Life-Span Developmental Psychology. Historical and Generational Effects

297 Pages·1984·4.5 MB·English

by  Kathleen A. McCluskey (Eds.)

Life-Span
Developmental Psychology
Historical and Generational Effects
Edited by
KATHLEEN A. McCLUSKEY
HAYNE W. REESE
Department of Psychology
West Virginia University
Morgantown, West Virginia
1984
ACADEMIC PRESS, INC.
(Harcourt Brace Jovanovich, Publishers)
Orlando San Diego New York London
Toronto Montreal Sydney Tokyo
COPYRIGHT © 1984, BY ACADEMIC PRESS, INC.
ALL RIGHTS RESERVED.
NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR
TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC
OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY
INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT
PERMISSION IN WRITING FROM THE PUBLISHER.
ACADEMIC PRESS, INC.
Orlando, Florida 32887
United Kingdom Edition published by
ACADEMIC PRESS, INC. (LONDON) LTD.
24/28 Oval Road, London NW1 7DX
Library of Congress Cataloging in Publication Data
Main entry under title:
Life-span developmental psychology.
(West Virginia University series on life-span develop-
mental psychology)
Includes index.
1. Developmental psychology—Methodology—Addresses,
essays, lectures. I. McCluskey, Kathleen A.
II. Reese, Hayne Waring, Date . III. Series.
BF713.5.L527 1984 155 84-11061
ISBN 0-12-482420-X (alk. paper)
PRINTED IN THE UNITED STATES OF AMERICA
84 85 86 87 9 8 7 6 5 4 3 2 1
Contributors
Numbers in parentheses indicate the pages on which the authors' contributions begin.
AARON ANTONOVSKY (143), Department of the Sociology of Health,
Center for Health Sciences, Ben-Gurion University of Negev, Beer-Sheva,
Israel
NANCY DATAN (143), Human Development, University of Wisconsin-
Green Bay, CHB ES-301, Green Bay, Wisconsin 54301-7001
LUTZ H. ECKENSBERGER (73), Universität des Saarlandes, Saarbrücken,
Federal Republic of Germany
GLEN H. ELDER, JR.1 (161), Departments of Human Development and
Family Studies and Sociology, Cornell University, Ithaca, New York
14853
BERNARD J. JAWORSKI (161), Graduate School of Business, University
of Pittsburgh, Pittsburgh, Pennsylvania 15261
ELISABETH KASPER (73), Universität des Saarlandes, Saarbrücken, Fed-
eral Republic of Germany
STEPHEN L. KLINEBERG (129), Department of Sociology, Rice Univer-
sity, Houston, Texas 77251
BERND KREWER (73), Universität des Saarlandes, Saarbrücken, Federal
Republic of Germany
GISELA LABOUVIE-VIEF (109), Department of Psychology, Wayne State
University, Detroit, Michigan 48202
JEFFREY K. LIKER (161), Department of Industrial and Operations En-
gineering, University of Michigan, Ann Arbor, Michigan 48109
BENJAMIN MAOZ (143), Center for Health Services, Department of
Psychiatry, Ben Gurion University of Negev, Beer-Sheva, Israel
KATHLEEN A. McCLUSKEY (17), Department of Psychology, West Vir-
ginia University, Morgantown, West Virginia 26506-6040
JOHN A. MEACHAM (47), Department of Psychology, State University of
New York at Buffalo, Buffalo, New York 14226
NINA R. NAHEMOW (249), Department of Sociology, Old Dominion
University, Norfolk, Virginia 23508
Present address: Carolina Population Center, University Square, Chapel Hill, North Caro-
lina 27514.
IX
X Contributors
ROSS D. PARKE (203), Department of Psychology, University of Illinois,
Champaign, Illinois 61820
HAYNE W. REESE (17), Department of Psychology, West Virginia Uni-
versity, Morgantown, West Virginia 26506-6040
K. WARNER SCHAIE (1), Individual and Family Studies, College of Hu-
man Development, The Pennsylvania State University, University Park,
Pennsylvania 16802
BARBARA R. TINSLEY (203), Department of Psychology, University of
Illinois, Champaign, Illinois 61820
Preface
The topic of the eighth West Virginia University life-span volume was
chosen to complete our coverage of the three broad categories of influences
on life-span development: normative age-graded influences were covered in
the fourth life-span volume, published in 1975 (Datan & Ginsberg, 1975);
nonnormative influences were covered in the seventh life-span volume, pub-
lished in 1983 (Callahan & McCluskey, 1983); and history-graded influ-
ences are covered in the present volume. History-graded influences are by
definition associated with definite time periods and affect significant por-
tions of the population in similar ways, although the effects may differ for
different age groups living at the time of such events. Examples of such
events include economic depressions, wars, movements toward moderniza-
tion, and changes in the demographic and occupational structure of a soci-
ety. Much of the research relevant to this topic has been conducted within
disciplines other than psychology—primarily within sociology. We have
therefore departed from the tradition of previous volumes in the life-span
series, in which most of the contributions were by psychologists, and have
included herein contributions by sociologists as well as psychologists in
about equal proportions.
