Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Adolescents' social attitudes: Genes and culture?
(USC Thesis Other)
Adolescents' social attitudes: Genes and culture?
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript
and there are missing pages, these w ill be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9" black and white
photographic prints are available for any photographs or illustrations appearing
in this copy for an additional charge. Contact UMI directly to order.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ADOLESCENTS’ SOCIAL ATTITUDES: GENES AND CULTURE?
By
Amy C. Abrahamson
Copyright 2000
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment o f the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)
May 2000
Amy C. Abrahamson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UM I Number: 1405203
___ ®
UMI
UMI Microform 1405203
Copyright 2001 by Bell & Howeil Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIV ERSITY O F S O U T H E R N C A L IFO R N IA
TH E GRA D U A TE S C H O O L
U N IV ER SIT Y P A R K
I-O S A N G ELES. C A L IF O R N IA > 0 0 0 7
This thesis, written by
under the direction of hJSL£Z.~Thesis Com m ittee,
and approved by all its m em bers, has been pre
sented to and accepted by the D ean of The
G raduate School, in p a rtia l fulfillm ent of the
requirem ents for the degree of
■Ma-s-tex— q.£.-A x . 1 l3.
Dt m m
Date.
2000
/
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ii
Table of Contents
List o f Tables ......................................................................................... iii
List o f Figures......................................................................................... iv
Abstract ......................................................................................... v
Introduction ......................................................................................... 1-6
Methods ......................................................................................... 6-14
Results 15-25
Discussion ......................................................................................... 25-28
References ......................................................................................... 29-30
Appendix ......................................................................................... 31-36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
iii
List o f Tables
Table 1. Predicted Covariances for All Biological and Adoptive Relatives ............. 12
Table 2. Means and Standard Deviations for Social Attitude Scales......................... 16
Table 3. Correlations Among Social Attitude Measures for Ages 12 to 15 Years ... 17
Table 4. Parent-Child and Sibling Correlations for Conservatism and
Religious Attitudes......................................................................................... 18
Table 5. Standardized Estimates of Genetic (A), Common Environmental (C),
Unique Environmental (E) and Gene-Environment Covariation (G-E)
Components of Variance and Covariance in Social Attitude Measures .... 22
Table 6. Model Fitting Results for Social Attitude Data at Ages 12 to 15 Y ears 24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV
List o f Figures
Figure 1. Multivariate path diagram model of familial resemblance for nuclear
families with two biologically related siblings ......................................... 9
Figure 2. Multivariate path diagram model of familial resemblance for adoptive
families with two adoptive siblings and their adoptive rearing parents ... 10
Figure 3. Multivariate path diagram model of familial resemblance for adoptive
families where one sibling is the biological offspring of the adoptive
parents ............................................................................................................. 11
Figure 4. Relative contributions o f additive genetic (A), common environmental (C),
unique environmental (E) and gene-environment covariance (A-C) factors
to variations in conservatism and religious attitudes ................................. 20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
Abstract
The present study investigated genetic and environmental sources o f variation
and covariation in social attitude measures obtained from 654 children in both
biological and adoptive family members participating in the Colorado Adoption
Project. The children annually from ages 12 to 15 completed social attitude measures
o f conservatism and religious attitudes years. These same attitudes were also
measured among parents during the 12-year old assessment. Model fitting results
from the present multivariate genetic analysis indicate that variations in both
conservatism and religious attitudes are strongly influenced by shared-family
environmental factors throughout adolescence. In contrast to previous findings from
twin studies, which suggest that genetic influence on social attitudes does not emerge
until adulthood, the present study detected significant genetic influence on variations
in the conservatism measure as early as age 12. However, there was no evidence of
genetic influence on religious attitudes during adolescence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Amy C. Abrahamson Laura Baker
ADOLESCENTS’ SOCIAL ATTITUDES: GENES AND CULTURE?
Abstract
The present study investigated genetic and environmental sources of variation
and covariation in social attitude measures obtained from 654 children in both
biological and adoptive family members participating in the Colorado Adoption
Project. The children annually from ages 12 to 15 completed social attitude measures
o f conservatism and religious attitudes years. These same attitudes were also
measured among parents dining the 12-year old assessment. Model fitting results
from the present multivariate genetic analysis indicate that variations in both
conservatism and religious attitudes are strongly influenced by shared-family
environmental factors throughout adolescence. In contrast to previous findings from
twin studies, which suggest that genetic influence on social attitudes does not emerge
until adulthood, the present study detected significant genetic influence on variations
in the conservatism measure as early as age 12. However, there was no evidence of
genetic influence on religious attitudes during adolescence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
Adolescent Social Attitudes: Genes and Culture?
INTRODUCTION
All individuals express certain sets of beliefs or attitudes towards various
political, social, moral and religious issues. Variations in these attitudes are evident at
least as early as young adolescence (Eaves et al., 1997), and by adulthood appear to be
relatively stable individual characteristics (Eaves, Eysenck and Martin, 1989). What
factors influence the development o f such attitudes? It is probably easy to imagine
how one’s social attitudes might be influenced by different experiences and
circumstances throughout life, or through social interactions with others such as
friends, family or teachers. Is it possible however, that one’s attitudes are also
influenced in part by one’s genes?
Indeed, several researchers have addressed this very question, and by
employing genetically informative designs, have examined the relative contributions
of genetic and environmental factors to variations in social attitudes. Surprisingly,
most of these studies have found evidence for genetic as well as environmental
components of variance in social attitudes (Eaves and Eysenck, 1974; Martin et al.,
1986; Waller et al., 1990). In addition, environmental influence on social attitudes
appears to be of the nonshared variety, experienced differently by children growing up
in the same family rather than a shared family environment that would be expected if
social attitudes were transmitted culturally via the family. However, most behavioral
genetic studies of social attitudes have focused primarily on attitudes in adulthood.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
In contrast to most findings for adults, a large twin study by Eaves et al.
(1997), reports that for subjects aged 20 years or younger, the effects of the shared
environment on a conservative dimension o f social attitudes are overwhelming
whereas genetic influence is negligible. Eaves’ study examined twin correlations for
conservatism scores among a large sample of identical (monozygotic; MZ) and non
identical (dizygotic; DZ) twin pairs aged 9 to 75+ years. Twin correlations were
substantial at all ages, although the pattern o f differences between MZ and DZ
correlations varied over the lifespan. Among older adults, MZ twins shared more
similar attitudes than DZ twins, suggesting genetic effects on social attitudes are
indeed expressed in adults. However, among twins aged 20 years or younger, DZ
twin correlations were roughly equivalent to those o f MZ twins, suggesting that twin
resemblance for conservatism at younger ages is entirely the result of nongenetic
factors and that these environmental influences were substantially shared by twins
growing up in the same family.
Thus, at first glance, the various findings might suggest that any genetic
influence in social attitude variation does not express itself until sometime during
adulthood. However, an important issue, which is difficult to resolve with data from
twin studies alone, concerns how to separate shared environmental or cultural effects
from the genetic consequences o f assortative mating. If positive assortative mating
(or parental similarity), for a genetically influenced trait occurs, parent-offspring and
sibling resemblance for that trait will be greater than expected under conditions of
random mating. In studies o f twins reared together, these effects are completely
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
confounded (Eaves, Eysenck and Martin, 1989). Assortative mating increases DZ
twin correlations relative to MZ twins, thus mimicking the effects of shared-
environment.
