are discouraged by such rela-
tives. High attachment to family may encourage migration
if the individual's intention to migrate is supported by
the significant members of his family. It should be re-
called that Crawford, in his study of New York rural youth,
found that the inhibiting effects of high attachment can
be overcome by support of the family for the act of migra-
tion.3

The inhibiting effects of family ties can also be
eclipsed by other conditions such as a call to military
service or a plan to acquire college or higher education
when the institutions of higher learning are located out-
side one's own community. It would appear that under
these conditions the effect of attachment would be lost
under the force of more powerful considerations of manda—
tory service in the army, or the unavailability of educa-
tional facilities in the home community, etc.

We shall also investigate the influence of rela-
tives in the migrant's choice of destination. As indi-

cated in the review of the literature, empirical evidence

 

3C. Crawford, "Family Attachment, Family Support
for Migration and Migration Plans of the Young People,"
Rural Sociology, 31, 1966, pp. 293-300.

 

/"

25

on the destination of migrants has shown that migrants
tend to go to places where their kin are already estab-
lished. Evidence of this phenomenon is found in Tilley
and Brown's study of Wilmington (Delaware) migrants,4
Schwarzweller's study of migration patterns of Kentucky
youth,5 and McDonald's study of Italian immigrants into
the United States.6

The above studies reported clusterings of migrants
in certain locations or destinations. Although this clus-
tering of migrants has been interpreted as an indication
of the influence of family and other non-economic factors
influencing the migrants' choice of destination, little
has been done to determine the factors within the family
system that lead migrants to those places where relatives
are located. It is the aim of this study to show that

relatives do influence the migrants' choice of destination

and that the level of attachment to family is directly

 

related to the migrants' tendency to choose destinations

where relatives reside. This will be tested by relating

 

4C. Tilley and H. Brown, "On Uprooting,Kinship,
and the Auspices of Migration?", International Journal of
Comparative Sociology, (8), Sept.—l967, p. I39.

5H. Schwarzweller and J. Brown, "Social Class
Origin, Rural Urban Migration and Life Changes," Rural
Sociology, (32), 1967), pp. 5-19.

6John and Leatrice McDonald, "Chain Migration,
Ethnic Neighborhood Formation and Social Network,"
Milbank Memorial Fund Quarterly, 42 (1964), pp. 82—97.

 

 

 

26

the migrants' choice of destination on their initial moves
to level of attachment to family as measured prior to the
act of migration.

It may be noted that a migrant may go to a place
where relatives reside because of emotional attachment, or
because he temporarily needs their help. Both cases can
be interpreted as indicating the influence of family in
directing the migrant to those places having relatives.
However, the factors motivating the migrant are somewhat
different in the two cases. In the one case the attach-
ment factor plays the significant part in the migrant's
decision and may be viewed as the primary influence. In
the other case the migrant is forced by circumstances to
seek refuge among his relatives, and the attachment may be
viewed as secondary.

It appears likely that the influence of the family
group does not operate in the same way on the migrant's
decision to return "home" as it sometimes does in the case
of out-migration. The returning migrant does not presum-
ably depend, in most cases, on family for his material
needs, yet the family group does affect his decision to
return "home."

It is suggested that family members have influence
on the migrant's decision to return through such psycho-
logical mechanisms as family sentiments and the general

nostalgia for the people and places one used to know before

27

migrating. It would be expected that persons with high
attachments to family are the ones most likely to experi-
ence deep sentiments for kin and the home community in
general, and hence will exhibit a greater tendency to re-

turn "home."

The Significance of the Problem

 

There are several factors that can lead an indi—
vidual to migrate or prevent him from migrating, one of
these factors being the individual's level of attachment
to the family. We propose here that migration carried out
under the influence of high attachment to family is most
likely to result in kin migrating together or going to
those places where other kin are already established.

Such migration is a case of "chain migration," which the
McDonalds contend is a major cause of ethnic or kinship
neighborhood formation...7

It is further suggested that when migration occurs
under the influence of "good" kinship relationships, the
resulting geographical separation may not necessarily lead
to the breakdown of the extended family. The fact that
family members are spatially apart does not necessarily
lead to social separation. For instance, members of a

family who are geographically isolated may maintain strong

 

7J. McDonald and L. McDonald, 2p. cit., p. 90.

28

ties as exemplified by regular visits, mutual aid, and
regular communication by mail or phone. Such a family may
be thought of as "closer" than the family which resides

in the same location but is characterized by conflict,
refusal of mutual aid, and lack of communication.

The significance of studying the influence of
family relations in the migration process rests on the
proposition that an understanding of the role family plays,
as a reference group, may also help our understanding of
such phenomena as chain migration, persistence of the ex-
tended family in industrial-urban areas, and return migra—

tion.

Conceptual Statement of the Hypotheses

 

Hypothesis 1

 

Migrants out of the community of origin will
exhibit a level of attachment to family not dis-
similar to that of non—migrants.

