Está en la página 1de 10

Parental Discipline and Abuse Potential Affects on Child Depression, Anxiety, and

Attributions
Author(s): Christina M. Rodriguez
Source: Journal of Marriage and Family, Vol. 65, No. 4 (Nov., 2003), pp. 809-817
Published by: National Council on Family RelationsNational Council on Family Relations
Stable URL: http://www.jstor.org/stable/3599892
Accessed: 26/11/2010 10:22

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ncfr.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content 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 support@jstor.org.

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
CHRISTINA
M. RODRIGUEZ University of Utah

ParentalDisciplineandAbusePotentialAffectson Child
Depression,Anxiety,andAttributions

Thecurrentstudyinvestigateddifferencesin chil- treatmenthas received heightened media and


dren's emotionalfunctioningas a productof their scholarly attention.Abuse estimates soared fol-
parents' reporteddisciplinarypracticesand child lowing the implementationof mandatoryreport-
abuse potential. Families with no knownhistory ing laws in the UnitedStatesduringthe 1960sand
of abuse were recruitedto ascertainwhetherde- 1970s (cf. Lindsey, 1994). Moreover,anonymous
pressogenic attributionalstyle and depressiveor surveys demonstratethe virtuallyuniversalprac-
anxious symptomatologywas evidentin children tice of parentalphysicaldisciplineof childrenand
of parentswho used harsherphysicalpunishment even more severe parent-childviolence.
and who had higher abuse potential. Forty-two The media typically spotlightthe most brutal
New Zealand children ages 8-12 participated and fatal instancesof child abuse,neglectingthat
with their parents. Child-reportmeasuresof de- most cases of maltreatment emergefrom the rou-
pression, anxiety, and attributionalstyle were tine practice of physical discipline strategies.
compared with parents' responses on physical Physical abuse often results from an unintended
discipline scenarios and child abuse potential. escalation while administeringphysical punish-
Children'sanxietysymptomswere higherin those ment for perceivedchild transgressions(Herren-
childrenwhoseparentsobtainedhigherabusepo- kohl, Herrenkohl,& Egolf, 1983), and maltreat-
tentialscores and had harsherdisciplinepractic- ment is often conceptualizedas on a continuum
es. Children's depressive symptomsand some with physicaldiscipline(Rodriguez& Sutherland,
components of maladaptive attributionalstyle 1999; Whipple& Richey, 1997). Althoughphys-
were also found infamilies withhigherabusepo- ical disciplineand child abuse can be considered
tential.Resultssuggest emotionaldifficultiessim- variantsof parentalaggression,abusivebehavior
ilar to those of maltreatedchildreneven without
is generallydeemedunacceptable,whereaspublic
identifiableabuse.
opinionregardingcorporalpunishmentis clearly
divided.This debateover the usefulnessof phys-
Since the recognitionin 1962 of the seriousness ical punishmenthas persistedfor generationsde-
of child physical abuse (Kempe, Silverman, spite supportfrom experts (e.g., Straus, 2001a,
Steele, Droegemueller,& Silver,1962), childmal- 2001b) regardingthe needto discontinueall forms
of family violence. In light of calls to end all le-
University of Utah, Departmentof EducationalPsychology, galizedformsof corporalpunishmenttowardchil-
1705 East Campus Center Drive, Room 327, Salt Lake dren (e.g., Hammarberg& Newell, 2002), re-
City, UT 84112-9255 (rodriguez@ed.utah.edu). searchers must address this controversy over
Key Words: child anxiety, child attributional style, child
physical punishmentby compiling furtherevi-
depression, child maltreatment,corporalpunishment,phys-
dence regarding the immediate and long-term
ical discipline. functioning of children receiving any physical

