DOCUMENT RESUME

ED 233 820 \ PS .013 823

AUTHOR Huesmann, L. Rowell; Yarmel, Patty WaEnickTITLE Intellectual Competence and Aggression.SPONS. AGENCY, National Inst. of Mental Health (DHEW), Rockville,

Md.,PUB DATE Aug 83GRANT MH-34410.NOTE - 25p.; Paper pre'ientedat the Annual Meeting of the

American Psychological Association (Anaheim, CA,,August 26-30, 1983).

PUBTYPE Reports - Research/Technical (143) --Speeches /Conference Papers (150)

MF01/PC01 Plus Postage.Adolescents; Adults; *Aggression; Children; FamilyInfluence; Grade 3; Individual Development;*Intellectual Development; Longitudinal Studies

EDRS PRICE .,

DESCRIPTORS

ABSTRACTUsing data from a broader longitudinal study, this

investigation explores within-subject and cross-generationalstability of intellectual competence and the relationship of suchstability to aggresgive behavior. Data were gathered three times(when subject's' modal age was, 8, 19, and 30 years). Initially,.subjects included the entire population enrolled in the third gradein Columbia County, New York. While aggression and IQ scores wereobtained for 632 subjects 8 years of age, samples available formultivariate analyses were much,smaller. All, however, were /

sufficiently large to provide reasonable statistical power. /

Additional measures of IQ or achievement were made when subjects were.19 and 30 years old. When subjects were 8 years old, mothers andfathers were interviewed;'-when 30 years of age, data were gatheredfrom the spouses and children. Measures relevant'to this study werethe California Test of Mental .Maturity,-school records of,IQ, theWide Range Achievement Test, Reitan's trail-making test, and a videocar race game. Results'suggested that aggression and intellectualcompetence are strongly related and that this strong relationship isdue partially tothe interference of aggressive behaviox patternswith .the development of intellectual competence. Intellectual,competence and.aggressive behavior showed siTilar patterns ofStability across generations and marriage - pair's and within subjects.Early IQ did not predict change in aggression from childhood toadulthood. (RH)

***********************************************************************Reproductions supplied by EDRS are the best that can be made

from the original document.***********************************************************************

milimmrriliesmann

Intellectual Competence and Aggression

L. Rowell' HueSmann and Patty Warnick Yarmel

. U.S. DEPARTMENT OF EDUCATION '1 NATIONAL INSTITUTE OF.EDUCATIONC,

University .of. Illinois at Chicago EDUCATIONAL RESOURCES INFORMATION_ CENTER (ERIC/CNI .

This document has been reproduced asCO received from the person or organizationoriginating itPr . 0 MinLx.franges have been made to improve

......,, ' reproduction quality.IVN

Points of view or opinions stated in this docu-CNJment do not necessarily represent official NIEposition or policy.

This paper was presented at the smposium on "Consistency of

Aggression and its Correlates over Twenty Years," at the meetings of

the Americ,an Psychological,,Association, Anaheim, California August

1983. ?The research wasSupported in part by grant MH-34410 from the

National Institute of Mental Health. Requests fqr reprints should be

sent to L. Rowell Huesmann, pepartment of Psychology, University of$

I.4d4is at Chicago, Box 4348, Chicago, Illinois.

n.

"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED BY

L RC.Huesm a,Imn

TO THE EDUCATIONAL RESOURCES. INFORMATION CENTEFUERIC)."

f2UebMd1111

-Intellectual' Competence and Agi?e'ssibn

I. Rowell'HUesmann and Patty Warnick YarMel.

Uni'versity of Tllinois at Chicago,

In the first presentation in this symposium, Eron argued,

compellingly that aggressiveness is a reasonably stable characteristic,

over the 20 yearsfrom childhood to middle adulthood. tFurthermore,

aggression is a characleristic that perpetuates itself within a family'

system. Aggressive people are more likely-to marry aggressive spouses

and raise aggressive children. However, aggression cannot be

considered in isolation from other developing behaviors and

charateristids. In particular, one must consider the relation

between'aggressiveness and intellectual competence.

While a number ofi'studies have revealed a correlation between

aggression and poor academic,performance or poor performanqe on

standardized tests of intellectual ability, the extent of this

relation and the processes responsible for it are not yet clear.

There are several psychological models that have been offered to

explain the relation between intellectual deficits and aggression.