Assortative mating effects are particularly problematic in twin studies o f social
attitudes. Unlike personality measures, which show only low to moderate spousal
resemblance, available evidence suggests that positive assortative mating for social
attitudes is quite strong (Feng and Baker, 1994; Martin et al., 1986). In an attempt to
address this problem, an analysis by Martin et al. (1986), re-examined data from two
separate studies of adult twins who had been reared together, in conjunction with data
gathered on their spouses. Once assortative mating effects, as estimated from spousal
correlations, were factored into the analysis, estimates o f any shared environmental
contribution to family resemblance in adult social attitudes were negligible.
Attempting to test further the claims by Martin et al. (1986), that all apparent
cultural effects on attitudes actually result from the genetic effects of assortative
mating, Truett et al. (1992) conducted a multivariate genetic analysis of attitudes,
church attendance and educational attainment. The authors hypothesized that if the
observed correlations among religion, education and attitudes derive solely from the
genetic consequences o f assortative mating for religion and education, then the
variable loadings of shared environmental factors on these variables should be
constant multiples of their loadings on genetic factors. Model fitting results indicated
that the shared environmental loadings could not be scaled relative to the genetic
loadings for these factors, thus arguing against a completely genetic explanation o f
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
familial influence. The authors further proposed that if cultural effects were in force,
the common environment may be mediated by church attendance and education.
Twin resemblance for church attendance was found to be almost entirely the result of
shared family influences. In addition, this shared family environmental factor also
explained a significant portion of family resemblance in social attitudes. Overall,
findings from this study suggest that some o f the attitude variation previously
attributed to the genetic consequences of assortative mating, may in fact be culturally
influenced.
Data from twin studies alone will not suffice in the effort to frilly resolve
questions concerning cultural transmission and assortative mating. The adoption
design, however, can serve as a powerful tool for identifying cultural and shared
environmental effects which are independent of any genetic resemblance between
family members. Since there is no genetic relationship between adopted children and
their adoptive parents, any observed parent-child similarities must derive solely from
common environmental influences, and will not be confounded with the genetic
consequences of assortative mating.
The present study investigated genetic and environmental sources of variation
in social attitude measures obtained from both biological and adoptive families. The
primary goal of the present research was to investigate whether previous findings for
strong common environmental influences on social attitudes during adolescence
continue to emerge when a more powerful test of such factors, the adoption design, is
employed. A multivariate genetic analysis of two social attitude measures, including a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
conservatism scale and a religious attitudes scale, was conducted to estimate the
relative influences of assortative marting, genetic, common (shared) environmental and
unique (nonshared) environmental factors on variations in these social attitudes. In
addition, the significance of paternal and maternal cultural transmission effects were
examined independently as forms o f common environmental influence. Cultural
transmission may be viewed as a comm on or shared environmental influence that
results specifically from the effects o f parents’ phenotypes (social attitudes) on the
environment of their children (Fulker, 1988).
Secondly, the multivariate design allows for covariation among any latent
genetic or environmental factors w hich may influence different social attitudes. As
such, in addition to examining the to tal contribution of genetic and environmental
factors to variations in each of the social attitude measures, sources o f covariation
• t-.
between measures were investigated in the present study. To illustrate more
specifically, it was hypothesized thatt conservative attitudes would be related to
religious attitudes. If so, what factorrs might account for this association? As noted
earlier, Truett et al. (1992) found tha_t shared family environmental factors which
influenced church attendance also exiplained a significant portion o f family
resemblance in social attitudes. The present study explored whether a similar findings
would emerge for shared environmental influences on conservatism and religious
attitudes. That is, do shared environmental factors which influence religious attitudes
also explain variations in levels of conservatism? Furthermore, the possible overlap
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
of genetic or unique environmental factors influencing these two attitude measures
was also examined.
METHODS
Participants
The sample includes families participating in the Colorado Adoption Project
(CAP), an ongoing, longitudinal study that has followed adoptive and non-adoptive
children and their families from infancy through young adulthood. The present study
examines social attitude data from 654 adopted and biological children (including 188
sibling pairs), and their rearing parents. Social attitude measures were not available
for the ‘birth’ parents of adopted children. Detailed descriptions of the CAP sample,
including subject recruitment and testing procedures, have been previously reported
(DeFries, Plomin and Fulker, 1994; Plomin and DeFries, 1985).
Procedures
Social attitude measures were initially assessed through interviews conducted
during a lab visit when the children were 12 years old. Parents were assessed by
questionnaire during the child’s age 12 lab visit only. A subset o f the parents
(145 mothers and 116 fathers) completed the questionnaire twice, once during the first
child’s age 12 assessment, and again when a younger sibling was 12 years old. This
subset is used to assess the reliability/stability o f each attitude scale over time,
however only parent ratings from the first available assessment are retained for
subsequent analyses. The average time lapse between first and second assessments for
these parents was 3 years. Continued assessments of children’s social attitudes were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
obtained annually by means of telephone interviews for ages 13-15 years. Social
attitude measures were not obtained from the biological parents o f adoptees.
Measures
The conservatism dimension o f social attitudes is derived from a 28-item scale
modeled after the Wilson-Patterson (1970) Conservatism scale. These items are
identical to those used by Eaves et al. (1996) to assess conservatism in twins aged
9-75+ years. This scale consists of single-word items such as death penalty, gay
rights, censorship, Republicans etc., to which participants are instructed to respond
along a five-point scale (1= “Strongly Agree” to 5= “Strongly Disagree”). Children
were instructed to tell the interviewer if they did not understand a given item so that
such items could be further explained to them. Conservatism scores among parents
were highly stable over time, with first and second assessment scores correlating .86
for mothers and .87 for fathers.
The religious attitudes measure was adapted from one scale o f Jessor and
Donovan’s Health Questionnaire (1985). This scale consists o f 5 items which
respondents rate in terms o f importance along a 5-point scale (1= “not at all
important” to 5= “very important”). Items on this scale question the importance of
believing in God, relying on religious counsel, relying on religious beliefs as a guide,
turning to prayer and church attendance. This scale also appears to be highly stable
over time among parents, with stability coefficients of .89 for mothers and .87 for
fathers. The individual questionnaire items for both attitude scales are included in an
Appendix.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
Analyse?
Structural equation modeling techniques were employed using the Mx
software package (Neale, 1997), to conduct a multivariate genetic analysis of social
attitudes during adolescence. An Mx model previously outlined by Neale et al.
(1994), for the multivariate genetic analysis of twin-family data was modified for use
with adoptive and nuclear family data. This model is based upon a factor model
described by Phillips and Fulker (1989), and allows for estimates o f genetic, common
environmental, unique environmental and assortative mating effects. In addition, the
model allows maternal and paternal cultural transmission parameters to be estimated
separately. These effects in turn contribute to the overall estimates o f common
environmental effects.