Journalof Marriageand Family65 (November2003): 809-817 809


810 Journal of Marriage and Family

punishment.The risk factorsare consideredsim- the disciplineexperiencedby the child, the more
ilar for both child physical abuse and physical aggressivelythe child behaves (i.e., externalizing
punishment(Straus,1983).Consequently,it is im- behaviorproblems;Weiss,Dodge, Bates,& Pettit,
portantto ascertainwhetherthe effects of physical 1992), althoughthese researchersdid not find a
abuse and physicaldisciplineoverlap. relationshipbetweenphysicaldisciplineandinter-
To date, the evidence indicatesthatchild mal- nalizing behaviorproblems.Anotherstudy sup-
treatmentis characterizedas detrimentalto vic- porteddisciplineas a significantpredictorof be-
tims, resultingin both short-termand long-term havior problems in children (Brenner& Fox,
sequelae in psychosocialand cognitive function- 1998). One dissertationindicatedhigher depres-
ing (e.g., Fantazzo, 1990; Starr& Wolfe, 1991). sion andanxietysymptomsfor thosechildrencat-
Thus physicallyabusedchildrenare more likely egorizedin high disciplinerisk groups(Dingwall,
than nonabused children to demonstratesuch 1997).
characteristicsas oppositionality,behaviorprob- Yet noticeablymoreresearch,as well as public
lems, depression,fearfulness,social withdrawal, discussion,has centeredon the influenceof phys-
and lower self-esteem.For example,abusedchil- ical disciplineleadingto aggressiveand external-
drenwerefoundto be morelikely to displayhigh- izing behaviorproblemsvia social learning.The
er levels of depressivesymptomatologyandhope- preponderance of researchliteratureties corporal
lessnessas well as lowerlevels of self-esteemthan punishmentto aggressivebehavior(e.g., Straus,
comparisongroups (Kazdin, Moser, Colbus, & 2001b), whereasthe associationof physicalpun-
Bell, 1985). Another study confirmedincreased ishmentwith internalizingproblemsremainsless
depression and hopelessness in children with apparentin the literature.Althoughthe connection
abuse historiesas well as a tendencytowardan between physical discipline and aggression is
external locus of control (Allen & Tarnowski, clearlymeaningfulto both family and society,re-
1989). Moreover,a more maladaptive,depresso- searchcontinuesto overlookits associationwith
genic attributionalstyle was reportedfor abused the emotionallife of the child. Anecdotally,we
children (Cerezo & Frias, 1994). Thus children hear of children'ssubjectivereportsof distressat
who are victims of abuse display many signs re- physicaldiscipline(e.g., Willow & Hyder,1998),
flective of internalizingdisorders. although actual empiricalresearchinvestigating
Some indications in the research, however, the child's internal experience of discipline is
pointto negativeoutcomesensuingfromcorporal lacking.
punishmentin the absence of parentalbehaviors Whereasboth child and adolescentboys re-
injuriousenoughto qualifyas abusive(e.g., Straus portedly demonstratemore disruptivebehavior
& Kantor,1994). Childmaltreatment researchhas and externalizingdifficultiescomparedwith girls
often relied on researchdesigns that assess chil- (e.g., Crijnen,Achenbach,& Verhulst,1997),sev-
dren afterthe fact when confirmedabuse has al- eral studies of prepubertalchildrenhave not de-
readyoccurred.Retrospectiveresearchstrategies, tected gender differencesin depressionor attri-
however,are typicallysubjectto recallbiasesand butional style (e.g., Joiner & Wagner, 1995;
errors. It remains unclear whether problematic Thompson,Kaslow, Weiss, & Nolen-Hoeksema,
symptomsmanifestedin some abusedchildrenac- 1998). Indeed,an interestingline of inquiryhas
tually appearin nonclinicalsamples of children begunto investigatethe emergenceof genderdif-
who have receivedless seriousforms of parental ferencesin depressionandanxietyduringadoles-
aggression,such as physical punishment.Theo- cence (Hayward& Sanborn,2002). Thus, com-
retically, some of the emotionaland behavioral paringthe internalizingprocessesfor prepubertal
difficultiesassociatedwith physical child abuse boys andgirls growingup withharshphysicaldis-
may develop primarilyas a functionof parents' ciplinarianswould be meaningful.
harsherdisciplinarystyles and attitudes.Thusthe Some componentsof the internalizingdomain
difficultiesexperiencedby childrenmay differde- mirroringthose drawn from the child maltreat-
pendingon whetherthey grow up with morever- mentliterature(includingdepression,anxiety,and
sus less physicaldiscipline. attributionalstyle) may be particularlyimportant
Researchhas indeed supporteda relationship to investigatein childrenreceivingphysicalpun-
betweenchildhoodhistoryof harshdisciplineand ishment. Although some researchhas included
adultpsychopathology(Holmes& Robins, 1988), symptomsof depressionand anxiety, depresso-
althoughby using retrospectiveresearchdesigns. genic attributionalstyle-a risk factorfor depres-
Anotherstudydemonstratedthatthe more severe sion as conceptualizedby Abramsonand col-
Parental Discipline and Child Functioning 811

leagues (Abramson,Metalsky, & Alloy, 1989; TABLE 1. DEMOGRAPHIC CHARACTERISTICS OF THE