Lower intellectual ability undoubtedly makes success in school more

difficult and leads to poorer achievement. Poorer achievement in.turn

probably reduces self-esteem and frustrates the child. When faced

with difficult social situations,_the frustrated child with lower

self-esteem is more likely to respond aggressively. In addition, the

child With diminished intellectual abilities would find it more

difficult to divise-alternative less direct strategies to qtain

his/her goals. Regardless of the 'success or lack.of success that the

low IQ child ha.,,s had with aggressive behavior, the child will tend to

repeat "the behavior if the child cannot lea,rn and retain alternative\ 0..

strategiet.

While socioeconomIc'factoraffect both intellQplual abilities. .

.

and a child's aggressiveness, a number of studies have shown that,. .- .

,.independently of socioeconomic status, IQ is a kignificant predictor'.

. , .

of-delinquency (Gibson & West,1970; Lowenstein,19?7; Hirschi.,Y

&

Hindelang, 1977;'Hindelang, Hirschi, & Weis, 1981). This relation may.

be somewhat inflated because bright delinquents may not be apprehended

as often. Also, the extent to which the relation with IQ due .to

performance failure in school instead of'intellectual incompetence per

se is dijficult to assess. The majority 'of studies om\this topic have

actually 'related achievement rather than competence to aggressLn.

Lowenstein (1977) and Olweus (1978),have shOwn that bullies in school'

are generally below average in academic skills. A large quantity of

recent evidence also exists relating poorJacademi'c achievement to

delinquency (Hogenson,1974; Loeber & Dishion,1983; Wadsworth,1979;

//' Andrew,1981). However, in .one study specifically directed at

r., separating out the effects of- IQ .and achievement, Feshback & Price

(1983) found only a negligible relation between IQ and aggression.

While no ene explanation for the relation between aggression and

intellectual competence has emerged as a dominant? theory, most

researchers have impl-icitly'adopted the assumption that the direction

of the effect is mostly, from intellectual incompetence or achievement

failure to aggression-.; Yet, Rnce one concedes that the relation is at

least partially due to the performance failure that stems.from i.,ow IQ,_-

one must also consider the possibility that aggression is a cause of

poor achievement and that the relation between intellectual cOmpetence

and aggression is bidirectional.

4.

huesmann

..The relation between the development of -aggression and the

development Of intellectual competence cannot be adequately explored

without'aiso investigating the relative stability of these

characteristigs ftom childhood to adulthood- Numerous studtes.have

revealed that adult intellectual competence is predictable from' middle

childhood intellectual, competence (Jensen, 1980). ,Similarlt, as'Eron. ,

argued in the first presentation, add as recent research has shOwn

(Huesmann, Eron, Lefkowitz& Walder,1983; Olweus, 1979), aggresiion is,

also stable. But how do these variables' stal3ili:t"ies,compare ilrf the

same sample? More importantly, assuming simijar stabilities,are-

chariges in a child's aggressive behavior from childhood to adukthood.

more predictable from the intellectual competencethan changes

in the child's intellectual Apmpetence)are,-predictable from'ihe

Vi 4'1child's aggression? 4'

To answer these quest ionswe will examine the stability of

intellectual competencS and its relation to aggressive behavior over

22 years from age 8 to age 30 in a VanTle of over 600 subjects.

Mehod

The longitudinal data were collected s part of a larger 22 yeart . 1

study that has been described in the fiist p esentation by Eron and in

detail elsewhere (Er'on, Lefk6witz, & Walden,1973; Eton,

' jWelder, & Huesmann, 1977; Huesmanixet al., 1983).,. ,

Subjects

The,Ntial lubj'ects comprised the entire populatiOn of youngsters'

enrolled in the 'third grade in a semi- rural county in ew York State.

Huesmann a

(COlumbia County). This included apprbximatelY.870 youngsterS whose

modal age at the time was 8 years. Thee youh9sters were tested in

their classrooms with a variety of.piOcedures. We also interviewed

personally approximately;75%'oftheir mdther and fathers. :Ten years

. later, vire reintervieWed '427 of the original subjects (41 boys, 216p.

