Modernsing Impact of Urbanisation

The Modernizing Impact of Urbanization: A Causal Analysis
Author(s): Allan Schnaiberg

Source: Economic Development and Cultural Change, Vol. 20, No. 1 (Oct., 1971), pp. 80-104
Published by: The University of Chicago Press
Stable URL: https://www.jstor.org/stable/1152806
Accessed: 29-10-2019 06:39 UTC
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/1152806?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and
extend access to Economic Development and Cultural Change
This content downloaded from 14.139.61.201 on Tue, 29 Oct 2019 06:39:27 UTC
All use subject to https://about.jstor.org/terms
The Modernizing Impact of Urbanization:
A Causal Analysis*
Allan Schnaiberg
Northwestern University
Introduction
Although the concepts of modernization and urbanization are closely
linked in the cognitive system of most social scientists, this linkage does
not rest on a satisfactory empirical basis. What we do know is that over
the long run urbanization and modernization have tended to develop over
time, both in the developed countries and in the underdeveloped world.1
But the processes by which this has occurred have been summarized only
in crude, soliptic terms. That is, regardless of the exact conceptualization of
"urbanization" or "modernization," we know that each affects the other, so
that a persistent covariation is found, whether one looks at cross-sectional
or longitudinal data. Yet this finding has become a trivial observation.
One means of attempting to unravel the underlying causal processes
is to focus on individuals undergoing social change rather than on the
institutions which are changing. Rather than focusing on the city, one can
study the city dweller; similarly, one can examine the degree to which
such individuals behave in a "modern" fashion, rather than how the
occupational, educational, and other institutions of societies change.
This approach has been most cogently put by Inkeles and his colleagues,
who have for some time been concerned with the process of "becoming
modern."2
* Revised version of a paper presented to the 1969 meetings of the Society for
the Study of Social Problems, San Francisco. The analyses reported here were carried
out at the Population Studies Center, University of Michigan, while the author was a
Population Council Fellow. Additional assistance at Michigan was provided by a
Ford Foundation grant to the center. The comments and assistance of David Goldberg,
James A. Palmore, Jr. and Janet Abu-Lughod are gratefully acknowledged. Supple-
mentary funds at Northwestern University have been generously provided by the
Center for Urban Affairs.
1 See, for example, the discussion in Daniel Lerner, The Passing of Traditional
Society: Modernizing the Middle East (Glencoe, Ill.: Free Press, 1958), pp. 69-75.
2 This is the title of a forthcoming book by Alex Inkeles and others (Becoming
Modern), based on the findings in a six-nation study of individual modernity. The
most recent report on this work appears in Alex Inkeles, "Making Men Modern:
On the Causes and Consequences of Individual Change in Six Developing Countries,"
American Journal of Sociology 75 (1969): 146-51.
80
Allan Schnaiberg
Of what import is this type of analysis? On a theoretical level, it

begins to systematize the approach to social change in a variety of develo-
ping societies. Moreover, it bears a strong relation to the task of interven-
ing in this social change, whether this intervention or "social engineering"
takes the form of family planning, community development, or eco-
nomic development programs. It is an assumption of this paper that
implicit in the actions of most development planners is the belief that
urban populations are much more responsive to innovative actions, for
example, the acceptance of contraception, because they are more modern,
however we define that quality. Yet the failures of many such programs in
urban areas of underdeveloped countries bear testimony to the inadequacy
of such a perspective. Conversely, a guiding principle of many develop-
ment planners is that the rural areas are the repositories of traditional and
"traditionalistic"3 modes of behavior and attitudinal structure. And, in
one sense, this leads to a "self-fulfilling prophesy" in that they are reluctant
to expend scarce resources for development where such obstacles exist,4
and thus rural areas do in fact lack sufficient opportunities for behavioral
development. That this need not be so is indicated in part by previous
work5 and in part from the observations and conclusions in the present
work.
This paper focuses on several questions: (1) To what extent does the
urban experience influence an individual's level of "modernism"?
(2) What is the mechanism by which such urban experience affects mod-
ernism? (3) To what extent can we conceive of modernism as a unitary
concept?
Source of Data
Unlike the Inkeles project,6 this study uses information from only one
developing country: Turkey. But the findings of the present study are
believed applicable to most developing countries, with due regard to
their degree of development or "industrialization." In part this view is
3 On the distinction between "traditional" and "traditionalistic" behavior, see
Wilbert E. Moore, "The Social Framework of Economic Development," in Tradition,
Values, and Socio-Economic Development, ed. R. Braibanti and J. J. Spengler (Durham,
N.C.: Duke University Press, 1961).
4 For a rather critical review of the "obstacles" to development, see A. O.
Hirschman, "Obstacles to Development: A Classification and a Quasi-Vanishing
Act," Economic Development and Cultural Change 13 (1965): 385-93.
5 A brief discussion of rural development work appears in Howard Schuman,
"Economic Development and Individual Change: A Social-Psychological Study of
the Comilla Experiment in Pakistan," Occasional Papers in International Affairs, no.
15, Harvard University Center for International Affairs, 1967. A longer detailed
study is Gayl D. Ness, Bureaucracy and Rural Development in Malaysia (Berkeley
and Los Angeles: University of California Press, 1967).
6 The Harvard study had a total sample of 6,000 men, with 1,000 cases drawn
from selected groups within each of the six societies. Further details are given in
David H. Smith and Alex Inkeles, "The OM Scale: A Comparative Socio-Psycho-
logical Measure of Individual Modernity," Sociometry 29 (1966): 353-77; and
Inkeles.
81
Economic Development and Cultural Change
substantiated by the high degree of similarity between the findings here

and those based on the Harvard six-nation cross-cultural analysis.
In 1966, data were gathered for the Ankara Family Study7 in Ankara
City and four villages in Ankara Province. The Ankara sample consisted
of 803 households, in which all married women were interviewed. A two-
stage sampling plan was used, and the resulting sample was a systematic,
stratified, and clustered one. The rural respondents, 335 in number, were
drawn from four villages in Ankara Province. They represent the universe
of married women in these villages. Selection of the four villages was
deliberate in an attempt to obtain some representative rural population
that could be interviewed with close supervision and limited resources.
Two of the villages (with 172 respondents) are classified as "accessible,"
as they were within reasonable reach of Ankara City, although at a much
greater distance than Lerner's village of Balgat.8 The remaining villages
are located about 120 kilometers (75 miles) from Ankara City.
Combining the sample from Ankara and the village universe, there
are 1,138 respondents in what is termed the "study population." This
study population is not representative of the married female Turkish
population, since it vastly overweights the proportion of urban women
and those who have migrated from rural to urban areas. This limitation
is not an overwhelming one because most of the analyses to be reported
here are specific to various types of urban experiences.9
Definitions of the Central Concepts

