Evidence from the Longitudinal Study of Generations Merril Silverstein PhD Professor of Gerontology and Sociology Davis School of Gerontology Department of Sociology University of Southern California ID: 133679
Download Presentation The PPT/PDF document "Understanding Family Complexity in the S..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Understanding Family Complexity in the Study of Intergenerational Relationships:Evidence from the Longitudinal Study of Generations
Merril Silverstein, Ph.D.Professor of Gerontology and SociologyDavis School of GerontologyDepartment of SociologyUniversity of Southern CaliforniaSlide2
Families Through Historical Time
Increased longevity means greater co-survival between generations and prolonged relationships. Possible kinship issuesFertility declineHigher prevalence of divorce, remarriage, step-familiesGeographic distance increasingWeaker sense of filial obligation
How to study social change in real time instead of using retrospective reports or using “proxy” evidence?
How to better approach families systemically?Slide3Slide4
Studies of Families and Social Change Using a single individual as informant about family process at one historical moment limits research questions that can be addressed
Use of retrospective reports has biasesCross-sectional comparisons regarding social change of interest (e.g., divorced vs. married) ignores socio-historical contextCohort studies in repeated cross-sections ignore intra-familial dependence and cannot address issues that require parent-child dataSlide5
5Generational-Sequential Design
Members of different generations in the same families measured at the same age but at different historical periods to test for effects of social conditions at a common life-stage.Useful for studying age-dependent processes where social conditions are also changing.Slide6
Comparison of Intergenerational Relations Across Historical Contexts
Historical/generational change in the quality of intergenerational relationshipsRequires early reports from parents and later reports from childrenHas the quality of older parent-child relations weakened over historical time?If so, is this related to:Increasing geographic distanceRising divorce ratesWeakening norms of familismSlide7
The USC Longitudinal Study of Generations (LSOG)
A multigenerational multi-time-point study, started in 1971 with repeated panels 2005.Consists of about 3,000 individuals from 374 three-generation families recruited within Southern California region.
Full families are surveyed: grandparents, parents, and grandchildren (16+), including siblings, spouses, former spouses.
Fourth generation added in 1991 (Fifth generation in 2010).Slide8
Design of LSOGSlide9
Multi-generational Family ClustersSlide10Slide11Slide12
Application of Generational Sequential DesignDo G3 children maintain less close relationship to their parents than G2 parents maintained with
their parents?Is so, does a G3-G2 difference persist after controlling for individual-level variables representing the “social change” of interest.Methodological individualism: characteristics of serial generations proxy the social change of interest by virtue of their unique historical/cohort experiences.Slide13
Sample & DesignData for this analysis from LSOG: 554 G2s in 1971 and their G3 children surveyed between 1991 2005.
G2s averaged 44 years of age in 1971.G3s reached the age of each parent somewhere between 1991-2005. For each G3 we use the survey that matches the closest to their parent’s 1971 age. Use multilevel modeling to estimate change in emotional closeness to parents over time in G2s and G3s, comparing (1) slopes and (2) levels at the historical time when they match in age.Slide14
Cross-Generational Comparisons in the LSOG
T1
T2
T3
T4
T5
T6
T7
T8
Year
1971
1985
1988
1991
1994
1997
2000
2005
G2
43
57
60
63
66
59
72
77
G3
20
34
37
40
43
46
49
54
G4
16
19
22
25
30Slide15
Cross-Generational Comparisons in the LSOG
T1
T2
T3
T4
T5
T6
T7
T8
Year
1971
1985
1988
1991
1994
1997
2000
2005
G2
43
57
60
63
66
59
72
77
G3
20
34
37
40
43
46
49
54
G4
16
19
22
25
30Slide16Slide17Slide18Slide19
Multi-level Regression Results Predicting the G3-G2 Cohort Gap
**
*
**
**
*Slide20
20
Cross Generational-Sequential Transmission of values, attitudes, beliefs, behavioral tendencies across age-matched generations within the same families.Multi-actor data?Causal direction?Research questions focusing on interdependencies and influence across family actors over time call for unique approaches.Slide21
Religion is a family affair.Children are socialized to religious traditions by parents and grandparents
Do grandparents influence the values, attitudes, and beliefs of their grandchildren beyond the influence of parents, synergistically with parents, and as mediated by parents?Slide22
LSOG Data: Lagged TriadsGrandparents in 1971
(mean age =44)G2 = 257Parents in 1988 (mean age = 40)G3 = 341Grandchildren in 2005 (mean age = 31)G4 = 565 Slide23
Measures of Religiosity
PracticeAttendance at religious services: “never” to “everyday”SalienceImportance of “a religious life” ranked among 13 social valuesIdentityHow religious are you?: “not at all” to “very religious”BeliefsStrength of conservative religious beliefs: agreement with statements
God exists in the form as described in the Bible
All people today are descendents of Adam and Eve
All children should receive religious training
Religion should play an important role in daily life
Additive scale (standardized factor score) computed for each generationSlide24
Nesting of Grandchildren in Two Three-Generational Families: Basis for Multi-level Modeling
Grandparent: Red
Grandparent: Green
Parent #1
Parent #2
Parent #3
Parent #1
Parent #2Slide25
Empirical Results from Multilevel Models Transmission of Religiosity
Grandparent Religiosity
1971
Parent
Religiosity
1988
Grandchild
Religiosity
2004
.10*
.38***
.32***Slide26
Parents’ direct influence is almost four times that of grandparents, but grandparents
do
directly influence their grandchildren net of parents.
Grandparents also indirectly influence their grandchildren through parents. Total influence of grandparents (.22) is 58% that of parents (.38).
Source: Copen & Silverstein, 2007,
Journal of Comparative Family Studies
.Slide27
Grandchildren are most religious when both their parents and grandparents are more religious.
Suggests that several generations together reinforce a family culture of religiosity.Slide28
Grandparents are better able to transmit their religiosity to grandchildren within intact families.
Parental divorce is associated with less religiosity in their children; grandparents do not compensate.Slide29
Measures of Gender Role AttitudeHusbands ought to have the main say in family matters [Disagree]
Women’s liberation ideas make a lot of sense to me [Agree]It goes against nature to put women in positions of authority over men [Disagree]Women who want to remove the word “obey” from the marriage service don’t understand what it means to be a good wife. [Disagree]Additive scale (standardized factor score) computed for each generationSlide30
Grandmother Gender Role Attitudes
1971
Mother Role Attitudes
1988
Grandchild
Gender Role Attitudes
2005
.11
.16**
.09**
Mother Contact with Grandmother
1988
.10*
Empirical Results from Multilevel Models Transmission of Gender Role AttitudesSlide31
31
Longitudinal Generational-Sequential Design in the LSOG Using 14 YearsSlide32
Summary
Generational-sequential designs provide useful tools for understanding how societal change is manifest in micro-family environments and across multiple family members.Generational differences can be investigated with GSD in terms of change across cohortsIntergenerational ties weakening over historical time.And in terms of cross-cohort continuityIntergenerational transmission occurring (and possibly changing) over historical time.