. modal age, 19). 'Fihally, in 19-81 we obtained somedata-,Oh 632

sesr, and children.of the'subjects, their s

hile we have age 8 aggression scores and age 8 IQ scores for all

63 of these subjects, longitudinal analyses will involve'somewhat

smaller niimbers. . At age 19 we were able to obtain` IQ data for_ony, .86

subject while at age 30 we obtained achievement scores got 294

'subjects. Furthermore,,we had interviewed S'oth parents ,of only 107 of

these subjectsin:1960. While these factors make the samples,

'available for so" multi:Variate analyses, much smaller=thah the total,

all important samples arg.large enough to provide reasonable,

statistica power.II

t .

Procedure add Measures

t,

The procedures employed and 'Measures have been destribed in',

1. io

detail. elseWhere..(Huesmann et al,',1983; Lefkowitz. et-al.,..

1977).. The..,

'measuus of pa'rticuiar concern for this paper are the s bjects age=,^r) ^

IQ which was measuved by the California Test of MentalvMaturitY.FI

(Sullivan, Clark & Tiegs, 1961), thl subjects' age 19 IQ which was

obtained; from school records, and the subjeCts'jeadigg, spelling and,,arithmetic achieN-7ement atage 30 which was measured by Jastak 4

Jastaks' (1'978) i?de Range Achievement Test (WRAT). Stores op these

tests were also combined`toyi.a WRATverbal scoreI ,

,

Huesmann

(st)elling+reading) and a WRAT total score(

(tpelling+reading+arithmetic); The children of the subjects were also.

given. WRAT,'s,.but their,osCores were standardized by age to. represent]

deviations from age norms. Twoother'measures of_intelleCtual

functioningused with'the age36 s 'ipjects were Reitan's trail-making ,

A.

a

4test in' which the. subject'cOnneCts- umbers and lines.aS rapidly as

"ossible,and a.video car xatOg me in which.the subject mUst keep,a

simulated Car on the roaa zs °rig a6'

Results

.1

. . .

The correlations of age 8 IQ 'with age 19 `dbd ape p measures of \

''.', .

intellectual competence,are shown in-Table'l. .-As one would expect,

age 8 IQ is significantly relateApto.schOol achievement and IQ at age

19 and to achievement Levels measured by the WRAT at age '30. The 22---%

yer relation is slightly 'higher f.or males than for° female's. These) .

cesults replicate-,in, our sample what many cphets

0have-found. More

.c

,

. , -.

unique are the correlations i'n Table 1 showing that age 8"IQ predicte.' '1

5).

,..9.I.

video game performance and .trailmaking time 22 years later:.

A

AppaTently the visual-motOr cooroinatiu component of age 8

predictive-of lilelOng 16rformance.' tillUsrates,hee 3

longitudinal relations.frOm age 8 to age 30-graphically. The mean

scores represented.by.these barographs were significantly different

substantiating the st/

ility of .intellectual competence over time. I

In Table the data are.presented showing thettability of

'intellectual competence-.actosS 3 genFrations in our subjfects'

,families. The greater was the subject's parents' eclucation, the

higher the subject's IQ,at age 8, and the higher thee subject's

Huesmann

achievement test scores and,,educat,ion at ao. The greater are these'

scares for.e subject, the higher are the subject's child's WRAT

'scare's.. These effects produce a weak but positive relation from

grandparents' educationto.grandchild's WRAT score in the small sample

(N=67) On whith these d.ata are available.

While the stability of intellectual competence within subjects

and across generations undoubtedly has genetic, physiological andp

environmental causes (Jellsen,.1980), Table 3 reveals a stability

families that carat be 'attributed to genetic or physiologic4

cet

fadars. People tend to marry people with similar educational

backgrounds' and comparable intellectual skills.

Taken together, the data presented thiis.far-cin be interpreted.in

one of two ways. They show that a person's intellectual competence as_

an adult can bqkpredicted to a significant extent from his IQ as a

child,,his family's intellectual. competence and 'his spouse's

competence. However, the percentages' of Variance left unexplained are

large; so it is also fair to say that a substantial portion of one's

intellectual competence is not predicted by these factors. Let us now

-examine,what other factors dopredict competence.

Table 4,:ihows.the correlations.between the measures' of

intelfectual competence at each age and several potential predictors.

