Trajectories of Antisocial Behavior and Psychosocial Maturity From Adolescence to Young Adulthood

15 Pages·2009·0.27 MB·English

tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
AA--PPDDFF PPaaggeeMMaasstteerr DDeemmoo.. PPuurrcchhaassee ffrroomm wwwwww..AA--PPDDFF..ccoomm ttoo rreemmoovvee tthhee wwaatteerrmmaarrkk
DevelopmentalPsychology ©2009AmericanPsychologicalAssociation
2009,Vol.●●,No.●,000–000 0012-1649/09/$12.00 DOI:10.1037/a0015862
Trajectories of Antisocial Behavior and Psychosocial Maturity From
Adolescence to Young Adulthood
Kathryn C. Monahan Laurence Steinberg
CenterforHumanDevelopmentandDisability, TempleUniversity
UniversityofWashington
S
Elizabeth Cauffman Edward P. Mulvey
UniversityofCalifornia,Irvine UniversityofPittsburghSchoolofMedicine
andWesternFPsychiatricInstituteandClinic
Most theorizing about desistance from antisocial behavior in late adolescence has emphasized the
importance of individuals’ transition into adult roles. In contrast, litOtle research has examined how
psychologicaldevelopmentinlateadolescenceandearlyadulthoodcontributesdesistance.Thepresent
studyexaminedtrajectoriesofantisocialbehavioramongseriousjuvenileoffendersfrom14through22
yearsofageandtestedhowimpulsecontrol,suppressionofaggression,futureorientation,consideration
of others, personal responsibility, and resistance to peer influence distinguished between youths who
persistedinantisocialbehaviorandyouthswhodesisted.DOifferentpatternsofdevelopmentinpsycho-
social maturity from adolescence to early adulthood, especially with respect to impulse control and
suppression of aggression, distinguished among individuals who followed different trajectories of
antisocial behavior. Compared with individuals who desisted from antisocial behavior, youths who
persistedinantisocialbehaviorexhibiteddeficitsinelementsofpsychosocialmaturity,particularlyin
impulsecontrol,suppressionofaggression,andRfutureorientation.
AQ:1 Keywords:psychosocialmaturity,antisocialbehavior
P
It is well established that antisocial and criminal activity in- creases in susceptibility to peer pressure, decreases in parental
creases during adolescence, peaks around age 17 (with the peak monitoring),lessisknownaboutthedeclineinantisocialbehavior
somewhatearlierforpropertythanforviolentcrime),anddeclines thatoccursduringthetransitiontoadulthood.Numerousexplana-
as individuals enter adulthood; evidence for t his so-called age– tions have been offered for this decline, including fatigue (Got-
crime curve has been found across samples that vary in their tfredson & Hirschi, 1990); entrance into social roles that make
A
ethnicity, national origin, and historical era (Farrington, 1986; continuedantisocialactivitydifficult,suchaswork,marriage,and
Piquero,2007;Piqueroetal.,2001).Althoughthereisasubstantial parenting(Laub&Sampson,2001;Uggen&Staff,2001);attain-
literatureonfactorsthatcontributetotheriseindelinquentactivity mentofadultstatus(Moffitt,1993);andpsychosocialmaturation,
that takes place during early andPmiddle adolescence (e.g., in- whichbringswithitincreasesinself-control,strongerresistanceto
peer influence, and the willingness to forsake immediate gratifi-
cation in order to achieve future goals (Steinberg & Cauffman,
KathrynC.Monahan,CenterforHumanDevelopmentandDisability, 1996;forareviewoftheoriesofdesistancefromantisocialbehav-
UniversityofWashington;LaAurenceSteinberg,DepartmentofPsychology, ior, see Mulvey et al., 2004). Empirical research on these propo-
TempleUniversity;ElizabethCauffman,PsychologyandSocialBehavior,
sitions is sparse, however, and much more is known about the
School of Social Ecology, University of California, Irvine; Edward P.
factors that lead individuals into delinquency and antisocial be-
Mulvey, University of Pittsburgh School of Medicine and Western Psy-
havior than about the factors that lead them out of it (Farrall &
chiatricInstituteandClinic.
Bowling,1999;Laub&Sampson,2001;Warr,1998).
The project described was supported by funds from the Office of
JuvenileJusticeandDelinquencyPrevention,NationalInstituteofJustice, TheinvertedU-shapedcurveinantisocialandcriminalbehavior
JohnD.andCatherineT.MacArthurFoundation,WilliamT.GrantFoun- characteristic of adolescence and young adulthood describes a
dation, Robert Wood Johnson Foundation, William Penn Foundation, robustgeneraltrend,butthereareexceptionstothisdevelopmental
Centers for Disease Control, National Institute on Drug Abuse Grant rule. That is, although the vast majority of individuals who are
R01DA019697, Pennsylvania Commission on Crime and Delinquency, involved in illegal activity during adolescence cease or diminish
andtheArizonaGovernor’sJusticeCommission.Wearegratefulfortheir
their antisocial behavior as they move into adulthood, not all do,
support.Thecontentofthispaper,however,issolelytheresponsibilityof
andevenamongthosewhodesistfromantisocialactivity,thereare
theauthorsanddoesnotnecessarilyrepresenttheofficialviewsofthese
variations in the timing and rate of their diminished antisocial
agencies.
behavior(Sampson&Laub,2003).Perhapsthemostwidelycited
CorrespondenceconcerningthisarticleshouldbeaddressedtoKathryn
C.Monahan,CenterforHumanDevelopmentandDisability,Universityof perspective on individual differences in trajectories of antisocial
AQ:2 Washington,Seattle,WA98195.E-mail:[email protected] behavioristhatproposedbyMoffitt(1993,2006),whohasdrawn
1
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
2 MONAHAN,STEINBERG,CAUFFMAN,ANDMULVEY
a distinction between individuals whose antisocial behavior is (regardless of when such behavior began) discontinue it as they
limitedtoadolescence(“adolescence-limitedoffenders”)andthose becomeadults,andonlyasmallproportionofdeviantadolescents
whose antisocial behavior starts at a younger age and continues willdevelopintodeviantadults.
Fn1 into adulthood (“life-course-persistent offenders”).1 Although the Moffitt’s taxonomy of offenders is not the only theory that is
antisocialbehaviorofthesegroupsisoftenindistinguishabledur- relevanttodesistancefromantisocialbehavior,however.Asnoted
ingadolescence,theunderlyingcausesoftheirantisocialbehavior earlier, one possibility, suggested by Steinberg and Cauffman
arehypothesizedtobequitedifferent.Adolescence-limitedoffend- (1996),isthatthegrowthofpsychosocialmaturitymoregenerally
ers,itisthought,engageinantisocialbehaviortoappearandfeel
(whichincludesimprovementsinself-control)underliesdesistance
moremature.AccordingtoMoffitt,thismotivehasintensifiedin
from antisocial behavior during the transition to adulthood. This
modern society as a result of an ever-widening distance between S
notion is consistent with one of the most influential theories of
the age of attainment of physical maturity and the age of attain-
antisocial behavior, Gottfredson and Hirschi’s general theory of
ment of adult status (what she refers to as the “maturity gap”).
crime (1990), which posits that deficiencies in self-control or in
Once these individuals have attained adult status, their antisocial
behavior stops. In contrast, life-course-persistent offenders are one’sabilitytorefrainFfromantisocialbehavioraretherootcause
ofallantisocialactivity.Thetheorysuggestsnumerousreasonsfor
hypothesized to suffer from neuropsychological and cognitive
the link between antisocial behavior and poor self-control: (a)
deficitsthat,incombinationwithearlyfamilydisadvantage,con-
individuals with low self-control pursue immediate gratification
tinuetoaffectfunctioningandunderpinantisocialbehaviorthatis
O
and are oriented to the “now” as opposed to the future; (b)
maintainedintoadulthood.
