This article was downloaded by: [New York University]

On: 18 April 2015, At: 09:16

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Clinical Child Psychology

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/hcap19

Development of a parenting alliance inventory

Richard R. Abidin & Jack F. Brunner
Published online: 07 Jun 2010.

To cite this article: Richard R. Abidin & Jack F. Brunner (1995) Development of a parenting alliance inventory, Journal of
Clinical Child Psychology, 24:1, 31-40, DOI: 10.1207/s15374424jccp2401_4

To link to this article: http://dx.doi.org/10.1207/s15374424jccp2401_4

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Ctin~ca!Child Psychology Copyeght 1995 by
1995, VoI. 24, No. 1 . 31-40 Lawrence Erlbagm Associates. tnc.

Development of a Parenting Alliance Inventory

Richard R.Abidin and Jack F, Brunner
o f Virginin
t!rri~~ersic

Reports the iriifiaifacforstrucrure, reiiabiii~,ar~nndpreiiminan.v a t i d i ~afihe Parent-

ing Alliance In1~entor.v( P A / ; .4birlin. 1'988).Tlre 20-item PA41assesses the degree ro
which parents believe that they h a w o soand workirrg reiariotrship wrdi rheir chiid's
other parent. Our sample consisled of512parenfs(321 mothers and I F Ifarhersj and
78 reacherxlchiid care providers. The P.41proved ro have high irrtenlal consisrerzcy
arid showed no significonr d€fferencebehvee~irhe resporrses qf merz and tsomen. The
PA41discriminated accumiely between couples who were nmrried, separate& ar;rl
dh~orced,In n nzanrler corzsistertr with its hypothesized naiure. the PA1 correlated
signzjicarerly with established measures of rnaritai sati~facrion,parenzirrg stress, and
Downloaded by [New York University] at 09:16 18 April 2015

parenting style. hi addition, the FAI correiared sign[ficant!\. wirh memures of the
target chiid'x positive adjusrmenr and sociai comperence. Aithough smrisricnily sip-
niflcarrt, the order of magnitude q f fhese reiationx are s m i i , arrd tiir results need to
be interpreted wirh caution.

This article describes the development of the Parent- issue of parenting. the] can continue to nurture thelr
ing Alliance fnventorj (FAI, .4bldln, 1988 r The P.M children after a ditorce or during a conflictual mar-
is a 20-~temself-report ialstrument that measures the nage. The case studles described b] W'eissman and
pnrentrng alliance. a concept created by Vi'elssman and Cohen suggest rhat a posithe parenting alliance tan
Cohen 11985) to describe the part of the manta! re!a- mitigate the harm of divorce. The measurement of the
tionship that rs concerned with parenthood and chiid parentnng alliance is. therefore. Important to rhe extent
rearing A sound parent~ngalliai~ce1s established if that .t relates to parents' abilitj to cooperate w ~ t heach
these conditions are met: "(a) each parent is invested Jn other bl nurturrnp and meeting the de\elopmeniaI
the child, (bj each parent values the other parent's needs of the child Such ameasure would ha\.e potential
involvement with the chiid, dc) each parent respects the udlit) in joint custod) decisrons. clicical work ulth
judgments of the other parent, and (d) each parent families. arrd research lnvestigatmp modeis that de-
desires to commun~cates i t h the other" t Mreissman 8r scribe the determinants of parenting behavior
Cohen. 1985, p. 25). Thus, the parenting alliance is The asscxiatlon between manta1 discord in the fam-
separate from the romantlc or sexual aspects of map- ilg and the deveIopment of behavioral problems in
riage and measures rhe degree of cornmftrnent and children has been suggested by a number of authors
cooperation between the husband and uife in chhd (Emery. I952: Gartiand & Day. 1992, Jour:Tes, Mu:-
reanng In contrast to other measures of marital sat:s- pP.1. & Fams. I991 1 Others hake demonstrated that
faction and addiustment(Locke & Wallace, 1959: Span- manis1 discord ns related to depression. fears and inse-
ler & Thompson. f 9821. the P.41 is a narrow-band cunty in the child and to long-tern1 negative effects.
measure of the type recommended by HoIden and Ed- including p m r academic performance and cf~fficulties
wards ( 1989) for second-generatnon family measures wlth the opposite sex (Dadds &L Powell. 1991. Katz &
Given our society's iarge number of divorced Gottman. 1933: Walierstein & Kell! , 1980 1. In the past
blended, arid intact famiiaes. measurement of the par- decade. howel er. the srrength of the relati~nbetu een
entlng alllance is potentlaHy usefur becau~e~t IS lrketj marria! dlscord and child behaylor problems has been
to relate to the coupie's success as parents versus the questioned Cslng a nonclrnrcai sample. E m e ~and
success of a couple. \Veissmar, and Cohen (1985) sue- O'Leary (1984) found that. although there u a s an as-
gested that. ~fparents have a strong alllance around the sociation between manta1 discord and chiid behavior
probiems. the re!atlon was fairl! weak Emerq and
We express our nppreciatron to Paticia Poizlen and Patsy Dass O'LearySs results are consnstent uith other studies that
f ~ their
r assistance in the collection and analysis of the data. used nonclin~calpoputatjons and found the relation
We are listed alphabeticdiy
Requests for reprints should be sent ia Richard R Ahidin. Curry
between marital discord and chiid behasior problems
Programs i r Clinical and School Psychology, Uni\.crsi:y of Virginiz. to be weak (Emen. Vierntraub, & IL'eaIe, 1982. \Yhi!-
405 Emmet Street, Chariottes\:ille. V A 12923-2495 taker Sr BQ, 1991)
 ABIDEN & BRUNNER