The sample sizes for the correlations vary as the different-,

significance levels suggest. A number of results arjas one would

expect, but several are notable. While the number of children in the

family was negatively related to the child's intellectual competence,

birth order was not. Of all the birth related variables investigated

(including additional variables not listed in the table),,) only

mother's age was a significant predictor of intellect. Older mothers

Al;

7

had more intelligent children, suggesting that a social class effectt

may be occurring, with poorer educated women !laving their children at

a ybunger age.

Probably the most important results'revealedj.n this table are

the strongLorelations betufeen thee child's identification with his4,1vi-

parents, the child',s social behaviors and the child's intellectual

competence both at age 8 and as an adult. The More intellectually' .

competent adult was a child who identified more with his/her parents,.

who was.less,aggressive and who was more popular. The more competenty

.adult also was rejected and punished less by his/her parents and was

restricted les by his/her mother. However, it is difficult to know

the extent towhich these. parent behaviors represent different child

rearitng4

styles or simply reflectchild behaViOrs.

In Table 5 the correlations of early intellectual competence with

adult aggvsion are directly compared to the correlations of early

aggression with adult intellectual competence. One can'see that for

both males and females early aggression is a much better'predictor of

adult intellectual achievement than early intellectual competence is

of adult aggression. The magnitude of the correlations from

peer-nominated aggression at age 8 to WRAt,scores at age 30 is about

the same as the magnitude from age 8 peer nominated aggression to

adult aggression. H er, the cbrrelations from a.ge 8 IQ to adult

dggression are much lower than the correlations to.adult WRAT.

Furthermore, as the partial correlations in Table 6 indicate, the

relation from early to later aggression cannot be ascribed to the

stability of intellect over time since the partial correlations

are 'about as high as the raw correlations.

These results suggest that childhood aggression is a more

Hue.Smann

\

important contributor to adult intellectual failure than has usually'

been hypothesized -- a More important contributOr- than" early"IQ is to

adult aggression. Perhaps the best analyeskto test this theory are

multiple regripssiOns in which adult aggression and academic0

achievement re predicted from early factors including aggression and

IQ. Such regressions are shown in Tables'q and 8.

In Table 7 age 8 IQ and peer-nominated aggression are predicted

from early socioeconomic and birth factors, then .from these plus, age 8

IQ or aggression, and then from, these plus relevant child rearingA

factors. The regressions show that-independently of socioeconomic,

birth and child rearirg factors, and aggression predict each other

at that time. Adding IQ or aggreSsion to the regression equations

significantsignificant increments in the multiple, correlation. -JOne-must

conclude that either a common factor not represented in these

equations is causing the development of both'aggressive behavior '

styles and'diMinished intellectual competence prior to,age 8 or that

one or both of these variables affects the other.

The regressions in Table.8 suggest thati-at.least between *age 8

and age 30, it is aggressive behavior patterns that are interfering

with'intellectuardevelopment more than kit is diminished intellectual

competence that is stimulating .4ggression. In `his table adult °WRAT

scores are first'regressed on IQ, birth and socioeconomic factors.4 t, 1,0V

Then, aggression is added as,a predictor, and'finally the parents'

child rearing behaviors are inserted. While early IQ is, of course,

the best predictor 15f adult WRAt scores, early aggression aidd's a very

significant contributiOneven when parental child rearing styles are

partialed out. On the other hand, when the procedure is'reversed .on.0

the right side of the tablet, nd adult aggression is. eedicted, early

i(A

10_

,

nucauIwwL

IQ does not make any significant contribution beyond what was-

predicted by early a'gVession.

"° NSummary and Conclusions

A.

The results from this longitudinal' study suggest that aggression

and intellectual competence are trongly related at least partially

Abecause aggression,interferes ;Wit the deveiopmenatoi intellectual

competence. Intellectual competence and aggressive behavior bear

strikingly similar patterns of stability from chIldhod to

middle-adulthood. In addiCion', both appear to perpetuate themselves.

froth grandparent,. to parent, to child; and both'perpetuate theMselves

within marriage pairs. While diminished intellectual abilities and

academic failure may well.stimulate aggressive responses in the young

child, our,

data did no+.Q

reveal any contribution of early IQ to

predicting change in aggression rom childhood.to adulthood. Of

course, thisdoe8 not mean that changes in academic performance

between childhood and.adulthood might not affect aggression. The rble

of academic success and failure should be important according to most

theories.of aggression. Qur data do demonstrate, however, that

d- regardless of a young child's IQ, his intellectual achievements

concurrently and in young adulthood are adversely affected.if he

consistently behaves aggressively.