Hypothesis 2

 

Migrants who choose a destination where their
relatives reside will exhibit higher attachment
to family than will migrants who choose destina-
tions where there are no relatives.

Hypothesis 3

 

Migrants who perceive the influence of having rela-
tives at destination as significant in their choice
of initial destination will exhibit a higher at-
tachment to family than will migrants who perceive
the influence of having relatives at destination as
not significant.

29

Hypothesis 4

 

Migrants who return to the community of origin
after migrating will exhibit a higher attachment
to family than will permanent migrants.

Hypothesis 5

 

Migrants who estimate their chances of returning
to the community of origin favorably will exhibit
a higher attachment to family than will migrants
who estimate their chances of returning unfavor-
ably.

Methodology

 

The Sample

 

The site for this study is Ontonagon County, an
economically depressed rural area in the Upper Peninsula
of Michigan. This area was selected by J. A. Beegle and
H. Goldsmith for a 1957 study dealing with the "initial
phase of voluntary migration." This study was done allow-
ing for the possibility of longitudinal restudies of the
same sample. The current Ontonagon County project is being
carried out under the direction of J. A. Beegle and J. H.
Reiger of Michigan State University who have collected the
data for a first decennial restudy of the original 1957
sample.

Data for the initial study were collected in the
spring of 1957 by means of a questionnaire administered
to all high school juniors and seniors in Ontonagon County,

resulting in a total sample of 269 subjects. Data for the

30

restudy has been collected by means of mail-back ques-
tionnaires.

In analyzing the influence of family in migration
behavior, we are using information from both 1957 and 1968
projects. The two questionnaires are the main sources of

data for this study.

Variables

 

1. Level of attachment to family.-—The measure of

 

the level of attachment to family was derived from the May
1957 questionnaire. Selected descriptive statements allow-
ing for responses in five categories from strong agreement
to strong disagreement were used. These were taken from a
belief check list concerning the subject's evaluation of
(1) his community subsequent to graduation from high
school; and (2) his relationship with his parents. Below
are the three statements which were used as an index of
the level of attachment to family. Scoring on these
statements was ordered in such a way that a higher score
always indicates higher attachment. The lowest possible

combined score was three and the highest was fifteen.

Attachment to Family Statements and Their Scoring:

Statement 1. There is real love and affection for me at

 

home.
Response Alternatives Score
Strongly disagree 1

Disagree 2

31

Undecided
Agree

Strongly agree
No response

Kimbw

Statement 2. As I have known it, family life is happy.

 

Strongly agree
No response

Response Alternatives ‘ Score
Strongly disagree 1
Disagree 2
Undecided 3
Agree 4

5

Y

Statement 3. After graduation your community will be a
good place to remain close to your rela-

 

Strongly agree
No response

tives.
Response Alternatives Score
Strongly disagree 1
Disagree 2
Undecided 3
Agree 4

5

Y

This study focused on five aspects of the migra-
tion process thought to be related to the level of attach-
ment to family. Each was treated as a variable and
measurement was by means of information gleaned from the
1968 restudy questionnaire.

2. Migration.-—This was determined by inspection

 

of the residence history provided by respondents in 1968.
Persons who had ever left Ontonagon County for one month
or more during the ten year period were categorized as
migrants and those who had not as non—migrants.

3. Initial destination where relatives reside.--

This was determined by an item in the 1968 questionnaire

32

asking whether the migrants had ". . . relatives living in
or near this place (or who were thinking of moving here)

at the time [they] were considering moving here." Respond-
ents who answered "yes" to this question were classed as
having an initial destination where relatives resided.

4. Influence of relatives on choice of initial

 

destination.--This was measured by a contingency item in

 

the 1968 questionnaire which elicited, for those who
answered "yes" to the question "did you have . . . rela-
tives living in or near this place . . .", the extent of
influence of this factor on the decision to go to that
place. The question asked was ". . . how much do you
think that fact affected your decision to move to this
place? Would you say it had: [ ] a major effect?

[ ] some effect? [ ] little or no effect?"

5. Return migration.-—This was measured by inspec-

 

tion of the residence history provided by respondents in
1968. Persons who had ever returned to live in Ontonagon
County during the ten year period, regardless of present
residence, were treated as return migrants.

6. Estimate of the chance of returning to

Ontonagon County.--This was measured for those persons

 

whose current address was outside the county by an item
in the 1968 questionnaire inquiring into the possibility
of such a return. The question was as follows: "How

would you estimate the chance that you will move back to

33

Ontonagon County? [ ] better than 75%; [ ] 50 to 75%;

[ ] about 50-50%; [ ] 25 to 50%; [ ] less than 25%.