One explanation of chis difference between clinical tua4 marrage exists are less Likely to have anpaired
and noncI:n~caIpopulations may be that marrtai drscord seEf-esteem. be torn between parents, or have negative
is most harmful when ~trs one of many siressors in the role modeis regarding ccmtlact resolution.
family, as may be the case In the clrnical samples. If The PM is offered as a more direct measure of a key
marital discord occurs in a familq that 1s function~ng component of a couple's shared commitment and com-
welt otherwise, the effects of marital discord may be municatlon regarding child rearlng. We hypothesized
weaker. The combination of several family stressors rs that thrs measure is more meaningful and likeIy to be
the most damoglng (Red B Cr~safu'uEIl,1990, Rutter, more directIy related to parenting behaviors because
1978). -4 posin~yeparentrcg alliance in a family with the focus is on bothparents' responses to their parenting
wrnrltal discord wouBd reduce the risk effect of the interactions, not on more remote aspects of the maritd
marital drscord. Another way of understanding the re- relationship We theorized that the common n m w e d
lation between mairai d~scsrcland chiIdhood malad- faerrs of :he questionnaire will reduce the *'noisem'
from
justment is that marital conflict in general 1s nor the other related and unrelated marital variables that ccln
problem, rather. 11 is pathologreal parental anteractions confound the relaclon between parenting bel~efs.par-
(e g., inconsrstent parentai discipl~ne)that create prob- enting behaviors. and chijd out-come.
lems (Becker. 1964: Jenkins B Smith, I99@).With
regard to this possrbility. Block, Block, and Morrison
f E 98 I :found that a disagrzerneamt among parents over
childrearrng was predictive of chlld behavlor problems
Downloaded by [New York University] at 09:16 18 April 2015

in school. More recently. Fauber, Forehand, Thomas.

