waller et al 1990 genetic and environmental influences on religious interests, attitudes, and values...

7/30/2019 Waller Et Al 1990 Genetic and Environmental Influences on Religious Interests, Attitudes, And Values - A Study of T

1/6

http://pss.sagepub.com/Psychological Science

http://pss.sagepub.com/content/1/2/138The online version of this article can be found at:

DOI: 10.1111/j.1467-9280.1990.tb00083.x

1990 1: 138Psychological ScienceNiels G. Waller, Brian A. Kojetin, Thomas J. Bouchard, Jr., David T. Lykken and Auke TellegenReared Apart and Together

Genetic and Environmental Influences on Religious Interests, Attitudes, and Values: A Study of Tw

Published by:

http://www.sagepublications.com

On behalf of:

Association for Psychological Science

can be found at:Psychological ScienceAdditional services and information for

http://pss.sagepub.com/cgi/alertsEmail Alerts:

http://pss.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

What is This?

- Mar 1, 1990Version of Record>>

at University of Nottingham on April 26, 2013pss.sagepub.comDownloaded from
http://pss.sagepub.com/http://pss.sagepub.com/http://pss.sagepub.com/content/1/2/138http://pss.sagepub.com/content/1/2/138http://pss.sagepub.com/content/1/2/138http://www.sagepublications.com/http://www.psychologicalscience.org/http://pss.sagepub.com/cgi/alertshttp://pss.sagepub.com/cgi/alertshttp://pss.sagepub.com/subscriptionshttp://pss.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsPermissions.navhttp://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://pss.sagepub.com/content/1/2/138.full.pdfhttp://pss.sagepub.com/content/1/2/138.full.pdfhttp://pss.sagepub.com/http://pss.sagepub.com/http://pss.sagepub.com/http://online.sagepub.com/site/sphelp/vorhelp.xhtmlhttp://pss.sagepub.com/content/1/2/138.full.pdfhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://pss.sagepub.com/subscriptionshttp://pss.sagepub.com/cgi/alertshttp://www.psychologicalscience.org/http://www.sagepublications.com/http://pss.sagepub.com/content/1/2/138http://pss.sagepub.com/


2/6

PSYCHOLOGICAL SCIENCE

Research Report .- rGENETIC AND ENVIRONMENTAL INFLUENCES ONRELIGIOUS INTERESTS, ATTITUDES, AND VALUES:A Study of Twins Reared Apart and Together

Niels G. Waller, Brian A. Kojetin, Thomas J. Bouchard, Jr.,David T. Lykken, and Auke TellegenUniversity ofMinnesota, Minneapolis

Abstract-The role of genetic and environmental factors in the expression ofreligious interests, attitudes, and valueswas examined using data from adulttwins reared apart and adult twinsreared together. The Religious Valuesscale of the Allport-Vernon-LindzeyStu(1y of Values, a Religious Interestscale from the Strong Campbell InterestInventory, the Wiggins Religious Funda-.mentalism scale from the MMPI, a Leisure Time Religious Interest scale, and asecond measure of Occupational ~ e l i - 'gious Interest were administered to 53identical and 31 fraternal twin pairs whohadbeen reared apart. The Leisure TimeReligious Interest Scale and the secondmeasure of Occupational Religious Interest were also administered to 458identical and 363 fraternal !Ivin pairswho had been rearedtogether. Biometricmodel fitting indicated that approximately 50% of the observed variance ofall five measures is genetically influenced.

In recent years, substantial empiricalattention has been directed toward understanding the relative importance ofgenetic and environmental influences inthe development (DeFries & Fulker,1986; Plomin, 1986) and manifestation(Goldsmith, 1983) of individual differences. Few studies, however, have considered the, irifluence of genetic factors inthe formation of religious interests, attitudes, and values. The paucity ofbehavior-genetic rese,archon these variables isperhaps due. to the belief, prevalent inboth lay and ac'ademic communities, thatreligiosity is primarily an environmentally influenced variable (Cavalli-Sforza,

Send correspondenceand reprint requeststo Niels G. Waller or Thomas J. Bouchard,Jr., Department of Psychology, University ofMinnesota, Elliott Hall, 75 East River Rd.,Minneapolis, MN 55455. .