Using these data, three concepts were measured. "Urbanism" is defined
as the degree to which an individual has spent various portions of her life
cycle in urban communities (those with larger population concentrations).10
The actual measures of urbanism are four: place of longest residence (1)
before the age of ten, (2) between age ten and the time of marriage, (3)
between the time of marriage and interview (i.e., for the duration of
marriage); and (4) the place of current residence (Ankara or the villages).
For each period, it was determined whether the respondent lived in
Ankara City, in Istanbul, or Izmir (the largest cities in Turkey), in other
provincial centers, in district centers, or in villages. With the exception of
the Ankara City category, the remaining locations represent the urban
7 Fuller details of the sampling and field procedures of the Ankara Family
Study, which was under the direction of David Goldberg, are given in Allan Schnai-
berg, "Some Determinants and Consequences of Modernism in Turkey" (Ph.D.
diss., University of Michigan, 1968), chap. 2.
8 Lerner, pp. 19-42.
9 In the larger study from which this report is drawn, a variety of techniques
were used to control for this weighting problem. The results reported here are sub-
stantiated by those more extensive and intensive analyses (see Schnaiberg, chap. 4).
10 The term "urbanism" is analogous, on the individual level, to Donald J.
Bogue's usage in "Urbanism in the United States, 1950," American Journal of Sociol-
ogy 60 (1955): 471-86. There is no consensus among urban sociologists as to usage
of terms like "urbanity," "urbanization," and "urbanism," as indicated by the
variations in the entire special issue of the March 1955 American Journal of Sociology.
82
Allan Schnaiberg
hierarchy of Turkey, in terms of size of place. Ankara's position is am-

biguous, since it has only recently become the second largest city in
Turkey11 and in many ways resembles the smaller provincial centers more
closely than it does Istanbul.
Turning next to the concept of "modernism, "12 it becomes imme-
diately apparent that whatever definition is to be offered will replicate only
a small share of the previous work done in this general area. For some, this
represents a purely psychological or at least attitudinal concept,13 for
others it denotes actual behavior,14 while in another group of studies it
reflects some mixture of attitudes and behavior.15 In this study, the
definition is more in the last group, although the predominant content of
modernism is one of action rather than attitudes. Another dispute in
delimiting the concept centers on whether shifts from traditional to
modern behavior are to be defined by the actors or the observer,16 while
a related problem is whether modern behavior is defined in some absolute
or ideal-typical fashion, or relative to the actual social structure that
preexisted.17 On these two dimensions, for the present purposes modern-
I' In 1935, Ankara City had a population of 122,720; by 1955 it had 451,241
inhabitants; and by 1965 had more than doubled to reach 905,660. Source:
24.10.1965, Census of Population by Administrative Division, publication no. 537,
Turkish State Institute of Statistics.
12 This is the same term used by Joseph A. Kahl, The Measurement of Modern-
ism: A Study of Values in Brazil and Mexico (Austin: University of Texas Press,
1968). As with "urbanism," there is little agreement on these labels.
13 For example, Leonard W. Doob, Becoming More Civilized (New Haven,
Conn.: Yale University Press, 1960), and "Scales for Assaying Psychological Moderni-
zation in Africa," Public Opinion Quarterly 31 (1967): 414-21.
14 See, for example, Remi Clignet and J. Sween, "Social Change and Type of
Marriage," American Journal of Sociology 75 (July 1969): 123-45; Arnold S. Feldman
and C. Hurn, "The Experience of Modernization," Sociometry 29 (1966): 378-95;
and Ian Weinberg, "The Concept of Modernization: An Unfinished Chapter in
Sociological Theory" (paper presented at the American Sociological Association
meeting, Boston, 1968).
1- Smith and Inkeles, Inkeles, and Kahl all deal with both attitudinal and behav-
ioral items, though with a heavy weighting on attitudinal items. To some extent,
this is partly true for the Feldman and Hurn work.
16 This has been most recently suggested by John B. Stephenson, "Is Everyone
Going Modern ? A Critique and a Suggestion for Measuring Modernism," American
Journal of Sociology 74 (November 1968): 265-75. This presentation has been criti-
cized by Alex Inkeles, "Comments on John Stephenson's 'Is Everyone Going
Modern ?'" American Journal of Sociology 75 (July 1969): 146-51; see also Stephen-
son's reply, "The Author Replies," American Journal of Sociology 75 (July 1969):
151-56.
17 This is a point stressed recently by Clignet and Sween, although S. M. Eisen-
stadt had earlier emphasized the importance of taking account of prior social structure
in predicting consequences of any given "modernizing" force in a society (e.g., in his
"Transformation of Social, Political, and Cultural Orders in Modernization,"
American Sociological Review 30 [1965]: 659-73). A somewhat similar critical view of
ideal-typical approach to modernization is contained in Oscar Lewis, "The Folk-
Urban Ideal Types," and Philip M. Hauser, "Observations on the Urban-Folk and
Urban-Rural Dichotomies as Forms of Western Ethnocentrism," both in The Study
of Urbanization, ed. P. M. Hauser and L. F. Schnore (New York: John Wiley &
Sons, 1965).
83
ism was defined by the observer and in some ideal-typical fashion,