Figure 1 shows a multivariate path diagram model o f familial resemblance for
nuclear families with two biologically related siblings. Figures 2 and 3 show
corresponding path diagrams, respectively, for adoptive families with two adoptive
siblings and their adoptive rearing parents, and adoptive families where one sibling is
the biological offspring of the adoptive parents.
Table 1 shows the predicted covariances between relatives derived from these
models. All variance components are specified as separate matrices in the model,
yielding an oblique factor structure for both genetic and common environmental
effects, and an orthogonal factor structure for the effects of unique environmental
factors. The structure and form o f all matrices in the model are provided by Neale et
al. (1994). The full Mx model for the present analyses is included in an Appendix.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
s s
Rc
Rc Ra Ra
Em Am A f a Cm
m
A s l
Esl As2 Es2
Ps2
Pm
PsI
Figure 1. Multivariate path diagram model o f familial resemblance for nuclear
families with two biologically related siblings. Parents’(PF and PM) and siblings’
(PSI and PS2) phenotypes are influenced by latent additive genetic (A), common
environmental (C) and unique environmental (E) factors. Matrix D represents
assortative mating parameters and matrices m and f allow for cultural transmission
from the parents’ phenotypes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
s s
Rc
Rc Ra Ra
Af Af
Mf
m
As 1
As2 Esl Es2
Psl Ps2
Figure 2. Multivariate path diagram model of familial resemblance for adoptive
families with two adoptive siblings and their adoptive rearing parents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
s s
Rc
Rc Ra
Am Af Cm Em
m
I
A sl
As2 EsI Es2
Ps2 P sl
Figure 3. Multivariate path diagram model o f familial resemblance for adoptive
families where one sibling is the biological offspring of the adoptive parents
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
Table 1
Predicted Covariances for All Biological and Adoptive Relatives
Covariance Matrix Predicted Covariance
Bioloeical Relatives:
Mother-child (Rpm’ + W f)c’ + (1 + RpD)T’(.5a’)
Father-child (Rpf + Wm’)c’ + (I + RpD)T’(.5a’)
Siblings .5a(Ra + ,5(T(D’ + D)T’))a’ + cRcc’ + asc’ +csa’
Adoptive Relatives:
Mother-child (Rpm’ + W f )c’
Father-child (Rpf + Wm’)c’
2 adopted siblings cRcc’
Siblings (1 natural) cRcc’ + csa’
Several modifications were made to the twin-family model to accommodate
data for the present analyses. First, two additional calculation groups were created to
estimate separate parent-offspring covariances between adopted children and their
rearing parents. Parent-offspring covariance estimates within biological families are
computed just as they would be for twins and their parents, and are a function of both
genetic and shared environmental pathways. For adoptive families however, there is
no genetic correlation between parents and their adopted children. As such, the
predicted covariances are a function of cultural or shared environmental pathways
only.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
In addition, the present model requires three separate groups to estimate
covariance matrices for sibling pairs. Predicted covariances between biological
siblings are calculated just as they would be for DZ twin pairs, and are a function of
both genetic and common environmental effects. For two adopted siblings, predicted
covariances are derived solely from shared environmental factors. For adoptive
sibling pairs where one child is biologically related to the parents, the predicted
covariances are not identical to those for two adopted siblings. In both instances, the
adoptive siblings are not genetically related. However, as shown in Figure 3, if one
sibling is genetically related to the parents, an additional pathway between the two
siblings arises. This pathway results from allowing gene-enviromnent correlations for
the natural child o f the adoptive parents. Thus, the common environment of the
adopted sibling may be indirectly related to the genes of the biological sibling. This
additional pathway must therefore be included to derive the predicted sibling
t
covariances for this group.
It is important to highlight the implications of distinguishing between these
two types o f adopted sibling pairs. In essence, it must be recognized that for the
purpose o f genetic analyses, not all adoptive sibling pairs are equal. If a given trait is
influenced by correlated gene-environment effects, adopted sibling pairs which
include one child who is the biological offspring o f the adoptive parents will be
expected to show greater trait resemblance than two adopted sibling pairs where
neither child is genetically related to the adoptive parents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Fourteen separate raw data groups were created to be read into the present
model. The first four groups include families with data for parents and a single
child only. Separate groups were created for adoptive and non-adoptive
families as well as male and female children. Families with data for parents and
two biological siblings were divided into two same sex sibling groups and 1
opposite sex siblings group. These same three groups were created for families
with data for parents and two adopted siblings. The final four data groups (two
for same-sex siblings and two for opposite-sex siblings) were created from
families where one child is the biological child o f the parents and one child is
adopted.
M aintaining separate sexes in each group allowed for preliminary model
fitting analyses to test for sex differences in parent-offspring or sibling
covariances. These analyses revealed no significant sex differences in covariance
structure, either within persons or between parent-offspring pairs. Consequently,
the genetic analysis was not expanded to test for sex-limited genetic or common
environmental effects. Findings for sex-limited effects in adult twin studies of
social attitudes have been mixed (Martin et al., 1986). As such, it would be
desirable to investigate the possibility o f such effects. Unfortunately, perhaps in
part due to the very limited number of same sex adoptive sibling pairs, sex-
limited effects may be difficult to detect in the present data set.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
RESULTS
Preliminary Analyses
Means and standard deviations for males and females at ages 12 through 15 on
both the conservatism and religious attitudes scales are presented in Table 2. A 2 x 2
analysis of variance (ANOVA) was performed on each of the social attitude scales at
all ages to test for mean sex differences. F values from the ANOVA procedures are
also shown in Table 2. Males scored significantly higher than females on the
conservatism scale at all ages. By contrast, females scored significantly higher than
males on the religious attitudes scale at all ages.
Mean scores for all parents on the social attitude scales are also shown in
Table 2. Consistent with the findings for adolescents, significant mean differences
were also found between mothers and fathers for conservatism and religious attitudes.
A paired-samples t-test o f mean differences was conducted on the two parent scores
i
for each of the social attitude measures. Fathers scored significantly higher on the
conservatism scale (T = 7.92) and mothers scored significantly higher on the religious
attitudes scale (T = -8.692).
Pearson correlations among the two social attitude scales across all ages are
presented in Table 3. Conservatism scores were highly stable over the four year
period with year-to-year correlations ranging from .60 to .78. Scores on the religious
attitudes scale were also highly stable with year-to-year correlations ranging from .73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
Table 2
Means and Standard Deviations for Social Attitude Scales
CONSERVATISM RELIGION
Males Females F Males Females F
Age 12 80.0 75.4 14.20** 17.3 18.7 12.24**
(SD) (8.9) (7.9) (5.3) (4.8)
N= 334 310 333 310
Age 13 78.5 74.7 27.61** 16.7 17.8 5.79*
(SD) (9.2) (8.3) (5.8) (5.4)
N= 306 273 306 274
Age 14 79.2 75.0 22.62** 15.6 17.6 12.38**
(SD) (9.4) (9.5) (5.9) (6.0)
N= 239 221 ■ 239 221
Age 15 79.3 74.1 24.49** 15.1 16.8 7.16**
(SD) (10.6) (10.2) (6.2) (6.1)
N= 198 203 198 204
Fathers Mothers I Fathers Mothers x
Parents 85.5 80.3 7.92** 18.2 20.3 -8.69**
(SD) (14.2) (13,9) (6.0) (5.4)
N= 334 371 333 372
*P<.05 **p<.01
to .81. Overall, the year-to-year stability of these measures appeared to increase as the
children grew older. Furthermore, at all ages, conservatism and religious attitudes
showed a significant positive within-person association. The correlation among these
measures ranged from .32 to ..38, both within and across all measurement periods.