antisocial acts provide easy or simple gratification of desires; (c)
Although Moffitt focuses on increased access to adult roles as
thechiefcauseofdesistanceamongadolescence-limitedoffenders, antisocial acts are exciting and risky, and individuals with low
it is also possible that declines in antisocial behavior during the self-control are active and physical; (d) because antisocial acts
transitiontoadulthoodareduetoincreasesinpsychosocialmatu- provOidefewlong-termbenefits,youthswithlow-selfcontroltend
rity. That is, if increases in antisocial behavior for this group tohaveunstablerelationshipsandhavelittleinterestinlong-term
reflectadesiretoappearandfeelmature,itstandstoreasonthat employmentopportunities;(e)youthswithlowself-controlexhibit
as youths become more psychologically mature in the course of little planning ability and instead use physical responses when
normative development, they will be less motivated to engage in frustrated; and (f) individuals with low self-control tend to be
R
antisocial activity. In contrast, because persistently antisocial self-centered,indifferent,orinsensitivetothesufferingandneeds
youths engage in antisocial behavior as the result of the lasting of others, in particular the victims of their antisocial activity
impactofearlyneurologicalandcontextualdisadvantage,thereis (Gottfredson&Hirschi,1990,p.89).Thegeneraltheoryofcrime
reasontoexpectthattheseindividualswillevincechronicdeficits has received much empirical support, and many studies have
P
in psychological functioning that will contribute to continued demonstratedthatlowself-controlisassociatedwithgreateranti-
antisocialbehavior. social behavior (e.g., Benson & Moore, 1992; Brownfield &
Research on Moffitt’s taxonomy of offending has identified Sorenson, 1993; Grasmick, Tittle, Bursik, & Arneklev, 1993;
individuals whose behavior is consistent with the life-course-
Paternoster & Brame, 1998; Piquero & Tibbetts, 1996; Pola-
persistentandadolescence-limitedpatterns,aswellasyouthswho
kowski,1994;Pratt&Cullen,2000;Winfree&Bernat,1998).
abstain from antisocial activity; however, mAany studies of trajec-
Gottfredson and Hirschi’s (1990) theory posits a number of
tories of antisocial behavior typically identify more patterns than
mechanisms,rangingfromsensationseekingtolimitedforesight,
these three. In a review of over 80 such studies, Piquero (2007)
underthebroadrubricofself-control.Amorefocused,anddevel-
found that, on average, three to five groups are identified in
opmental, theoretical formulation, presented by Steinberg and
trajectory analyses and that slighPtly more groups are found in
Cauffman (1996), maps onto Gottfredson and Hirschi’s theory.
studies that used self-reports of antisocial behavior than in those
SteinbergandCauffmansuggestthatduringadolescenceandearly
thatusedofficialarrestrecords.ConsistentwithMoffitt’stheory,
adulthood,youthsdevelopincreasing“temperance”(impulsecon-
studiestypicallyidentifythosewhoabstainfromantisocialbehav-
ior,anadolescent-peakpaAtternofantisocialbehavior(althoughthe trol and suppression of aggressive behavior), “perspective” (the
abilitytoconsiderthefutureconsequencesofactionsandtoview
specific peak age varies from study to study), and a chronic
antisocialbehaviortrajectory.(Individualsinthistrajectorytendto one’sactionsfromthevantagepointofothers),and“responsibil-
declineintheirantisocialbehavioratsomepointinadulthood,and ity”(theabilitytotakepersonalresponsibilityforone’sbehavior
thisfactsuggeststhatpersistentandtermsfortrajectoriessuchas and to resist the coercive influence of others). Consistent with
adolescent-peak are relative, not absolute.) In addition to these predictions derived from Gottfredson and Hirschi, youths with
patterns, studies identify individuals who consistently engage in lower temperance, perspective, and responsibility are more in-
moderatelevelsofantisocialbehavior,alate-onsetchronicgroup
(individuals who begin antisocial behavior in middle-to-late ado-
1Althoughweacknowledgethattherearedifferencesbetween“offend-
lescence and engage in antisocial acts at a steady rate into adult-
ing,” “delinquency,” and “criminal behavior,” some of which pertain to
hood),andagroupofindividualswhoareantisocialaschildrenbut
technicalratherthanbehavioraldifferencesamongthem(e.g.,shoplifting
not as adolescents or adults. Although the discovery of these
is considered delinquency when committed by a minor but criminal be-
additional trajectories has led to refinement of Moffitt’s frame-
haviorwhenitisdonebyanadult;carryingafirearmisanoffenseduring
work(Moffitt,2006;Moffitt,Caspi,Harrington,&Milne,2002), adolescencebutnotnecessarilyduringadulthood),weusethetermanti-
one fundamental tenet of her viewpoint remains incontrovertible: social behavior to refer to all types of offending, in keeping with the
Mostindividualswhoengageinantisocialbehaviorinadolescence traditionwithinthefieldofdevelopmentalpsychology.
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
ANTISOCIALBEHAVIORANDPSYCHOSOCIALMATURITY 3
clined to engage in antisocial behavior (Cauffman & Steinberg, youths who desist from antisocial behavior should show more
2000). rapidincreasesinpsychosocialmaturityduringthistransitionthan
Gottfredson and Hirschi (1990) argued that self-control is de- shownbypeerswhocontinuetheirillegalactivity.
termined early in life and is stable across development. But sta- Thepresentstudyexaminedtherelationbetweentrajectoriesof
bility (which refers to individuals’ relative standing on a given antisocialbehaviorandthedevelopmentofpsychosocialmaturity
characteristic) and change (which refers to individuals’ absolute in a sample of serious juvenile offenders (i.e., adolescents who
levels of that characteristic) are not the same thing. Height, for havebeenadjudicatedofaseriouscrime).Onechallengeinherent
example, is a trait that is characterized by high stability but in the study of adolescence-limited versus life-course-persistent
significant change over time. Perhaps because of their interest in offendersisthatchronicoffendingisrelativelyrare,asonlyabout
thestabilityofindividualdifferencesinself-control(ratherthanin 5% of adolescents persist in anStisocial behavior into adulthood.
changes in self-control over time), Gottfredson and Hirschi, al- Because there is some evidence that individuals who engage in
though they acknowledged that antisocial behavior declines after more serious offenses are more likely to persist in antisocial
adolescence, did not provide an especially satisfying account of behaviorovertime(Wierson&Forehand,1995),studyingasam-
whyindividualsdesistfromantisocialbehaviorduringthetransi- pleofseriousoffendersFhelpsincreasetheprobabilityofincluding
tion into adulthood. They suggested only that because desistance persistent offenders and thereby ensures sufficient power with
from antisocial activity “cannot be explained by change in the whichtocomparethisgroupwiththeiradolescence-limitedcoun-
person [italics added] or by his exposure to anti-criminal institu- terparts.Althoughitisbynomeansanormativesample,agroup
O
tions,weareleftwiththeconclusionthatitisduetotheinexorable of serious juvenile offenders constitutes an ideal one in which to
aging of the organism” (1990, p. 141). It is not clear, though, compare adolescents who desist from antisocial behavior with
within the general theory of crime, exactly what it is about this thosewhocontinuetheirantisocialbehaviorintoadulthood.