although the actual operationalization of the measures used does introduce
some relational components in terms of the cutting points for "modern" as
opposed to "traditional" coding.18
One final issue must be raised before we turn to the conceptualization
and operationalization of modernism here. That is the question of the
unity of the concept itself: are we referring to some single theoretical
and empirical element, or as Bendix19 and others have suggested, does
the use of a single term such as "modernism" reflect conceptual and oper-
ational casualness which is all too common in social science? From a
perspective of Wissensoziologie, the dilemma and questions put forth by
the many observers in this area, as noted in the preceding paragraph
suggest that there does not exist empirically any unified perspective o
what constitutes modernism. Unfortunately, there are no standard
around which consensus exists to judge the theoretical and empirica
basis for determining the unity of a concept, so no straightforward resolu-
tion of this problem is possible at this point.
However, with this perspective in mind, I have attempted to defin
and operationalize modernism in such a way as to leave the question
somewhat open. For this study, a woman is modern who: (1) participat
in mass-media consumption;20 (2) is relatively free of extended fami
ties;21 (3) plays a somewhat egalitarian role within the nuclear
18 Smith and Inkeles note that their cutting points differ for the different societie
in their sample, as they used the median for each country as the cutting point, whereve
possible (see p. 360 and n. 13, pp. 360-61). This in fact makes modernism a measur
of change relative to the previous social structure, a point which is somewhat latent
in most earlier discussions by the Harvard group, as well as by Kahl. There are man
statistical problems, however, related to the choice of cutting points for dichotomie
few of which have been stressed in the Harvard study, or in Kahl's (with the excepti
of n. 6, p. 211 in Inkeles) (see Hubert M. Blalock, Jr., Causal Inferences in Non-
experimental Research [Chapel Hill: University of North Carolina Press, 1961],
pp. 32-34).
19 Reinhard Bendix, "Tradition and Modernity Reconsidered," Comparative
Studies in Society and History 9, no. 3 (1967): 292-346 offers the most detailed critiqu
of a singular concept of modernity and of related theories which stress uniform
development in all institutional sectors of society. For a shorter critical view, see al
Joseph Gusfield, "Tradition and Modernity: Misplaced Polarities in the Study
Social Change," American Journal of Sociology 72 (January 1967): 351-62. To some
extent, Marion Levy recognizes the variation around the concepts of modernity a
modernization, yet he chooses to emphasize the oneness of the concept and processes
(see his definition on p. 11 of Modernization and the Structure of Societies [Princeton
N.J.: Princeton University Press, 1966]; also Inkeles, p. 211).
20 The role of mass media in development has been stressed especially by Lerner
and by Wilbur Schramm, Mass Media and National Development (Stanford, Calif
Stanford University Press, 1964). Smith and Inkeles, in their chart 1 on p. 354, have
a similar "theme," mass media evaluation (MM); Kahl, in his table 1, pp. 30-34,
also has a mass media participation (VIII) index (items in parentheses refer to the
authors' designations of these indexes).
21 See the excellent summary of findings and speculations on the changing
structure of the extended family in Robert F. Winch and R. Blumberg, "Societal
Complexity and Familial Organization," in Selected Studies in Marriage and the
84
Allan Schnaiberg
family;22 (4) does not participate too actively in religious functions;23 (5)
has a rather broad view of the environment from which to draw values and
norms;24 and (6) engages in low levels of home production of non-
durable goods while consuming larger amounts of manufactured goods.25
Six separate indexes of modernism were computed: mass-media partici-
pation, extended family relations, nuclear family role structure, religiosity,
environmental orientation, and production/consumption behavior.26
Each of these, it was felt, constituted one area of behavior that social
scientists had included under varying definitions of modernism or moder-
nity. The index value was computed from the responses to a number of
behavioral and attitudinal questions, with each response being classified
as "modern" (scored as 1) or "traditional" (scored as 0). For any individ-
ual item, there appeared little ambiguity as to the direction of classifica-
tion: only the cutting points for the dichotomization are arbitrary. In
general, criteria of maximizing the variance in index scores without doing
violence to any obvious cutting points were followed; this means cutting
the distribution at the median category, wherever possible. For some items
responses had initially been dichotomized in the original coding, so that
no decision as to cutting point was required. The index value, then,
became the "number of modern responses" and was subsequently treated
as an interval measure in all analyses.27 The criteria used for establishing
cutting points are similar to those employed in other modernism studies.
Furthermore, it should be noted that such procedures of maximizing that
variance of each item in fact take account of the preexisting social structure
Family, ed. R. F. Winch and L. W. Goodman, 3d ed. (New York: Holt, Rinehart &
Winston, 1968). Smith and Inkeles include extended kinship obligations [KO(1)] and
kinship obligation to parental authority [KO(2)] in their thematic list; Kahl has a
measure called low integration with relatives (V).
22 For example, see William J. Goode, "The Role of the Family in Industrializa-
tion," in Social Problems of Development and Urbanization, vol. 7 (Washington,
D.C.: Agency for International Development, n.d.). This has some overlap with
Smith and Inkeles's women's rights [WR(1)] and coed work and school [WR(2)],
while Kahl includes family modernism (XII).
23 A brief discussion of this appears in Milton Singer, "The Modernization of
Religious Beliefs," in Modernization: The Dynamics of Growth, ed. M. Weiner (New
York: Basic Books, 1966). In Smith and Inkeles, see religious-secular orientation
[RE(2)], and in Kahl, low religiosity (XIII).
24 See chap. 2 in Lerner for a discussion of some aspects of this concept, which
falls partly under what he is calling "empathy"; Smith and Inkeles have some overlap
in their occupational aspirations [AS(2)] and women's rights [WR(1)]; Kahl has
elements similar to mine in his low stratification of life chances (II), low occupational
primacy (IV) and anti-big-companies (IX).
25 One of the few studies in this area is Deborah Freedman, "The Role of Con-
sumption of Modern Durables in a Developing Economy: The Case of Taiwan"
(Ph.D. diss., University of Michigan, 1967). On the theory of declining home pro-
duction, see Talcott Parsons and R. F. Bales, Family Socialization and Interaction
Process (New York: Free Press, 1955). Smith and Inkeles cover part of this theme in
their consumer aspirations [CO(2)].
26 The items used are listed in the Appendix.
27 This is quite similar to the assumptions made by both Smith and Inkeles and
Kahl in their major studies about the interval nature of their measures.
85
(i.e., the traditional structure), in that the operational definition of mod-

ernism is a relative one although the conceptualization is generally not
stated in that way. This creates problems which go beyond the scope of
this paper but which are nonetheless important and need further thought.
At this point, it is important to specify the overlap of this definition
of modernism with some of the recent important work in the area.28
It should be stressed first that the study was not intended as a replication
of these earlier approaches, and thus some of the overlap is fortuitous;
the remainder of the overlap reflects common theoretical and empirical
backgrounds of the several researchers. Thus, in general, the items listed
in the Appendix are not the same as those used by either Kahl or Inkeles,
but they are clustered under some of the same general behavioral areas.
Next, the differences in forming the several modernism indices are critical.
Whereas both Kahl and Inkeles used statistical criteria of association to
form their several indices, I have used the items as defined a priori from
previous theoretical and empirical literature. Obviously, then, the scales
or indices constructed in this fashion will have lower scale reliabilities29
than will those in which are included only those items meeting a given
item-scale correlation criterion, or some equivalent screening. However,
this was a deliberate choice in my part, as I wished to test the adequacy of
prior definitions of modernism included in the sociological literature, as
well as propositions relating to such conceptualizations.30
Therefore, differences between the six modernism indices may reflect
differences in scale reliabilities as well as valid behavioral differences.
Yet I am prepared to argue that most of such differences may be attrib-
utable to weaknesses in the theoretical framework from which our
propositions here are drawn. This assumes that we have operationalize
modernism correctly, of course, but this is always a matter of som
conjecture; it is a more plausible assumption for the present study than th
counter-assumption that it is the operationalization that is defective an
the theory valid. The plausibility is increased by the fact that both Kahl
and Inkeles found31 that a sizable fraction of their indices did not "fit"
28 Particularly Smith and Inkeles, Inkeles, and Kahl.

29 Although Inkeles (p. 210) and Smith and Inkeles (pp. 361, 362 ff.) discuss and
tabulate the reliabilities of their various measures, there are many problems involved
in assessing reliability; e.g., Smith and Inkeles rely on the Spearman-Brown formula,
which is subject to numerous difficulties (see Helen Walker and J. Lev, Statistical
Inference [New York: Holt, Rinehart & Winston, 1953], pp. 302, 308-11). For one
thing, scale reliabilities may be drastically affected by the skewness of some of the
items which reflect some "lagging" or "leading" of change in particular social
institutions.
30 Smith and Inkeles (pp. 367-69) discuss the limitations of their "criterion
group" method of determining scale contents in that the factors used as criteria can
then no longer be used in testing propositions concerning their relation to modernity
(except by tautological reasoning). Since my primary purpose is to test the empirical
basis for such criteria being accepted, this method is clearly not appropriate.
31 For example, Smith and Inkeles (n. 18, p. 361) indicate that their scale OM-1,
consisting of seventy-nine items, has higher reliability than their OM-2 scale of 119
86
Allan Schnaiberg
together under the rubric of "modernism" or "modernity." But their

concerns were with those elements that "fit," whereas, in testing the ade-
quacy of our theoretical systems,32 mine are both with those that "fit"
and those that do not. Obviously, neither extreme position can be validated:
observed differences are always attributable in part to instrument problems
and in part to inadequate explanatory systems; I merely stress the latter.
A third concept, that of "socioeconomic attainment," was defined
as the degree to which the respondent has been able to attain a high
level of social and economic status, either directly or through marriage
and sharing her husband's status. This was operationalized in terms of
the respondent's education, her husband's education, and the family
income at the time of the interview. The role of this concept will become
apparent shortly.
Causal Model and Hypotheses

