Clayton's Premarital Intercourse a Substantive Test of the Contingent Consistency Model

8/2/2019 Clayton's Premarital Intercourse a Substantive Test of the Contingent Consistency Model

1/4

Clayton's "Premarital Intercourse: A Substantive Test of the Contingent Consistency Model"

RevisitedAuthor(s): David W. Grimes and Roger W. LibbyReviewed work(s):Source: Journal of Marriage and Family, Vol. 35, No. 1 (Feb., 1973), pp. 9-11Published by: National Council on Family RelationsStable URL: http://www.jstor.org/stable/351090 .Accessed: 10/04/2012 05:34

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

National Council on Family Relations is collaborating with JSTOR to digitize, preserve and extend access to

Journal of Marriage and Family.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=ncfrhttp://www.jstor.org/stable/351090?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/351090?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=ncfr


2/4

L e t t e r s to t h e E d i t o rCLAYTON'S"PREMARITALINTERCOURSE:A SUBSTANTIVETESTOF THECONTINGENTCONSISTENCYMODEL"REVISITED

In a recent article (Journal, May, 1972)Richard R. Clayton attempted to apply thecontingent consistency perspectiveto premari-tal intercourse in order to delineate therelationshipbetweenattitudesandbehavior.Hesimultaneouslyassessedsubjects'beliefs towardpremarital permissiveness,religiosity, percep-tion of referencegroup premaritalpermissive-ness, and self reportsof premaritalntercourse.Overall, a higher degree of attitude-behaviorconsistency was found than has been reportedin past studies (cf., Wicker,1969). Noting thatcontingent factors(i.e., referencegrouppermis-siveness)had little uniformeffect as predictorsof behavior,Clayton suggestedthe contingentconsistency approachrequires urtherclarifica-tion. However, there are certain difficultiesinthe Clayton study that call into questionwhether he actually had "a substantivetest ofthe contingentconsistencymodel."Clayton divided his contingent consistencymodel into three major sections: attitudes,contingent factors, and behavior.Attitudes andcontingent factors were designatedas indepen-dent variables o predict premaritalntercourse.We will comment first on his treatment ofattitudes.Clayton asserts"there are at least two levelsat which [attitudes] ... operate" (p. 274).Generalattitudes "deal with content which issomewhat abstract and sometimes rather farremoved from any kind of behavioralorienta-tion" (p.274). Behavior-specificattitudes, onthe other hand, are similar in content tosubstantivebehavior toward a particularatti-tude object-in the present case, premaritalintercourse.Clayton's index of "behavior-specific" tti-tudes consisted of five items from the Reiss(1967) PremaritalPermissiveness cales. How-ever, in spite of Clayton's conceptualizationonly two of the five items utilized are actuallybehavior-specificowardpremaritalntercourse.The other three items deal with kissing andpetting-a far cry from coitus. Thus, Clayton'spersonal attitudinal sexual permissiveness(P.A.P.) scale turns out to be a generalscaleofpremarital permissiveness of doubtful unidi-mensionalitywith regardto the attitudeobjectof premaritalntercourse.

The five P.A.P. items constituted six Gutt-man-scale types which were used to dichoto-mize respondents into low and high attitudecategories. The three "less difficult" scaletypes, i.e., (0) no endorsement;(1) kissing-noaffection; and (2) petting-engaged,were classi-fied as low permissiveness nd the three"moredifficult" scale types, i.e., (3) petting-noaffec-tion; (4) full sexual relations-engaged; nd (5)full sexual relations-strong ffection, were clas-sified as high permissiveness.Thus, subjectswho endorsed items up to and including"petting-no affection" and rejected the twoitems dealingwith coitus wereclassifiedashighon permissiveness.The implications of this"cookbook" groupingprocedure on the atti-tude-behavior elationshipareas follows: Thosesubjects who expressed agreementtoward thekissingand the two petting items and disagree-ment toward both the coital items were(wrongly) classifiedas behaviorally onsistentifthey reported previous engagement in coitus.Further,those subjectswho similarlyagreed othe kissing and the two petting statementswhile rejecting both coital items, were (againwrongly) classifiedas behaviorally nconsistentif they reportedno previoussexualintercourse.Clearly, the logical division of lows and highson the P.A.P. scale should have been betweenthose who agreedto either one or both of theintercourseitems and those who did not, i.e.,between scale types 3 and 4, not 2 and3. ThusClayton in effect treatedcongruenciesbetweenattitudes and behavior as inconsistencies andincongruenciesas consistencies.Thesecriticismssuggest that any interpretation of Clayton'sresults must be viewed with considerablecaution.In additionto questioning he P.A.P. scaleasa valid measure of behavior-specificattitudes,we disagreewith its statusas a valid indicatorofpersonal attitudes. Thus, item three reads: "Ibelieve that petting is acceptable for the male(female) before marriagewhen he (she) is notparticularlyaffectionate toward his (her) part-ner." To make it an item of personalattitudes,the readershould reread the statement, substi-tuting "me," "I am," and "my" for "male(female)," "he (she) is," and "his (her),"respectively. Clearly, such a rephrasing de-scribes a personal standard which Claytonintended to assess. The original P.A.P. itemselicit beliefs about the acceptanceof same-sexsocial standards o which a respondentmay ormay not havepersonallysubscribed.