Feldman, Chen, & Dornbush, 1982). Forexample, Plomin (1989) in his recent review of the behavior genetic literatureconcluded that "Religiosity and certainpolitical beliefs . . . show no geneticinfluence" (p. 107).Although few twin or adoption studies have directly examined the degree ofgenetic and environmental influence onreligiosity, this variable has sometimesbeen included in behavior genetic research as a control variable because ofthe presumption that religiosity is predominantly determined by common family environments rather than genetic influences (Goldsmith & Gottesman, 1977;Loehlin & Nichols, 1976; Pogue-Geile &Rose, 1985). These studies, however,have relied almost exclusively on youngtwins who may still have been stronglyinfluenced by their families. In adults,genetic factors may playa more important role in the expression of religiosity.For example, Martin, Eaves, Heath, Jardine, Feingold, and Eysenck (1986), using a large cross national sample of adulttwins, examined the heritability of several social attitudes, including attitudestowards sabbath observance, divine law,church authority, and Bible truth, andfound that all of these religious attitudeshad a significant genetic component.The present investigation exploredthe role of genetic and environmentalfactors in the expression of religious interests, attitudes, and values using multiple and converging measures gatheredfrom adult twins reared apart and adulttwins reared together.

METHODSubjectsOur samples included 53 pairs ofmonozygotic twins reared apart (MZA)and 31 pairs of dizygotic twins rearedapart (DZA) who participated in the Min-

nesota Study of Twins Reared Apart between 1979 and 1989. The mean age forthe 40 male and 66 female MZA twinswas 36.77 (SD = 12.99) and 41.65 (SD =11.29) years, respectively; the mean agefor the 18 male and 44 female DZA twinswas 45.05 (SD = 14.44) and 42.50 (SD =9.41) years. Details of the MZA andDZA twin recruitment are reported inBouchard (1984, 1987) and referencescited therein. The median age of separation was .24 years and ranged from birthto 4.5 years. Separation time was arbitrarily set at zero for twin pairs who werereared in different homes but who hadperiodic contact during childhood. Thelength of separation (number of yearsfrom separation until first contact)ranged from 0 to 69 years. Zygosity ofthe reared apart twins was determinedby a procedure that results in a misdiagnosis rate of less than .001 (Lykken,1978). .

Data for two of our measures of religiosity (the Religious Leisure Time Interests sca le and the Religious Occupational Interests scale) were also availablefrom 1,642 twins who participated in theMinnesota Twin Registry. The 312 maleMZTs and 604 female MZTs had meanages of 35.86 (SD = 7.3) and 35.98 (SD= 7.31). The 246 male DZTs and 480 female DZTs had mean ages of 37.27 (SD= 7.57) and 37.72 (SD = 7.91). All DZTtwins in this sample were same-sexpairs. Details of twin recruitment and zygosity determination by questionnairefor the rea red togethe r twins are reported elsewhere (Lykken, Bouchard,McGue, & Tellegen, 1989).

Because twins are correlated perfectly for age, and same-sexed twins forsex as well, twin similarities can be positively biased if the twin pair differenceson these variables are not partialled outofthe variables of interest. In the presentstudy, all data were age and sex corr ec ted with a procedure described in,McGue and Bouchard (1984).

138 Copyright 1990 American Psychological Society VOL. 1, NO.2, MARCH 1990at University of Nottingham on April 26, 2013pss.sagepub.comDownloaded from
http://pss.sagepub.com/http://pss.sagepub.com/http://pss.sagepub.com/http://pss.sagepub.com/


3/6


N.G. Waller et al.

Table 1. /lltercorrelations ofand internal consistency reliabilities fo; fivemeasures of religious interests, attitudes, and values

a The AVL Religious Values Scale is an ipsative scale, consequently an internalconsistency reliability cannot be computed. The number reported is a test-retest reliabilityfrom the AVL manual (Allport et aI., 1%0) based on a sample of 53.n = 75 to 155 twins reared apart.

Scales 2 3 4 51. Religious Leisure Time Interests .912. Religious Occupational Interests .64 .863. Wiggins Religious Fundamentalism .75 045 .824. Strong Campbell Vocational InterestsInventory: Religious Interests .77 .67 .60 .895. Allport-Vernon-Lindzey Religious Values .67 040 .69 .69 .93 3