Uncontrolled Variables

 

It is recognized that differences in the cultural,
social and economic backgrounds may affect the behavior
of individuals in the migration process. However, most
subjects in the sample seem to have come from similar
backgrounds. Two hundred sixteen of the 269 subjects
lived with both their own parents. An additional 42 lived
with one parent or a combination of one parent and other
relatives. Only 1 did not live with either parent, but
with other relatives such as an aunt or uncle.

The subjects' estimates of their parents' annual
income in 1957 was between 3000 and 5999 dollars for 107
subjects, less than 3000 dollars for 50 and more than
4999 dollars for 26, while the remainder failed to provide
this information. A comparison of the estimated 1957 in-
come of parents of the return and permanent migrants re-
vealed no significant differences, which indicated a lack
of relationship between return migration and parental
economic position. About 56% of the returnmigrants had
estimated their parents' income in 1957 to be less than
$5,000 while 57% of the permanent migrants had estimated
their parents' income to be in the same category. Approx-
imately 21% of the permanent migrants had estimated their

parents' 1957 income to be $5,000 or more.

34

There are other uncontrolled variables in this
study that there is reason to believe are directly related
to migration behavior. Among these are (1) sizes of
families, (2) ages of parents, (3) migration history of
the family including length of residence in Ontonagon
County and (4) the types of new families formed by the

subjects through marriage.

Operational Statement of the Hypotheses

Hypothesis 1

 

Migrants out of Ontonagon County will exhibit a
mean index score on attachment to family not
significantly different from that of non-migrants.

Hypothesis 2

 

Migrants who, on their initial move out of
Ontonagon County, choose destinations where rela-
tives reside will exhibit a higher mean index
score on attachment to family than will migrants
who choose destinations where there are no rela-
tives.

Hypothesis 3

 

Migrants who indicate that having relatives at
destination had "some effect" or a "major effect"
in their choice of initial destination will ex-
hibit a higher mean index score on attachment to
family than will migrants who indicate that having
relatives at destination had "little or no effect."

Hypothesis 4

 

Migrants returning to live in Ontonagon County will
exhibit a higher mean index score on attachment to
family than will permanent migrants.

35

Hypothesis 5

 

Migrants who estimate their chances of returning
to live in Ontonagon County at 50% or above will
exhibit a higher mean index score on attachment to
family than will migrants who estimate their
chances of returning at less than 50%.

Method of Analysis

 

In the initial analysis’of data, contingency tables
were constructed showing proportions of migrants and non-
migrants, permanent and return migrants etc., exhibiting
high and low levels of attachment to family as measured by
a ten item index. Chi-square tests of significance did
not reveal any differences between the groups being com-
pared. This analysis is presented in Appendix A.

A feeling that X2 analysis was perhaps not sensi-
tive enough led us to try the difference-of-means test (t
test). We were also aware that although using many items
normally improves the reliability of“a measurement, it may
also increase its "impurity." For the t test, therefore,
it was decided to select from the original ten item at-
tachment to family index three items which were thought
to most precisely reflect the subjects' feelings towards
their families. We then compared the three-item mean
index scores of migrants and non—migrants, migrants with
relatives at destination and those without any relatives,
return and permanent migrants, and so on. "Results dis-
cussed in Chapter III are based on computation of these

mean index scores and the use of t tests.

CHAPTER III
HYPOTHESIS-TESTING AND RESULTS

Level of Attachment t9 Family and
The Migration Process

 

This chapter reports the results of testing the
five hypotheses on the significance of attachment to
family as a factor in migration. The following format
will be used: (1) a statement of the hypothesis, and (2)
a descriptive analysis of the results. Further discussion
of the results will be presented in Chapter IV.

In general, the data have supported one hypothesis
and are at least consistent with the other four without
being strong enough to individually satisfy the statis-
tical tests of significance.

Hypothesis 1: Migrants out of Ontonagon County will ex-
hibit a mean index score on attachment to
family not significantly different from
that of non-migrants.

The results, as shown in Table 1, support the
above hypothesis. (See page 37 for Table 1.)

Although non-migrants exhibited a slightly higher

mean index score than migrants on attachment to family,

this difference was not statistically significant. The

36

37

TABLE l.--Migration by Level of Attachment to Family

 

 

Migration Status Number Percent Level of Attachment
Mean Score

 

Migrants 234 88.97 10.99
Non-migrants 29 11.03 11.55
Total 263* 100.00 11.02

 

*Because there were six persons on whom there was
no information beyond that available from the 1957 ques-
tionnaire, there are 263 subjects in the total instead of
269.
total sample mean index score was 11.02. The score for
migrants was 10.99 while that for non-migrants was 11.55.
Hypothesis 2: Migrants who, on their initial move out of

Ontonagon County, choose destinations where
relatives reside will exhibit a higher mean
index score on attachment to family than
will migrants who choose destinations where
there are no relatives.