10

f

ve

1'

References

Andrew, J.M. Delin, quency: Correlating variables.,.

Journal of Clinical Chidd Psychology, 19-81, 10, 136.'140.

Eron, L.D., 44alder, L.O. & Lefkowitz, M.M. Learning of Aggressiona

in Children, 1971, Boston: Little, Brown.

Feshbach, S., Price, J. The development of cognitive competencies

and the control of aggression, Paper presented at the

-InternatiOnal Society for the- Study of,Behi'vior Development.

Munich./ August, 1983.

Gibson, N.B. & West, D.J. Social and intellectual handicaps

as precursors of early delinquency: British.Journal of

.Criminology, 1970, 10,.21-32.

Hindelang, M J., Hirschi, T., & Weis, O.G. Measuring. Delinquency.

.Beverly Hills, London: Sage, 1981.

F

Hirschi, T., & Hindeldng, M.J. Intelligence and delinquency: A4

revisionist view. American Sociological Revrew,1977,42,57,1-587.

42, 571-587.

-

'Hogenson, D.L.- Readin2 faihire and juvenile delinquency.

Bulletin of the Orton Society, 1974, 24, 164-169.

Huesmann, L.R., Eron, L.D., LefkowitE, M.M., & Walder, L.O.

The Stability of aggression over time and generations.

Technical Report, University of Illinoissst Chicago,- 1983.

12

Huesmann'

Jastak, J.F.__&Jastak, S. Wide Range Achievement Test. Manual

1g

of Instructions. Revised Editkon. Wilmington, Delaware:

Jastak Aspciates, 197Q.

Jensen; A. Bias in Mental Testing. New York: Free Press, 1980.

,Lefkowitz Rron, L.D4-, Walder,. & Huesmann,

Growing up to be violent: A longitudinal study, of the

development.of aggression. PergaMon press, 19774

Locher,: R., Dishion, T. Early predigtors of male delinquency:

A review. Psychological Bulletin, 1983, 94, 1, 68-99. /Pl.

Lowenstein, L.F. Who is the bully? Home and School,

1977, 11, 3-4.

Olweus, D. Aggression in the Schools. New,..Yorkl Wiley, 1.97

Olweus, D. The stability of aggressive reaction patterns

human males; a review. Psychological Bulletin, 1979,

85,. 852-875. =

Sullivan, E.T., Clark, W.W., & E.W. California Short

Form of Mental Maturity. Los Angeles: California Test.

Bureau, 1957.

Wadsworth, M. RootsOf-delinquebcy: Infancy, adolescence, and

crime. NeWYork: Harper and Row, 1979.

13

Huesmann

Table 1

CCrrelations of Age.8 IQ with Intellectual Competence

at Age 19 and Age 30

Meas.ures of Intellectual

Competence Males

Age 19

Subject IQ age 19

Subject Achievement age 19

Number of Books in House

Age 30

Subject WRAT Verbal .56***

'1Subject WRAT Quantitative .55***

Subject Education .33***

Subject Trailmaking Time -..40***

Subject Race Game Score .25**

+p<.10 *p<.05 **p<,01 ***p<.001

Age 8 IQ

FeMales. Overall

.62*** .58***

.66*** .55***

'.19***

.47*** .51***

.42*** .48***

.c29*** .31***

-.23** -.27***

.28*** .25***.

Table c2. .

Correlations of IntelleAUal2ComPetence ACross Generations

PARENrSINTELL'ECTWAL.COMPEtNCB-. SUBJECTS INTELLECTUAL COMPETENCE

AGE 30? AGE 8 AGE 30

EDUCATIONWRAT 'WRAT

IQ EDUCATION VERBAL' QUANTITATIVE

'ARENT'S INTELLECT/

EDUCATION :.32*** ..31***

/MILD'S INTELLECT

DRAW A,PBRSON IQ.. --- .17/

%WRAT VERBAL --- .17 .31**

WRAT,QUANTITATIVE .16 28.1c1c .48***

*p<,.05 **p<.01 ***p<.001

.28*** .18**

.19 .21+

.29**- .34**.

.45*** .51***

Thrents' education was reverse coded;. so the signs of its correlationshave been reversed for consistency with the 'other variables.