and Wierson (1990) asserted that manta1 conflrct is not The total sample consisted of 5 12 parents: 52 I moth-
the primar? cause of childhood Eaiadjustmect. ri rs the ers and 191 fathers. Seventy-eight teachers/ehild care
disruption of parentinsg caused by the marital drscord providers prov~dedcriterion validit: data The total
that is responsible sarnpie was collected in rwa waves. The first sampie
Despite Fauber el ai,'s (19901 siudj. other research- cocs~stedof 99 mothers, 61 fathers. and 78 teachers1
ers continue to beiieve that marital conflict does hahe child care providers. AH 61 fathers in the firs[ sample
a direct effect on child adjustnnent through factors such had their child's other parenr in the sample. For the 99
as creating a high tevel of tenslon ln the home or mothers. 6 I had partners in the sample, 16 had partners
causing the child to internalize cenaia~nodeled hus- who dndn't participate, and 22 were ~ ~ i i h opartners
ut in
band and wrfe te~teractions(KIlne, Johnston. & the home The second sample consisted of 222 nothers
Tschann. 1991) Cox, Owen, Lewis, and Henderson and 130 farhers. The parents were recrujted from child
:I9891 stated that, even when differences In psycholag- care centers located In mrd and urban areas of Virginia,
~ c a ladjustment were %ken ~ n i oaccount, mothers are a group peb~atricpractice, the Ronald h4cDonald
wlamer and more sensitive .n ~ t htheir infants and fa- House. and a public recrearlonai facility The sample
thers hold more positrve attitudes tcmx-3 their infants represents a wide ranage of racial and socioeconomic
and their robes as parents when they are in close:con- ISES) backgronnds. Nineteen percent sf the sample
fiding marriages. EIowes and Markman 1,1989; ex- consisted of minority parents, Including African
tended ehrs conciusion and asserted that rnaritai quality Americans, Hispanics. and Asran t2merncans. A w9de
borh before and after the ckrid.~birth are h ~ h l related
y range of total fami!y incomes was represented, mith
to the chird's Pinnctioning-speclficaIiy, to the child's 2 I % of the farniiles earning beHo~*$20.000, 17% e m -
patterns of attachment and sociability. The paths of ~ n gbetween %20,000and $29,999, 27% e m m g be-
influence of parent51 behaylor on the chrld 3re likely to tweeil $30.000 and $39,999, 18% earning between
be complex enough to include both i:nes of reasonang. [email protected] $49,999.9% earnrng berweei~650,000and
As Goidberg and Easterbrooks (19841 suggested. the $59,999 and 7% earnlwg $60.000 07 more. The median
paths of influence between n~wtdqualrty. parenting, family income for Virginia was $34,WQ. The derno-
and chiid outcome are likeiy to be conpHex. bidirec- g r q h c s of the sample are presented in Tatsle I and
tronai, and circular :ndicate that the ninorlty scores are inadequate to draw
The relative absence of negative effects on children canclusions about those populat~ons.
ir! noncllnacai sampies may be the result of the bufferrng We drd not record :he refusals of subjects ap-
that occurs due to the qualit? of the parenaneg aBI:ance. proached, thus, refusal rates are unavailable. Compll-
It is reasonable to assume that parents nho are commit- ance after agreemen! to serve as a subject bas 89% for
ted to rnaintairning a posit:\ e parentrng alliance, despite women and 91% for men.
being in conflicted and unsatrsfying marriages, will
have chiIdren who are better adj;justed than chiidren
from similar famiiies in which a weak or neganke
parenting alilaance exists. In the presence of a positwe Wave f parents were contacted t,kough a letter sent
parenting alllance children in families where a conflic- to those parents whose children attended the cooperat-
32
 THE P&RENTING 4Ll-IANCE I N V E N T O R j
Table 1, Farnil? Demographic Srar~stzcs The packets rncluded t h e e sets of questionnazres
one for the mother. one for the father, and one for the
cia\ a r e pro\~der/tedcher The mother and father sets
of quest~onnaireswere identical and consisted of the
Sex of Target Chiid foilowrng Instruments a demographic sheet, the hfsr-
~Wde
Femaie kowe-Crow ne Social Desirab~I~t> Scale (M-CSDS
Age oi Child Crow ne & Marlowe, f 9501. the Plil (ftbidm, 1988'1.the
Parental Attitudes F o u a r d Chtldrearing Scale
(PATCS, Goldberg tit Easterhrooks. 1984). the Parent-
~ n Stress
g Index (PSI, .LLbndin, 1983:. the Earl3 School
Behavior Scale t ESBS, Caldwell & Pranta. 199I!, and
theReblsed Manta1 Adjustment Test IRMAT. KimmeI
Agz of Mother & Van Der Veen. "74\
20-29
The teacherkhrid care probiders' packet cons~stedof
30-33
40- rhe ESBS (Caidwell (4i Piaarra. 1991 I. the B e h a ~ l o r
Age of Fainei A~ademicSelf-Esteem Scale iBASES, Coopersmith &
20-29 Gilberts. I982 1. and the Cahforcla Preschool Social
30-39
Cornpetencq Scale ICPSCS: Le\ me. E?ze\. & Leu 1s.
30 +
1969: Wave I of data collectror! produced a sample of
Downloaded by [New York University] at 09:16 18 April 2015