SAMPLE(N = 42)
Abramson, Seligman, & Teasdale, 1978)-has
been relativelyneglectedin researchon the phys- Characteristics % or M n or SD
ical discipline of children. Maladaptiveattribu-
Parentgender
tions, as delineatedin the learned helplessness
model of depression,could potentiallydevelopin Mothers 93% 39
Fathers 7% 3
response to uncontrollablephysical discipline,
which may in turnlead to depressiveor anxious Parentage 38.67 5.43
Child gender
symptomatologyin children.Further,of particular
interestin the currentstudy was the element of Girls 33% 14
internallocus of control,given thatfindingsin the Boys 67% 28
maltreatment literature(Allen & Tarnowski,1989) Child age 10 years, 1 year,
11 months 3 months
suggestthatphysicaldisciplineis likely to be per-
ceived as outside of the child's control.In addi- Ethnicity
tion, researchon attributionalstyle has concen- Pakeha 88.1% 37
Maori 2.4% 1
tratedon children'sexplanationof both positive Other 9.5% 4
and negative events in their lives, with support
Partnered
acrosssamplesthatboth are relevantto children's
Yes 76% 32
depression(Gladstone& Kaslow, 1996; Joiner& No 24% 10
Wagner,1995; Thompsonet al., 1998). Both are No. of children in home 2.81
relevantas well in maltreatedchildren(Kress & 1.06
Annual family income ($)
Vandenberg,1998; Runyon& Kenny,2002).
The currentstudy investigatedseveral symp- <14,999 10% 4
15,000-29,999 34% 14
toms linked to internalizingdimensionsin an at-
30,000-44,999 24% 10
tempt to determinewhethera nonclinicalsample >45,000 32% 13
of childrenreceiving physical disciplinedemon-
stratesdifficultiestypically associatedwith child
abuse victims. Childrenwhose parentshold more
physicallyabusiveattitudesandthose who engage gradelevel receivingconsentforms.Caregiversin
in harsherdiscipline were comparedwith lower this largerstudy were parentswho returnedcon-
risk parents.Familieswith no establishedhistory sent formssenthomefromtheirchild'sclassroom,
of abuse were selectedin orderto assess whether with abouthalf of all distributedforms returned
increaseddepressiveor anxious symptomatology to the school. All familiesin the largerstudywith
was detected in children whose parents had a a child betweenthe ages of 8 and 12 wereinvited
harsher discipline style and abusive attitudes. to participatein the currentstudy.Nearly every
Moreover,to examinehow disciplineattitudesand child in the defined age range was availablefor
practices may relate to how children explain participationat the time of his or her parents'par-
events in their own lives, children'sattributional ticipation.Sampledemographicsappearin Table
style was also examined,specificallytheirexpla- 1, with the sampleinvolvingpredominantly moth-
nations for both positive and negative events as ers of Pakeha(New Zealandersof Europeande-
well as how they internalizeresponsibility,thatis, scent) children.Approximatelyone fourthof the
locus of control. childrenin the samplewere living in single-parent
homes (parentswithouta partner).The obtained
METHOD
sample is comparableto the ethnic distribution
and family composition expected for Dunedin,
and the annualfamily income, in New Zealand
Participants dollars, is comparablewith the nationalincome
Forty-twochildrenand theirparentswere recruit- distributionreportedby the New ZealandCensus
ed fromschoolsin Dunedin,New Zealand,as part (Departmentof StatisticsNew Zealand,1992).
of a largerstudyof parentsnot identifiedas abu-
sive (see Rodriguez& Sutherland,1999);the fo- ParentMeasures
cus was on disciplinebeliefs and abusepotential.
Two randomlyselected schools in Dunedinwere The Child Abuse Potential Inventory (Milner,
approached,with one classroomper appropriate 1986) is a 160-itemself-reportmeasureinvolving
812 Journal of Marriage and Family

attitudesand beliefs believed to be predictiveof child as blameless (e.g., accidentallydropping


physical child abuse potential.One of the most toys in a newly cleanedarea).Whenchildrenare
frequentlyused measuresto screen for physical consideredto be misbehaving,they aremorelike-
abuse, the Child Abuse PotentialInventoryas- ly to be blamed for disciplinedecisions (Muller
sesses parental characteristicsassociated with et al.). Scenarioswerepurposelyconstructedto be
physicallyabusiveparents,althoughthe measure gender-neutral with regardto both the child and
does not tap their specific behavioralresponses adultin the scenario,as this factorcan effect rat-
while deliveringdiscipline.Each item presentsa ings (e.g., Herzberger& Tennen,1985).Two sam-
statementwith which subjectsindicateagreement ple scenariosare as follows:
or disagreement.Of these items, 77 contributeto
the abuse scale score and its six factors,and the * Borderline abusive/nonculpable:A child is
remainingitems serve either as distractors/fillers helping a parentwash dishes. While the child
or providea meansof detectingdistortionbiases. is dryingone of the nice chinaplates,the soapy
With respectto internalconsistencyof the Child dish slips and falls, breakingon the floor.The
Abuse PotentialInventory,correctedsplit-halfre- parent strikes the child several times on the
liabilitywas reportedas .96, and Kuder-Richard- back and buttockswith a belt, sayingthe child
son reliability coefficient was reported as .92 shouldbe more careful.
(Milner).Retestreliabilitiesrangedfrom .91 after * Mild/culpable:A parentis watchingtelevision
I day to .75 after3 months(Milner).As a screen- and the kids are bickeringon a nearby sofa.
ing tool for abusive behavior,studies have sug- Havingalreadyaskedthemto stop fighting,the
gested a correct classificationrate of 89.2% of parent turns around and slaps the children's
confirmedchild abusersand96.3%of controlsub- knees, telling them to stop fighting.
jects (Milner,Gold, & Wimberley,1986).
Parentswere asked to rate each scenarioon a
Discipline Scenarios. To supplementthe Child 7-point Likertscale, reportingon how frequently
Abuse PotentialInventorywith a measurespecif- they use similarphysicalpunishmenton theirown
ically gearedtowarddisciplinebehavior,parents children(Practices),from I (not at all) to 7 (of-
also read 12 briefscenariosdepictingphysicaldis- ten). To generatea total score acrossthe 12 sce-
cipline of a young child 8 years or younger.Sce- narios, subscores on each of the three levels
narioswere developedfor anotherstudy(see Rod- (based on the four scenarios per level) were
riguez& Sutherland,1999),basedon consultation weighted parallelingthe 7-point severity level;
with New Zealandchild protectiveserviceswork- mild scenarioswere weighted1, moderatescenar-
ers, and were designedto vary in termsof three ios weighted4, and borderlineabusivescenarios
specified severity levels of discipline, with two weighted7. In this manner,highertotal scoreson
punishmenttechniquesat each level: mild(slapon practicesrepresentmore severe disciplinebehav-
hand,pokingthe child);moderate(spanking,pull- ior.
ing the child up by the arm);and borderlineabu-
sive (hittingwith an objectsuchas a belt or wood-
en spoon).The finalcategory,borderlineabusive, ChildMeasures
reflecteda severitylevel with perceivedambiguity The Children'sAttributionalStyle Questionnaire
for parentsaboutwhetherthe behaviorwould be (Kaslow,Tannenbaum, & Seligman, 1978; Selig-
abusive although child welfare indicated these man et al., 1984) is a 48-item forced-choicemea-
wouldprobablywarrantfurtherinvestigation.Sce- sure designedto assess attributionalstyle in chil-
narios were intentionallycreatedto portraydis- dren ages 8 to 18 years. Childrenselect one of
cipline techniquesthat would not be obviously two optionsthatbest explainswhy a hypothetical
abusive,whichseveralearlierreportshaveutilized situationin each item would have happenedto
(e.g., Muller,Caldwell,& Hunter,1993), because them. The hypotheticalsituationsvary along the
nearlyall respondentswouldconsiderextremevi- three attributionaldimensionsof internality,sta-
olence (e.g., burning)inappropriate,therebyre- bility, andglobality,with half of the itemsinvolv-
ducing variabilityand increasingsocially desir- ing negativeoutcomesandhalfpositiveoutcomes.
able responses.In addition,half of the scenarios For example, there are eight internal-positive
depicted the children as misbehaving(i.e., per- items andeightinternal-negative items.This mea-
ceived culpability),such as punchinga sick sib- sureyields dimensionalscoresacrosspositiveand
ling, whereasthe othersix scenariosportrayedthe negative events (InternalTotal,Stable Total,and
Parental Discipline and Child Functioning 813