“inexorableaging”thatcontributestodesistance.Webelievethat Inthepresentstudy,weemployedgroup-basedtrajectorymod-
cluescanbefoundinrecentresearchonpsychosocialdevelopment elingOto identify distinct patterns of antisocial behavior by age
in late adolescence and early adulthood that indicate that signifi- within a sample of juvenile offenders who were followed for 4
cant improvements in future orientation, planning, and impulse years,fromages14to18untilages18to22.Becausewebegan
controltakeplaceduringthisperiodofdevelopment(Steinberget withasampleofindividualswhowereknowntobeantisocial,by
al.,2008,2009). definition we had no genuine “abstainers” in our sample and
R
Inthisarticle,wearguethatdesistancefromantisocialbehavior therefore had eliminated one group that is commonly found in
amongadolescence-limitedoffendersisinfactduetoincreasesin studiesofantisocialbehavior(i.e.,youthsinourstudywereeither
psychosocial maturity and that the reason life-course-persistent persistent offenders or adolescence-limited offenders). Although
offenderscontinuetoengageinantisocialbehaviorasadultsisthat Moffitt’s theory also differentiates between individuals who ex-
P
they do not experience the normative increases in psychosocial hibit antisocial behavior before adolescence (and who are more
maturitythattypicallytakeplaceasindividualsmatureintoadult- likelytobeantisocialacrossthelifespan)andindividualswhoare
hood.Thisview,whichintegratesnotionsborrowedfromMoffitt not antisocial until adolescence (and who are more likely to be
(i.e., that desistance from antisocial behavior during late adoles- adolescence-limited offenders), differences in age of onset of

cence is normative), Gottfredson and Hirschi (i.e., that the root antisocialbehaviorarenotthefocusofthepresentanalysis.
cause of antisocial behavior is deficient seAlf-control), and Stein- After identifying trajectories of antisocial behavior from ado-
berg and Cauffman (i.e., that the development of psychosocial lescenceintoadulthood,weexaminedthedevelopmentofpsycho-
maturity accounts for reductions in problem behavior), is consis- social maturity in the various trajectory groups. The central hy-
tent with recent developmental research that has elucidated the pothesis in the present study was that different trajectories of
neurobiological underpinnings ofPchanges in behavior commonly antisocial behavior would be distinguished by different levels of,
seeninadolescence,inparticular,thenormativematurationinlate andpatternsofchangein,psychosocialmaturity.Generallyspeak-
adolescenceandearlyadulthoodofbrainsystemsresponsiblefor ing, we expected that individuals who exhibited higher levels of
self-regulation(Steinberg,2008). psychosocial maturity would demonstrate lower levels of antiso-
Itisimportanttonote,AasMoffitt(1993)suggested,that“onthe cial behavior. We predicted that adolescents whose antisocial
basis of ... commonly used indexes of adolescent delinquency,” behaviorsignificantlydiminishedastheytransitionedtoadulthood
life-course-persistent and adolescence-limited offenders are “in- wouldbemorelikelythanpeerswhoseantisocialbehaviordidnot
distinguishable”andthatduringadolescencethereis“noeffective declinetoshowrelativegainsinpsychosocialmaturityand,more-
meansfordiscriminatingbetweentheseriouscareeroffendersand over, that the degree of decline in antisocial behavior over time
nonserious offenders” simply on the basis of their behavior (p. would be correlated with the degree of gain in psychosocial
678). If our speculation is correct, however, adolescence-limited maturity.Incontrast,wehypothesizedthatyouthswhoseantisocial
and life-course-persistent offenders should show very different behavior did not decline into adulthood would show little or no
patternsofpsychosocialdevelopmentduringadolescence.Thatis, growthinpsychosocialmaturityovertime.
although the two groups should be indistinguishable in adoles-
cencewithrespecttotheirantisocialbehavior,persistentoffenders
Method
would be expected to evince relatively lower levels of psychoso-
cialmaturityconsistentlyovertime,whereasadolescence-limited Participants
offenders would be expected to become increasingly mature as
theyage.Thisisnottosaythatpersistentoffenderswillshowno ParticipantsweremaleadolescentsenrolledinthePathwaysto
increaseinmaturityastheymoveintoadulthoodbut,rather,that Desistancestudy(seeMulveyetal.,2004),aprospectivestudyof
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
4 MONAHAN,STEINBERG,CAUFFMAN,ANDMULVEY
seriousjuvenileoffendersinPhoenix(n(cid:1)565)andPhiladelphia pant had committed. Honest reporting was strongly encouraged,
(n(cid:1)605;seeSchubertetal.,2004,forcompletedetailsofstudy andinterviewswereconductedoutofearshotofotherindividuals
methodology; the sample did not include a sufficient number of whenever possible. All recruitment and assessment procedures
young women with which to conduct the analyses used in the wereapprovedbytheinstitutionalreviewboardsoftheparticipat-
presentreport).Adolescentswereeligibleforstudyparticipationif ing universities, and adolescents were paid $50 for their partici-
they were between the ages of 14 and 17 and had been charged pationinthebaselineinterview(whenallowedbyfacilityrules).
withafelonyorsimilarlyseriousnonfelonyoffense(e.g.,misde- Each of the follow-up interviews was completed in one 2-hr
meanorweaponsoffense,misdemeanorsexualassault).Becausea session,andparticipantcompensationincreasedateachtimepoint.
large proportion of offenses committed by adolescents are drug Participants were reinterviewed every 6 months for 3 years fol-
offenses, the proportion of enrolled males whose enrollment of- lowing the baseline interview; aSfter 36 months, participants were
fensewasadrugoffensewascappedat15%ofthesampleateach interviewed annually. Follow-up interviews were conducted only
of the sites. Thus, only 15% of the sample at each site could be if completed within 6 weeks of the scheduled date. Participant
enrolled on the basis of a drug offense. All youths whose cases payments for the follow-up interviews were increased gradually
were being considered for trial in the adult system and had been with each contact, in oFrder to minimize attrition; retention of the
arraignedwereeligibleforenrollment.Ofeligibleyouths,67%of sample was excellent. From baseline interview to the 48-month
those whom we located and invited to participate in the research follow-up,841individualsincludedinthepresentanalyses(76%)
agreedtoenrollinthestudy(N(cid:1)1,170). completed all 8 interviews; 162 youths (15%) completed 7 inter-
O
Compared with youths who declined to participate, enrolled views; 67 youths (6%) completed 6 interviews; 32 youths (3%)
participantshadmorepriorarrestsleadingtoformalcharges(2.1 completed 5 interviews; and 3 youths ((cid:2)1%) completed 4 inter-
vs.1.5fornonparticipants),weresomewhatyoungeratfirstarrest views. The number of youths incarcerated at a given interview
(13.9 years vs. 14.2 years for nonparticipants), were somewhat timepointvaried(baseline(cid:1)47.1%incarcerated,6-monthfollow-
youngeratadjudication(15.9yearsvs.16.1yearsfornonpartici- up(cid:1)O43.9%incarcerated,12-monthfollow-up(cid:1)34.6%incarcer-
pants), and were somewhat more likely to be non-Hispanic Cau- ated,18-monthfollow-up(cid:1)28.9%incarcerated,24-monthfollow-
casian (25% vs. 20% for nonparticipants). Although statistically up (cid:1) 27.7% incarcerated, 34-month follow-up (cid:1) 25.5%
significant,thesedifferencesaremodestinmagnitude. incarcerated, 36-month follow-up (cid:1) 25.8% incarcerated, and 48-
Thebaselineinterviewwasconductedanaverageof36.9days monthfollow-up(cid:1)25.7%).Tocreateuniformtimemeasurement
R
(SD (cid:1) 20.6) after participants’ adjudication (for those in the forpurposesofthepresentanalyses,wecombineddatafromthe6-
juvenilesystem)or,ifparticipantswereeligibleforprosecutionas to36-monthsemiannualfollow-upinterviewsintoyearlonginter-
anadult,theirdecertification(i.e.,waiver)hearinginPhiladelphia vals, by averaging (in the case of psychosocial variables) or by
or their adult arraignment in Phoenix. The present analyses are counting the variety of endorsed offenses (in the case of self-
P
limited to the 1,105 males in the sample who completed at least reportedantisocialbehavior)fromthe6-monthand12-month,the
half of the interviews administered during the 5-year period cov- 18-month and 24-month, and the 30-month and 36-month inter-
ered by the present analyses (see below). At the time of the views,respectively.Thepresentanalysesthereforeincludeatotal
baselineinterview,thisgroupofparticipantswas,onaverage,16.5 of5timepoints,each1yearapart.Individualshadtoprovidedata

years of age (SD (cid:1) 1.11) and predominantly of lower socioeco- at both time points to have valid data for any annual period; this
nomic status. Less than 4.5% of the particAipants’ parents held a requirementresultedinthelossof14participantsfromtheanalytic
4-year college degree, and 40% of participants’ parents had less sample. Because the design of the study is an accelerated cohort
than a high-school education. The ethnic backgrounds of partici- design, there was a different number of participants at each age-
pants were as follows: 41% African American, 35% Hispanic groupfrom14to22years(14years,n(cid:1)141;15years,n(cid:1)344;
American,20%non-HispanicCauPcasian,and4%other. 16 years, n (cid:1) 658; 17 years, n (cid:1) 969; 18 years, n (cid:1) 1,034; 19
years,n(cid:1)893;20years,n(cid:1)673;21,n(cid:1)386;22,n(cid:1)84).