There exists a considerable body of literature from which to draw infer-
ences concerning the development of attitudes and behavior in individuals
living in developing countries. Much of the relevant discussion has been
cogently summarized by Inkeles33 in terms of "early socialization" and
"late socialization" models of individual change. Early socialization
forces, as cited by both Inkeles and Kahl,34 include such things as resi-
dental experience and educational exposure, while late socialization
influences might be occupational, work place, situs, levels of affluence, and
so on. The focus in the present work is on the early socialization processes,
in part because the sample is composed of females, most of whom are not
subject to the direct influence of many of the late or adult socialization
factors listed above. The objective here is to tie together in a parsimonious
model several of the early socialization influences, so as to both under-
stand the interrelations of these separate factors and, more important, to
permit intervention to accelerate the development of modernism.
Starting from an early socialization perspective, the following
results were predicted. First, there would be substantial differences in the
levels of modernism of urban and rural women, the urban women being
substantially more modern. Much of the observed differential would be
due to the type of early residential experience, so that women growing up
in rural areas were expected to carry on traditional cultural values and
items, which is the scale reported on later by Inkeles. Kahl (p. 37, table 3) finds only
seven of his fourteen indices are included in the "core" of modernism.
32 Kahl is quite open in cautioning the reader that the "apparent neatness
between the theoretical discussion above, and the empirical results.. . is somewhat
artificial" (p. 28).
33 Inkeles, pp. 212-16; see also Smith and Inkeles, and an earlier piece by
Inkeles, "The Modernization of Man," in Modernization: The Dynamics of Growth,
ed. M. Weiner (New York: Basic Books, 1966), as well as Inkeles's introduction to the
Schuman study of the Comilla experiment cited earlier.
34 Inkeles, "Making Men Modern," pp. 212-16; Kahl, chaps. 3, 4.
87
social action, while those being reared in urban places tended to be

socialized in more modern ways. The second hypothesis was that most
of the effects of early residential experience were produced by differential
access to education in rural and urban areas, with the consequence that
rural girls would be uneducated and would tend to marry men with little or
no education, thus creating a family of procreation with relatively low
economic resources, bound to a localized traditional environment. This
low level of "socioeconomic attainment" of the woman and her husband
was postulated as the primary variable intervening between her urban
background and her modernism. Both these hypotheses are tested, with
substantial confirmation, in the balance of this paper. Graphically, the
model described above can be represented as in figure 1.35
EARLY RESIDENCE R'S EDUCATION ) R'S MODERNISM
FAMILYRINCOME
HUSBAND'S
EDUCATION
FIG. 1.--Proposed early socialization model for the developmen

female modernism.
In addition to the two hypotheses raised above, there remains one

closely related issue which must be dealt with. That is the question of
whether the model is equally applicable to all the dimensions of modern-
ism that we have defined. If we view the model as explicating or elabo-
rating on some more general evolutionary theory of social change,36 we
thus shift our attention from the microsocial level to the macrosocial
superstructure in which it is nested. One of the simplest macrodevelop-
mental models is one which predicts that all interrelated areas of social
action will change at about the same rate, so that individuals who will be
modern on one dimension of behavior should likewise be modern (or
more appropriately, equally modern) in all the other behavioral areas.
That is, if all the major institutional spheres change at the same rate, so
should the microsocial level of action. Thus, on this basis, the model
proposed above should be equally applicable or powerful for the predic-
tion of all six dimensions of modernism. We will subject this to just one
test: the comparability of the predictability of each of the modernism
indices, using the causal path noted in figure 1. If there appear to be
substantial differentials in the utility of the model for the several indices,
3 There is some similarity here to Kahl's chart 1, p. 86.
36 An excellent discussion of such "monistic" theories of social change may be
found in Wilbert E. Moore, Social Change (Englewood Cliffs, N.J.: Prentice-Hall,
Inc., 1963), chap. 2, especially pp. 23-24 and 33-44. Much of Levy's work could be
classified as "monistic."
88
Allan Schnaiberg
this will cast still more doubt on the theory of near-uniform evolution of
all major social institutions.37 Under such circumstances, we would also
have reason to question the utility of a single concept of modernism.38
The Effects of Urbanism on Modernism

The first task of the analysis was to establish the existence of substantial
modernism differentials for women with varying urban experience
(between birth and age ten). For purposes of comparability, the index
scores are presented in both raw form, and in standardized, Z + 3,
transformations in table 1. The latter was computed to provide compara-
TABLE 1
DIFFERENTIALS IN MEAN MODERNISM SCORES, BY EARLY RESIDENCE
MEAN MODERNISM SCORE
Environ- Produc-
Extended Nuclear ment tion/
WHERE LIVING Mass Family Family Relig- Orienta- Consump-
BEFORE AGE TEN Media Ties Roles iosity tion tion
A. Raw Scores
Ankara......... 3.31 3.84 10.39 1.22 1.99 4.58

Istanbul/Izmir... 3.91 4.23 12.67 1.87 2.46 5.67
Provincial center. 3.05 4.00 11.17 1.05 2.04 4.86
District center... 2.27 3.48 8.92 0.97 1.81 4.26
Village......... 0.83 3.03 5.36 0.64 1.40 3.18
B. Standardized Scores (Z + 3)
Ankara......... 3.89 3.29 3.55 3.38 3.33 3.34

Istanbul/Izmir... 4.26 3.55 4.04 4.12 3.86 3.91
Provincial center. 3.73 3.39 3.72 3.17 3.38 3.49
District center ... 3.25 3.04 3.24 3.08 3.12 3.18
Village ......... 2.37 2.74 2.74 2.70 2.67 2.61
bility among indexes with varying numbers o

ranges. As may readily be observed, there d
urban differences on modernism, although t
varies considerably among the dimensions of m
uphold uniform institutional change theory.
To measure the impact of the total urban
on modernism, table 2 presents standardized
for various patterns of urbanism. Each wom
summarized by the pattern, which incorpora
for the successive life-cycle stages (roughly, ch
adolescence, early marriage, and current). A
substantial differentials in modernism appear. I
37 These differentials might be a product, of cours
reliability (how well items cohere in a single scale or
surement error leading to unreliability in any item or
38 Moore, Social Change, pp. 33-44.
89
0
TABLE 2
DIFFERENTIALS IN MEAN MODERNISM SCO
MEAN MO
RESIDENTIAL Mass Extended Nuclear En