Global Total);a score across all positive events internalizingthanexternalizingbehaviors,and its


for a Positive Total;a score across all negative internalconsistencyis reportedat .82 (Reynolds,
events for a Negative Total; as well as a Total 1982; Reynolds& Richmond,1985).
Composite score calculated by subtractingthe
NegativeTotalscorefromthe PositiveTotalscore. Procedure
Lower Positive Total scores and higherNegative
Total scores correspondto more maladaptiveat- A convenient time for a session in the child's
tributionalstyles; lower InternalTotalscores cor- home was arrangedby telephoneupon receiptof
respondto externalizinglocus of control.The Pos- a consent form. For parents,instructionsand in-
itive and Negative totals indicate attributions dividualitems for both the ChildAbuse Potential
dependingon the valenceof the situation.Forthe Inventoryand the Practicesmeasure were pre-
present study, the Children'sAttributionalStyle sentedon a computerscreen,withthe 12 scenarios
PositiveTotaland NegativeTotalscoreswere ex- appearingin randomorder.Parentsenteredall re-
amined, along with the more specific index for sponses to questionsanonymously,and to further
locus of control,InternalTotal. facilitateprivacy,their responsesdid not appear
With respect to psychometriccharacteristics, on the computerscreenas they enteredthem.The
moderateinternalconsistencyhas been reported computerizedprocedureswere implementedin or-
for the Children'sAttributionalStyle Question- der to maximizeparticipants'reportingaccuracy
naireTotalComposite,Positive Total,and Nega- and to minimize social desirabilityresponding.
tive Total scores (.73, .71, and .66, respectively; While parents completed their portion of the
Seligmanet al., 1984). Temporalstabilityover 6 study,childrenwere takento a quietplace in their
monthsrangesfrom .71 for the PositiveTotaland home to completethe measures.The threeques-
.80 for the Negative Total(Seligmanet al.). The tionnaireswere administeredin a counterbalanced
TotalComposite,PositiveTotal,andNegativeTo- order.Items were read aloud to the child respon-
tal has been correlatedwith indices of depression dents while they read along and wrote their an-
(e.g., Thompsonet al., 1998), consistentwith the swers privatelyon a separateform. After com-
learnedhelplessnessmodel. pleting the forms, the childrenwere given either
The Children'sDepressionInventory(Kovacs, a $2 video rental coupon or a $2 video game
1983, 1985) is a 27-item self-reportmeasure,the voucheras a token of appreciationfor their par-
most widely used instrumentgauging childhood ticipation.
depressivesymptoms,and is suitablefor children
ages 8-17 years. Each item presentsthree state- RESULTS
mentsrepresentinggradedlevels of depressivese-
verity,valuedfrom0 to 2. Highertotal scoresare All analyseswere conductedusing the SPSS 11.0
indicative of more severe depressive symptom- for Windowsstatisticalpackage.Meansand stan-
atology. Kazdin (1990) reportedthat the Chil- darddeviationswerecalculatedfor the totalscores
dren'sDepressionInventoryhas moderatetest-re- of the parentreportmeasures(Child Abuse Po-
test stability, high internal consistency, and tential Inventoryand the Practicesmeasure),as
concurrentvalidity with other depressionmea- well as the child questionnaires(Children'sDe-
sures. High coefficientalphashave been reported pression Inventory,Children'sManifestAnxiety
for both childrenand adolescents,rangingfrom Scale-Revised andChildren'sAttributional Style
.83 to .94 (Saylor,Finch,Spirito,& Bennett,1984; Questionnaire).All obtainedmeanscoreson these
Smucker,Craighead,Craighead,& Green, 1986). measureswere withinnormallimits (see Table2).
The Children'sManifestAnxiety Scale-Re- An examinationof demographicdifferenceson
vised (Reynolds & Richmond,1978, 1985) is a the child reportmeasuresindicatedno significant
37-item self-reportmeasureof anxietysymptoms genderdifferenceson any of the three question-
for childrenages 6-19 years. Each item is pre- naires(all p > .05). Similarly,no significantas-
sented in a yes/no format.The total score, indic- sociationswere found between any of the three
ative of the overall anxiety level, is convertedto child reportmeasuresand child age (all r > .05).
a standardizedT-score that adjustsfor age and Insufficientvariabilityby ethnicitydid not allow
gender differences.Nine items contributeto the analysesfor this variable.Consequently,the full
Lie scale, designedto detectsocial desirabilityre- sample was utilized without any covariatesre-
sponses.The Children'sManifestAnxietyScale-- quiredin orderto examinedifferencesin parental
Revised total score correlatesmore highly with groupdisciplineattitudesand practices.
814 Journal of Marriage and Family