History-graded influences reflect biological as well as environmental de-
terminants of development. They are associated with historical time and
therefore with changing contexts. They produce so-called cohort effects,
which in turn often become functionally autonomous as cohort-graded in-
fluences. As Eckensberger, Krewer, and Kasper point out (Chapter 4 in this
volume), the study of history- and cohort-graded influences demands ex-
plicit investigation of "the developing individual in a changing world,"
which has come to be the implicit motto of life-span developmental psychol-
ogy and life-course sociology. Viewed in this way, history- and cohort-
graded influences take on special significance in these disciplines, in that age-
graded influences are by their nature resistant to historical change, and
nonnormative influences are by definition related more to individual differ-
ences than to norms—whether they reflect age-graded stability or history-
graded change. However, in spite of being fundamentally important for an
understanding of life-span development, history- and cohort-graded influ-
ences are the least well-investigated category of influences. The chapters in
XI
XU Preface
this volume should facilitate expansion of relevant research by providing
theoretical and methodological frameworks and by providing exciting
examples of what can be learned.
REFERENCES
Callahan, E. J., & McCluskey, K. A. (Eds.). (1983). Life-span developmental psychology:
Nonnormative life events. New York: Academic Press.
Datan, N., & Ginsberg, L. H. (Eds.). (1975). Life-span developmental psychology: Normative
life crises. New York: Academic Press.
Eckensberger, L. H., Krewer, B., &c Kasper, E. (1984). Simulation of cultural change by cross-
cultural research: Some metamethodological considerations.
Acknowledgments
The success of the eighth West Virginia University Conference on Life-
Span Developmental Psychology was facilitated by a large number of indi-
viduals. Stanley Wearden, Interim Dean of the College of Arts and Sciences,
helped in providing university funds. Kennon A. Lattal, Acting Chairman of
the Department of Psychology, was very generous with departmental re-
sources and also helped to oversee the organization of the conference. Car-
rie Koeturius, in her role as director of the West Virginia University Confer-
ence Office, worked with us from the beginning to ensure the smooth
operation of the entire conference. Dr. and Mrs. Franklin Parker deserve
special recognition for their very charming and personal account of life
events in the past three decades given in their keynote address. Thanks are in
order for the graduate students in developmental psychology at West Virgi-
nia University who worked with the co-chair from the inception of this
conference through the very end: Mary Kay August, Donna Barré, Linda
Holt, Beverly Hummel, Jim Killarney, Nancy Meek, Dennis Papini, Gale
Richardson, Rosellen Rosich, C. J. R. Simons, and Jeanne Thomas. Ann
Davis provided many extra hours of secretarial work, for which we are
grateful.
The contributors to this volume and the other participants who attended
the conference deserve recognition for creating an invigorating intellectual
climate that aided the creation of the essays published in this volume.
CHAPTER 1
Historical Time and Cohort Effects
K. WARNER SCHAIE
INTRODUCTION
Historical accounts of the life-span psychology movement are likely to
point out that one of its major contributions may have been to shift the
focus of concern from the search for purely developmental patterns of a
normative nature to the context within which development occurs. This
context, of course, not only refers to the characteristics of place and culture,
but also includes as a major parameter the historical time during which
development occurs. Concerns about the influence of historical period,
however, have emerged largely from the study of adults. Children, as
emergent organisms, might reasonably be assumed to possess some charac-
teristics that are constant throughout history because those characteristics
are involved in the establishment of behavioral competencies essential for
survival. But few, if any, of these characteristics are important for develop-
ment during much of adulthood, even though survival-relevant behaviors
might again merit concern for the study of advanced old age. Therefore, I
have in the past argued that scientific interest in age-related behavioral
change recedes for those variables in which a behavioral asymptote is
reached in young adulthood. Instead, close attention needs to be given to
cohort and period effects (Schaie, 1973a, 1977).
In retrospect, it seems that much of the concern with methodologies
designed to separate age, cohort, and period effects has stemmed from our
preoccupation with the role of age as the independent variable of prime
interest to developmentalists. Not unlike the early experimental psychol-
ogists who saw individual differences as a primary source of unwanted error
variance, developmentalists have often treated historical time and genera-
LIFE-SPAN DEVELOPMENTAL PSYCHOLOGY 1 Copyright © 1984 by Academic Press, Inc.
HISTORICAL AND GENERATIONAL EFFECTS All rights of reproduction in any form reserved.