PATTERNt REF.t N Media Family Ties Family Roles Rel
A-A-A-A .............. 1 108 3.89 3.29 3
A-A-I-I ................ 2 51 4.24 3.45 4
A-I-I-I ................. 3 23 4.21 3.51 4.
A-A-Pc-Pc ............. 4 106 3.81 3.41 3
A-Pc-Pc-Pc............. 5 72 3.60 3.33 3
A-A-Dc-Dc ............ 6 98 3.30 3.04 3
A-Dc-Dc-Dc ........... 7 49 3.22 2.91 3
A-A-V-V............... 8 158 2.68 3.06 2
A-V-V-V............... 9 105 2.49 2.89 2
(AV)-V-V-V ........... 10 163 2.29 2.59
(IV)-V-V-V............. 11 161 2.03 2.44 2
* Standardized (Z + 3) scores for each modernism
t Pattern consists of (left to right): current residen
before age ten. A = Ankara City; I = Istanbul-Izmi
sible village; V = Village.
T Reference number for each residential pattern, to
? This reversal is due to the fact that one of the i
to be very religious. For example, the village has no m
Allan Schnaiberg
early residence, the woman's modernism is also affected by her timing of

migration as well as her community of destination. This indicates the
relevance of both early and late socialization models for individual
modernization. In order to present a stronger case for our focus on early
socialization, it becomes necessary to attempt to evaluate the relative
importance of the two paths to modernization. For this reason, the data in
table 2 have been rearranged (except for the indigenous Ankara City
residents) in tables 3 and 4.
In table 3, the data are arranged so as to control for the variation
in life-cycle timing of migration to Ankara City.39 This is an attempt,
then, to measure the impact of residence in the origin community, in the
earlier part of the respondent's life cycle. Internally, in table 3, the fact
that the ranges listed in the upper panel (those migrating later) are
generally larger than those in the lower panel indicates that life-cycle
timing of migration does have some effect on subsequent modernism.
But the fact that such differences for any modernism dimension are quite
small, relative to the actual range, suggests that the timing and therefore
adult or later resocialization influences are not overly important.
A more direct measure of these effects of timing of migration appears
in table 4. Here the community of origin is controlled, and modernism
differences due to timing of migration to Ankara are computed. If one
contrasts the magnitudes of differences in modernism indicated by the
ranges in table 3 with these differences in table 4, one general observation
can be made. Regardless of the type of community a woman spent her
early life in, the impact of having spent these "formative" years in such an
environment appears very strong and persistent. It is persistent in that
there appears to be relatively small attrition of this effect even if the
woman moves before her marriage to a large urban area (Ankara). This
very strongly argues for the predominance of early socialization in the
factors affecting modernism.40 However, in their covariation with the
woman's urbanism, there are clear differences among the six modernism
dimensions. First, there are substantial differentials in the ranges of
variation tabulated in table 3 (upper and lower panels). The largest
differentials (or greatest covariation with urbanism) are for the mass-
media consumption, nuclear family role structure, and religiosity indices.
One of the lower covariants is the production/consumption index. From
other analyses,41 it appears quite certain that there are factors associated
with the current community of residence which strongly influence home
9 For this analysis, we exclude all those born in Ankara City, as well as those
who have moved to one of the study villages from some other village, since there is no
possibility of separating effects of early and late residence.
40 Inkeles, "Making Men Modern," finds that education is the single most
important determinant of modernism, which lends additional weight to our findings.
41 See my "Rural-Urban Residence and Modernism: A Study of Ankara Prov-
ince, Turkey," Demography 7, no. 1 (February 1970): 71-85; and "Some Determi-
nants and Consequences of Modernism in Turkey," chap. 3.
91
t.N>
TABLE 3
EFFECTS OF PREVIOUS RESIDENCE (FOR MIG
MEAN MO
MIGRATED TO Mass Extended Nuclear En

ANKARA FROM REF.t N Media Family Ties Family Roles Re
After Midmar
Istanbul/Izmir ........... 3 23 4.21 3.51

Provincial center ......... 5 72 3.60 3.35
District center ........... 7 49 3.22 2.91
Village ................. 9 105 2.49 2.89 2
Range (Istanbul-Village) ... ... 1.72 0.62
After M
Istanbul/Izmir ........... 2 51 4.24 3.45

Provincial center ......... 4 106 3.81 3.41
District center ........... 6 98 3.30 3.04
Village ................. 8 158 2.68 3.06 2
Range (Istanbul-Village) . . ... 1.56 0.39
* Standardized (Z + 3) scores for each moder
t See table 2 for details of the residential pat
TABLE 4
EFFECTS OF TIMING OF MIGRATION (FOR MIG
MEAN MO
MIGRATED TO ANKARA Mass Extended Nuclear

FROM AND WHEN REF.t N Media Family Ties Family Roles R
Istanbul/Izmir:
a) After midmarriage... 3 23 4.21 3.51
b) After midadolescence. 2 51 4.24 3.4
Difference (a - b).... ... ... -0.03 +0.06 +0
Provincial center:
Difference (a - b).... ... ... -0.21 -0.08 +0
District center:
Difference (a - b)....... ... -0.08 -0.13 -0.
Village:
Difference (a - b).... .... ... -0.19 -0.17 -0
* Standardized (Z + 3) scores for each modernism index.
t See table 2 for details of the residential patterns to which these reference numbers
production and consumption. In particular, one might cite the unavail-

ability of substitutes for home-produced goods in many rural areas and
small towns as explaining this phenomenon, although for the present
purposes these factors are extraneous to our analysis. Other interpretations
may be offered to explain the lower covariation shown by the extended
family relations and environmental orientation indices.42
The overall interpretation drawn from these findings is that modern-
ism is very closely related to a woman's urban experience. Furthermore,
the most important aspect of this experience appears to be that of early
childhood, which, as I will presently argue, strongly determines the
future socioeconomic attainment of females in developing societies. But
this interpretation is not equally valid for all dimensions of modernism:
for production/consumption behavior, current residence is an important
factor. For extended family relations, weaker relationships exist with both
past and current residence, indicating once more how little we know about
the determinants and consequences of extended family structures.43
Testing the Model