Table 3'

C,,rrelations Between Subject's and Spouse's Intellectual Competence

SPOUSE INTELLECTUAL

SUBJECTS' INTELLECTUAL COMPETENCE

AGE 8 AGE.30

COMPETENCE AT ABOUT WRAT, WRATAGE 30 IQ EDUCATION VERBAL QUANTITATIVE

EDUCATION

WRAT VERBAL

WRAT QUANTITATIVE

.28*** .52*** .27 * ** .24***.

.22** '.36*** . .18* .20*

.26*** .21** .16+

+p<.10 *p<.05 **p<.01 ***p<.001

.11.94111.

hl'w

leman Jam aqq LIM huaqsTsuoD Joj POSJBADJ usaq ;al

suomalmo s4T 30 su2Ts 611 os ipapoo asauu SUM uoTponpa ,squexed

100%4*** 10'>d#* 50'>d* 011X+

....' +L I' .8/Y. 1...., $ MO MI H3H1V3 30 S33NSAILOIRISH

... ' Or* ... **Or" 4i"

... EKON JO SENSAIIONIS311

16L. 101"- **Ll'-' .

***Er- -- **iv- aouoaraa

*ior ***91,e- ***ZZ'- 7f762' .91- ***LE' INDITINDd

S30I1 Veld DNI8V2R QIIHD IN3INd

***61e- ***OZ'

***()E. ***tn.

***OE'

***KI ***EP ***6' ***6r IHV7QdOd GIVNINONJOEd ,

***EP- *Sr- ***6r- N0ISS380DV QRIVNINON'Od

a' ***621 mg?' mos RIM NOIIVOI3I1N3QI

*Sr +?ek 444q* SHIOW HIIM

,

MIME QNV NONVOIEINSQI MHO

9l- wrr

tE14: *Z1' +El' tE1' +1Z' *11.'

o

- ,

IDONDSEd 30 HIDN37'

IDEfiluR

'30 HIHI5 IV SDV SEHION

Q2S0 SIRIONNI HIUIH

maao &DUE(

SHO10V3 HZHIfi

411. 0. ft 11 '- mid JUIN NI MIMED 30 11301QN

**Ll'- .+60'- SDIViS IvNouval000,Sa2103

.**NLE' *1181' '***9e ***62' ***W N01111303 ISINSHVd.

SHOD ovs OIN0N0030I00S

MOEN 9 30V

SHOOS 31,1V0 31411 NOLIVOIKIS, 24IIVITINVilb 1VE:1311. OI bI

30V4 0N125111111E,

30N3I3dit0 Timm' og RDY 61 SO 9 HY

geicleinA Ana 414Tm

OE 0211 qt °Dumdum° TemocaPI siloacqn jo suoTpielm0

t

Table 5

at ioNupt:PPAT,n,no .0Age dr.Aiz r.eltsian.,an ,Ag

With Aggression and IntelleAtual Competence at ,Age 30

I

AGE 30 MEASURES

MALES

'4

AGE AGE 8

AGGRES ION IQ

N R

FEMALES.

AGE 8 AGE 8

AGGRESSION IQ

R

AGGRESSION

' HUI SCALES F+4+9

SPOUSE ABUSE f

PUNISHMENT OF CHILD BY SUBJECT

CRIMINAL JUSTICE CONVICTIONS

SERIOUSNESS OF CRIMINAL ACT

MOVING TRAFFIC VIOLATIONS

- DRIVING WHILE INTOXICATED

SiLF-RATING OF PHYSICAL AGGRESSION

INTELLECTUAL COMPETENCE

WRAT SPELLING

WRAC; READING

WRAT ARITHMETIC

4.1)(,1*p<.a5 **p<41 ,***p<,001

190 .30*** -.19**

88 .27** ONO

63 .24*

335 .24*** -.15**

332 .21*** -,14**

322 .21*** , MI

322 .29***

193 .25***,

(.29*,**)

0

136 -.30*** .54***,

136 -.20* 56***

136 -.20* .55***

a

209 .16*

.20**)

74 (°.

0.0 0

.24** .21**

207

.201

209

OW

Wawa.

IP NI

1 158 -.35*** .44***

158 -.37*** .47***

.35*** .42***158

The correlations in parentheses are those that changed >.03 with a skew-correcting transformation.