Prior to the model fitting analyses, parent-offspring and sibling correlations on
each attitude scale were computed for both adoptive and biological relatives. The
correlations among these relatives at all ages are presented in Table 4. For
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
conservatism scores, the correlations among biological relatives are larger than those
found for adoptive relatives. This trend was evident at all ages, and suggests that
some genetic influence is likely to be involved in the development of these attitudes
during adolescence. Significant positive correlations among adoptive relatives for
conservatism scores however, suggests that common environmental influences are
also present.
Parent-child correlations for the religious attitudes scale were equally high
among both adoptive and biological relatives, indicating that these attitudes may be
culturally transmitted from parents to children. By contrast, at ages 12 and 13 years,
r
biological sibling correlations for the religious attitudes scale were notably higher than
Table 3
Correlations Among Social Attitude Measures for Ages 12 to 15 Years
Year 12 Year 13 Year 14 Year 15
Consrv Relig Consrv Relig Consrv Relig Consrv Relig
Yr 12 Consrv
Relig .33**
Yr 13 Consrv .60** .34** -
Relig .32** • • .73*t- .34** —
Yr 14 Consrv .62** .29**. .72** .31** —
Relig .32** .70^* .33** .79** .32** —
Yr 15 Consrv .57** .31** .66** .33** .78** .38** —
Relig .33** .60** .32** .70** .33** .81** .38**
*p<.05 **p<.01
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
Table 4
Parent-Child and Sibling Correlations for Conservatism and Religious Attitudes
ADOPTED BIOLOGICAL
Father-
Child
Mother-
Child
Sibling Father-
Child
Mother-
Child
Sibling
Yr 12 Consrv .30** .37** .20 .54** .60** .56**
Relig .41** .41** .20 .47** .56** .63**
(n)
(240) (262) (95) (254) (305) (96)
Yr 13 Consrv .31** .29** .14 .59** .60** .62**
Relig .42** .44** .15 .47** .57** .64**
(n)
(208) (229) (66) (231) (274) (73)
Yr 14 Consrv .33** .30** .45** .54** .60** .62**
Relig .43** .44** .31 .47** .56** .52**
(a)
(162) - (178) (29) (197) (230) (45)
Yr 15 Consrv .32** .25** .29 .52** .59** .53**
Relig .36** ' -.-42** .44* .43** .48** .41*
(n)
(154) •(168) (28) (163) (193) (37)
*p<.05 **p<.01
for adoptive siblings, suggesting that some genetic influence may be present.
However, by age 15, sibling correlations were roughly equal among both groups.
Overall therefore, these preliminary findings suggest that common environmental
factors are likely to be the primary determinant of family similarity for religious
attitudes during adolescence.
Model Fitting Results
The full factor model depicted in Figures 1-3 was employed to fit social
attitude data for ages 12 to 15 years. Estimates of the relative influence o f genetic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
(A), common environmental (C), unique environmental (E), and gene-environment
covariance (A-C) effects on variations in the two social attitude measures, as obtained
from this full unconstrained model, are presented in Figure 4. Overall, these results
suggest two relatively distinct patterns of influence on variations conservatism and
religious attitudes.
Familial influences on both conservatism and religious attitudes appear to be
substantial during adolescence. For conservatism, both genetic and common
environmental factors accounted for substantial proportions o f variance at all ages.
•i*tn \ r
Heritability estimates ranged from .16 to .24. For ages 12 to 14, common
environmental influences accounted for almost half of the total variation in
conservative attitudes. However, there is some indication that common environmental
influences may begin to decline, and genetic influences may start to become stronger
around age 15. Nonetheless, additional data points beyond age 15 would be needed to
discern such a trend. Variations in religious attitudes also appear to be highly
influenced by common environmental factors, however in contrast to conservatism,
estimates of genetic influence were negligible, ranging from .00 to .06.
The standardized estimates of genetic (A), common environmental (C), unique
environmental (E), and gene-environment (A-C) covariation components of variance
and covariance in the two social attitude measures, obtained from fitting the full
unconstrained factor model at each age, are presented in Table 5. The diagonal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CONSERVATISM
20
Age 12
□ E
E3A-C COV
0 C
E3A
Age 13 Age 14 Age 15
RELIGION
Age 12 Age 1 3
□ E
□ A-C COV
HC
□ A
Age 14 Age 15
Figure 4. Relative contributions of additive genetic (A), common environmental (C),
unique environmental (E) and gene-environment covariance (A-C) factors to
variations in conservatism-and religious attitudes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
elements in Table 5 represent the total amount of variance in each o f the two social
attitude measures explained by the above factors (A, C, E and A-C). As such, the
vertical sum of these elements for each attitude measure is equal to one. The off
diagonal elements depict the components of covariation between the two factors. The
sum o f these elements is equal to the estimated phenotypic correlation between the
two social attitude measures.
The phenotypic correlation between conservatism and religious attitudes may
therefore be derived in this manner. As expected, the correlation between these two
measures is notably high, ranging from .37 to .44 at each age. Genetic factors
; i-
accounted for little o f the covariation between these attitude measures. As shown in
Table 5, the genetic components of covariation between conservatism and religious
attitudes ranged between .01 and .04 at all ages, thus accounting for only 2-10% of the
covariation between these two measures. By contrast, common environmental factors
appear to have a much greater impact on the relationship between conservatism and
religious attitudes. Common environmental factors accounted for 32-81% o f the
covariance between these two measures across all ages.
Table 6 provides a summary of the results obtained when alternate models
were fit to the social attitude data at each year. When structural equation models are
fit to raw data, Mx computes minus twice the log-likelihood of the data, with an
arbitrary constant that is a function of the data. As such, no absolute measure of fit is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Table 5
Standardized Estimates of Genetic TAT Common Environmental fCY Unique
Environmental fF> and Gene-Environment Covariation fG-Ei Components of
Variance and Covariance in Social Attitude Measures
Year 12 Year 13 Year 14 Year 15
Genetic Factors (A)
Conservatism .16 .18 .17 .24
Religion .04 .05 .03 .06 .03 .00 .01 .00
Common Environment (C)
Conservatism .45 .42 .46 .11
Religion .22 .47 .33 .37 .30 .43 .14 .25
Unique Environment (E)
Conservatism .22 .24 .22 .49
Religion .05 .40 -.04 .49 -.04 .55 .21 .74
Gene-EnvCov (G-E)
Conservatism .16 .17 .16 .16
Religion .11 .08 .10 .09 .08 .02 .08 .01
* All estimates derived from fitting the full unconstrained factor model.