Procedures
Measures
ThejuvenilecourtineaAchlocaleprovidedthenamesofeligible
adolescents (based on age and adjudicated offense). Interviewers Of interest in the present report are measures of antisocial
then attempted to contact each eligible juvenile and his parent or behaviorandameasureoftheamountoftimetheadolescentspent
guardiantoascertainthejuvenile’sinterestinparticipationandto in the community, as opposed to in an institutional placement,
obtain parental consent. Once the appropriate consents had been during each interval (this measure was used as a covariate in the
obtained, interviews were conducted in a facility (if the juvenile analyses, because it affects opportunity to engage in antisocial
was confined), in the juvenile’s home, or at a mutually agreed- behavior). Also of interest were six elements of psychosocial
uponlocationinthecommunity. maturity: impulse control, suppression of aggression, consider-
The baseline interview was administered over 2 days in two, ation of others, future orientation, personal responsibility, and
2-hrsessions.Interviewersandparticipantssatsidebysidefacing resistance to peer influence (see Table 1 for means, standard T1
acomputer,andquestionswerereadaloudtoavoidcomprehension deviations,andintercorrelationsofkeyvariables).
problems caused by reading difficulties. Youths were informed Antisocial behavior. Involvement in antisocial activities was
thattheonlyexceptionstoconfidentialitywereifchildabusewas measuredwiththeSelf-ReportofOffending(Huizinga,Esbensen,
suspected or if a participant expressed plans to hurt himself or &Weiher,1991).Participantsreportediftheyhadbeeninvolved
someoneelse,describedaspecificplantocommitacrimeinthe inanyof22aggressiveorincome-generatingantisocialacts(e.g.,
future,disclosedthatsomeonewasinjailforacrimethepartici- “Takensomethingfromanotherpersonbyforce,usingaweapon,”
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
ANTISOCIALBEHAVIORANDPSYCHOSOCIALMATURITY 5
Table1
Means,StandardDeviations,andBivariateCorrelationsofKeyVariables
Variable BaselineM(SD) 48-monthM(SD) 1 2 3 4 5 6 7
1.Antisocialbehavior 4.77(4.19) 1.28(2.30) — (cid:4).28,(cid:4).38 (cid:4).32,(cid:4).39 (cid:4).14,(cid:4).30 (cid:4).09,(cid:4).22 (cid:4).08,(cid:4).22 (cid:4).02,(cid:4).11a
2.Impulsecontrol 2.95(0.95) 3.25(0.95) — .56,.60 .12,.19 .21,(cid:4)29 .30,.45 .16,.27
3.Suppressionofaggression 2.77(0.98) 3.05(0.92) — .17,.22 .11,.20 .26,.38 .08,.16
4.Considerationofothers 3.45(0.89) 3.71(0.81) — .39,.40 .07,.22 .05,.13b
5.Futureorientation 2.32(0.55) 2.65(0.54) — .16,.37 .10,.26
6.Personalresponsibility 3.00(0.47) 3.23(0.45) — .31,.45
S
7.Resistancetopeerinfluence 2.96(0.58) 3.32(0.53) —
Note. Allcorrelationsaresignificantatthep(cid:5).05levelunlessotherwisenoted.
aResistance to peer pressure was not significantly correlated with antisocial behavior at the 5-year follow-up. bResistance to peer pressure was not
significantlycorrelatedwithconsiderationofothersatthebaselineinterview,1-yearfollow-up,2-yearfFollow-up,and5-yearfollow-up.
“Carryingaweapon,”“Stolenacarormotorcycletokeeporsell,” Psychosocial maturity. Steinberg and Cauffman’s (1996)
“Used checks or credit cards illegally”). At the baseline and model of psychosocial maturity consists of three elements—
O
48-month interviews, these 22 questions were asked with the temperance, perspective, and responsibility—each of which has
qualifyingphrase“Inthepast12months,haveyou...”Atthe6- twocomponents.Inthepresentarticle,weexamineeachofthese
through 36-month follow-up interviews, these questions were six components independently. For temperance, we examine im-
asked with the qualifying phrase, “In the past 6 months, have pulse control and suppression of aggression; for perspective, we
you ...” examOine consideration of others and future orientation; and for
Responsesweresummedacrosssemiannualtimepointstocre- responsibility, we examine personal responsibility and resistance
ateannualvarietyscoremeasuresofantisocialactivity.Forexam- to peer influence. Four measures, described below, were used to
ple,thenumberofoffensesendorsedacrosstimewascounted,but create these six indices: the Weinberger Adjustment Inventory
the same offenses (e.g., “Carrying a weapon”) could count only (Weinberger & Schwartz, 1990), which includes subscales that
R
onceinagivenyearlongrecallperiodifanindividualendorsedthe assess impulse control, suppression of aggression, and consider-
item at two subsequent 6-month intervals. Thus, we created a ationofothers;theFutureOutlookInventory(Cauffman&Wool-
count of the total number of different antisocial acts that an ard,1999),whichwasusedtoderiveameasureoffutureorienta-
individualendorsedacrossayearlonginterval. tion;thePsychosocialMaturityInventory(Greenberger,Josselson,
P
Variety scores, a count of the number of different types of Knerr, & Knerr, 1974), which includes a scale that assesses per-
antisocial acts that an individual endorsed, were calculated for sonalresponsibility;andtheResistancetoPeerInfluencemeasure
each annual interval. Variety scores are widely used in crimino- (Steinberg&Monahan,2007).