ISBN 0-12-482420-X
2 K. Warner Schaie
tional effects as confounds to be controlled and explained away. Thus, it is
not without a good deal of justification that Rosow (1978) could argue that
the work on the sequential strategies (e.g., Bakes, 1968; Schaie, 1965)
treated the effects other than age as nuisances and that any information
developed on them was at best incidental.
Such a position, of course, has never been true for all developmen-
talists. Riegel, throughout the latter portion of his work, vehemently argued
for a dialectic interplay of historical events with life-stage and cohort effects
(1972, 1975, 1976). More recently, Sinott (1981) has considered implica-
tions of the theory of relativity for the study of development, which may
provide considerable metatheoretical support for what we are about to
consider. It should also be noted that several sociologists have addressed the
interface of life stages and cohorts theoretically as well as substantively (e.g.,
Carlsson & Karlsson, 1970; Elder, 1974; Ryder, 1965). In my own work I
have paid just as much attention to the estimation of period and cohort
effects as to those of chronological age, perhaps at times even de-emphasiz-
ing the role of the age variable (e.g., Schaie, 1979, 1982b). But it remains
quite true that not much has been done thus far to go beyond the description
and identification of period and cohort effects, although attention has been
called to the potential importance of these effects for fields as diverse as
mental health, adult education, and the professional problems of engineers
at midcareer (Schaie, 1978, 1981, 1982; Schaie & Willis, 1978).
The time has come then to remedy the lack of substantive attention
given to the specific meaning of historical time and cohort. If this is to be
done most constructively, it is not enough to appeal to sociologists and
historians to provide definitions for what might be the most relevant soci-
etal changes and cohort boundaries for ordering developmental phe-
nomena. It might be more constructive to begin by sketching out a frame-
work that behavioral scientists can apply to the study of historical events
that may be relevant to life-span development. To do this effectively, it may
be necessary to broaden the concepts of cohort and period, to suggest
methods for scaling the possible impact of historical events upon behavioral
phenomena, and to suggest ways in which individual differences in position
on space-time templates for diverse attributes might permit more creative
uses of age as a dependent variable. In the course of this attempt, much of
which is still highly speculative and provisional, it is possible that new
insights may emerge on how the apparent stalemate in the estimation of age,
cohort, and period effects could perhaps be finessed. But before we get too
ambitious, let us begin far more humbly by recalling the kind of data that
have persuaded at least some of us to leave the comforts of a static and
ahistorical approach to the study of human development.
1. Historical Time and Cohort Effects 3
HISTORICAL TIME AND COHORT EFFECTS
IN PSYCHOLOGICAL DATA
Some concern had been expressed earlier regarding the impact of social
change upon behavioral variables (e.g., Kuhlen, 1940), but fairly little for-
mal attention had been given by developmental psychologists to the impact
of generational or historical events. This lack of attention began to change
when my concern with the discrepancies between cross-sectional and longi-
tudinal findings on changes in adult intelligence led to the publication of my
paper on the general developmental model (Schaie, 1965). What I had
noted, essentially, was the fact that when I compared data from two cross-
sectional samples drawn from the same parent population 7 years apart, the
mean values on ability measures for the later sample exceeded those for the
earlier sample with great regularity (Schaie & Strother, 1968). In addition,
the overall mean for subjects at all ages also differed positively over time.
The implication of these findings suggested that there could either be
the phenomenon of a unique period effect active across all cohorts studied,
or that there was a long-term trend involving successively higher perfor-
mance asymptotes in young adulthood (Schaie, 1983). It soon became ob-
vious that to resolve these two alternatives it would be necessary to conduct
an additional data collection, to permit construction of what we now call a
longitudinal sequence (Schaie & Bakes, 1975). That is, data were needed
that allow comparison of two or more cohorts followed over the same age
range, a procedure that requires a minimum of three measurement points.
This is what we did, and we were then persuaded that we were not faced
with a period trend unique to the original time span, but that we were
actually faced with substantial cohort differences (Schaie & Labouvie-Vief,
1974).
It will not surprise anyone that at the time of that study we lacked
reasonable hypotheses with respect to the substantive meaning of either
period or cohort effects (Schaie & Labouvie-Vief, 1974). The time period
studied was an artifact of the timing of research funding; the cohort bound-
aries (and consequently the age ranges) were arbitrarily fixed to be equiv-
alent to that time period. Because our initial interest was indeed the control
of confounds for the age variable, this approach made sense. It simplified
numerical analyses and permitted comparison of the magnitudes of variance
components. In fact, this approach (perhaps inordinately directed toward
attaining methodological sophistication) also gave rise to a number of meth-
odological controversies (see Adam, 1978; Botwinick & Arenberg, 1976;
Schaie & Hertzog, 1982), which may turn out to be of little substantive
interest. For example, comparison of age and cohort effects, using equal