TABLE 2. MEANS AND STANDARD DEVIATIONS FOR TOTAL SAMPLE AND BY GROUPS AND GROUP DIFFERENCES ON
CHILD REPORT MEASURES (N = 42)

Total Sample High CAPI Low CAPI High Practices Low Practices
M (SD) M (SD) M (SD) M (SD) M (SD)

CDI total 9.29 (6.50) 11.86 (6.03) 6.71 (6.03)** 11.38 (7.76) 7.19 (4.14)*
CMAS-R total 50.00 (11.14) 55.24 (8.93) 44.76 (10.81)*** 54.05 (11.83) 45.95 (8.94)**
CASQ Negative total 4.57 (4.49) 8.29 (2.87) 7.71 (2.70) 8.05 (2.60) 7.95 (2.99)
Positive total 12.69 (3.29) 11.76 (3.24) 13.62 (3.15) 12.38 (2.60) 13.00 (3.91)
Internaltotal 7.98 (1.94) 7.29 (1.42) 8.67 (2.17)** 7.90 (1.55) 8.05 (2.31)
CAPI Abuse scale 100.38 (89.67)
Discipline practicesi 67.22 (21.17)
Note: CAPI = Child Abuse PotentialInventory;CDI = Children'sDepressionInventory;CMAS-R = Children'sManifest
Anxiety Scale-Revised; CASQ = Children'sAttributionalStyle Questionnaire.
aMeanscore based on weighted sum across 12 scenarios.
*p? -.05. **'p - .01. ***p <- .001.

Parentswere dividedby mediansplitbasedon variables were significantly different between


theirabusepotentialscores into two groups:high High Practicesand Low Practicesgroupsdid not
Child Abuse PotentialInventoryand low Child findany significantdifferencesdue to attributional
Abuse Potential Inventoryscorers, representing style scores (all p > .05). Analysesindicatedsig-
parentsat high and low risk in termsof their at- nificant differences between the two practices
titudesandbeliefs predictiveof physicallyabusive groups, however, on children's anxiety total
behavior. A multivariateanalysis of variance scores, F(1, 40) = 6.26, p < .01, and on chil-
(MANOVA) was conducted to determine be- dren'sdepressiontotal scores,F(1, 40) = 4.76, p
tween-groupdifferencesfor the following depen- ? .05.
dent variables:Children'sDepressionInventory
total, Children's Manifest Anxiety Scale-Re- DIscussION
vised total, and the Children'sAttributionalStyle
Questionnairedimensionsof interest,PositiveTo- The currentstudyexaminedwhetherparentswith
tal, Negative Total, and InternalTotal. Results more physicallyabusiveattitudesandharsherdis-
from this MANOVAwere significant,F(5, 36) = cipline practiceshad childrenwith more signs of
498.7, p - .001. Subsequentanalyses(see Table internalizingsymptomsthanchildrenof lowerrisk
2 for means)indicatedthatthe two abusepotential parents.Forty-twoparent-childdyadsfroma New
groups differed on the children's anxiety total Zealandsampleof childrenwere assessedin their
scores,F(1, 40) = 11.7,p .001;depressiontotal homes.A nonclinicalsamplewas targetedin order
scores,F(1, 40) = 7.64, p <- .01; andattributional to determinewhetherdifferencesin childhoodde-
style InternalTotalscores, F(1, 40) = 5.94, p : pression,anxiety,and attributionalstyle wouldbe
.01. Additionally,the two Child Abuse Potential evidentin families with no establishedhistoryof
Inventorygroupswere marginallydifferenton the maltreatment.Currentfindingssuggest some in-
attributional style PositiveTotal,F(1, 40) - 3.54, terestingdifferencesin childrenof parentswho
p = .06. No significantdifferencesbetweenabuse use moreseverephysicaldisciplineandholdmore
potentialgroupswere observedon the attribution- attitudessupportiveof physicalabuse.
al style NegativeTotalscores (p > .05). Most consistencywas found in the resultson
Parentswere then also dividedby mediansplit children'sself-reportedanxiety.Childrenof par-
scores on their discipline scenarios practices ents with more physically abusive attitudesand
scores,whichrepresentsa behavioralindicationof beliefs, as measuredby the ChildAbusePotential
actualdisciplinepractices.High Practicesscorers Inventory,reportedmore anxioussymptomatolo-
were those who engaged in harsherdiscipline gy thanchildrenwith parentswho had low scores
practicescomparedwith parentsin the Low Prac- on the ChildAbusePotentialInventory.Similarly,
tices group. A MANOVAexaminingdifferences childrenof parentsusing less severe physicaldis-
between these two groups of parentsacross the ciplinetechniquesreportedloweranxietythandid
six dependentvariableswas significant,F(5, 36) childrenof harsherdisciplinarians.Similarfind-
= 517.0,p - .001. A closerexaminationof which ings of higheranxiety in childrenwere obtained
Parental Discipline and Child Functioning 815