In analyzing the data on the role of family in
directing migrants to places where relatives are located,
we excluded persons who migrated for military or educa-
tional reasons on the grounds that the presence or ab-
sence of relatives at the place of destination probably
has relatively little relevance to these situations.

Table 2 shows that 102 persons left for military or

educational reasons, of which only 21 claimed to have had

relatives at destination.

38

TABLE 2.--Reasons for the Initial Move Out of Ontonagon
County by sex, in Rank Order

: m

 

 

 

 

Reason for Men Reason for Women

Moving Number Percent Moving Number Percent
Work Work

Related 39 33.6 Related 40 33.9
Education 35 30.2 Family 37 31.4
Military 33 28.5 Education 31 26.3
Family 8 6.9 Military 3 2.5
Other 1 .8 Other 7 5.9
Total 116 100.0 Total 118 100.0

 

Results bearing on Hypothesis 2 are presented in

Table 3. The expected relationship between the level of

TABLE 3.—-Destinations with Relatives Present or Absent
by Level of Attachment to Family, for Persons
who Migrated for Reasons Other Than Military
Service or Education

 

 

:-' f
I _7 -

Relatives Present/Absent Number Percent Level of Attach—
at Destination ‘ ment--Mean Score

 

Relatives present at

destination 81 62.3 11.19
No relatives at

destination 49 37.7 10.71
Total 130*. 100.0 11.01

 

*Because two persons failed to give information
concerning presence or absence of relatives at destination,
the total sample here is less than the 132 indicated in
Table 2.

39

attachment to family and migrants' choice of destination
where relatives are present is supported in the sense that
although the difference is not statistically significant,
it is at least in the direction of the hypothesis. Mi-
grants with relatives at destination exhibited a slightly
higher mean index score on attachment to family than mi-
grants without relatives at destination (11.19 compared
to 10.77).
Hypothesis 3: Migrants who indicate that having rela-
tives at destination had "some effect" or
a "major effect" in their choice of initial
destination will exhibit a higher mean in-
dex score on attachment to family than will
migrants who indicate that having rela-
tives at destination had "little or no
effect."
Hypothesis 3 was meant to complement Hypothesis 2
in elucidating the role of attachment to family as a
factor influencing migrants to select destinations where
other kinsmen are located. As with Hypothesis 2, the re-
sults provide support at least in direction. Data in
Table 4 show that persons who felt that relatives had had
a "major effect" in their choice of destination exhibited
a mean index score of 11.25 on attachment to family, which
was slightly higher than the means of 11.10 and 11.08
respectively for persons who felt that relatives had had
only "some effect" or "little or no effect."

Of this total of 81 persons who migrated to places

where relatives were located, it was incidentally noted

40

TABLE 4.--Effect of Relatives on Choice of Initial Des-
tination by Level of Attachment to Family; Mi-
grants with Relatives at Destination who Mi-
grated for Reasons Other Than Military Service
or Education

-‘
J

Effect of Relatives on Number Percent Level of Attachment

 

Choice of Destination Mean Score
Major effect 44 54.3 11.25
Some effect 25 30.9 11.10
Little or no effect 12 11.8 11.08
Total 81 100.0 11.19

 

that 33 moved back to Ontonagon County at some later time.
We thought it would be interesting, parenthetically, to
examine the level of attachment to family of this subgroup
as it related to the effect of relatives on the choice of
initial destination out of Ontonagon County. Results are
reported in Table 5.

As can be seen from the table, there is a differ-
ence of 1.71 between the mean index scores of those per-
sons who felt that relatives had had a "major effect" on
their choice of initial destination and those who felt
that relatives had had "little or no effect." The means

difference is significant at the .005 level.

41

TABLE 5.--Effect of Relatives on Choice of Initial Des-
tination by Level of Attachment to Family; Mi-
grants with Relatives at Destination who Mi-
grated for Reasons Other Than Military Service
or Education, and who Later Returned to
Ontonagon County

Effect of Relatives on Number Percent Level of Attachment

 

Choice of Destination Mean Score
Major effect 14 42.4 12.21
Some effect 13 39.4 10.92
Little or no effect 6 18.2 10.50

Total 33 100.0 11.39

 

Hypothesis 4: Migrants returning to live in Ontonagon
County will exhibit a higher mean index
score on attachment to family than will
permanent migrants.

The expected relationship between the level of
attachment to family and the migrants' propensity to re-
turn to Ontonagon County is supported in that the differ-
ence between the means is in the direction predicted.

As can be seen from Table 6, persons who had come back to

live in the county exhibited a mean index score on attach-

ment to family of 11.10, which is slightly higher than

that of 10.87 for the permanent migrants. The difference

was not statistically significant.

42

TABLE 6.--Return/non-Return to Ontonagon County by Level
of Attachment to Family

 

Migration Status Number Percent Level of Attachment
Mean Score

 

Return migrants 124 53.0 11.10
Permanent migrants 110 47.0 10.87
Total 234 100.0 10.99

 

Hypothesis 5: Migrants who estimate their chances of re-
turning to live in Ontonagon County at 50%
or above will exhibit a higher mean index
score on attachment to family than will
migrants who estimate their chances of re-
turning at less than 50%.