logicalresearchbecausetheyarehighlycorrelatedwithmeasures ThreesubscalesoftheWeinbergerAdjustmentInventorywere

of seriousness of antisocial behavior yet are less subject to recall used:ImpulseControl(e.g.,“Isaythefirstthingthatcomesinto
bias than are self-reports of the frequencyAof antisocial behavior, my mind without thinking enough about it”), Suppression of
which yield unreliable estimates for higher frequency behaviors, Aggression (e.g., “People who get me angry better watch out”),
such as drug-selling. Hindelang, Hirschi, and Weis (1981) have and Consideration of Others (e.g., “Doing things to help other
arguedthatvarietyscoresandfrequencyscoresrepresentthesame people is more important to me than almost anything else”). The
antisocial propensity, and givenPthe problems associated with measure asks participants to assess how accurately a series of
frequencymeasures,varietyscoresrepresentapreferredmethodof statements matched their own behavior in the previous 6 months
measuring antisocial behavior, especially in a sample with high (ona5-pointscale,fromFalsetoTrue).Eachsubscalewasfound
ratesofantisocialbehavior. tohaveadequatereliability(asindexedbyCronbach’salpha)and
Exposuretime. BecauAseincarcerationcanlimitopportunityto good fit to the baseline data (as indicated by confirmatory factor
engageinantisocialacts,failuretoaccountforthiscanaffectthe analysis): Impulse Control (8 items; (cid:3) (cid:1) .76, normed fit index
identificationoftrajectoriesofantisocialbehavior(Piqueroetal., [NFI](cid:1).95,comparativefitindex[CFI](cid:1).95,root-mean-square
2001). Youths reported on a calendar the number of days during errorofapproximation[RMSEA](cid:1).07);SuppressionofAggres-
the recall period that they had been in a detox/drug-treatment sion (7 items; (cid:3) (cid:1) .78, NFI (cid:1) .96, CFI (cid:1) .97, RMSEA (cid:1) .06);
program, psychiatric hospital, residential treatment program, or ConsiderationofOthers(7items;(cid:3)(cid:1).73,NFI(cid:1).98,CFI(cid:1).99,
secureinstitutions.Accordingly,allanalysescontrolledforexpo- RMSEA(cid:1).04).
sure time, operationalized as the proportion of time in a year an The Future Outlook Inventory is a 15-item measure that in-
individual was in the community and not in these four secure cludes items from the Life Orientation Task (Scheier & Carver,
settings.Becausethisinformationwasnotavailableatthebaseline 1985), the Zimbardo Time Perspective Scale (Zimbardo, 1990),
interview, all baseline values for this variable were set to 1, a and the Consideration of Future Consequences Scale (Strathman,
method consistent with other work on antisocial behavior that Gleicher,Boninger,&Edwards,1994).Theinventoryaskspartic-
utilizes exposure time as a covariate (e.g., Mulvey et al., 2008). ipantstorankthedegreetowhicheachstatementreflectshowthey
The amounts of exposure time reported for each 6-month period usually act, on a scale of 1 (Never True) to 4 (Always True). A
were averaged to derive the exposure time covariate for each future orientation score is calculated based on the mean of eight
annualinterval. items from the scale (e.g., equivalent to “I will keep working at
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
6 MONAHAN,STEINBERG,CAUFFMAN,ANDMULVEY
difficult,boringtasksifIknowtheywillhelpmegetaheadlater”). similarpatternsofantisocialbehavioracrossage.Becauseanaly-
The scale showed good reliability and an excellent fit to the seswerebasedoncountdata(numberofdifferentantisocialacts
baselinedata((cid:3)(cid:1).68,NFI(cid:1).96,CFI(cid:1).97,RMSEA(cid:1).03). endorsed),weusedzero-inflatedPoissonmodelingtoaccountfor
ThePsychosocialMaturityInventory(Greenbergeretal.,1974) the clustering at zero (Lambert, 1992). We estimated the proba-
includesa30-item,reverse-scoredsubscalethatassessespersonal bilitythateachindividualbelongedtoagivengrouponthebasis
responsibility(e.g.,“Ifsomethingmoreinterestingcomesalong,I of the data and simultaneously derived maximum-likelihood pa-
willusuallystopanyworkI’mdoing”).Individualsrespondona rameter estimates associated with membership in each of the
4-point scale, from Strongly Disagree to Strongly Agree. An definedtrajectories(i.e.,posteriorprobabilitiesofgroupmember-
overall personal responsibility score is calculated as the mean ship). On the basis of posterior probabilities, individuals were
acrossall30items.Themeasureshowedexcellentreliabilityand assignedtotheirmostlikelygroSuptrajectory.Antisocialbehavior
an adequate fit to the baseline data ((cid:3) (cid:1) .89; NFI (cid:1) .82, CFI (cid:1) was examined across five measurement points, with a total age
.87,RMSEA(cid:1).04). range of 14 to 22. Because we are interested in developmental
Finally,themeasureofResistancetoPeerInfluence(Steinberg changes in psychosocial maturity that covary with antisocial be-
& Monahan, 2007) assesses the degree to which adolescents act havior, we conductedFgroup-based trajectory analyses by age.
autonomously in interactions with their peer group. Participants However, if there were no expected developmental differences,
arepresentedwithtwoconflictingstatements(e.g.,“Somepeople such as the impact of antisocial sanctions on antisocial behavior,
go along with their friends just to keep their friends happy” and onecouldalsomodelthedatabytime(e.g.,notgroupingpeople
O
“Other people refuse to go along with what their friends want to onthebasisofage;Mulveyetal.,2008).
do, even though they know it will make their friends unhappy”) Dataweretestedfordifferentnumbersoflatentclasses,andthe
andthenareaskedtochoosethecharacterizationthatmostclosely fitofdifferentmodelswascomparedwiththeBayesianinforma-
reflects their behavior. Next, participants are asked to rate the tioncriterion(BIC;Jones,Nagin,&Roeder,2001).Mixturesofup
degree to which the statement is accurate (i.e., “sort of true” or to seOven latent classes were considered. The best trajectory solu-
“really true”). Each item is scored on a four-point scale, ranging tionwasdeterminedbythreecriteria:thelowestBICvalueacross
from1(ReallyTrue)forthecharacterizationindicatinglessresis- models, a conceptually clear model, and a model in which each
tance to influence to 4 (Really True) for the characterization group included at least 5% of the sample. We decided on the
indicatingmoreresistancetoinfluence;answersof“SortofTrue” numberofclassesandthendeterminedtheformofthepolynomial
R
are assigned a score of 2 (if associated with the less resistant (e.g.,linear,quadratic)usedtocapturetheshapeofeachtrajectory.
option)or3(ifassociatedwiththemoreresistantoption).Tensuch Thehighestsignificantpolynomialtrendwasincludedinanalyses.
itemsarepresentedtoparticipants.Eachitemexploresadifferent AlthoughtheBICvaluesindicatedthatasix-groupsolutionbest
dimensionofpeerinfluence(e.g.,goingalongwithfriends,saying fit the data, a five-group solution was selected because the six-
P
things one doesn’t really believe), and one resistance to peer group solution did not add substantially to the understanding of
influencescoreiscomputedforthismeasurebyaveragingscores differentgrouppatterns(seeTable2).Furthermore,theadditional T2
on the 10 items. The measure showed excellent reliability and subgroupinthesix-groupsolutionwasdistinctneitherinshapenor
adequatefittothebaselinedata((cid:3)(cid:1).73,NFI(cid:1).92;CFI(cid:1).94; in level of antisocial behavior when compared with the groups

RMSEA(cid:1).04). identified in the five-group solution, and one trajectory group in
A the six-group solution consisted of less than 5% of the sample.