February1973 JOURNAL OF MARRIAGEAND THE FAMILY 9


3/4

Turningnow to Clayton'sindices of generalattitude, we are somewhatconfused as to whyhe designatedreligiosity as an attitude towardpremarital sex. True, religious attitudes havebeen empiricallyrelated to permissivenessreli-giosity has been related to manyvariables),butit is difficult to conceive of religiosity as ageneral attitude toward coitus, premaritalorotherwise. Moreover, we seriously doubt itsrelevance as an independentvariable n testingthe contingent consistency postulate.In a priorpaper reporting on the same study (1971),Clayton's purposewas to investigatethe effectof religiosity on premaritalpermissiveness tti-tudes.Therefore,the more recentpaperappearsto be a post hoc reanalysisof dataintended forother purposes,rather than a study designed otest the relationshipbetween verbal attitudesandovert behavior.The second section of Clayton's researchpurports to deal with contingent factors.However, they are nothing more than arephrasing f the five-itemP.A.P.scale.Whereaseach P.A.P. item begins with the phrase, "Ibelieve . . .," the two contingenI variables,normative referencegroup (N.R.G.) and com-parison reference group (C.R.G.), begin with,"They believe that.. ." and "Most people oncampusbelieve that...," respectively.We findit rather arbitrary o considerP.A.P. items as.ttitudes and N.R.G. and C.R.G. measuresascontingent factors. The N.R.G. and C.R.G.items just as clearly elicit cognitions or beliefsas the P.A.P. items. To the present authors allthree scales seem to elicit a perceived normconcerning same-sex category. And like theP.A.P. scale, tle two referencegroupscalesarelimited to same-sex groups, whereas the pre-dicted behavior involves a heterosexual ph-e-nomenon. Unfortunately, Clayton failed toprovide opposite-sex reference group informa-tion. Moreover,the same criticismsappliedtothe P.A.P. scale concerning ts generalcontentand the procedureof dividingsubjects nto lowand high categoriesalso apply to the Guttman-type N.R.G. andC.R.G. scales.A further biasing limitation of Clayton'soperationalization of both reference groupindices involves their restrictionto the campusnormativesystem (cf., Clayton, 1971). Consid-ering that approximately47 per cent of thesubjects were freshmen who had been oncampus less than one semester, it is doubtfulthat the campusnormativesystem providedanappropriatebase for selection of salientcontin-gent factors (cf, Linn, 1965). The relativelylow relationshipbetween referencegroup dataand reportedcoitus may have been due in part

to the fact that this large percentageof newarrivalswere not familiar with the prevalentcampus attitudes and sanctions governingsex-ual permissiveness. n addition, if respondentshad been provided the opportunity, many ofthem might have designatedtheir most impor-tant referencegroupas off-campus e.g., family,friendsat home, etc.).Clayton claims "thereis the logicalnecessityof assumingthat the attitudinalandcontingentfactor variablesareantecedent to the behaviorunder study" (p. 274, italics added).However,it is surprising o find his researchdesign s thevery antithesis of this "logical necessity."Clayton'sbehaviorcriterionwas reportedcoitusduring the calendaryear prior to the measure-ment of attitudes andcontingentfactors.Thus,it would appear that Clayton has actuallyinadvertently tested the effect of previousbehavioron currentlyheld attitudesand contin-gentfactors.Clayton maintains that "behavior shouldalways be very similar to the content of thebehavior-specificattitudes and could refer toeither overt behavior or an attitudinaldisposi-tion" (p. 274). We have alreadydocumentedthe dissimilarcontent of the P.A.P. and thesubstantive behavior indices. The concept"overt behavior"as utilized in the contingentconsistency literature mplies a commitmenttoengage in behavior or actual engagement inbehavior. The concept clearly omits mereretrospectiveverbal self reports of one's priorbehavior(Wicker,1969). By referring o behav-ior as an attitudinal disposition, Clayton isactuallydesignatingbehavioras a component ofattitudes, an assumptionwhich the contingentconsistency postulate seriouslychallenges War-ner andDeFleur, 1969).Although Clayton did attempt to measureattitudes toward premarital intercourse, hecompletely neglected to recognize that atti-tudes and behavior differ in.situational aspects(cf. Figa-Talamanca,1972). All of his indiceswere situationallysimilaras they were assessedin the same anonymoussetting.Since behavior,attitudes, and contingent factors were simulta-neously assessed,the issues of reactivity, disso-nance reduction, social desirability, demandcharacteristics, tc., areproblematic.On the basis of his interpretation of thecontingent consistency perspective, Claytonconcludes "it was expected that reportedbehavior would be the result of additiverelationships..." (p. 274, italics added). How-ever, studies which have operated under thisperspectivestrongly suggestthat the realkey topredicting behavior lies in discovering inter-

