APC Models and Methods for Crosssectional Research Designs

1
APC Models and Methods for Cross-sectional
Research Designs
 

• Comments on the State of the Art
• Duane Alwin
• Sociology and Demography
• Penn State University

2
APC Models and MethodsIssues to address

• (1) How important is theory?
• (2) What are the optimal designs?
• (3) How best do we explore available repeated
  cross-sectional surveys?
• (4) Substantive areas as examples of application
  of APC ideas

 
3
Guidelines for sense makingwith respect to APC
models
 

• Theoretical considerations concerning aging
  processes, impacts of historical events and
  processes, and the phenomena of cohorts and
  generations.
• Research design and data moving beyond
  cross-sectional designs.
• The role of exploratory methods in making sense
  of repeated cross-sections and the importance of
  looking at the data.

4
Theoretical assumptions of APCs

• Theoretical concept
• Aging age-related within person change
• Historical events eras, epochs and times
• Generations, youth movements and social change

• APC Operationalization
• Agebetween person differences in age
• Periodeffects tied to the date of the survey
• Cohortshistorical location versus historical
  participation

 
5
Theoretical Limitations of APCs

• Theoretical concept
• Aging age-related within person change
• Historical events eras, epochs and times
• Generations, youth movements and social change

• APC Limitations
• Between person differences are not the same thing
• Periodan effect tied to the date of the survey
  is a limiting concept
• Cohortshistorical location is not the same thing
  as historical participation

 
6
Beyond APC models

• Theoretical concept
• Aging age-related within person change
• Historical events eras, epochs and times
• Generations, youth movements and social change

• Remedies
• Latent curve models using longitudinal data
• Read W.H. Sewells Logics of History
• Cohorts vs. generationspay better attention to
  classical writings of Mannheim and Ryder

 
7
What is Aging?

• Changes within persons associated with the
  passage of biographic time
• Changes linked to species-specific biological
  (life cycle) and/or neurological processes of
  maturation and development (gains and losses)
• Changes linked to culturally constructed
  age-graded experiences, i.e. life stages or phases

8
Example of Word FluencyGSS

• Alwin, D.F. (1991). Family of Origin and Cohort
  Differences in Verbal Ability. American
  Sociological Review, 56, 625-638.
• Alwin, D.F. McCammon, R.J. (1999). Aging vs.
  Cohort Interpretations of Intercohort Differences
  in GSS Verba Scores. American Sociological
  Review, 64, 272-286.
• Alwin, D.F. McCammon, R.J. (2001). Aging,
  Cohorts and Verbal Ability. Journal of
  Gerontology Social Sciences, 56B, S151-S161.
• Alwin, D.F. (2008). History, Cohorts, and
  Cognitive Aging. In H. Bosworth C. Hertzog
  (Eds.), Cognitive, Social and Psychological
  Development Essays in Honor of Warner Schaie.
  Washington, D.C. American Psychological
  Association.

9

10

11
(No Transcript)
12
(No Transcript)
13
Example of Immediate Word RecallHRS

• Alwin, D.F. and others. (2008). Population
  Processes and Cognitive Aging. In S.M. Hofer
  D.F. Alwin (Eds.), Handbook of Cognitive
  AgingInterdisciplinary Perspectives (pp. 69-89).
  Thousand Oaks, CA Sage Publications.
• Alwin, D.F. (2008). The Aging Mind in Social and
  Historical Context. Unpublished research
  monograph.

14
(No Transcript)
15
Logic of Growth Curve Analysis

• Multiple group models where each group is a
  different birth cohort
• Begin with an age-based model this is the
  conventional convergence model that assumes no
  cohort effects
• Remove constraints on cohort-specific intercepts
  and slopes
• Examine plots of the data and goodness-of-fit
  information

16
(No Transcript)
17
(No Transcript)
18
Historical Events and Processes
 

• How do events and processes occurring in
  historical time affect the lives of individuals?

19
Three ways to think about the effects of
historical time

• Episodic Events e.g., wars, recessions,
  depressions, political crises
• Eras, epochs or historical periods e.g., the
  60s, the sexual revolution, the cultural
  revolution, the womens movement, the Civil
  Rights era
• Transitions or gradual social change

 
20
Examples

• Research on the
• Aging of the Baby Boomers
• Alwin, D.F. (1998). The Political Impact of the
  Baby Boom Are there Generational Differences in
  Political Beliefs and Behavior? Generations, 22,
  4654.

 
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
 
25
 
26
Effects of Cohort

• Enduring attributes of persons associated with
  unique placement of birth cohorts in historical
  time
• Result from the intersection of biographical and
  historical time
• Linked to culturally constructed generational
  experiences

27
Generation as historical participation

• An element of historical time
• Common experience of participation in movements
  brought to the individual through historical
  events
• No necessary convergence with other concepts of
  generation
• Persons are nested in social movements and social
  organizations that are nested within time periods
  and groups of cohorts

28
References

• Alwin, D.F. McCammon, R.J. (2003).
  Generations, Cohorts and Social Change. In J.T.
  Mortimer M.J. Shanahan (Eds.), Handbook of the
  Life Course (pp. 23-49). New York Kluwer Acadmic
  / Plenum Publishers.
• Alwin, D.F. McCammon, R.J. (2007). Rethinking
  Generations. Research in Human Development, 4,
  219-237.
• Alwin, D.F. (2008). Whos Talking About My
  Generation? In M. Silverstein R. Giarrusso
  (Eds.), From Generation to Generation Continuity
  and Change in Aging Families, Baltimore, MD
  Johns Hopkins University Press.