Hypothesis 5 applies to the 152 persons who were
living outside Ontonagon County at the time they responded
to the 1968 re-study questionnaire. On the expected rela-
tionship between the level of attachment to family and mi-
grants' estimate of the probability of their returning to
live in Ontonagon County, the results lend suggestive sup-
port. Table 7 shows that the 15 persons who estimated
their chances of returning at 50% or above had a mean in-
dex score of 11.0 on attachment to family compared to
10.90 for the 122 persons who estimated their chances of
returning at below 25 percent. But the persons who esti-

mated their chances of returning to be between 25 and 50

percent had the highest mean score, 11.30. The small

43

TABLE 7.--Migrants' Estimates of the Chances of Returning
to Ontonagon County by Level of Attachment to

 

 

Family
Estimated Changes Level of Attachment
of Returning Number Percent Mean Score

to Ontonagon County

 

50% and above 15 9.9 11.00
25 to 49% 15 9.9 11.30
Below 25% 122 80.2 10.90
Total 152 100.0 10.95

 

number of persons in the upper categories and the statis-
tically insignificant differences between the mean index
scores of the groups render this interpretation highly

speculative, however.

Summary

Using data collected from a sample of Ontonagon
County 1957 and 1958 high school graduates at the time
they were still in school and ten years thereafter, this
study has sought to (1) demonstrate that kinship plays an
important role in the migration process, and-(2) investi-
gate specifically the "level of attachment to family" as
a factor in the migration decisions and behavior of rural

youth.

44

The results in this study have demonstrated that
kinship did play an important role in the migration of
Ontonagon County youth. On the importance of attachment
to family as it relates to various strategic migration
decisions by individuals, the results are highly sugges-
tive. Although the hypotheses have individually re-
ceived weak support in the sense that differences between
means were not statistically significant, the fact that
the differences were consistently in the direction pre-
dicted is highly encouraging.

Results showed that almost two thirds (63%) of the
migrants who initially moved out of Ontonagon County for
reasons other than military service or education went to
places where relatives resided. And about 85 percent of
these persons claimed that relatives were influential in
their choice of destination. Migrants who had relatives
at destination had also a higher mean index score on at-
tachment to family than those without relatives. Return
migrants also exhibited a higher mean index score than did
permanent migrants.

The next chapter further discusses these results
and some of the possible reasons for the outcomes observed,
as well as the general issue of family influence in the

migration process.

CHAPTER IV

DISCUSSION AND CONCLUSION

Introduction

 

In this chapter we shall discuss the results of
testing the hypotheses and attempt to evaluate the sig-
nificance of family influence in the migration process.
Finally, a few suggestions will be offered for future re-
search on the role of attachment to family as a factor

in the migration behavior of rural youth.

Evaluation of the Results

In the original design of this study, it was in-
tended that the data be subjected to Chi-square analysis,
results of which are presented in Appendix A. The Chi-
square analysis supported Hypothesis 1 which stated that
migration out of Ontonagon County was unrelated to the
level of attachment to family. In the other four hypoth-
eses, the results were generally consistent with the
hypotheses but not statistically significant. The out-
comes of the t tests, reported in Chapter III, were very

similar to those based on X2 tests. Again the data were

45

46

consistent with the hypotheses but differences were not
statistically significant.

Lack of statistically significant differences in
the results may raise some doubts as to the importance of
"level of attachment to family" in the migration process.
On the other hand, the consistency of the results of this
study allow for a reasonable speculation that attachment
to family is perhaps a significant factor but the methods
used to measure it may have been less sensitive than would
be desirable.

The original 1957 study items from which measure-
ments of attachment to family were derived may have cer-
tain defects and limitations in measuring the level of
attachment to family as a predictor of migration behavior.
The items used actually measured only the subjects' ex-
pressed attitudes toward certain selected aspects of their
homelife. Such arbitrary items may or may not get at the
most sensitive aspects of family ties, and, for that mat-
ter, may contain considerable impurity.

Other possible factors accounting for the weak
statistical relationships observed could be the number of
uncontrolled variables not addressed in this study. Not
the least of these is the fact that the types of family
relationships experienced by the subjects in 1957, as they

responded to the original questionnaire, have been subject

47

to change with the passage of time and may thereby pro-
foundly affect subsequent events.