fromhigh disciplineparentgroupsin previousre- ior. Consequently,futureresearchdesigns should


search (Dingwall, 1997). Therefore,the current incorporatemultiple approaches,examiningpar-
findings provide evidence that childrengrowing ents' disciplinarybehavioras well as theirphilo-
up in families with more severe discipline may sophical approach.For example, perhapsdiffer-
experiencemore anxietysymptoms. ences in child depressionbased on the parent's
Similarly,parentswho held more physically Child Abuse PotentialInventoryscore imply that
abusiveattitudesalso hadchildrenwho weremore childrenare more depressedin the context of a
likely to reportdepressivesymptomsthanchildren largerbelief systemadoptedby theirparents,rath-
of parentswith low child abusepotential.Consis- er thanthat theirdepressionis relatedto the par-
tent with this finding,parentswho werepracticing ents' concretebehaviors.Unfortunately,such nu-
moreharshdisciplinealso hadchildrenwithhigh- ances can only be evaluated if future studies
er depressionscores. Such results parallelthose employ comprehensiveassessments of parents'
encounteredin the maltreatmentliterature(e.g., disciplineapproaches.
Kazdinet al., 1985), suggestingthatchildrenev- Futureresearchshould also addresssome of
idence signs of depressionand anxiety,both in- the limitationsof the currentstudy, and should
ternalizingdifficulties,even when physicalabuse especiallyinvolve a larger,moregender-balanced
has not been identified. sample to confirmthe absence of gender differ-
With respect to depressogenic attributional ences in internalizingprocesses,as found in the
style, which is theorizedas a potentialrisk factor currentstudy as well as in earlierresearch(e.g.,
for depression,the findingswere more complex. Thompsonet al., 1998). The currentvolunteer
No differencesemergedon children'sattributional sample likely representsthe most internallymo-
style based on their parents'classificationon se- tivatedfamilies,and thus moreheterogenous,po-
verity of parentalpracticeof discipline.Children tentiallyexternallymotivated,samplesshouldbe
did appearto externalize-that is, feel less in con- sought. Moreover,because these measureswere
trol-if they were growingup with a parentwho designed for North American use, this work
had more physically abusive attitudes.Although shouldbe replicatedwith a U.S. sample,as there
not statisticallysignificant,an interestingtrend may be limitationson the use of such measuresin
was observed:the high risk parentshad children a non-U.S. sample(Rodriguez& Pehi, 1998).
with more maladaptiveattributions,particularly The present sample intentionallyinvolved a
for positive events. Although attributionalstyle group of families not identifiedas abusive. An-
for both positive and negative events has been other researchstep could comparechildrenwho
shown to relate to children's depression (e.g., have neverbeen identifiedas abusedwithchildren
Joiner& Wagner,1995; Thompsonet al., 1998), for whom abuse has been substantiated.Using a
in the currentstudy the maladaptiveattributional suitable comparisonsample and controllingfor
tendencytowardnegative events did not appear demographicvariables,researcherscouldexamine
relevant.The presentfindingssupportthose of an whetherthese childrendifferwith regardto inter-
earlierstudyinvolvingabusedchildren,indicating nalizing problems.Ideally, two sources of infor-
that these children exhibit an externallocus of mationwould againbe utilized,separatingthe in-
control (Allen & Tarnowski,1989). Moreover, dices of child internalizingdifficultiesfromthose
such an externallocus of controlcould contribute assessing parentdiscipline,in orderto minimize
to a sense of powerlessness and hopelessness sourcebias.
(Abramsonet al., 1989). Resultsfrom the current Finally, the currentstudy is limited by diffi-
studysuggestthatwhenfaced with uncontrollable culties intrinsicto its researchdesign.Decadesof
physical discipline, childrenmay be inclined to researchon the associationbetweencorporalpun-
develop maladaptiveexplanatorycognitions.Giv- ishmentandaggressionhave sufferedfromsimilar
en the mixed resultson attributional style, further drawbacksusing cross-sectionalsamples(Straus,
investigation of the of
characteristics locus of con- 2001b). Although some internalizingdifficulties
trol andwhatexplanatoryapproachis relevantfor weredetectedin thoseusingharsherparentingdis-
childrenreceivingphysicalpunishmentwould be cipline practicesin the currentresearch,it is the-
intriguing. oretically possible that parents respond more
The presentfindingsreflect some differences harshlyas a consequenceof their children'sex-
in resultsbetweenparentalattitudesand parental hibiting internalizingproblems,or as a conse-
practices,consistentwith the truismthatattitudes quence of some other,unidentifiedvariable.As
are not necessarilyrepresentative of actualbehav- the investigationof internalizingprocessesin chil-
816 Journal of Marriage and Family