InstrumentsFive measures were employed. TheReligious Fundamentalism scale devisedby Wiggins (1966) is composed of twelveitems, taken from the Minnesota Mul

tiphasic Personality Inventory (Hathaway & McKinley, 1951), that tap variousfacets of fundamental religious beliefs.The Religious Occupational Interestsscale (Waller, Lykken, & Tellegen, inpress) is a four-item factor analyticallyderived scale that measures interests inreligious occupations, such as minister,priest, or rabbi, working as a missionary,or becoming a professional writer on religious subjects. Our Strong-Campbellscale is slightly modified from the Religious Activities Basic Intere'st scale fromthe S t r o n g ~ C a m p b e l l Interest Inventory(Hansen & Campbell, 1985) and is composed of nine items (29, 83, 137 , 177 ,228,239,245,266,318) that measure interests in religious occupations, schoolsubjects, and activities. The ReligiousLeisure Time Interests scale (Waller,Lykken, & Tellegen, in press) is a fiveitem, factor analytically derived scalethat measures expressed interests in various religious leisure time activities, suchas attending religious services, doingwork for the church of synagogue, orpursuing religious studies. The AllportVernon-Lindzey Religious Values scale(Allport, Vernon, & Lindzey, 1960), oneof six scales from the Allport-VernonLindzey Study of Values, measures thesalience of a religious value system. Table I presents the intercorrelations andthe reliabilities of the five measures ofreligious interests, attitudes, and values.

Analytic ProceduresWith data available from both identical and fraternal twins reared apart andreared together,' the analysis of geneticand environmental"influences can be ac

complished in several ways. One methodinvolves direct comparison of the intraclass correlatioris.:of MZA and MZTtwins. Because MZA twins are genetically identical but share essentially noenvironmental influences, the MZA correlation is a direct estimate of heritability. The MZT twins share all of theirgenes and a common family environment, thus the difference between theMZA and MZT correlations estimatesthe influence of common family environ-

ment. A second method is based on thedifference between the MZT and DZTintraclass correlations. Because MZtwins share all of their genes and DZtwins share 50% of their genes on average, the percentage of observed variancethat is due to additive genetic influences(i.e. , narrow heritability) can be estimated as . twice the difference betweenthe MZT and DZT correlations (Falconer, 1981). A third approach, and theone used in the present investigation, involves specification and comparison ofalternative biometric models (Eaves, Eysenck, & Martin, 1989; Jinks & Fulker,1970) applied to the within and betweenmean squares of all the twin data. Thislatter method provides maximum likelihood estimates of the genetic and environmental parameters as well as a chisquare test of the degree of fit of eachmodel.To obtain estimates of the genetic and.environmental parameters, several bio.metric models were tested and evaluatedfor their ability to account parsimoniously for the observed data. The modelsexamined were: (a) a purely environmental model, (b) a purely genetic model,and (c) a full model that included bothenvironmental and genetic parameters.For the two scales on which we had datafrom all four twin groups, the ReligiousLeisure Time Interests scale and the Religious Occupational Interests scale, theenvironmental variance parameter wasfurther broken down into shared-familialand non-shared environmental components. Our models assume (a) the MZAand DZA correlations are due entirely togenetic factors, (b) the DZA and DZTgenetic correlations are half the MZA

and MZT genetic correlations (i.e., weassume a complete additive model), (c)the absence of selective placement (forthe reared apart twins), (d) the absenceof assortative mating, and (e) no interaction effects. Selective placement doesnot appear to be a problem with ourreared apart twins. All reared apart twinsfilled out the Family Environment scale(Moos & Moos, 1986), a widely used instrument designed to tap the major dimensions of the rearing family environment. The intraclass correlation for theMoral Religious Orientation scale of thisinventory was .03 for the pooled sample,.09 for the MZAs and - .07 for theDZAs. The data from the twins rearedapart suggests that there is positive assortative mating for religiosity (medianspousal correlation on the five measureswas .56); however, due to the small number of spouses available in our rearedapart sample we did not include this parameter in our models. Failure to includeassortative mating parameters in biometric modeling typically results in geneticestimates that are biased downward sothat our genetic estimates are probablyconservative. We are currently collecting spouse datafor these traits on a muchlarger sample. Due to varying samplesizes for the five measures it was notpossible to conduct a joint analysis ofthese measures to estimate a commonphenotypic- or a commongenetic-factor.

RESULTSTable 2 reports the intraclass correlat ions, the between- and within-meansquares and the sample sizes for the fivescales for each zygosity and rearing condition.

VOL. I , NO.2, MARCH 1990 139at University of Nottingham on April 26, 2013pss.sagepub.comDownloaded from


4/6


Influences on Religious Interests and Values

Table 2. Intraclass correlations, numbers ofpairs of twins, between and within mean squares for five measures of religiousinterests, attitudes, and values

The MZA and MZT correlations areuniformly higher than the correspondingDZA and DZT correlations, suggestingthat genetic factors playa significant rolein the expression of this trait. The meansquares were used in the biometricmodel fitting, the results of which are reported in Tables 3 and 4. Based on thedegree of statistical fit of the competingmodels it is clear that neither the purelyenvironmental nor the purely geneticmodels account adequately for the observed data.For four of the five measures the general model, whose parameters are reported in the tables, cannot be rejected.