computed. However, relative measures o f fit may be obtained by comparing the
differences in fit function for two alternative models. Calculating the difference in fit
function (-2Ln) between two submodels, along with the change in the degrees of
freedom, provides a chi-squared test of the change in fit. A significant chi-squared
(p<.05) indicates that the alternate model being tested results in a significantly worse
fit of the data. In this manner, the significance of dropping various parameters from a
given model may be tested.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
Genetic factors. The significance of dropping all genetic parameters from the
model may be determined by comparing the difference in fit function between model
II and the full unconstrained factor model (model I). Model II sets all additive genetic
(A) and gene-environment covariance (S) parameters equal to zero. As shown in
Table 6, dropping all genetic parameters from the model results in a significantly
worse fitting model (p<.05) at all ages. Model Ha allows for genetic influence on
conservatism only, and fixes all genetic and gene-environment covariance parameters
contributing to religious attitudes at zero. These constraints produce no significant
change in overall fit relative to model I. However, allowing for genetic influence on
religious attitudes alone, while dropping all genetic and gene-environment covariance
parameters for conservatism, does result in a significantly worse fit of the data relative
to the full unconstrained factor model (model I) at every age (p<.01). As such, genetic
factors do appear to have a significant influence during these years, but only on the
conservatism measure.
Common environmental factors. The significance o f dropping all common
environmental parameters was tested by comparing the overall fit o f model H I, which
sets all common environmental (C), gene-environment covariance (S), and maternal
and paternal cultural transmission parameters equal to zero, to the full unconstrained
model (model I). As noted earlier, cultural transmission effects contribute to overall
estimates of common environmental influence. Dropping all common environmental
parameters resulted in a significantly worse fitting model at every age (p<.01). In
model Hla, the full model was refit to the data for each year setting both maternal and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Table 6
Model Fitting Results for Social Attitude Data at Ages 12 to 15 Years
12 Years 13 Years 14 Years 15 Years
I. Full Model; ACEMFD
[estimated parameters=42]
19366.1 18472.7 17416.9 16765.3
n. Drop A; A=S=0
[df=7]
19387.6
[21.5]**
18494.3
[21.6]**
17431.0
[14.1]*
16781.6
[16.3]*
a. Drop A,S for Religion
[df=4]
19369.3
[3.2]
18475.0
[2.3]
17417.2
[0.3]
16765.5
[0.2]
b. Drop A,S for Conservatism
[df=4]
19386.5
[20.4]**
18493.0
[20.3]**
17430.9
[14.0]**
16781.5
[16.2]**
in. Drop C; M=F=C=S=0
[df=15]
19533.8
[167.7]**
18596.4
[123.7]**
17509.0
[92.1]**
16820.3
[55.0]**
a. Drop Cult Trans; M=F=S=0
[df=12]
19504.1
[138.0]**
18580.2
[107.5]**
17495.9
[79.0]**
16815.7
[50.4]**
b. Drop C,M,F,S for Religion
[df=8]
19479.7
[113.6]**
18566.0
[93.3]**
17474.4
[57.5]**
16806.7
[41.4]**
c. Drop C,M,F,S for Consrv
[df=8]
19520.8
[154.7]**
18591.8
[119.1]**
17505.4
[88.5]**
16818.0
[52.7]**
IV. Drop Assortment; D=0
[df=4]
19736.5
[370.4]**
18873.4
[400.7]**
17819.2
[402.3]**
17137.2
[361.9]**
a. Drop Assortment for Relig
[df==3]
19499.3
[133.2]**
18587.7
[115.0]**
17530.3
[123.4]**
16871.5
[106.2]**
b. Drop Assortment for Consrv
[df=3]
19607.3
[231.2]**
18720.8
[248.1]**
17654.7
[240.8]**
16978.6
[213.3]**
* p<.05 **p<.01 Chi-square test for change in fit relative to the full unconstrained model (Model I)-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
paternal cultural transmission parameters to zero. These constraints also produced a
significant change in fit relative to model I (p<.01). Model Hlb tests the significance
of dropping all common environmental parameters for religious attitudes only, and
model file drops all common environmental parameters for conservatism only. Both
of these sub-models also proved to be a significantly worse fit o f the data as compared
to model I (p<.01). Overall therefore, it appears that all common environmental
parameters, including maternal and paternal cultural transmission, are required to
explain variations in both social attitude measures.
Assortative mating. Finally, model IV sets all assortment parameters (D) to
zero. As shown in Table 6, this constraint produces the most dramatic change in fit
relative to the full unconstrained model. In addition, as shown in models IVa and
IVb, constraining the full model to allow for assortment on conservatism alone or
religious attitudes alone, also results in a significantly worse fitting model (p<.01).
Combined, these results indicate that significant spousal assortment is present for both
conservatism and religious attitudes.
DISCUSSION
Overall, findings from the present analyses suggest that two relatively distinct
patterns of genetic and environmental influence contribute to variations in the
expression of conservatism and religious attitudes during adolescence. In summary,
familial influences appear to be substantial for both conservatism and religious
attitudes. Model fitting results suggest that family resemblance for conservative
attitudes arises from both genetic and common environmental factors. By contrast,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
familial influence on religious attitudes derives almost entirely from shared-family
environmental factors.
The conservatism scale for the present study was comprised of 28 items
identical to those used by Eaves et al. (1996). Eaves’ reports that among adolescents,
the effects of the shared environment on variations in conservative attitudes are
overwhelming, and that twin resemblance for this dimension of social attitudes is
entirely the result of non-genetic factors. In contrast to these findings, results from the
present analysis suggest that family resemblance for conservatism is not entirely the
result of shared-environmental factors. While common environmental effects were
found to be substantial, genetic factors also accounted for a significant proportion of
the variance in conservatism scores at all ages. By age 15, genetic influences even
outweighed the effects of the shared environment. Estimates of common
environmental influence were markedly lower at age 15, with unique environmental
factors accounting for a greater proportion of the variance in conservatism scores.
One possible explanation for these disparate findings is that as noted earlier, in
studies of twins reared together, shared-environmental effects are confounded with the
genetic consequences of assortative mating. That is, positive assorative mating
increases DZ twin correlations relative to MZ twins, thus mimicking the effects of the
shared environment. Consistent with findings from earlier studies (Martin et al.,
1986; Feng & Baker, 1994), significant assortment for both social attitude measures
was found among parents in the present study. Such strong assortment for social
attitudes may have contributed to the relatively high DZ twin correlations found
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
among adolescents. Since there is no genetic relationship between adoptive family
members, estimates of shared-environmental influences in the present design are
independent of the genetic consequences o f assortative mating.
It should also be noted that the common environment shared by adoptive
family members may not be identical to that shared by twins. Estimates of common
environmental influence in the current adoption design derive from patterns of
similarity between adopted children and their parents, as well as adoptive siblings o f
varying ages. There may be important differences between adoptive family
relationships and traditional family relationships which have an impact on how
shared-family environmental influences operate. On the other hand, many unique
features may also contribute to the common environment shared by twins. Twins are
necessarily the same age and may be similarly influenced by a number of
circumstances and experiences not shared by other traditional or adoptive family
members.