Plan of Analyses Thus, the five-group solution was selected because it had a low
BICvalue,aconceptuallyclearmodel,andanadequatepercentage
Analyseswereconductedintwosteps.First,mixturemodeling
ofthesampleineachtrajectorygroup.
was used to identify trajectoriesPof antisocial behavior by age. Figure 1 shows the final five-group antisocial behavior trajec- F1
Group-based trajectory modeling is an exploratory, data-driven
tory solution. Group 1 (the low antisocial behavior trajectory)
analytic technique that derives patterns of antisocial behavior
consisted of 37.3% of the sample. Individuals in this trajectory
basedonclustering,notaprioriideas.Second,patternsofchange
wereinvolvedinverylittleantisocialbehavioroverthecourseof
(e.g., growth curves) in tAhe six components of psychosocial ma- 5years.IndividualsinGroup2(themoderateantisocialbehavior
turity were compared among individuals who followed different
trajectory;18.7%ofthesample)engagedinaslightlyhigherrate
trajectoriesofantisocialbehavioridentifiedinthemixturemodels.
Inparticular,averagelevelsofpsychosocialmaturityandchanges
in psychosocial maturity among adolescents who persisted in
Table2
antisocial behavior were compared with data for those who de-
BayesianInformationCriterion(BIC)and2log(cid:6)B (cid:7)a ofthe
sisted from antisocial behavior during adolescence. Thus, in the e 10
ModelsConsidered
Resultssection,weidentifypatternsofantisocialbehaviorwithin
our data and subsequently explore how psychosocial maturity
No.groups BIC Nullmodel 2log(B )
variesasafunctionoftrajectorygroupmembership. e 10
1 (cid:4)5,124.44
Results 2 (cid:4)5,145.47 1 42.06
3 (cid:4)6,238.54 2 2,186.14
Trajectories of Antisocial Behavior 4 (cid:4)4,794.47 3 2,888.14
5 (cid:4)4,766.23 4 56.48
Weusedgroup-basedtrajectorymodeling(Nagin,2005;Nagin 6 (cid:4)4,759.78 5 12.90
7 (cid:4)4,765.34 6 11.12
&Land,1993)toidentifysubgroupsofindividualswhofollowed
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
ANTISOCIALBEHAVIORANDPSYCHOSOCIALMATURITY 7
12 each element of psychosocial maturity. Moreover, unlike other
dataanalytictechniques(suchasdualtrajectoryanalysis),growth
ng 10
di curve modeling permitted us to identify patterns in the develop-
en 8 ment of psychosocial maturity that are associated with a given
Off 6 trajectory group, rather than examine of psychosocial maturity
y of 4 within the whole sample. Individuals were classified into age-
riet groupsbasedontheirageatenrollmentintothestudy(e.g.,14,15,
Va 2 16, 17). For purposes of analysis, age was centered at 18 years,
0 which was approximately the average age across all of the time
14 15 16 17 18 19 20 21 22 pointsaswellasatransitionalpSointfromadolescenceintoadult-
Age hood.
Low Group Moderate Group
First, we conducted unconditional growth models to examine
Desisting Group Adolescence-Peak Group
the average pattern of each component of psychosocial maturity
Persister Group
overtime.UnconditionFalmodelsdeterminetheaveragepatternof
Figure1. Trajectoriesofantisocialbehavior. changeovertimeandwhetherthereissignificantvariabilitywithin
thesampleinlevelofpsychosocialmaturity(intercept)andchange
inpsychosocialmaturitywithage(slope).Ifsufficientvariability
O
ofantisocialbehaviorthandidthoseinthelowantisocialbehavior exists in either intercept or slope, antisocial behavior trajectory
trajectory and were consistent in antisocial behavior across time. membershipisusedtopredictthisvariance.
ThoseinGroup3(themid-adolescence-peaktrajectory;14.6%of Afterwehaddeterminedthegeneralpatternofdevelopmentin
the sample) displayed increased antisocial behavior through ado- unconditional models, if sufficient variability was found in inter-
lescence,peakingaroundage16,andlessenedantisocialbehavior ceptOorslope,weestimatedconditionalmodelsinwhichtrajectory
thereafter.Group4(thesteadilydesistingtrajectory;23.7%ofthe groupmembershipwasusedtopredictdifferencesintheintercept
sample) consisted of youths who were involved in high levels of and/or slope of the component of psychosocial maturity. For all
antisocial behavior at younger ages but rapidly decreased their intercept and slope terms for which antisocial behavior group
involvementinantisocialbehaviorovertime.Finally,membersoRf trajectorymembershippredictedheterogeneityaroundtheparam-
Group5(thepersistingtrajectory;5.7%ofthesample)consistently eters, we conducted planned contrasts to compare persistently
engaged in high levels of antisocial behavior from adolescence antisocial individuals with those in each of the other trajectory
intoyoungadulthood. groups. Because it is possible that growth in one domain of
Posterior probabilities, which estimated the likelihood of an psychosocialmaturitywashighlycorrelatedwithgrowthinother
P
individual’s being a member in each of the five groups, were domains of psychosocial maturity, when we examined change in
calculated.Ingeneral,posteriorprobabilitiesareanindexofhow one component of psychosocial maturity (e.g., impulse control),
wellindividualsarematchedtotheirassignedgroup.Ideally,each wecontrolledforsimultaneouschangeinthefiveotherdomainsof
individualshouldhaveaveryhighprobability ofbelongingtothe psychosocial maturity (i.e., suppression of aggression, consider-
group to which he is assigned and a very low probability of ation of others, future orientation, personal responsibility, and
belonging to other groups. To assess howAwell individuals have resistancetopeerinfluence).Thus,wealwaysexaminedgrowthin
been matched to their respective groups, one averages posterior one domain of psychosocial maturity independently of growth in
probabilities across all individuals within each group. Posterior otherdomains.
probabilitiesabove.70indicatethatindividualsarewellmatched Impulse control. Results indicated linear growth in impulse
to groups and that an adequate groPup solution has been achieved controlwithage(seeTable3);boththeinterceptandslopeterms T3
(Nagin, 2005). In the present analyses, posterior probabilities were significant, and there was significant heterogeneity around
indicated that, on average, individuals were well matched to the both.Accordingly,conditionalmodelswereestimatedthatallowed
groups to which they were assigned (average posterior probabili- bothinterceptandslopetermstovary;antisocialbehaviortrajec-
tieswereasfollows:lowAantisocialbehaviorgroup(cid:1)80%,mod- torymembershipwasusedtopredictthisvariance.