In tabulations not shown here,44 very strong relationships were found
between the woman's socioeconomic attainment, particularly her educa-
tion, and her levels of modernism. This, of course, is a necessary but
insufficient condition for our second hypothesis to be sustained. Several
techniques were used to test the hypotheses more directly. The first was
the use of the incremental explained variance results from a multiple
classification analysis,45 with both urbanism and socioeconomic attain-
ment as the independent variables, and the modernism indexes as the
dependent variables. These results are summarized in table 5, using
earliest residence as the sole measure of urbanism, and the additive
effects of respondent's education, husband's education, and family income
as the measure of socioeconomic attainment. What we discover is that
most of the effects of earliest residence operate through socioeconomic
attainment to affect the levels of modernism. The additive effect of earliest
42 See nn. 18 and 29 above and discussion in the earlier section on the statistical
and substantive problems associated with skewness of dichotomous variables.
43 In addition to the Winch and Blumberg presentation, see for example Thomas
K. Burch, "The Size and Structure of Families: A Comparative Analysis of Census
Data," American Sociological Review 32 (1967): 347-63; and Morris Zelditch, Jr.,
"Family, Marriage and Kinship," in Handbook of Modern Sociology, ed. R. E. L.
Faris (Chicago: Rand McNally, 1964). The recent literature ranges from "unheralded"
to "overheralded" discoveries of the viability of some form of extended kinship tie in
developed societies, and a strong question of its prevalence in preindustrial societies.
44 Schnaiberg, "Some Determinants and Consequences of Modernism," chap. 5.
45 The basic program is described in Frank Andrews, James Morgan, and John
Sonquist, Multiple Classification Analysis: A Report on a Computer Program for
Multiple Regression Using Categorical Predictors (Ann Arbor, Mich.: Survey Research
Center, Institute for Social Research, 1967). The uses of stepwise MCA are covered in
Peter M. Blau and 0. D. Duncan, The American Occupational Structure (New York:
John Wiley & Sons, 1967), pp. 163-77.
94
TABLE 5
VARIANCE IN MODERNISM EXPLAINED BY EARLY R
FORWARD AND REVERSE MODELS
(70)
MODERN
Mass Extended Nuclear

INDEPENDENT VARIABLES Media Family Ties Family Roles
A. Forward model:
R's residence before age ten........... 49.4 9
Increment for:
Respondent's education ............ 22.3 3
Husband's education ............... 1.2
Family income ... ............. 1.3 2
Increment for socioeconomic
attainment*t .................. 24.9 5.4
B. Reverse model:
Family income........................ 47.0 7.1
Increment for:
Respondent's education ............. 11.4
R's residence before age ten......... 2.6
Explained by all independent
*
variablest.............
This is s
income.
t Increments may not add to total because of rounding error.
ul
residence is quite small, once the three socioeconomic variables have

been used to predict modernism (reverse model).46 On the other hand,
there are substantial additive effects of socioeconomic attainment, after
earliest residence has been taken into account (forward model).47 This
very strongly suggests that socioeconomic attainment is much more
closely tied into the modernizing process and that the urbanism effects
are observed only because of the close linkage between rural residence and
low educational and economic opportunities in developing areas.48 The
strong exception here is production/consumption behavior, which we
suggest is more closely related to urban or rural economies. Again, it is
important to note the variation among the indices in terms of the level of
explanation of the model.
If we wish to extend our basic model as drawn in figure 1 to include
other portions of the woman's residential history, aside from earliest
residence, a rather more complex diagram, such as figure 2, might be
sketched. Birthplace was substituted for residence between age ten and
marriage in order to provide more early socialization factors in the model.
With the nature of the statistical techniques and the data we have used, it
is impossible to measure all the "paths" laid out, so that our analysis
should be recognized for a more limited covariance (multiple classifica-
tion) analysis and not a formal path analysis.49 Therefore, figure 2 should
be treated heuristically rather than literally in attempting to follow subse-
quent tabulations. Two approaches are used; in table 6, the various
measures of urban experience at each stage of the life cycle are interposed
with the socioeconomic attainment factors in a temporal and presumably
causal ordering. The significant role of the woman's education and socio-
economic attainment in general is apparent here, as it was in the previous
tabulation.
An alternative approach to incorporating all aspects of urbanism
into the model is to treat both the residential pattern listed in table 2
and socioeconomic attainment together in some analysis. Since this
creates problems in terms of temporal ordering of these variables, a
46 Blau and Duncan, pp. 131 if, describe "forward" and "reverse" partitioning
of explained variance from MCA.
47 The most lucid discussion of partitioning variance appears in Otis Dudley
Duncan, D. L. Featherman, and B. Duncan, Socio-Economic Background and
Occupational Achievement: Extensions of a Basic Model, final report, project no.
5-0074 (EO-191) (Washington, D.C.: Office of Education, 1968), sec. 2.8.
48 On the differentials in urban-rural access to education in Turkey, see Frederick
W. Frey, "Education: Turkey," in Political Modernization in Japan and Turkey, ed.
R. E. Ward and D. A. Rustow (Studies in Political Development, no. 3 [Princeton,
N.J.: Princeton University Press, 1964]); Andreas M. Kazamias, Education and the
Quest for Modernity in Turkey (Chicago: University of Chicago Press, 1966); and
Richard D. Robinson, The First Turkish Republic: A Case Study in National Develop-
ment (Cambridge, Mass.: Harvard University Press, 1963).
49 For the similarities and differences between covariance and path analysis, see
James Fennessey, "The General Linear Model: A New Perspective on Some Familiar
Topics," American Journal of Sociology 74 (July 1968): 1-27.
96
BEFORE AGE 10 EDUCATION AFTER MA
R'S RESIDENCE RESPONDENT'S RES
R'S PLACE HUSBAND
OF BIRTH EDUCATION
FIG. 2.-Proposed earl
Oo
TABLE 6
VARIANCE IN MODERNISM EXPLAINED BY URBANISM
FORWARD AND REVERSE MODELS
(%)
MODERN
Mass Extended Nuclea

INDEPENDENT VARIABLES Media Family Ties Family Roles
A. Forward model:
R's birthplace ....................... 50.7 8.2
Increment for:
R's residence before age ten .. 2.4
R's education ..................... 19.1 3
Residence after marriage .. 1.3 1
Current residence .................. 0.4
Family income. ................... . 0.9
B. Reverse model:
Family income ...................... 47.0 7.1
Increment for:
Current residence .................. 5.7 3
Residence after marriage . 3.1 1
R's residence
R's education before
......................
age 10 .. 10.0
1.0 2
R's birthplace ..................... 0.4 0
Explained by all independent
variables* .................... 75.8 16.5
* Increments may not add to total
Allan Schnaiberg
more useful analysis was carried out as follows. First, the three socio-
economic factors were run in the same multiple classification analysis,
and the "net effects" (the equivalent of the regression coefficients, treating
each category of the variable as a separate variable50) were recorded.
Then, each urbanism or residential pattern was cross-tabulated with each
of the three socioeconomic variables, and a "predicted" modernism score
was computed for each urbanism pattern. The computation was based
on the sum of the cross-products of the relative frequencies of each of the
categories of the socioeconomic variables mutliplied by the "net effect"
for that category. Each modernism index was treated separately, as usual,
so that we wind up with predictions for each residential history, across all
six modernism indices.
The results of this prediction operation are listed in table 7. What we
have done, in effect, is to test the linear additive model, in which all the
effects of urbanism operate through socioeconomic attainment. If the
predictions are accurate, as compared with the actual modernism scores
for each urbanism pattern, then we can be fairly confident that this
parsimonious model we have proposed does explain most of the modern-
ism differentials among urbanism patterns. The differences between our
predicted and the actual scores reflect two things, therefore: (1) the
additive effects of urbanism (those not explained by socioeconomic attain-
ment); and (2) the interactions among the socioeconomic variables, and
between them and the urbanism pattern. If we turn to table 7, the striking
thing to be observed is the very high level of accuracy in our predictions;
the differences unexplained are small and do not reflect any strong con-
sistent bias from one dimension of modernism to another. Thus the
explanatory power of our simple additive model appears fairly impressive.
Discussion
What are the inferences to be drawn from the findings above ? The most
important is that we can begin to understand a variety of research findings
on characteristics of rural and urban populations in developing areas.
Where cities offer access to educational achievement and subsequent
occupational roles that afford some degree of economic security, we should
expect to find high levels of modernism and high acceptance of innovations.
If urban agglomerations offer few such opportunities, or if a substantial
portion of the population in such areas has little of such access, we should
expect to find large areas of traditional behavior, a phenomenon elsewhere
termed that of the "urban villager."5' Indeed, many underdeveloped
societies contain such nonindustrialized large urban places, and many
development programs founder in such areas because of the underlying
assumption of urban modernity as a universal phenomenon.52 For
5o The most readable discussion of this appears in Beverly Duncan, "Education
and Social Background," American Journal of Sociology 72 (January 1967): 363-72.
51 See Herbert J. Gans, The Urban Villagers (Glencoe, Ill.: Free Press, 1962).
52 Lewis and Hauser both provide a detailed critique of this assumption.
99
0I~
0
TABLE 7
PREDICTION OF MEAN MODERNISM SCORES (STANDAR