In contrast to the current findings for conservatism, both preliminary and
model fitting results suggest that genetic factors have no significant impact on
variations in religious attitudes during adolescence. Common environmental effects
however, were found to be significant across all four ages. In addition, estimates
derived from fitting the full factor model at each age suggest that these common
environmental influences also account for a substantial proportion o f the observed
covariation between religious attitudes and conservatism. These findings are
consistent with the multivariate genetic analysis reported by Truett et al. (1992).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Among adults, twin resemblance for church attendance was found to be almost
exclusively the result o f shared family influences. Furthermore, this common
environmental factor also accounted for a significant proportion o f family
resemblance in social attitudes.
Ultimately, results from the present analyses indicate that genetic factors may
actually exert an influence on social attitudes much earlier than previously indicated.
Significant genetic influences on variations in conservatism in this study were
discovered as early as age 12. Consistent with previous findings, the present study
provides further evidence that shared environmental factors also contribute to
variations in social attitudes during adolescence. It is noteworthy however, that unlike
previous estimates o f shared environmental influence obtained from twin studies,
estimates of shared family influences in the present analysis are free from any
confounding genetic effects of assortative mating. This strength o f the adoption
design is especially important given the high degree of positive assortment that is
found for social attitudes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
References
DeFries, J. C., Plomin, R., & Fulker, D.W. (1994). Nature and nurture during
middle childhood. Cambridge, MA: Blackwell.
Donovan, J. E., lessor, R., & Jessor L. (1985). Structure of problem behavior
in Adolescence and Young adulthood. Journal o f Consulting and Clinical Psychology.
Z L 890-904.
Eaves, L. J. & Eysenck, H. J. (1974). Genetics and the development of social
attitudes. Nature. 249. 288-289.
Eaves, L. J., Eysenck, H. J., & Martin, N. G. (1989). Genes. Culture and
Personality: An empirical approach. Academic Press, London.
Eaves, L., Martin, N., Heath, A., Schieken, R., Meyer, J., Silberg, J., Neale,
M., & Corey, L. (1997). Age changes in the causes of individual differences in
conservatism. Behavior Genetics. 27. (2), 121-124.
Feng, D. & Baker, L. (1994). Spouse similarity in attitudes, personality and
psychological well-being. Behavior Genetics. 24. (4), 357-364.
Fulker, D. W. (1988). Genetic and cultural transmission in human behavior.
In B. S. Wei, J. Eisen, M. Goodman, & G. Namkoong (Eds.), Proceedings of the
second international conference on quantitative genetics, (pp. 318-340).
Massachusetts: Sinaur Associates Inc.
Martin, N. G., Eaves, L. J., Jardine, R., Heath, A. C., Feingold, L. F., &
Eysenck, H. J. (1986). Transmission of social attitudes. Proc. Natl. Acad. Sci. USA.
£2* 4364-4368.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Neale, M. C., Walters, E. E., Eaves, L. J., Maes, H. H., & Kendler, K. S.
(1994). Multivariate genetic analysis of twin-family data on fears: Mx models.
Behavior Genetics. 24. (D. 119-139.
Neale, M. C. (1997). Mx: Statistical Modeling. Box 126 MCV, Richmond,
VA23298: Department o f Psychiatry. 4th Edition.
Phillips, K., & Fulker, D. W. (1989). Quantitative genetic analysis of
longitudinal trends in adoptive designs with application to IQ in the Colorado
Adoption Project. Behavior Genetics. 19. 621-658.
Waller, N. G., Kojetin, B. A., Bouchard, T. J., & Lykken, D. T. (1990).
Genetic and environmental influences on religious interests, attitudes and values: A
study of twins reared apart and together. Psychological Science. 1. 138-142.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Appendix A
Conservatism Scale Items
1. capital punishment 15. integration
2. astrology 16. capitalism
3. x-rated movies 17. segregation
4. modem art 18. moral majority
5. women’s liberation 19. passivism
6. foreign aid 20. censorship
7. federal housing 21. nuclear power
8. democrats 22. living together
9. military drill 23. republicans
10. the draft 24. divorce
11. abortion 25. school prayer
12. property tax 26. unions
13. gay rights 27. socialism
14. liberals 28. bussing
Religious Attitudes Scale Items
“How important is it to you...?
1. to be able to rely on religious counsel or teachings when you have a problem?
2. to believe in God?
3. to rely on your religious beliefs as a guide for day-to-day living?
4. to be able to turn to prayer when you’re facing a personal problem?
5. to attend religious services regularly?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
Appendix B
Mx Script For Full Unconstrained Factor Model
Group 1: Mother-Father covariance
Data Calculation NGroups=41
Matrices
D FU 2 2 free
P Symm 2 2 free I within person cov (Rp)
Compute P*D*P/
MAP
50
.01 10
Option No_output
End
Group 2: Calculate T Gen-Phen covariance
Data Calculation
Matrices
C Low 2 2 free I common env paths
G Sym 2 2 free I additive genetic cov
A Low 2 2 free I additive genetic paths
S Full 2 2 free I A-C covariance
Compute G*A' + S*C' /
Start 1 G 2 2 G 2 2 G 1 1
MAA
3
.01 1
MAC
1
.01 1
Option No_output
End
Group 3: Mother-Child covariance (BIO)
Data Calculation
Matrices
D Full 2 2 =D(1)
C Low 2 2 =C(2)
F Full 2 2 free
H Full 1 1
I Iden 2 2
A Low 2 2 =A(2)
M Full 2 2 free
P Sym 2 2 =P(1)
T Full 2 2 =%E2
W Full 2 2 =%E1
Compute (P*M' + W*F')*C‘ +
(l+P'DrT^H®^) /
Matrix H 0.5
End
assortative mating
common env paths
paternal cultural trans
scalar, .5
identity matrix
additive genetic paths
maternal cultural trans
within person cov (Rp)
A-P covariance
spouse covariance
Group 4: Father-Child Covariance (BIO)
Data Calculation
Matrices
D Full 2 2 =D(1)
CLow 2 2 =C(2)
F Full 2 2 =F(3) I paternal cultural transmission
H Full 1 1 =H(3) ! scalar .5
I Iden 2 2
A Low 2 2 =A(2)
M Full 2 2 =M(3) I maternal cultural transmission
P Symm 2 2 =P(1)
T Full 2 2 =%E2
W Full 2 2 =%E1
Compute (P*F + W*M’)*C’ +(I+P*D,)*T,*(H@A') /
Option No_output
End
Group 5: Mother-Child covariance (ADOPT)
Data Calculation
Matrices
D Full 2 2 =D(1)
CLow 2 2 =C(2)
F Full 2 2 =F(3)
H Full 1 1 ! scalar, 0 for unrelated kids
I Iden 2 2
ALow 2 2 =A(2)
M Full 2 2 =M(3)
P Symm 2 2 =P(1)
T Full 2 2 =%E2
W Full 2 2 =%E1
Compute (P*M’ + W*F)*C' + (l+P*D)’ T*(H@A') /
Matrix H 0.0
End
Group 6: Father-Child Covariance (ADOPT)
Data Calculation
Matrices
D Full 2 2 =D(1)
CLow 2 2 =C(2)
F Full 2 2 =F(3)
H Full 1 1 =H(5) ! scalar 0 for unrelated kids
I Iden 2 2
ALow 2 2 =A(2)
M Full 2 2 =M(3)
P Symm 2 2 =P(1)
T Full 2 2 =%E2
W Full 2 2 =%E1
Compute (P*F' + W'*M')*C' +(I+P*D')*T*(H@A') /
Option No_output
End
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
Group 7: Biological Sibling Covariance
Data Calculation
Matrices
R Sym 2 2 Free ! sibling common env cov
D Full 2 2=D(1)
CLow 22 =C(2)
G Sym 2 2 =G(2)
H Full 1 1 =H(3) ! scalar, .5 for bio siblings
A Low 2 2 =A(2)
S Full 2 2 =S(2)
T Full 2 2 =%E2
Compute F^A^G+HOCHD'+DrnrA' +
C*R*C' +A*S*C' + C*S'*A' /
Option No_output
End
Group 8: Adopted Siblings Covariance
Data Calculation
Matrices
R Sym 2 2 =R(7)
C Low 2 2 =C(2)
Compute C*R*C7
Option No_output
End
Group 9: Natural/Adopt Siblings Covariance
Data Calculation
Matrices
R Sym 2 2 =R(7)
C Low 2 2 =C(2)
ALow 22 =A(2)
S Full 2 2 =S(2)
Compute C*R*C' + C*S'*A' /
Option No_output
End
Group 10: Genetic Constraint
Data Constraint Nlnput=2
Matrices
D Full 2 2 =D(1)
G Symm 2 2 =G(2)
H Full 1 1 =H(3) ! scalar, .5
I Iden 2 2 ! to form segregation variance, .5 1
T Full 2 2 =%E2 !