erate antisocial behavior group (cid:1) 79%, mid-adolescence-peak Controllingforconcurrentchangeinotherdomainsofpsycho-
group(cid:1)75%,steadilydesistinggroup(cid:1)73%,persisters(cid:1)85%). social maturity, we tested whether antisocial behavior trajectory
membership predicted differences in the intercept and slope of
Patterns of Psychosocial Maturity Over Time as a impulsecontrol(seeTable4).Ashypothesized,antisocialbehav- T4
iorgrouptrajectorymembershippredicteddifferencesinboth(see
Function of Trajectory Group Membership
Figure2).Plannedcontrastswereusedtoexaminespecificdiffer- F2
Because we were interested in comparing patterns of develop- encesintheinterceptandslopeofimpulsecontrolasafunctionof
mentalchangeinvariousaspectsofpsychosocialmaturityacross antisocial behavior trajectory group membership. Contrasts indi-
trajectorygroups,ratherthanexaminingoverallchangeinmaturity cated that individuals in the low antisocial, moderate antisocial,
within the entire sample, we conducted separate growth models steadily desisting, and mid-adolescence-peak groups reported
examining impulse control, suppression of aggression, consider- greaterimpulsecontrolthandidyouthsinthepersistingtrajectory
ation of others, future orientation, personal responsibility, and group(thecontrastbetweentheadolescence-peakgroupandper-
resistance to peer influence as a function of age. Growth curve sistent offenders reached borderline significance). Although they
modelingiswellsuitedtothepresentanalyses,becauseitallows showeddifferencesinaveragelevelsofimpulsecontrol,youthsin
a test of differences between groups in both level and change in thelowantisocial,moderateantisocial,andsteadydesistergroups
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
8 MONAHAN,STEINBERG,CAUFFMAN,ANDMULVEY
Table3
UnconditionalGrowthModelsofImpulseControl,SuppressionofAggression,andConsiderationofOthers
Model1 Model2 Model3
Impulsecontrol Suppressionofaggression Considerationofothers
Effect B SE B SE B SE
Fixedeffects
Intercept 3.08(cid:1)(cid:1) 0.02 2.84(cid:1)(cid:1) 0.02 3.51(cid:1)(cid:1) 0.02
Linearslope 0.03(cid:1)(cid:1) (cid:5)0.01 0.03(cid:1)(cid:1) (cid:5)0.01 S0.04(cid:1)(cid:1) (cid:5)0.01
Randomeffects
Intercept 0.49(cid:1)(cid:1) 0.02 0.48(cid:1)(cid:1) 0.02 0.33(cid:1)(cid:1) 0.02
Linearslope 0.01(cid:1)(cid:1) (cid:5)0.01 0.01(cid:1)(cid:1) (cid:5)0.01 0.01(cid:1)(cid:1) (cid:5)0.01
Level1error 0.37(cid:1)(cid:1) 0.01 0.39(cid:1)(cid:1) 0F.01 0.35(cid:1)(cid:1) 0.01
Modelfit
(cid:4)2loglikelihood 18,834.0 19,116.4 17,896.7
AIC 18,846.0 19,128.4 17,908.7
BIC 18,876.0 19,158.4 17,938.7
O
Note. AIC(cid:1)Akaike’sinformationcriterion;BIC(cid:1)Bayesianinformationcriterion.
(cid:1)(cid:1)p(cid:5).01.
O
did not differ from the persisters in the pattern of change in Suppression of aggression. Growth in suppression of aggres-
impulse control with age. However, persisters and individuals in sion from adolescence to early adulthood was examined with a
the mid-adolescence-peak antisocial behavior trajectory showed similar strategy. Analyses indicated linear growth in suppression
opposite patterns of change with age; those whose antisocial of aggression with age; both the intercept and slope terms of the
behavior peaked in mid-adolescence and then declined showedRunconditional model were significant and indicated significant
increasesinimpulsecontrolacrossadolescenceandyoungadult- individualvariabilityinbothcoefficients(seeTable3).
hood,whereaspersistersshoweddeclines.Thus,individualsinthe The conditional model revealed that trajectory group member-
persistent antisocial behavior group showed diminished self- ship predicted differences both in the intercept and the slope of
controlatage18anddeclinesinself-controlovertimeP. suppression of aggression (see Table 4 and Figure 3). Planned F3
Table4
ConditionalGrowthofImpulseControl,Suppr essionofAggression,andConsiderationofOthersonOffendingTrajectoryGroup
AModel1 Model2 Model3
Impulsecontrol Suppressionofaggression Considerationofothers
Effect B SE B SE B SE
P
Fixedeffects
Intercept 0.59(cid:1)(cid:1) 0.09 0.54(cid:1)(cid:1) 0.09 2.15(cid:1)(cid:1) 0.09
Trajectorygroup F(4,6198)(cid:1)12.87(cid:1)(cid:1) F(4,6198)(cid:1)16.87(cid:1)(cid:1) F(4,6198)(cid:1)15.96(cid:1)(cid:1)
Impulsecontrol A — — 0.48(cid:1)(cid:1) 0.01 (cid:2)0.01 0.01
Aggressionsuppression 0.46(cid:1)(cid:1) 0.01 — — 0.01 0.01
Considerationofothers (cid:2)(cid:4)0.01 0.01 0.01 0.01 — —
Futureorientation 0.07(cid:1)(cid:1) 0.02 0.01 0.02 0.36(cid:1)(cid:1) 0.02
PSMI 0.21(cid:1)(cid:1) 0.02 0.19(cid:1)(cid:1) 0.02 0.08(cid:1)(cid:1) 0.02
RPI 0.07(cid:1)(cid:1) 0.01 0.01 0.02 0.02 0.02
Linearslope(age) 0.03 0.01 0.02(cid:1)(cid:1) 0.01 0.03(cid:1)(cid:1) 0.01
Trajectorygroup F(4,6198)(cid:1)4.08(cid:1)(cid:1) F(4,6198)(cid:1)3.69(cid:1)(cid:1) F(4,6198)(cid:1)4.29(cid:1)(cid:1)
Randomeffects
Intercept 0.25(cid:1)(cid:1) 0.01 0.26(cid:1)(cid:1) 0.01 0.23(cid:1)(cid:1) 0.01
Linearslope (cid:2)0.002(cid:1)(cid:1) (cid:2)0.01 0.004(cid:1)(cid:1) (cid:2)0.01 0.004(cid:1)(cid:1) (cid:2)0.01
Level1error 0.29(cid:1)(cid:1) 0.01 0.30(cid:1)(cid:1) 0.01 0.34(cid:1)(cid:1) 0.01
Modelfit
(cid:4)2loglikelihood 16,077.8 16,418.5 17,174.9
AIC 16,115.8 16,456.5 17,212.9
BIC 16,210.9 16,551.6 17,308.0
Note. Dashes indicate that term was not estimated. PSMI (cid:1) Psychosocial Maturity Index; RPI (cid:1) Resistance to Peer Influence; AIC (cid:1) Akaike’s
informationcriterion;BIC(cid:1)Bayesianinformationcriterion.
(cid:1)(cid:1)p(cid:5).01.
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
ANTISOCIALBEHAVIORANDPSYCHOSOCIALMATURITY 9
4
3.5
ol
r
ont 3
C
e
puls2.5
m
I
2
S
1.5
14 15 16 17 18 19 20 21 22
Age
F
Low Moderate Desister Adolescence-Peak Persister
Figure2. GrowthinImpulseControl(cid:8)AntisocialBehaviorTrajectoryGroupMembership.