FOR SELECTED RESIDENTIAL P
MODERNISM
ACTUAL A
RESIDENTIAL REF.T OR Mass Extended Nuclear Enviro
PATTERNt N PREDICTED Media Family Ties Family Roles Religio
A-A-A-A ...... 1A-P

P108
3.64A 3.89
3.24 3.29
3.513.53.
+ 0.25 + 0.05 + 0.04 +
A-A-I-I........ 2 51 A 4.24 3.44 4.00 4
P 4.06 3.45 3.93 3.
A-P +0.18 -0.01 +0.07 +
A-A-Pc-Pc ... 4 106 A 3.82 3.40 3.66
P 3.67 3.25 3.56 3.
A-P +0.15 +0.15 +0.10 -
A-A-Dc-Dc. ... 6 98 A 3.31 3.04 3.26
P 3.21 3.08 3.14 2.
A-P +0.10 -0.04 +0.12 +
A-A-V-V....... 8 158 A 2.68 3.06 2.60
P 2.65 2.79 2.70 2.
10 A-P +0.03 +0.27 -0.10 +0.0
V-V-V-V ....... 11 324 A 2.16 2.51 2.39
P 2.32 2.70 2.45 2.
A-P -0.16 -0.19 -0.06 -
Absolute average deviation ... 0.15 0.12 0.0
* (Z + 3) transformations of the modernism scores.
t Ordering within the pattern (left to right): current resi
birth to age ten. A = Ankara; I = Istanbul/Izmir; Pc = Pr
I Reference numbers are the same as those for residentia
Allan Schnaiberg
example, family-planning programs have often suffered from such un-

founded assumptions because they have failed to link rural-urban differen-
tials in fertility with socioeconomic differentials in one model. Yet our
analyses have indicated that urbanism is closely "locked in" with socio-
economic attainment in such societies, that the latter is the factor leading to
modernization, and that it is this modernization which is closely related to
demographic change.53 If we fail to understand that the primary deter-
minant of modernism is socioeconomic attainment, and that urbanism is a
less proximate factor, we fallaciously assume some homogeneity of
potential for demographic innovation in urban areas.
The long-term solution to development, presuming that modernism
is a prerequisite for social, economic, and demographic change in these
areas, is to provide a wider opportunity structure in both rural and urban
areas. In particular, educational opportunities for females may be very
critical for the success of many programs,54 along with occupational
improvement for their families.55 Although this suggests increased eco-
nomic inputs and a reversion to the usual "vicious cycle" of underdevelop-
ment, an alternative perspective may be found. It is true that future
development in part feeds on or requires earlier development,56 but once
the first stages of development have taken place, our findings strongly
suggest that permanent value and behavioral changes will provide the
basis for an upward spiral of development. As Blau and Duncan57 have
53 Modernism is closely linked to low fertility expectations and family planning

practice in the study population (e.g., Schnaiberg, "Some Determinants and Conse-
quences of Modernism in Turkey," chap. 7, and David Goldberg and Greer Litton,
"Family Planning: Observations and an Interpretive Scheme," Turkish Demography:
Proceedings of a Conference, Hacettepe University Publications no. 7, pp. 219-40.
These findings are similar to those of Kahl, chap. 5).
54 Whereas Bogue ("Attitude Change and Adoption of New Behavior," in Mass
Communication for Birth Control, ed. Donald J. Bogue and B. Straits [Chicago:
Community and Family Studies Center, 1967]) believes family planning programs
can be successful even in traditional rural society, I tend to be more skeptical. It is
becoming apparent that we need to resocialize women in order to reduce fertility
sharply, rather than simply provide information and supplies. Education may serve
that purpose better than a simpler family planning program and may open up eco-
nomic opportunities leading to lesser needs for children.
55 It should be stressed that this study deals primarily with females, and a simple
extension of the model for males would perhaps be inappropriate. That is, there may
be significantly less "locking in" of rural backgrounds and low socioeconomic
attainment for males, especially when we consider the possibility of self-selection of
rural migrants to urban destinations. Those who are most likely to overcome their
rural handicaps may be the most likely to migrate. But the broad outline of the model
may not be an unreasonable first approximation to the process of male moderniza-
tion. This view is reinforced by the fact that (a) many of the modernism items used
in the present study refer to the husband-wife dyad and not solely to the wife's
behavior; and (b) there is a fairly close relationship between husband's and wife's
background due to the process of homogamy in mating (e.g., the gamma coefficient
for the association between husband's and wife's education was +.85.)
56 With appropriate measurement over time, this example of "reciprocal causa-
tion" can be handled with our recursive statistical models (Blalock, pp. 55-57).
57 Blau and Duncan, especially pp. 404-5.
101
so cogently argued, this is not what we should designate with the title of
a "vicious cycle," for once we overcome the initial "locking in" of social
or residential origins and educational attainment, the influence of these
origins virtually disappears. And this appears to be as true of Turkish
women as American men.
On the level of sociological propositions or theories of development,
how do our findings alter the picture? Considering the differences in
sampling procedures, cultures, and sex of respondents, the results are
supportive of many of the conclusions drawn by Inkeles and Kahl.58 This
study suggests some fairly clear socialization and stratification models
for the development of modernism, pulling together elements which had
to some extent been separately discussed in the previous work.59 However,
there appear to be strong doubts about the validity of anything approach-
ing a uniform developmental theory across all major institutions, doubts
which have been raised on many earlier occasions.60 Whereas the primary
focus in both Kahl and Inkeles has been the measurement and use of some
single concept of modernism, the present study has stressed the lack of
support for such a conceptualization. Indeed, a close examination of the
earlier work brings forth findings very similar to ours, but the emphasis
in most of such studies has been on those aspects of "modernity" which
do fit very closely together.61 But it is equally important to delimit those
areas in our theories which fail to be substantiated empirically. Thus,
though we may indeed speak of "psychic unity," 62 there does not appear
to be such firm evidence for "social unity" or institutional integration in
the modernization process.
58 In particular, the role of early socialization factors appears well documented
in both the Harvard study and Kahl. There is less comparability for the later
socialization factors, since their samples are male (i.e., within the labor force) and
mine is female (with relatively low proportions in the labor force). Though none of
the studies were intended as replications, as indicated by the variation in themes,
indices, items, and sample designs, the results are strikingly similar.
59 Both Inkeles and Kahl discuss urbanism and education as factors influencing
modernism, but there is little analysis linking the two determinants together in a
developmental model. This is also somewhat true of the earlier work of Melvin M.
Tumin and Arnold S. Feldman, Social Class and Social Change in Puerto Rico
(Princeton, N.J.: Princeton University Press, 1961), although some tabulations there
come closer to the present model.
60 For example, by Bendix, Gusfield, Tumin, and Feldman. Moore (Social
Change, chaps. 2, 6) is a critic of such theories, and yet he feels compelled to ac-
knowledge some long-term utility to such theories (e.g., p. 116). The ambivalence in
Moore's position is reflected in much of the work, including the present study.
61 Although the reports of the Harvard group have stressed "overall modernity,"
there are occasional references to a "profile" of respondents on the several dimensions
of modernism (e.g., Smith and Inkeles, pp. 358-59, especially n. 10, p. 359; n. 22,
pp. 362-63). Thus, on the one hand, there is a heavy emphasis on the unity of mod-
ernism, but with the escape clause of the multidimensionality to come. As was the
case with Moore (see n. 60), there is a high degree of ambivalence underlying expres-
sions of the unity since there are no theoretical criteria for determining this, and even
the few statistical criteria, such as reliability, are subject to considerable misinterpreta-
tion, as I have noted earlier.
62 Smith and Inkeles, p. 377; Inkeles, "Making Men Modern," p. 212.
102
Allan Schnaiberg
Appendix
Items Included in the Modernism Indices
Mass Media Index
1. Attends a movie at least once or twice a month.
2. Reads a newspaper daily.
3. Listens to the radio daily.
4. Has read at least one magazine in the last month.
5. Has read at least one book in the last month.
Extended Family Relations Index