On the whole, however, the data seem to suggest
that family was very influential in the migration process
of these Ontonagon County youth. Nearly two thirds of the
migrants who left for other than educational or military
reasons actually went to places where other kinsmen were
residing. This would appear to be a very strong indica-
tion of relatives' influence in the choice of destinations
by migrants in situations where they can be considered
relatively "free" to choose. And support for this view is
given by the migrants themselves. More than half (54.3%)
of the 81 migrants with relatives at destination said that
that fact had had a "major effect" on their decision to go
to that place, and another 31% said it had had "some ef-
fect." Only 12 persons thought that having relatives at
destination had had "little or no effect" on their choice.
There were even 20 persons who hadn't any relatives at
destination who nevertheless claimed that relatives were
influential in their choice of destination.

Among females further evidence of family influence
in migration decisions comes from the fact that data on
the reasons for the initial move out of Ontonagon County
show that the number of those who left for family reasons
(31.4%) ranked second only to those who left for work

reasons (33.9%).

48

It also appears that family was influential in
migrants' decisions whether or not to return "home." Of
the 130 persons who originally left the county for rea-
sons other than military service or education, the number
living outside the county (66) at the end of the ten year
period was virtually equal to the number who had moved
back by that time (64). Of the 66 permanent migrants,

47, or about 71%, had had relatives at their initial des-
tination outside Ontonagon County. Of the 64 return mi-
grants only 35, or 54.7%, had had relatives at their ini-
tial destination. These facts lead to the conclusion that
migrants who had relatives at their destinations were less
likely to return home than migrants who did not have rela-
tives so located.

Of the 124 migrants who ever returned to live in
the county, 58% were males and 42% females. It appears
that marriage could have been part of the explanation for
the lower rate of return for females. As of 1968, among
the 234 persons who had migrated, there were 23 men, but
only 7 women, who had never been married. Seventeen of
the single males and 4 of the single females had returned
to live in the county by the end of the ten year period.

The high incidence of return migration is an ex-
tremely interesting phenomenon. One hundred twenty-four
or 53% of those who ever left returned to live in Ontonagon

County for some period after their initial departure,

49

among whom 82 were still there at the time of the decennial
re-study. Why did these 124 persons return to Ontonagon
County, an economically depressed, rural county? Although
it is true that there has been a slight improvement in the
economic condition of the area in the last few years, a
more likely explanation of these people's return would ap-
pear to be in the influence of social relationships, in
particular, the family. Since measurements of family at-
tachment revealed that these subjects had significant and
important family ties (assuming that the measurements were
valid), the kinship factor was almost certainly influen-
tial in the migrants' decisions to return. It would be
expected that the level of family attachment would be lower
among permanent migrants, and this is what the present

data have indicated.

Direct evidence of family influence in migrants'
decisions to return was supplied during the 1968 re-study
in a separate check-list of "reasons for returning" filled
out by the return migrants. Sixty-one out of eighty-two
(about 74%) of these persons in the county agreed with
the statement "I enjoy being near my relatives and wanted
to remain close to them." These data lend strong support
to the notion of the importance of family in the migrants'
decisions to return to Ontonagon County.

There was often more than one reason stated by

migrants for their initial move out of Ontonagon County,

50

and it would appear that there is no theoretical basis

for expecting any one factor, be it economic, family, etc.,
to be the sole cause for return of these migrants. In-
stead, we should expect that a number of factors may be
operating in this situation, and certainly, something even
as simple as nostalgia for the places and people one.
used to know may not be totally insignificant.

When the facts discussed above are taken into con-
sideration, it would appear that family plays a much more
important role in the migration process than it is com-
monly given credit for. Even in cases where persons do
migrate for primarily economic reasons, their family rela-
tionships probably still influence the decision. This
would be the case at least partly because evaluation of
alternative opportunities is a subjective process, and
persons making such decisions are likely to perceive the
desirability of possible destinations at least partly in
terms of the social relationships involved, and not
solely in terms of the objectively "good" economic oppor-
tunities available in those places.

As was mentioned in the literature review, recent
sociological studies have challenged the notion that family
in industrial-urban settings has lost importance as a
social unit. This study set out to demonstrate that the

family does play a significant role in the migration

51

process, and thereby contribute to a more comprehensive

View of the contemporary family.

Conclusion and Recommendations
for Further Research

 

 

The outcome of the results on attachment to family
as a factor motivating the individuals to refer to their
families in their migration plans and behavior suggests a
need for a better index of "family attachment." It may
be that attitudinal measurements of attachment used in
this study could be combined with measurements of the
amount and character of interaction among family members.
A combination of these two measures should give us a bet-
ter index of family attachment than either one of these
taken singly. Future research should thus focus not only
on the individual migrant, but also on the members of his
family unit. This would give us a better proof of the
reciprocal aspects of the interaction patterns which we
suppose to be operating in the influence of the family

group upon the individual actor.

BIBLIOGRAPHY

10.

BIBLIOGRAPHY

Axelrod, Morris, "Urban Structure and Social Partic-
ipation," American Sociological Review (21),
1956, pp. 13-18.

Blumberg, L. and Bell, Robert, "Urban Migration and
Kinship Ties," Social Problems, 6 (1959), pp.
328-333.