Race of mother
African American 99 mothers. 61 fathers. an3 78 teachers/chrld care pro-
Hispanic \ iders. The target child n as betn een 4 and 6\. ears
As~anAmerican
To ensure adequate sample slze an conduct the factor
Caucasian
Empioymenr !Fuli-Time and rel~abilityanalyses of the first phase of the data
Mother analysis. addntlonal subjects ( W a . ~ c21 iiere recm~ted
Yes froir. a public recrearrondl facilit) and the outpatient
No pedia:ric clinic of the Un~~e:slt:vof Virginia Medical
Father
Yes
Center The sample consisted of 222 mothers and :30
No fathers. These subjects completed a srnalier set of ques-
Fam!y mcome tio~~naires consisting of a demographic sheet. the PA1
$0-$20.093 Abldin. i988i. 2nd Questions I and I4 taken from the
320.WXJ-$24.999
RMAT (Kirnrriet & \'an Der Veen. 197.1) Cross and
$30.0.31,-$39.993
S40,03€~-S-r9,993 Sharpley t 198Ii showed that these tvro questions pre-
950 OtW-$59,999 d ~ c tthe results of the RhIAT as we11 as the scale as a
$60,3204 \\hoie, thus they serve as a suitable sutsstltute for the
Mother's Eaucatlonai Levei
entre test. For Wa\e 2 subjectb. the corre;dt:on be-
Grades ! Through E
Grades 9 Through 12 tween the sum of Quest~ons1 and :4 and the rota1
\'ocatronai'Some Coiiege RM.4T score u a s 89 KO star~st~rally signiGcanr dif-
College Graduate ferences were f o u n d hetneen W a \ e I anct U'a5e 2
Graduate 'Profess~onalSchoo samples I n etcher the demagraph~s\ariaDles or the tesr
Fatheis Educanonal Level
Grades 1 Through 5
data
Grades 9 Through 12
\.xational:Sorne Coilege
College Graduate
Graduate Professional Schw

The P.41, developed h j Abidln 11988), consisted of

ing agencies f i e letterexplained the purpose nfthe study an initial pool of 80 items written bj farnil) therapists
and asked the parents for their parilcipation. Particrpa- and psychologists, uslnp the four drmenslon, of a sound
tion in the studj was voluntap and involved answering parenting allrance cWeissman tit Cohen. 19851 as a
zt packet of questionnaires that wouid require approxi-
gu~de.The items were submitted to a panek of fve
matel) 1 ih, hr to complete. In addition, the parents were farnll! therap~stsand five clrnical ps>choiog~siswho
asked to select an individual outbide the farnll! who are in\ crlved in farn11) research for e l ajuation First. the
observed their child on a regular basis and M ho would panel rnembers rated each ltern to deten~lrneu hlih of
also esaluate the child Such an indl~iduatcould be a the four dimension\ of the parenting aI?ianze t h e be-
teacher or chiid care pro\ ider The parents were asked Ileved the item described. Second. thel rated each itern
to give these rndividuals apacket of questionnaires and on a 5-point Likeri scale ranging from I idcic*.sn' f niea-
a signed form, givnng permission to the child care sirre dln:etzsrrri ) to 5 isrxoriai) nrecsures dtr?ensio~t!to
provader or teacher to complete the questionnaire. deterrnlne hovl weil the!, belle! ed the ntem measured
33
Downloaded by [New York University] at 09:16 18 April 2015
 THE P4RENTING .4tLIA?if E IN\-EYTORY

Data AnaIysis were avarlable The product-moment correlations of

the 61 couples' scores indicated that ihe PA1 scores of
There were tarophases of data analysls rn thrs stud: this nonci~nacalsample were moderate!! cerrelated ( r
Phase 1 was designed to establish the statistical charac- =.SO.p = 011.
teristncs ofthe FA1 (Abid~n,1958 1 Phase 2 ~c\,olvedan
investigation of preliminq concurrent and construct
validity data for [he PAT In the form of correlations Cornparison of the PA1 and tile R3Z4T
between the P N and otherrerated measures used In the
study. To examine the relataon of the parenting ali~ance
to that of marital adjustment. conelatlons were run
between the PA1 and she M4.49 for both n~orhersand
Results fathers. For both mothers ( n = 93 1 2nd fathers fr; =
61). PAX scores and RMAT scores uere significant1)
correlated r r = 20. p = .05, and r = 44. p = .Xlk.
Phase 1-The Statistical respective!) ? Hower er. the relatlon for fathers was
Characteristicsof the PAI stronger Glven rhls degree of relation betweer. the
PAX and the RMAT. correlations were run to corn-
The PAI responses of 321 mothers and 191 fathers pare their respective concurrent vaiid~ryIn relat~on
(,V = 51 2 parents) were used :n Phase I analyses. A to the criterion \ arrables.
Downloaded by [New York University] at 09:16 18 April 2015