dren experiencingcorporalpunishmentmatures, (1997). Comparisons of problems reported by parents


futureresearchshould incorporatea longitudinal of children in 12 cultures: Total problems, external-
izing, and internalizing. Journal of the American
design, assessing changes in emotionalfunction- Academy of Child and Adolescent Psychiatry, 36,
ing of childrenover time, as well as changes in 1269-1277.
parentaldisciplinein responseto child behavior. Department of Statistics New Zealand. (1992). 1991
New Zealand Census. Te Tari Tatua, Wellington, New
Obviously, such designs are burdensomefor a Zealand: Author.
numberof reasons, althoughultimatelysuch an
Dingwall, J. A. (1997). Perception of facial expressions
avenuecould respondto the realitiesof how pa- of emotion by children with severe, moderate and low
rentalphysicaldisciplineevolves. levels of physical discipline. Unpublished doctoral
Nonetheless,the results of the present study dissertation, California School of Professional Psy-
chology, San Diego.
provide preliminary indications that parents' Fantazzo, J. W. (1990). Behavioral treatment of the vic-
physicalpunishmentrelatesto some aspectsof a tims of child abuse and neglect. Behavior Modifica-
child's emotionalwell-beingeven withoutidenti- tion, 14, 316-339.
fied physicalabuse.Hencethe currentfindingsex- Gladstone, T. R. G., & Kaslow, N. J. (1996). Depression
tend the researchon corporalpunishmentinto the and attributions in children and adolescents: A meta-
realmof internalizingdifficultiesbeyondwhathas analytic review. Journal of Abnormal Child Psychol-
ogy, 23, 597-606.
alreadybeen implicatedfor childrenlearningto Hammarberg, T., & Newell, P. (2002, July). Ending le-
become violent adults. Longitudinalresearchin gal violence to children: Accelerating the end of cor-
the future may reveal that nonabusivephysical poral punishment. Paper presented at the Internation-
al Congress on Child Abuse and Neglect, Denver,
disciplinemay adverselyinfluencethe short-and Colorado.
long-termemotionalfunctioningof children;so- Hayward, C., & Sanborn, K. (2002). Puberty and the
phisticated research may even pinpoint which emergence of gender differences in psychopathology.
symptomsof psychopathologyappearonly in re- Journal of Adolescent Health, 30, 49-58.
sponse to physicalchild abuse incidents.Contin- Herrenkohl, R. C., Herrenkohl, E. C., & Egolf, B. P.
ued researchon parents'use of physicaldiscipline (1983). Circumstances surrounding the occurrence of
child maltreatment. Journal of Consulting and Clin-
has implicationsfor moreseriousformsof family ical Psychology, 51, 424-431.
violence, given that corporalpunishmentis often Herzberger, S. D., & Tennen, H. (1985). "Snips and
a precursorfor child abuse.Futureworkclarifying snails and puppy dog tails": Gender of agent, recip-
these issues would provide guidance to profes- ient, and observer as determinants of perceptions of
sionals and parentsalike as we continuethe de- discipline. Sex Roles, 12, 853-865.
Holmes, S. J., & Robins, L. N. (1988). The role of pa-
bate over the use of corporalpunishmentin the rental disciplinary practices in the development of de-
home. pression and alcoholism. Psychiatry, 51, 24-35.
Joiner, T. E., & Wagner,K. D. (1995). Attributional style
and depression in children and adolescents: A meta-
NOTE
analytic review. Clinical Psychology Review, 15,
Thanks are extended to the Otago Children and Young 777-798.
Persons Service of Dunedin, New Zealand, for their in- Kaslow, N. J., Tannenbaum, R. L., & Seligman, M. E.
valuable assistance in constructing the scenarios used in P. (1978). The KASTAN-R:A Children's Attributional
this study. Style Questionnaire (KASTAN-R-CASQ). Unpub-
lished manuscript, Department of Psychology, Uni-
REFERENCES versity of Pennsylvania, Philadelphia.
Kazdin, A. E. (1990). Childhood depression. Journal of
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Child Psychology and Psychiatry, 31, 121-160.
Hopelessness depression: A theory-based subtype of Kazdin, A. E., Moser, J., Colbus, D., & Bell, R. (1985).
depression. Psychological Review, 96, 358-372. Depressive symptoms among physically abused and
Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. psychiatrically disturbed children. Journal qf Abnor-
(1978). Learned helplessness in humans: Critique and mal Psychology, 94, 298-307.
reformulation. Journal of Abnormal Psychology, 87, Kempe, C. H., Silverman, E N., Steele, B. E, Droege-
49-74. mueller, W., & Silver, H. K. (1962). The battered
Allen, D. M., & Tarnowski, K. J. (1989). Depressive child syndrome. Journal of the American Medical As-
characteristics of physically abused children. Journal sociation, 181, 17-24.
of Abnormal Child Psychology, 17, 1-11. Kovacs, M. (1983). The Children's Depression Inven-
Brenner, V., & Fox, R. A. (1998). Parental discipline tory': A self-rated depression scale for school-aged
and behavior problems in young children. Journal of youngsters. Unpublished manuscript, University of
Genetic Psychology, 159, 251-256. Pittsburgh School of Medicine, Pittsburgh, PA.
Cerezo, M. A., & Frias, D. (1994). Emotional and cog- Kovacs, M. (1985). The Children's Depression Inven-
nitive adjustment in abused children. Child Abuse and tory (CDI). Psychopharmacology Bulletin, 21, 995-
Neglect, 18, 923-932. 998.
Crijnen, A. A. M., Achenbach, T. M., & Verhulst, E C. Kress, E, & Vandenberg, B. (1998). Depression and at-
Parental Discipline and Child Functioning 817