Constraint G-(H@(G+H@(T*(D'+D)*T')+I)) /
Option Residual
End
Group 11: A-C Constraint diagonal and below
Data Constraint Nlnput=2
Matrices
D Full 22 =D(1)
F Full 2 2 =F(3)
H Full 1 1 =H(3) I scalar, .5
M Full 2 2 =M(3)
P Sym 2 2 =P(1)
S Full 2 2 =S(2)
T Full 2 2 =%E2
Constraint S-(H@T*(M,+F+D’P*M'+D,*P*F)) /
Option Rsidual
End
Group 12: A-C Constraint - Part 2, above diagonal
Data Constraint Nl=1
Matrices
D Full 2 2 =D(1)
F Full 2 2=F(3)
H Full 1 1 =H(3) I scalar. .5
M Full 2 2 =M(3)
P Sym 2 2 =P(1)
S Full 2 2 =S(2)
T Full 2 2 =%E2
X Zl 1 2 I Partitioned Zero|ldentity matrix (n-1 x n)
ZIZ 2 1 ! Partitioned Zero|ldentity matrix (n x n-1)
Constraint X*(S-(H@T*
(M’+F’+D*P*M'+D,*P*F,)))'*Z /
Option Rsidual
End
Group 13: Phenotypic Variance Constraint
[Parents and Bio Kids (Equation 5)
Data Constraint NI=2
Matrices
R Symm 2 2 =R(7)
CLow 22 =C(2)
G Symm 2 2 =G(2)
J Low 2 2 free ! specific environment paths
ALow 22 =A(2)
P Symm 2 2 =P(1)
Constraint P-(A*G*A' + C*R*C' + J*J'+ A‘S*C’ +
C*S'*A') /
Option Rsidual
MAJ
8
.01 5
End
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
Group 14: Common Env Const (Equation 4)
Data Constraint Nl=2
Matrices
B Iden 2 2 ! common env residual variance
R Sym 2 2 =R(7)
F Full 2 2 =F(3)
M Full 2 2 =M(3)
P Sym 2 2 =P(1)
W Full 2 2 =%E1
Constraint R-(M*P*M' + F*P*F + M*W*F +
F*W'*M' + B) /
Option Rsidual
End
Group 15: Within person covariance matrix
lAdopted Kids
Data CA
Matrices
R Symm 2 2 =R(7)
C Low 2 2 =C(2)
G Symm 2 2 =G(2)
J Low 2 2 =J(13)
A Low 2 2 =A(2)
S ZE 2 2 ! GE cov is zero for adopted kids
CO (A*G*A' + C*R*C' + J*J'+ A*S*C'
+ C*S'*A‘ ) /
Option Rsidual
End
Group 16: read in adoptive data - Single Child
DA Nl=6 NO=1500
VL Fl= C:\Data\ad1.dat
X Full 1 9
A Full 2 2 =%E5 ! Adopt Mother-child covs
B Full 2 2 =%E6 ! Adopt Father-child covs
D Full 2 2 =%E1 ! Spouse covariances
E Symm 2 2 =P1 ! Within person covs
F Symm 2 2 =%E15 ! w/in pers covs adoptee
M1 X/
M2 ((E | D * | B )_
(D|E | A)_
(B '| A' | F ))' /
SP x
200 201 202 203 204 205
S t7 5 x 1 1 x 1 3 x 1 5
st 20 x 1 2 x 1 4 x 1 6
Option Rsidual IT=4000
End
Group 17: Nuclear family data -Single Child
DA Nl=8 N0=1500
VL FI=C:\Data\bio1.dat
Matrices
X Full 1 9
A Full 2 2 =%E3 ! Bio Mother-child covariances
B Full 2 2 =%E4 ! Bio Father-child covariances
D Full 2 2 =%E1 ! Spouse covariances
E Symm 2 2 =P1 ! W ithin person covariances
M 1 X/
M2 ((E | D' | B )_
(D | E | A )_
(B'| A' | E )) /
SP x
200 201 202 203 204 205 206 207
st 75 x 1 1x13x15x17
st 20 x 1 2 x 1 4 x 1 6x 1 8
Option Rsidual IT=4000
End
Group 18: Adopt Family: 2 Adopted Siblings
DA Nl=8 N0=1500
VL FI=C:\Data\ad2.dat
Matrices
X Full 1 12
A Full 2 2 =%E5 ! Adopt Mom-child covariances
B Full 2 2 =%E6 ! Adopt Father-child covariances
C Full 2 2 =%E8 ! Adopted sibling covariances
D Full 2 2 =%E1 ! Spouse covariances
E Symm 2 2 =P1 I W ithin person covariances
F Symm 2 2 =%E15 ! w/in person covs adopted
M 1 X I
M2 ( (E | D' | B | B )_
(D | E | A | A )_
(S'| A' | F | C )_
(B'| A' | C'| F )) /
SP x
200 201 202 203 204 205 206 207
st 75 x 1 1x13x15x17
St20x12x14x16x18
Option Rsidual IT=4000
End
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
Group 19: Nuclear Family Data: 2 biological sibs
DA Nl=12 NO=1500
VL FI=C:\Data\bio2.dat
Matrices
X Full 1 12
A Full 2 2 =%E3 ! Biological Mother-child covs
B Full 2 2 =%E4 ! Biological Father-child covs
C Full 2 2 =%E7 ! Biological sibling covs
D Full 2 2 =%E1 I Spouse covariances
E Symm 2 2 =P1 ! Within person covariances
M1 X/
M2 ((E | D11 B | B )_
(D[E | A | A)_
(B'| A' | E | C )_
(B’[ A' | C'| E )) /
SP x
200 201 202 203 204 205 206 207 208
209 210 211
st75x1 1x14x17x1 10
St 20 x1 2 x 1 3 x 1 5 x 1 6 x 1 8 x 1 9x1 11x1 12
Option Rsidual IT=4000
End
Group 20 1 Naural/ 1 Adopted Child
DA Nl=12 NO=1500
VL FI=C:\Data\ad3.dat
Matrices
X Full 1 12
Q Full 2 2 =%E3 I Bio Mother-child covs
R Full 2 2 =%E4 ! Bio Father-chilcf covs
C Full 2 2 =%E9 ! Adopted/Bio sibling covs
D Full 2 2 =%E1 ! Spouse covariances
E Symm 2 2 =P1 ! Within person covariances
A Full 2 2 =%E5 ! Adopted Mother-child covs
B Full 2 2 =%E6 ! Adopted Father-child covs
F Symm 2 2 =%E15 !w/in pers covs adoptee
M1 XJ
M2 ((E | D' | B | R )_
(D|E | A | Q )_
(B'| A11 F | C )_
(R’| Q' | C'| E )) /
SP x
200 201 202 203 204 205 206 207 208
209 210 211
st 75x1 1 x1 4x1 7x1 10
st20x1 2x1 3x15x16x18x1 9x1 11x1 12
Option Rsidual IT=4000
End
Group 21: Calculate Expected Phenotypic
Standard Deviations
Data Calculation
Matrices
H Full 1 1 =H(3) Iscalar, .5
I Iden 2 2 lidentity matrix
P Sym 2 2=P(1) Iwithin person covariance
Compute (I.P)A H/
Option Rsidual
End
Group 22: Standardize Assortment Matrix