significant mean difference ( r test, p < .05\ was found SlgniGcant negatlve conelatlons between the P.41
between mothers' and fathers' responses to the P.41. and the F34AT scores u ere found ir: relation to the total
Each data set was subsequently factor analyzed by sex stress score of the PSI. The three ~nterpersonalsub-
using n principal factor analysis with varimax rotations scales of the PSI and their correlations with the P.4I and
Both soIutions suggested a two-dimensional scale that RhitAT ,are reported in Table 5 These PSI subscales
can be reduced to the same 20 items. Each of these items
has a factor loading of .50 or greater on their respectwe
scales (see Table 2). The reduction in the length of the Table 2. Resulrs ojFactor Arzd~sisqf P,.if b! SEXqf
scde was done for practical purposes and to eliminate Parent
the few items with very low factor loadings. The factor
analysls of the 28-item scale accounted for 66.3% and Mother Father
64.3%of the variance for the mother and father data,
respectively. The shortened PtU displayed high inter- Elgenvaiue
Factor 1 i? 0 I! 5
nal consistency, as indicated by an alpha reliablliiy of Factor 2 1; 14
34. Percentage of Yanmce
Significant differences were found among the PAT Factor 3 600 5- 5
responses of married. separated, single, and divorced Factor 1 63 68
Tord 60 3 6-43
wornear Of the four marital groups, women's mean
scores followed a marked 2nd distinct progression Mother Father
Mamed women have the highest mean score on the
P.&l Item Factor t Factor 2 Factor 1 Factor Z
PAI. followed by separated women. then single
women. and finally divorced women. who have the
lowest mean score. A comparable analysis of men's
responses could not be adequate13 conducted because
there was an insufficient number of men in each of the
categories of separated, single. and divorced. A s i c i f -
acant mean difference was found betweel: married rnen
and the psoBed data of the other three men's groups (see
Tabie 3).

Phase %Concurrent and Construct

Validity

The responses of 99 mothers and 61 fathers frorr.

families obtained in Wave 1 of data colIecihon were
used during Phase 2. These were all cases in which
ieachers/ch:ld care providers' dependent variaOIe data
 ABIDlN & BRLTNNER

Table 3, i Tesr and Planned Comparisons of PA] Score5 b~ hdarital Status for Men and Women

Womena iUellb

Diuorced;
Singlei
hiarried Separated Married Separated

Marital Status M SD LM SD .4lpha M SD M S Alpha

14.0 I?.! 57.9 17.6 .XI] 86.3 9,0 70.1 15.0 .Ol
Separated Versus Singie 05.4 17.6 50.7 21.9 "06 -C
Singie Versus Divorced 60.9 21.9 52.4 16.4 "08 -C
Married Versus Single 84.0 3 60.7 21.9 .001 -C
Marr~edVersus Divorced 84.0 13.1 52.4 16.4 .W1 -<
Separated Versus Divorced 65.9 15.6 52.4. 16.4 ,001 -. L

-
"Married, c = 221; separatec?., n = 32: singie, r; 34; divorced. rt = 34 ( N :: 321\.
' ~ a r r i e d . ii = 172; separated, n = 9: singie. r; = 3; divorced. n = 7 {N = 191).
'inadequate sarnpie sizes for cornpansox.

specifically address the relationship berween the parent correlated slgntficantly with heir RMAT and P.41
Downloaded by [New York University] at 09:16 18 April 2015