For the Wiggins Religious Fundamentalism scale, the negative DZA correlationis inconsistent with the assumptions ofthe additive model. However, the derived parameters are consistent withthose estimated for the other four scales.Thus, our findings indicate that individual differences in religious attitudes, interests and values arise from both genetic and environmental influences.More specifically, genetic factors account for approximately 50% of the observed variance on our measures. Forthose variables whose environmentalvariance parameter is broken down intoa shared-familial and a non-shared envi-

ronmental component (Table 3), our dataalso suggest that the bulk of the environmental experiences that influence the expression of religiosity, like those that influence the expressionof personality andtemperament (Bouchard & McGue, inpress; Plomin& Daniels, 1987; Tellegen,Lykken, Bouchard, Wilcox, Segal, &Rich, 1988), are experiences not sharedby family members. In the case of theReligious Leisure Time Interests scale,twins reared together are considerablymore similar than twins reared apart,suggesting an effect of shared family environment. The biometric model, however, assumes that such effects will in-

Table 3. Genetic and environmental variance estimates from the full model biometric model fittingusing the fOil'. group designaEnvironmental p value for test of

Non Shared General No Genetic No EnvironmentalS c a l ~ Genetic Shared Familial Model Effect Effect

Religious Leisure 047 042 .11b .600


5/6


N.G. Waller et al.

Table 4. Genetic and environmental variance estimates from the full model biometric model fittingusing the two group designp value for test of

No NoGeneral Genetic EnvironmentalScale Genetic Environmental Model Effect EffectWiggins Religious .46 .54 .019 .004


6/6


Influences on Religious Interests and Values .. r

Loehlin, J.C., & Nichols, R.C. (1976). Heredity, environment andpersonality: A study of850 setsof twins. Austin: University of Texas Press.Lykken, D.T. (1978). The diagnosis of zygosity intwins. Beha"ior Genetics, 8; 437-473.Lykken, D.T. (1982). Research with twins: The concept of emergenesis. Psychophysiology, 19,

361-373.Lykken, D.T., Bouchard, T.J., Jr., McGue, M., &Tellegen, A. (1989). The Minnesota Twin Registry. Paperdelivered at the Sixth InternationalTwin Congress, Rome, Italy.Martin, N.G., Eaves, L.J., Heath, A.C., Jardine,R., Feingold, L.M., & Eysenck, H.J. (1986).Transmission of social attitudes. Proceedingsof the National Academy of Sciences, USA.83. 4364-4368.McGue, M., & Bouchard, T.J., Jr. (1984). Adjust-

142

ment of twin data for the effects of age andsex. Beha"ior Genetics, 14. 325-343.Moos, R.H., & Moos, B.S. (1986). Manual: FamilyEm'ironmentScale. Palo Alto: Consulting Psychologists Press.Plomin, R. (1986). Development. genetics, and psychology. Hillsdale, NJ: Lawrence Erlbaum.Plomin, R. (1989). Environment and genes: Determinants of behavior. American Psychologist.44(2), 105-111.Plomin, R., & Daniels, D. (1987). Why are childrenin the same family so different from one another? BehQl'ioral and Brain Sciences. 10. 1-59.Pogue-Geile, M.F., & Rose, R.J. (1985). Developmental genetic studies of adult personality.Developmental Psychology, 21. 547-557.Scarr, S., & Weinberg, R.A. (1981). InS. Scarr(Ed.).

Race. social class and individual differences. in I.Q. Hillsdale, NJ: Lawrence Erlbaum.Tellegen, A., Lykken, D.T., Bouchard, T.J. Jr. ,Wilcox, K.J., Segal, N.L., & Rich, S. (1988).Personality similarity in twins reared apart andtogether. Journal of Personality and SocialPsychology, 54, 1031-1039.Waller, N.G., Lykken, D.T., & Tellegen, A. (inpress). Occupational interests, leisure time interests and personality: Three domains or one?Findings from the Minnesota Twin Registry.In R.V. Dawis & D. Lubinski (Eds.), Assessing indil'idual differences in human behavior:New methods. concepts andftndings. Minneapolis: University of Minnesota Press.Wiggins, J.S. (1966). Substantive dimensions of selfreport in the MMPI item pool. PsychologicalMonographs, 80.(RECEIVED 8nJ89; ACCEPTED 9/11189)

VOL. I, NO.2, MARCH 1990at University of Nottingham on April 26, 2013pss.sagepub.comDownloaded from

waller et al 1990 genetic and environmental influences on religious interests, attitudes, and values...

Documents