29
(No Transcript)
30
(No Transcript)
31
Cohort replacement theory

• Human beings tend to develop their beliefs and
  orientations early in adulthood.
• Differences are likely to develop between cohorts
  due to differences in historical
  location/participation.
• Beliefs and orientations remain relatively stable
  over the adult life span.
• Social change results from personnel replacement

 
32
Karl Mannheim in The Problem of Generations (1952)

• Even if the rest of ones life consisted of one
  long process of negation and destruction of the
  natural world view acquired in youth, the
  determining influence of these early impressions
  would still be predominant.

33
Ryder on Individual development

• (p. 851) Implicit in the foregoing account of the
  interdependency of social change and cohort
  differentiation is the assumption that an
  individuals history is highly stable or at least
  continuous. The model dominating the literature
  on human development presents life as a movement
  from amorphous plasticity through mature
  competence toward terminal rigidity.

34
Ryder on Stability

• The continuity of individual life (p. 856)A
  persons past affects his present and his present
  his future. Persistence is enhanced by the
  tendency to structure inputs, so that each will
  disturb as little as possible the previous
  cognitive, normative or even esthetic design,
  and, in the extreme, to reject dissonant items.
  An individuals life is an organic entity, and
  the successive events that constitute it are not
  random but patterned.

35
Ryder on Stability

• The role of early socialization(p. 856)every
  society seizes upon the circumstances of birth as
  modes of allocating status, limiting the degrees
  of freedom for the persons path through life.
  Virtually every subsequent occurrence will depend
  upon the characteristics present at birth sex,
  race, kinship, birthplace and so forth.

36
Ryder on Stability

• The role of commitment to a way of life(p.
  857)Beyond the age of noncommitment, the new
  adult begins a process of involvement in the
  various spheres of life, in which his actions and
  those of others progressively reduce the degrees
  of freedom left to him in the societal scheme.
  Within each role, once allocated, he forms a
  growing commitment to a line of activity.
  Conformity to (social structures) implies
  resistance to change.

37
Ryder on Stability

• As life takes on a steadier tempo, routinization
  predominates(p. 858)Routines are barriers to
  change because they limit confrontation with the
  unexpected and the disturbing. Older people learn
  to exercise greater control over a narrower
  environment, and avoid risks of venturing into
  unstructured situations. The feasibility of
  personal transformation is probably limited more
  by restricted membership than by psychological
  aging.

38
Alternative theory

• Humans do acquire their beliefs and orientations
  in young adulthood, but cohorts do not differ
  substantially in their experiences.
• There is a great deal of heterogeneity in the
  generational experiences of a given cohort
• Persons tend to change their positions on issues
  across the life span and remain relatively
  flexible over time.
• Intra-cohort change, not cohort replacement, is
  responsible for the bulk of social change.

 
39
Kenneth Gergen in Life-Span Development (1980)

• For any individual the life course seems
  fundamentally open-ended. Existing developmental
  patterns appear potentially evanescent, the
  unstable result of the particular juxtaposition
  of contemporary historical events. Even with
  full knowledge of the individuals past
  experience, one can render little more than a
  probabilistic account of the broad contours of
  future development.

40
Peter Berger in The Homeless Mind (1974)

• Modern identity is peculiarly open Not only
  does there seem to be a great objective capacity
  for transformations of identity in later life,
  but there is also a subjective awareness and even
  readiness for such transformations.

41
Peter Berger in A Rumour of Angels (1971)

• But even when the world of everyday life retains
  its massive taken-for-granted reality it is
  threatened by the marginal situations of human
  experience the haunting presence of
  metamorphoses.

42
Frank Musgrove in Margins of the Mind (1977)

• The evidence suggests that adults are capable
  of more fundamental change than many
  psychologists will admit, but that
  consciousness, identity, and the self are
  far more resilient and resistant to change than
  important schools of sociology and social
  psychology will concede. We are not, in fact,
  chameleons.
•  

43
Optimal Designs

• Theoretical concept
• Aging age-related within person change
• Historical events eras, epochs and times
• Generations, youth movements and social change

• APC Operationalization
• Repeated observations on the same persons
• Historical and archival data
• Supplemental data on cohort experiences

 
44
The Classic Identification Problem

• There is a linear dependence among variables that
  quantify aging, period and cohort (A, P and C)
• Age Period - Cohort
• Period Cohort Age
• Cohort Period Age
• If you know any two of these variables you know
  the third
• By removing one of the factors we can more
  clearly see the natural confounding of the other
  two

45
Implications of the Identification Problem

• Cross-sectional comparisons of age groups include
  effects of both aging and cohort
• Cross-time comparisons within cohorts include
  effects of both aging and period

46
(No Transcript)
47
(No Transcript)
48
Final caveatMortality Selection

• Mortality is obviously selective and to the
  extent that selectivity is linked to factors
  associated with levels of outcome variables, then
  mortality selection is a potential explanation
  for many findings of cohort effects. Indeed, it
  might be the case that such performance-linked
  selectivity in survivorship might be masking a
  true cohort effect that favors more recent born
  cohorts. Therefore a strong argument can be made
  that differential age-specific mortality rates
  should be taken into account when examining age
  differences in health-related outcomes, or when
  comparing cohorts in patterns of age-related
  within-cohort change.

49
Conclusions

• The reality is that for many developmental
  phenomena there are plausible expectations for
  the simultaneous influence of aging, period and
  cohort factors
• The complete understanding of human behavior must
  consider the plausibility of all of these
  influences
• The natural confounding of the effects of aging,
  cohorts and periods in longitudinal data makes it
  difficult (if not impossible) to detect their
  independent effects

50
Conclusions (continued)

• Historically analysts of longitudinal data have
  not made their assumptions about the effects of
  aging, cohort and period effects explicit
• Early solutions to the identification problem in
  developmental research produced futile results
• There are no ready made solutions to the
  confounding of aging, period and cohort effects,
  and no substitute for careful and informed
  analysis of longitudinal data