and the child; thus, they are germane to this study, scores.
which examines paents' social competence in the mar- Another PSI subscale consndered rs the Child Re-
ital and pwental relationship as ir relates to the chiId's inforces Parent subscale, which measures how much
social relationships and social competence. The the parent experiences the chiid as a source of posi-
parentsv marital relationship and their parenting aili- tile reinforcernenr. For both mothers and fzthers.
ance represent modeIs of behavior for their children. Er thex Chdd Reinforces Parent subscaie scores corre-
is reasonzble to assume tha.: these ineeracaons ~vil! laced h~gnificant!) @ = "01) with therr PAX scores.
influence their children's social functioning. However. these scores were not significantly corre-
Both the RMAT and the PA1 woeld be expected to Iared b3th ~helrRMAT scores, which supports the
measure a construct similar to the Relationship With independence of the P.41 construct from that of over-
Spouse PSI subscale. As expected. both the mothers' al1 marital adjustment.
and fathers' Relationeship With Spouse slrbscale scores The final PSIre!at,ionshlip-related subscake examined
was Parent Attachment, which measures the degree of
artachmeni in the parent-child interaction from the
Table 4. Correiattsris Bemeen the PAI, the RMAT, parents' perspective The fathers' Parent ~~t:achrnenl
and the Crilerion L'oriables subscale score correlated significantly with both the
fathers' ChrEdRernforces Parent subscaIe score Cr = .32.
Mother Father p = "01) and his Relationship With Spouse subscale
score jr = -.23, p = 05). Note, hswrever, the mothers'
Criterion PAIa WAT8 ?Alb RMAT~ Parent Attachment subscale score failed to correlate
-

PA1 - .20* - ;t;*** significantly eith either her PAI or M A T scores,

PSI suggesting greater independence of the morher-ehild
PSITOP - .&
?fr** -. ,2~** - ,29** - ,2;* relaaonshlp from the marital relationship and t h par-
RFPA - .-?d**
. - .09 - .-?$** - ,19
ATTMT -.I3 - .09 - 32** -,20
entlng alliance.
RELNT - ,45*** - S9*** -.39*** -,44***
PATCS
WARM .&
38** .04 3gw* :7 P.41 and RMAT Relations to the
STRlCT -.I? - .06 .I 4 .U1 PATCS
CI-0NFL.T - .I8* - .0: - .?4* - .Oi
Noic: Sei.ent:-eigh: percent of mothers and 1M!% of fathers were Both fathers' and mothers' warn. parenting style
married. PA1 = Parenting Alliance Inventoy; RMAT = Revised
l Authoritative subscale of the PATCS) co~eIatedslg-
Marabi Adjustment Tes:; PSI = Parenting Stress Index; PS:TOT
= Parenting Stress lndex tntai score; RFPA = Child Reinforces nificantly w t h the P-ml scores ( r = "58.p = "001:r =
Paren: {PSIsubscale): ATTKT = Parent Atracheni (PSI subscaie): .28, p = "1301). respectivelj. In contrast. neither
RSLNT = Relationship With Spouse {PSI subscale); PATCS = mothers' nor fsthers' RMAT scores correlated with any
Pa~entaiAttimdes Toward Chiidrearing Scaie: WARM = Warm
of the subscales of the PACR scores (Table 4). This
Parenting S ~ i (PATCS
e subscaie). STRICT = Strict Parenting Styie
{PATCS subscaie:; COMFLT =- Conbic: (PATCS suhscaie:. finding provides posrtjve evidence that the F.4I may
% = 99.'~= 61. make a separate contribution ta understanding parental
*.D a .a5 **p < .01 ***p < "031. attitudes and behavior.
 TFIE PARENTING 4LtlANCE INVENTOR',

Comparison of the Predictive Vaiidity and RMhT scores drsplayed a dlstinctiy different pat-
of the PA1 and the WMRT ten. The fathers' PAMscore was slpnificmtl> related to
the teachers' ratlng of the child's self esteem (BASES)
The n o s t striking finding was the pattern of signifi- and s m a l competence I CPSCP). u hereas the fathers'
cant correlations of mothers' and fathers' P.41 scores RMAT score was rndependent of both.
correlating with each other's rating of their child. Their Kote that both the men and women's. scores on the
RhZAT scores were not related io the other parent's PA1 nelther correlated significantly with their M-
evaluation of their child's ad-justment (Table 5). An- CSDS scores nor with the demograph~cvariables
other significant indication of the possible greater pre- (Table 6).
dlctive validity of the PAI versus the RMAT is that the
father's ELnIAT was not significantly related to any of
the 3 mothers' ratings and 6 teachersichild care Discussion
pro\,iders' ratings of the child's functioning. whereas
the PiU was significantly correiated with 6 of the I0 The parenting alijance is v i e ~ e das a central \ arrable
ratings. The pattern of correlation of the mothers' in understanding the determinants nf parenting behav-
scores for the PA1 and RMAT in relation ta the ciitenon ior iitbldn. 1990. BeHskj, PIenzog, & Rex me. 1986;
variables was somewhat more variable Nelther the M'eissman & Cnhen, 1985) The concept suggests that.
mothers' P N nor their RIMAT scores correlated with although the parenting alliance may reiated to vari-
their ratings of their child"s behavior ESBS scores. ables such as parental personalit! and manta1 satisfac-
Downloaded by [New York University] at 09:16 18 April 2015