1. Not living with parents or in-laws now; or, if both sets are dead, neve
lived with them in the past.
2. Feels closer to friends, rather than relatives.
3. Has no relatives among three closest friends.
4. Has had a love marriage rather than an arranged one.
5. Husband is not a relative.
6. Children have no obligation to provide help, money, or material aid to
parents.
7. No expectation of old-age support from children.
Nuclear Family Role Structure Index

1. Disapproves or only mildly approves that males should make decisions.
2. Disapproves or only mildly approves that there should be a separation o
male and female work.
3. Disapproves or only mildly approves that women are irresponsible.
4. Disapproves or only mildly approves that men should not do housework.
5. Disapproves or does not care whether men are free to go out alone.
6. Approves or does not care whether mothers should be free to be active.
7. Husband permits wearing of short dresses.
8. Husband permits sitting together with male visitors.
9. Husband permits shopping alone.
10. Husband permits talking to men not known by husband.
11. Husband permits not wearing head covering.
12. Husband permits visiting women not known by husband.
13. Husband permits going to parties alone.
14. Egalitarian or wife's decision on choice of couples to visit.
15. Egalitarian or wife's decision on choice of relatives to see.
16. Egalitarian or wife's decision on purchase of major household items.
17. Egalitarian or wife's decision on budget for food.
18. Egalitarian or wife's decision of uses of savings and earnings.
19. Wife has some leisure activities outside the home.
Religiosity Index
1. Couple has had a civil marriage only.
2. Wife prays less than five times a day.
3. Wife does not fast for the entire period of Ramadan.
Environmental Orientation Index

1. Believes that daughters need the same education as do sons, or even more
education.
103
2. Believes that success is based more on hard work than on luck, or that
both are equally important.
3. Desired occupational sphere in corporations or profession.
4. Accurately perceives that the polar regions, Japan, Southeast Asia, or
western United States are the farthest points from Turkey.
Production/Consumption Index
1. Owns a radio
2. Owns a sewing machine.
3. Produces half or less of all the sweaters worn.
4. Produces half or less of all the dresses worn.
5. Produces half or less of all the soup consumed.
6. Produces half or less of all the tomato paste consumed.
7. Produces half or less of all the pickles consumed.
8. Does not produce home preserves very frequently.
104

Modernsing Impact of Urbanisation

Uploaded by

Copyright:

Available Formats

Modernsing Impact of Urbanisation

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modernsing Impact of Urbanisation

Uploaded by

Copyright:

Available Formats

The Modernizing Impact of Urbanization: A Causal Analysis

Author(s): Allan Schnaiberg

Of what import is this type of analysis? On a theoretical level, it

substantiated by the high degree of similarity between the findings here

Definitions of the Central Concepts

hierarchy of Turkey, in terms of size of place. Ankara's position is am-

ism was defined by the observer and in some ideal-typical fashion,

(i.e., the traditional structure), in that the operational definition of mod-

28 Particularly Smith and Inkeles, Inkeles, and Kahl.

together under the rubric of "modernism" or "modernity." But their

Causal Model and Hypotheses

social action, while those being reared in urban places tended to be

EARLY RESIDENCE R'S EDUCATION ) R'S MODERNISM

FIG. 1.--Proposed early socialization model for the developmen

In addition to the two hypotheses raised above, there remains one

The Effects of Urbanism on Modernism

DIFFERENTIALS IN MEAN MODERNISM SCORES, BY EARLY RESIDENCE

MEAN MODERNISM SCORE

Ankara......... 3.31 3.84 10.39 1.22 1.99 4.58

Ankara......... 3.89 3.29 3.55 3.38 3.33 3.34

bility among indexes with varying numbers o

DIFFERENTIALS IN MEAN MODERNISM SCO

RESIDENTIAL Mass Extended Nuclear En

early residence, the woman's modernism is also affected by her timing of

EFFECTS OF PREVIOUS RESIDENCE (FOR MIG

MIGRATED TO Mass Extended Nuclear En

Istanbul/Izmir ........... 3 23 4.21 3.51

Istanbul/Izmir ........... 2 51 4.24 3.45

EFFECTS OF TIMING OF MIGRATION (FOR MIG

MIGRATED TO ANKARA Mass Extended Nuclear

production and consumption. In particular, one might cite the unavail-

Testing the Model

Mass Extended Nuclear

residence is quite small, once the three socioeconomic variables have

FIG. 2.-Proposed earl

Mass Extended Nuclea

* Increments may not add to total

PREDICTION OF MEAN MODERNISM SCORES (STANDAR

A-A-A-A ...... 1A-P

example, family-planning programs have often suffered from such un-

53 Modernism is closely linked to low fertility expectations and family planning

Extended Family Relations Index

Nuclear Family Role Structure Index

Environmental Orientation Index

You might also like