10 JOURNALOF MARRIAGEAND THE FAMILY February1973


4/4

action effects of attitude and situationalvari-ables ratherthan their separateadditiveeffects(Acock andDeFleur,1972; Wicker,1969).Consistent with Clayton's hypothesis ofadditivity was his utilization of Coleman's(1964) additivetechniqueof multivariateanaly-

sis for dichotomous variables.He also appliedColeman'stest for interactioneffects. It wasonthe basis of comparing the results of thesestatistical tests that Clayton claims "ratherstrongsupport" for an additivemodel. Unfor-tunately, as Coleman (1964) and Goodman(1972) caution, measuring and interpretinginteraction effects among ndependentvariablesusingColeman'stechniqueremainsan unsolvedproblem. Further,not all forms of interactionare handled by the Coleman model. Theseconsiderations plus Clayton's failure to meetnecessary statistical assumptions and proce-dures, i.e., homoscedasticity, equal marginals,unweightedestimates, etc., cast doubt that anadditive nterpretations appropriate.Since attitudes and behaviorwereassessedatthe same time, in the same setting, and in thesamemanner,one would expect an evenhigherconsistency thanClayton reported.The reasonsfor lack of a higher correspondencebetweenattitudes and behaviorweredocumented n ourcomments on serious conceptual and method-ological problems. In carryingout a secondaryanalysisof his data which were obviouslynotappropriate to test a contingent consistencymodel, Clayton cannot be credited with asubstantive test of the attitude-behaviorrela-tionship. Hopefully, future studies will developa closer fit between theory andresearchdesign.

DAVIDW.GRIMESROGERW. LIBBYWashingtonState University

REFERENCESAcock,AlanC. andMelvinL. DeFleur1972 "A configurational pproach o contingentconsistency in the attitude-behavior ela-tionship."AmericanSociologicalReview37(December): orthcoming.Clayton,RichardR.

1971 "Religiosityand premarital exual permis-siveness:elaborationof the relationship nddebate." Sociological Analysis 32 (Sum-mer):81-96.1972 "Premaritalexual ntercourse: substantivetest of the contingentconsistencymodel."Journal of Marriageand the Family 34(May):273-281.Coleman,JamesS.1964 Introduction to MathematicalSociology.NewYork:McGraw-Hill.Fig'a-Talamanca,rene1972 "Inconsistencies f attitude andbehaviornfamily-planning tudies." Journal of Mar-riageandthe Family34 (May):336-344.Goodman,Leo A.1972 "A modified multiple regressionapproachto the analysisof dichotomousvariables."AmericanSociological Review 37 (Febru-ary):28-46.Linn,LawrenceS.1965 "Verbal attitudes and overt behavior: astudy of racial discrimination." SocialForces43 (March):353-364.Reiss,Ira1967 The Social.Context of PremaritalSexualPermissiveness.New York:Holt, RinehartandWinston.Warner, yleG. andMelvinL. DeFleur1969 "Attitude as an interactional oncept:socialconstraintand social distanceasinterveningvariables between attitudes and action."American SociologicalReview 34 (April):153-169.Wicker,AllenW.1969 "Attitudesversusaction: the relationship fverbal and overt behavioralresponses toattitude objects." The Journal of SocialIssues25 (Autumn):41-78.

February1973 JOURNALOF MARRIAGEAND THE FAMILY 11

Documents

Clayton's Premarital Intercourse a Substantive Test of the Contingent Consistency Model