O
contrastsindicatedthatlowantisocialandsteadydesisterindivid- wellasinthepatternofchangeinthisaspectofmaturitywithage
ualsshowedgreatersuppressionofaggressionthandidpersistent (see Table 4 and Figure 4). Planned contrasts examining differ- F4
offenders. Examination of patterns of change in suppression of ences among trajectory groups indicated that, compared with in-
aggression over time indicated that individuals in the steadily dividOualsinthepersistentantisocialbehaviortrajectory,individu-
desisting and mid-adolescence-peak trajectory groups showed als in every other trajectory group (low antisocial, moderate
more rapid increases in suppression of aggression with age than antisocial, steadily desisting, and mid-adolescence-peak) showed
didpersisters,who,aswithimpulsecontrol,declinedinsuppres- greaterconsiderationofothers(thedifferencebetweenindividuals
sionofaggressionfromadolescencetoadulthood.Therewereno
in the mid-adolescence-peak and persistent groups reached bor-
differencesbetweenpersistersandeithertheloworthemoderateR
derline significance). Surprisingly, however, individuals in the
antisocial behavior group in patterns of change in suppression of
persistent antisocial behavior trajectory increased more rapidly
aggressionwithage.Again,youthswhodidnotdesistinantisocial
with age in their consideration of others than did individuals in
behavior across the transition to adulthood showed diminished
either the low or the moderate antisocial behavior trajectory
suppressionofaggressionbothatage18andovertimeP.
groups.(Theloweraveragelevelofconsiderationofothersshown
Consideration of others. Unconditional models showed that
by persisters is attributable to their substantially lower scores on
considerationofothersshowedlineargrowthfromadolescenceto
this variable at younger ages; as Figure 4 indicates, by early
adulthood, with intercept and slope significant, as well as signif-
icantheterogeneityaroundtheseparameters(se eTable3).Conse- adulthoodtheyhavecaughtupwiththeothergroups.)Therewere
quently,bothtermswereallowedtovary,andweusedantisocial no differences between individuals in the persistent antisocial
behaviortrajectorygroupmembershiptoprAedictthisvariabilityin behavior group and either the steadily desisting or the mid-
theinterceptandslopeofconsiderationofotherswhilecontrolling adolescence-peak group in patterns of change in consideration of
forchangeinotheraspectsofmaturity. others with age. Thus, although there were differences among
Antisocial behavior trajectory group membership significantly groups at age 18, with persistently antisocial youths showing
P
predicteddifferencesinaveragelevelofconsiderationofothersas diminishedconsiderationofothersatage18,persistersincreased
A
3.2
n
o
ssi 3
e
r
g
Ag 2.8
of
n 2.6
o
si
s
re 2.4
p
p
u
S 2.2
2
14 15 16 17 18 19 20 21
Figure3. GrowthinSuppressionofAggression(cid:8)AntisocialBehaviorTrajectoryGroupMembership.
tapraid5/z2p-devpsy/z2p-devpsy/z2p00409/z2p2443d09z xppws S(cid:1)1 4/13/09 4:20 Art:2008-0698
10 MONAHAN,STEINBERG,CAUFFMAN,ANDMULVEY
3.8 18, and there is no evidence that they caught up, or further
rs3.6 declined,infutureorientationovertime.
he3.4 Personal responsibility. Unconditional growth models indi-
Ot
of 3.2 cated significant linear growth in personal responsibility during
on 3 adolescencebutasignificantdecelerationingrowthasindividuals
rati2.8 transitionedintoadulthood;thispatternwassimilartothatseenfor
Conside22..46 fvuatruiarebiolirtiyenatraotuionndt(hseeeinTtearbcleep5t)a.nTdhleinreeawrsalsopsiegnpiafriacmanetteirnsdbivuitdnuoatl
2.2
aroundthequadraticparameter,andthisindicatedthatallindivid-
2 ualsslowedintheirdevelopmenStofpersonalresponsibilityatthe
14 15 16 17 18 19 20 21
samerate.Insubsequentmodels,therefore,theinterceptandlinear
Figure 4. Growth in Consideration of Others (cid:8) Antisocial Behavior slopewereallowedtovary.
TrajectoryGroupMembership. Average level of personal responsibility did not vary as a
F
functionoftrajectorygroup,butthereweresignificantdifferences
between groups in patterns of change with age (see Table 6 and
morerapidlyintheirconsiderationofothersandreachedthelevel Figure 6). Planned contrasts indicated that individuals in the low F6
ofconsiderationreportedbyothersbyage22. antisocial, modOerate antisocial, and steadily desisting groups did
Futureorientation. Acrossthesample,therewasasignificant not differ from persistently antisocial youths in the pattern of
linear increase in future orientation during adolescence but a growth in personal responsibility with age. However, and unex-
deceleration in growth as youths transitioned into adulthood (as pectedly, compared with persisters, adolescence-peak individuals
T5 indexedbyasignificantquadratictrend;seeTable5).Significant reporOtedslowerlineargrowthinpersonalresponsibilityovertime.
heterogeneitywasfoundaroundtheinterceptandslopetermsbut Resistance to peer influence. In general, resistance to peer
not around the quadratic term (i.e., all individuals decelerated at influenceshowedapatternsimilartothatseeninpersonalrespon-
thesamerate).Consequently,conditionalmodelspredicteddiffer- sibility:alinearincreaseacrossadolescence,withgrowthslowing
encesininterceptandlinearslopeoffutureorientationbutnotin asyouthstransitionedintoadulthood(seeTable5).Althoughthere
quadraticgrowth. Rwas significant heterogeneity in the intercept and linear slope of
Antisocialbehaviortrajectorygroupmembershippredicteddiffer- resistance to peer influence, there was no variability in the qua-
ences in individuals’ average level of future orientation but not in dratic term (e.g., all individuals slowed growth at the same rate).
T6,F5 changesinfutureorientationwithage(seeTable6andFigure5). Consequently, subsequent models tested if antisocial behavior
Planned contrasts indicated that low antisocial, moderaPte antiso- group membership predicted differences in intercept or slope,
cial, and steadily desisting adolescents evinced significantly controlling for concurrent development in other domains of psy-
greater future orientation than did persistently antisocial individ- chosocialmaturity.
uals;therewerenodifferencesinfutureorientationbetweenindi- Conditionalmodelsrevealedthatthisvariabilityintheintercept
viduals in the mid-adolescence-peak and persi stent antisocial be- andslopeofresistancetopeerinfluencewasunrelatedtotrajectory
haviortrajectories.Thus,individualsinthemid-adolescence-peak group membership (see Table 6). That is, once other aspects of
A
and persister groups showed the lowest future orientation at age psychosocial maturity were taken into account, the trajectory
Table5 P
UnconditionalGrowthModelsofFutureOrientation,PersonalResponsibility,andResistancetoPeerInfluence
Model4 Model5 Model6
A
Futureorientation Personalresponsibility Resistancetopeerinfluence
Effect B SE B SE B SE
Fixedeffects
Intercept 2.50(cid:1)(cid:1) 0.01 3.09(cid:1)(cid:1) 0.01 3.09(cid:1)(cid:1) 0.01
Linearslope 0.06(cid:1)(cid:1) (cid:2)0.01 0.03(cid:1)(cid:1) (cid:2)0.01 0.06(cid:1)(cid:1) (cid:2)0.01
Quadraticslope (cid:4)0.01(cid:1)(cid:1) (cid:2)0.01 (cid:4)0.01(cid:1)(cid:1) (cid:2)0.01 (cid:4)0.01(cid:1)(cid:1) (cid:2)0.01
Randomeffects
Intercept 0.14(cid:1)(cid:1) 0.01 0.11(cid:1)(cid:1) 0.01 0.15(cid:1)(cid:1) (cid:2)0.01
Linearslope 0.003(cid:1)(cid:1) (cid:2)0.01 0.002(cid:1)(cid:1) (cid:2)0.01 0.002(cid:1)(cid:1) (cid:2)0.01
Level1error 0.16(cid:1)(cid:1) (cid:2)0.01 0.11(cid:1)(cid:1) (cid:2)0.01 0.16(cid:1)(cid:1) (cid:2)0.01
Modelfit
(cid:4)2loglikelihood 11,278.3 7,894.3 11,199.9
AIC 11,292.3 7,908.3 11,213.9
BIC 11,327.4 7,943.4 11,248.9
Note. AIC(cid:1)Akaike’sinformationcriterion;BIC(cid:1)Bayesianinformationcriterion.
(cid:1)(cid:1)p(cid:5).01.