Data Calculation
Matrices
D Full 2 2 =D(1) lassortative mating delta paths
S Full 2 2 =%e21 Iphenotypic standard devs
Compute S*D*S7
Option Rsidual
End
Group 23: Standardize Maternal Cultural Trans
Data Calculation
Matrices
R Symm 2 2 =R(7) Icommon env covariance(Rc)
H Full 1 1 =H(3) Iscalar, .5
I Iden 2 2 lidentity matrix
M Full 2 2 =M(3) IMatemal cultural transm ission
S Full 2 2 =%e21 Iphenotypic standard deviations
Compute ((I.R)A H)~*M*S/
Option Rsidual
End
Group 24: Standardize Paternal Cultural Trans
Data Calculation
Matrices
R Symm 2 2 =R(7) Icommon env covariance(Rc)
H Full 1 1 =H(3) Iscalar, .5
I Iden 2 2 lidentity matrix
F Full 2 2 =F(3) IPatemal cultural transm ission
S Full 2 2 =%e21 Iphenotypic standard devs
Compute ((I.R)A H)~*F*S/
Option Rsidual
End
Group 25: Standardize Genetic Factors (AGA’)
Data Calculation
Matrices
G Symm 2 2 =G(2) ladditive genetic covs (Ra)
A Low 2 2 =A(2) ladditive genetic paths
S Full 2 2 =%e21 Iphenotypic standard devs
Compute S— *(A*G*A')*S~/
End
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
Group 26: Standardize CRC
Data Calculation
Matrices
R Symm 2 2 =R(7) Icommon env covariance (Rc)
C Low 2 2 =C(2) Icommon environment paths
S Full 2 2 =%e21 Iphenotypic standard deviations
Compute S-*(C*R*C')*S~/
Option Rsidual
End
Group 27: Standardize JJ'
Data Calculation
Matrices
J Low2 2=j13 [specific env paths
S Full 2 2 =%e21 Iphenotypic standard deviations
Compute S-*(J*J')*S~/
Option Rsidual
End
Group 28: Standardize ASC'
Data Calculation
Matrices
A Low 2 2 =A(2) ladditive genetic paths
C Low 2 2 =C(2) Icommon environment paths
S Full 2 2 =s2 IA-C covariance ’
Z Full 2 2 =%e21 Iphenotypic standard deviations
Compute Z-*(A*S*C')*Z-/
Option Rsidual
End
Group 29: Standardize CS'A'
Data Calculation
Matrices
A Low 2 2 =A(2) ladditive genetic paths
C Low 2 2 =C(2) Icommon environment paths
S Full 2 2 =s2 IA-C covariance
Z Full 2 2 =%e21 Iphenotypic standard deviations
Compute Z~*(C*S'*A')*Z~/
Option Rsidual
End
*The 14 groups originally employed to read in data separated by both sex and adoptive status have been
reduced to 5 for purpose of simplifying the script outlined above.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Head injury and dementia: A co-twin control study of Swedish twins
PDF
History of depression, antidepressant treatment, and other psychiatric illness as risk factors for Alzheimer's disease in a twin sample
PDF
Cognitive functioning and dementia following cancer: A Swedish twin study
PDF
Alcohol expectancies and consumption: Age and sex differences
PDF
Five quantile test: A comparison of independent groups
PDF
Intragroup evaluations, attitude source, and in-group member derogation
PDF
Ethnic identity, acculturation, self-esteem and perceived discrimination: A comparison study of Asian American adolescents
PDF
Hispanic culture, acculturation, and distress among caregivers of dementia patients
PDF
Hyperactive symptoms, cognitive functioning, and drinking habits
PDF
Emergent literacy differences in Latino and African American children: Culture or poverty?
PDF
Articulated thoughts regarding cognitions toward older adults
PDF
Invariance to changes in contrast polarity in object and face recognition
PDF
Effects of threat and self-focus on consensual bias in majorities
PDF
Insights into the nature of phonological and surface dyslexia: Evidence from a novel word learning task
PDF
Effects of personal resource sufficiency on perceived difficulty and desirability of earthquake preparedness
PDF
Depression as a risk factor for dementia
PDF
Affection and conflict in family relationships
PDF
Dyslexic's development of word recognition and phonological skill over time
PDF
Depression and suicidality in Latino adolescents: A study of acculturation and gender role beliefs
PDF
Depressed children and the social and behavioral attributes of their best friend
Asset Metadata
Creator
Abrahamson, Amy C.
(author)
Core Title
Adolescents' social attitudes: Genes and culture?
School
Graduate School
Degree
Master of Arts
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,psychology, developmental,psychology, social
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Baker, Laura A. (
committee chair
), [illegible] (
committee member
), Gatz, Margaret (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-34366
Unique identifier
UC11341261
Identifier
1405203.pdf (filename),usctheses-c16-34366 (legacy record id)
Legacy Identifier
1405203.pdf
Dmrecord
34366
Document Type
Thesis
Rights
Abrahamson, Amy C.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
psychology, developmental
psychology, social