 

Bott, Elizabeth, Family and Social Networks, "Roles,
Norms and External Relationships in Ordinary
Urban Families," Tavistock Publications Ltd.,
London, W1, 1957.

 

Brown, James, Schwarzweller, H. K., and Mangalam,
J. J., "Kentucky Mountain Migration and the Stem
Family: An American Variation on a Theme by Le
Play," Rural Sociology, (28), 1963, pp. 48-69.

Choldin, Harvey, "Kinship Networks in the Migration
Process," unpublished article, Sociology Depart-
ment, Michigan State University, East Lansing,
1968.

Cooley, Charles, "The Primacy of Primary Groups," in
Mass Society in Crisis, Social Problems and
Social Pathology, Rosenberg (editor); New York:
Macmillan Company, 1964.

 

Crawford, Charles, "Family Attachment, Family Sup-
port for Migration and Migration Plans of the
Young People," Rural Sociology, (31), 1966, pp.
293-300.

Gans, Herbert, The Urban Villagers, New York: The
Free Press, 1962.

George, Hillery A., "Definitions of Community," Rural
Sociology, (20), June, 1955.

 

Goldsmith, Harold and Beegle, J. Allan, "The Initial
Phase of Voluntary Migration," East Lansing
Michigan Agricultural Experiment Station, Rural
Sociology Studies Nos. 1, 1962.

52

ll.

12.

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

53

Goode, W. J., WorldRevolution and Family Patterns,
New York: Free Press, 1963.

 

Gugler, Joseph, "The Impact of Labor Migration on
Society and Economy in Subsahara Africa. Em-
pirical Findings and Theoretical Considerations."
University of Zambia Institute of Social Re-
search, Nos. 6, December, 1968.

Jitodai, Ted, "Migration and Kinship Contacts,"
Pacific Sociological Review, Fall, 1963, pp. 49-
54.

 

Kelley, Harold, "Two Functions of Reference Groups,"
Readings in Social Psychology, (G. E. Swanson,
et, al., editors), Holt, New York, 1952.

 

Killian, Lewis, "The Adjustment of Southern White
Migrants to Northern Urban Norms," Social Forces,
(32), 66-69, 1953.

Kiser, Clyde, Sea Island to City, A Study of St.
Helena Islanders in Harlem and other Urban
Centers, New York: Columbia University Press,
1952.

 

Litwak, "Geographical Mobility and the Extended Fam-
ily Cohesion," American Sociological Review,
(25), 1960, pp. 385-394.

McDonald, John and Leatrice, "Chain Migration, Ethnic
Neighborhood Formation and Social Network,"
Milbank Memorial Fund Quarterly, 42: 82-97, 1964.

 

Merton, R. and Kitt, A., "Contribution to the Theory
of Reference Group Behavior" in Continuities in
Social Research Studies in the Scppe andlMethods
of_“Tbe American Sbldier," R. Merton anleEul
Lazersfeld Teditors), New York: Free Press, 1950,
pp. 40-105.

 

Park, Robert, Old World Tpaits Transplanted, Illinois:
University of Chicago Press, 1925.

 

Riley J. and Riley, Matilda White, "Mass Communica—
tion and the Social System," in Sociology Today,
R. Merton, et al., pp. 537-578, New York: Basic

Books, Inc., 1962.

Roggers, E., Diffusion of Innovations, New York:
The Free Press, 1962.

23.

24.

25.

26.

27.

28.

29.

54

Schwarzweller, H., and Brown, J., "Social Class Ori-
gin, Rural Urban Migration and the Life Chances,"
Rural Sociology, (32), pp. 5-19, 1967.

 

Schwarzweller, K. H., Socio-Cultural Origins and
Migration Patterns of Young Men From Eastern
Kentucky, Bulletin 685, December, 1963, Universi-
ty of Kentucky Agricultural Experiment Station,
Lexington.

Sussman, Marvin, "The Help Patterns in the Middle
Class Family," American Sociological Review, Vol.

Thomas, Brinley, Migration and Economic Growth, A
Study of Great Britaip and the Atlantic Economy,
Cambridge University Press, 1954.

Tilley, Charles and Brown, H., "On Uprooting, Kin—
ship, and the Auspices of Migration," Inter-
national Journal of Comparative Sociology, (8),
September,”l967.

Tonnies, Ferdinand, Community and Society:
Gemeinschaft und Gesellschaft, Translated by C.
p. Loomis, New York: Harper and Row, 1963.

 

Young, M., and Wilmont, Peter, Family apd Kinship in
East London, London, Routledge and Paul Kegan,
1957, pp. 57-68.

 

APPENDIX

APPENDIX A

Measurement of Attachment to Family

The following statements were the original items

used in a ten item index of level of attachment to family:

There is real love and affection for me at
home.

It is hard for me to feel pleasant at home.
My parents criticize me too much.