tributionin abused childrenand their nonoffending Seligman,M. E. P., Peterson,C., Kaslow,N. J., Tanen-
caregivers. Psychological Reports, 83, 1285-1286. baum,R. L., Alloy, L. B., & Abramson,L. Y. (1984).
Lindsey,D. (1994). Mandatedreportingandchildabuse Attributionalstyle and depressivesymptomsamong
fatalities:Requirementsfor a system to protectchil- children. Journal of Abnormal Psychology, 93, 235-
dren. Social Work Research, 18, 41-54. 238.
Milner, J. S. (1986). The Child Abuse Potential Inven- Smucker,M. R., Craighead,W.E., Craighead,L. W., &
tory:Manual(2nd ed.). Webster,NC: Psyctec. Green, B. J. (1986). Normativeand reliabilitydata
Milner,J. S., Gold, R. G., & Wimberley,R. C. (1986). for the Children'sDepressionInventory.Journalof
Predictivevalidity of the Child Abuse PotentialIn- Abnormal Child Psychology, 14, 25-40.
ventory. Journal of Consulting and Clinical Psychol- Starr,R. H., Jr.,& Wolfe, D. A. (1991). The effects of
ogy, 54, 865-866. child abuse and neglect: Issues and research. New
Muller,R. T., Caldwell,R. A., & Hunter,J. E. (1993). York:GuilfordPress.
Child provocativenessand genderas factorscontrib- Straus,M. A. (1983). Ordinaryviolence, child abuse,
and wife beating:Whatdo they have in common?In
uting to the blaming of victims of physical child D. Finkelhor,R. J. Gelles, G. T. Hotaling,& M. A.
abuse. Child Abuse and Neglect, 17, 249-260.
Straus (Eds.), The dark side offamilies: Currentfam-
Reynolds,C. R. (1982). Convergentand divergentva-
lidity of the Revised Children'sManifest Anxiety ily violence research (pp. 213-234). Newbury Park,
Scale. Educational and Psychological Measurement, CA: Sage.
42, 1205-1211. Straus, M. A. (2001a). Beating the devil out of them:
Corporal punishment in American families and its ef-
Reynolds,C. R., & Richmond,B. 0. (1978). What I
thinkand feel: A revisedmeasureof children'sman- fects on children.New Brunswick,NJ: Transaction.
ifest anxiety. Journal of Abnormal Child Psychology,
Straus,M. A. (2001b).New evidencefor the benefitsof
never spanking.Society,38, 52-60.
6, 271-280.
Straus,M. A., & Kantor,G. K. (1994). Corporalpun-
Reynolds,C. R., & Richmond,B. 0. (1985). Revised ishmentof adolescentsby parents:A risk factor in
Children's Manifest Anxiety Scale (CMAS) Manual.
Los Angeles:WesternPsychologicalServices. epidemiologyof depression,suicide, alcohol abuse,
child abuse,and wife beating.Adolescence,29, 543-
Rodriguez,C. M., & Pehi, P. (1998). Depression,anx- 561.
style in a New Zealandsample
iety, and attributional Thompson,M., Kaslow, N. J., Weiss, B., & Nolen-
of children. New Zealand Journal of Psychology, 27, Hoeksema,S. (1998). Children'sAttributionalStyle
28-34. Questionnaire-Revised:Psychometricexamination.
Rodriguez,C. M., & Sutherland,D. (1999). Predictors Psychological Assessment, 10, 166-170.
of parents' physical disciplinary practices. Child Weiss, B., Dodge, K. A., Bates, J. E., & Pettit, G. S.
Abuse and Neglect, 23, 651-657. (1992). Some consequencesof earlyharshdiscipline:
Runyon,M. K., & Kenny,M. C. (2002). Relationship Child aggressionand maladaptivesocial information
style, depression,andposttraumadis-
of attributional processing style. Child Development, 63, 1321-1335.
tressamongchildrenwho sufferedphysicalor sexual Whipple,E. E., & Richey, C. A. (1997). Crossingthe
abuse. Child Maltreatment, 7, 254-264. line from physical discipline to child abuse: How
Saylor,C. E, Finch, A. J., Spirito,A., & Bennett,B. much is too much? Child Abuse and Neglect, 5, 431-
(1984). The Children'sDepressionInventory:A sys- 444.
tematicevaluationof psychometricproperties.Jour- Willow, C., & Hyder,T. (1998). It hurts you inside:
nal of Consulting and Clinical Psychology, 52, 955- Children talking about smacking. England: National
967. Children'sBureauEnterprises.