19

Table 6

Partial Correlations of Age ;'8 Aggvession andAge 30 Aggression With'

Age 8 IQ Partialed Out

AGE 8 PEER

AGE 30 _,MEASURES MALES

MMPI SCALES F+4+9 .27***

SPOUSE ABIISE .24*

PUNISHMENT QF CHI4D BY SUBJECT .27*

CRIMINAL JUSTICE CONVICTIONS .21***'

SERIOUSNESS OF CRIMINAL' OFFENSES .19***

MOVING TRAFFIC VIOLATIONS .19***

DRIVING WHILE INTOXICATED .30***

SELF-RATING 0 Ft PHYSICAL AGGRESSION .24***A

+p<.10 *p<'.05 **p<.01 ***p<.001

4

NOMINATED AGGRESSION....

7.)

FEMALES

.15*

.19*

.13

Table 7

Regressions Predicting Age 8 Aggression and Intellectual Competence

,PREDICTORS

AGE 8 IQ

STANDARDIZED

REGRESSION COEFFICIENTS

tlAGE 8 AGGRESSION

STANDARDIZED

REhRESSION COEFFICIENTS

BIRTH & SOCIOECONOMIC FACTORS

PARENTS' EDUCATION .38***

FATHER'S OCCUPATIONAL STATUS

NUMBER OF CHILDREN IN FAMILY -.11+

MOTHER'S AGE v BIRTH OF SUBJECT ,19** ,20*** ,20 * **

.35*** .27***

Nam

W011.1 001101

AGE 8 CHILD FACTORS

IQ

AGGRESSION

AGE 8 CHILD REARING FACTORS

CHILD IDENTIFICATION WITH MOTHER

CHILD IDENTIFICATION WITH FATHER

PARENTAL PUNISHMENT OF CHILD

PARENTAL REJECTION OF CHILD

MOTHER'S RESTRICTIVENESS

Multiple Correlation Squared

Degrees of Freedom

F-value

+p<,10 *p<,05 **p<,01 ***p< 001

- ,25 * ** -.17**

.25***

NOW

.165 .223 .283

4,235 5,234 10,229

11.60*** 13.45*** 9.02***

-.15* wwft .11.M

'-.14* -.13+ -.15*

.14* .11+ .10+

-.28*** -.17**

-.16*

WOO

.057 .1232,7

4,235 5,234 10,229

3,58** 6.59*** 7:32***

2

2The signs of the coefficients for parents' education were reversed to reflect its reverse coding.

J

Tablo tf

.

Regro8,ion5 redIcting Ag 30 Aggre88i6 ami tnt8liectual Compotenec

PREDICTORS

4 4 M M 11. 04,

INTELLECTUAL COMPEENC (WA) IAGAG

0

T GRESSION (NNPI F +4 +9)

STANDARDIZED ST NDARDIZED

REGRESSION COEFFICIENTS REGkESSI N CQEFFICIENTS

tL\BIRTH & SOCIOECONOMIC FACTORS

\ PARENTS' EDUCATION

\ FATHER'S OCCUPATIONAL STATUS

\NUMBER OF CHILDREN IN FAMILY

ifOTHER'S AGE AT BIRTH OF SUBJECT

AGE $ CHILD FACTORS

k

IQ's,

AGGRESSION

AGE 8 ILD REARING FACTORS

CHILD NTIFICATION WITH MOTHER

CHILD IDENTIFICATION WITH FATHER

PARENTAL PUNISHMENT 0? CHILD

PARENTAL REJECTION OF CHILD

01.40 04. rrr

1.00 MS

0.11110 OM.

,54 * ** .49*** ,48 ***

NW. MO&

. 00 00. 000

O 0 .00 ..M.

100 000.

WON OW.'

..21*** AO*** .30*** ,34***

ow.

.17+

(MOTHER'S RESTRICTIVENESS -.17*

i!,

MULTIPLE CORRLLATION SQUARED .257 .372 .454

DEGREES OF FREEDOM. 4,101 5,100 10,95

F-VALUE 10033*** 9,90*** 7,18***

24

+p(.10 *p<.05 **p<.01 ***p<.601

.105 .105

4,142 50141

3,32 ** 2.75*

wrw

MOP

.139

10,136

2.00*

The signs of the, coefficients for parents' education were reversed to reflect its reverse coding

40,