As far as my ideas are concerned, my parents
and I live in two different worlds.

My friends have happier homes than I do.

Too often my parents compare me unfavorably
with other children.

As I have,known it, family life is happy.

My parents expect too much of me.

After graduation your community will be a
good place to remain close to your relatives.
My parents try to understand my problems and
worries.

Responses were in five categories from strong

disagreement to strong agreement and scoring was designed

so that the highest possible total score was 50 and the

lowest was 10. The actual mean score for the 269 subjects

was 37.7.

as follows:

Levels of attachment to family were worked out

scores of 41 and above meant a high attach-

ment to family, scores of 36 to 41 meant medium attachment,

while scores of 35 and below meant low attachment.

55

56

The following hypotheses were tested:

Hypothesis 1: The proportion of persons with high index
scores on attachment to family who migrated
out of Ontonagon County will not be sig-
nificantly different from that of persons
with low index scores on attachment to

family.

TABLE l.--Migration by Level of Attachment to Family

 

 

 

 

 

 

Migration Level of Attaghment
Status High Medium Low Total
N % N % N % N %
Migrants 75 90.4 86 86.0 73 91.3 234 89.0
Non-migrants 8 9.6 14 14.0 7 8.7 29 11.0

Total 83 100.0 100 100.0 80 100.0 263 100.0

 

57

Hypothesis 2:- A larger proportion of migrants with high
index scores on attachment to family will
on their initial move out of Ontonagon
County go to places where relatives are
located than will be the case for migrants
with low index scores on attachment to
family.

TABLE 2.--Destinations with Relatives Present or Absent
by Level of Attachment to Family, for Persons
who Migrated for Reasons Other Than Military
Service or Education

r
4—

 

 

Relatives Present/ Level of Atpachment
Absent High Medium Low Total
at Destination N % N % N % N %

 

Relatives present
at destination 27, 65.9 31 64.6 23 56.1 81 62.3

No relatives
present at des-

tination 14 34.1 16 33.3 18 43.9 48 36.9
No response 1 2.1 l .8
Total 41 100.0 48 100.0 41 100.0 130* 100.0

 

*Because two persons failed to give information
concerning presence or absence of relatives at destination,
the total sample here is less than the 132 indicated in
Table 2, Chapter III.

Hypothesis 3:

58

A larger prOportion of migrants with high

index scores on attachment to family will
indicate that having relatives at destina-
tion had a "major effect" on their choice
of destination than will be the case for
migrants with low index scores on attach-
ment to family.

TABLE 3.--Effect of Relatives on Choice of Initial Destina-
tion by Level of Attachment to Family; Migrants
with Relatives at Destination who Migrated for
Reasons Other than Military Service or Education

Effect of Relatives
on Choice
of Destination

Level of Attachment
High Medium Low Total
N % N % N % N %

 

 

 

Major effect

Some effect

Little or no effect
No response

Total

18 66.7 16 51.6 14 60.9 48 59.3

7 25.9 10 32.3 7 30.5 24 29.6
2 7.4 5 16.1 1 4.3 8 9.9
2.8 1 4.3 1 1.2

27 100.0 31 100.0 23 100.0 81 100.0

 

59

Hypothesis 4: A larger proportion of migrants with high
index scores on attachment to family will
have returned to live on Ontonagon County
than will be the case for migrants with
low index scores on attachment to family.

TABLE 4.--Return/non-Return to Ontonagon County by Level
of Attachment to Family

 

Level of Aptachment

 

 

 

Migration Status High Medium Low Total
N % N % N % N %
Return migrants 40 53.3 48 55.8 36 49.3 124 53.0

Permanent migrants 35 46.7 38 44.2 37 50.7 110 47.0

Total 75 100.0 86 100.0 73 100.0 234 100.0

 

Hypothesis 5:

60

A larger proportion of migrants with high
index scores on attachment to family will
estimate their chances of returning to live
in Ontonagon County at 50% or above than
will be the case for migrants with low in-
dex scores on attachment to family.

TABLE 5.--Migrants' Estimates of the Chances of Returning

 

 

 

 

 

to Ontonagon County by Level of Attachment to
Family Jnfi
P
= «.
Estimated Chances Level of Attachment {-3
of Returning to ’High Medium Low Total I '
Ontonagon County . N % N. % N % N %

 

Better than 75%
50 to 75%

50% to 50%

25% to 50%

Less than 25%
No response

Total

4 7.8 3 5.5 4 8.7 11 7.2
6 11.8 5 9.0 3 6.5 14 9.2
36 70.6 42 76.4 37 80.4 115 75.7
3 5.9 4 7.3 l 2.2 8 5.3

51 100.0 55 100.0 46 100.0 152 100.0

 

NOV 21 I969

HICHIGRN STATE UNIV. LIBRRRIES

III III
9

III III llllllll II
6

312 31032 7518