Mothers' PAE scores but not their RMAT scores were, tion. it makes a unique contribution to parenting
however. significantly correlated with fathers' rating of behanor and ulttmatelg to children's adjustment
their child's total adjusted ESBS; the reverse was found The dekeiopment of a measure of the parenting
for their RIMAT scores. alliance was the focus of this study. The PAI's de.ieE-
For mothers. both the PAI and RMAT scores dis- aprnent was guided b> the recommendation of
pHayed the same pattern of relat~onto rhe teachersichild Holden and Edwards (1SiS9'i regarding the need to
care providers' judgment of their child's self-esteem create narrow-band. parent self-report measures in
and social competence. In both cases, the chid's social the family assessment area. The P a l ' s high alpha
competence was independent of the PA1 and Rh4AT reliability and [\yo-factor strucfnre suggest some de-
scores: in both cases, the PA1 and RMAT were corre- gree of success t o ~ a r dthe creation of a focused
lated Ir = "20,p = 05) with self-esteem Fathers' PA1 narrow -band measure

Dependent 1-erirnbie MPAP h[K\IAP FPAP" FWIAT~

MPOSADJ
M 4XX
MCOMPTNC
MCONDUCT
MM-CSDS -

FPOSADJ
FANX
FCOMPTNC
FCONDUCT
FM-CSDS
17* -;
TPOSADf 1.5 .20* .i : .-A

TANX ?c**
.-- .16 .22* .I1
TCOMP'FNC .04 i5 .ZI
--
.h
TCONDUCT .r37 .15' I0 .07
:-.
TBASE .20* 20% .19* . &z.
.M ., .
.7
TCPSC .I5 .3.***
,Vote Prefixes: M = mother; F = fahe;: T = teacher <includingchiid care provider^. Variabir levels. POSADJ = Fsit!ce ad-b~stment
is the sum of the three ESBS subscales of Competence, . h e n !reversed), and Conduct (:eversell: ILNX = h u e ? !ESBS suhscaie): COlMPTNC
= Competence {ESBS subscale); CONDUCT = Conduct (ESBS subscale). M-CSDS = hiailawe-Crowne S x i J Des:rability Scde: BASES
= Behsviord Academic Self-Esteem Scale: CPSCS = California Preschoo! Social Competency Scale.
'ns = 99. 60, and 78 for conelations with mother, father. an6 teacherichild care provider, respec:ivei:,. ' n s = 6i. Kl, an6 52 foi conela-
rions with riother. father, and teacher'ckld care provider. respectiirely.
*p < .05. **p < .0l. ***p c .KI!
 ABIDIK B BRUNNER

Table 6 . PAi Correlations Wirh the Demographic FATCS scores and marital adjustment. It appears that
i*'a/ariabtes and kf-@SDS for Men an$ warm parenting style is related to the quality of the
Women paenticg alliance bit not to the overall quality of the
marriage for both men and women. The data also sug-
Men \~i.omen gest that the father's parenting alliance may be a
stronger predictor of the w m t h of his parenting be-
Chiid Age .Oe .06
.05 .06
havior than is the case for mothers. The absence of a
Parent Age
Educatior, 09 13 significant relation between mothers' ratings of their
Income children's adjustment and their PAMscores supports the
M-CSDS ~ndependenceof the P M measure from the standpoint
.?;ore: All correlations were not slgnificanr of a response bias on the part of mothers. Further
support of the PPeI's relative independence from re-
The prekiminary validity data for the PM. although sponse biases is that the P N scores of both mothers and
weak, was mostly positive. That the P.M was able to fathers are independent of their scores on the M-CSBS.
discriminate between the responses of married, sepa- In contrast to mothers' PM scores, fathers' PAMscores
rated, and divorced coupies is logical, because one were consistently related to children's adjustment
would assume that the parenfng alliance is likely to scores regardless of the source of the chiid rating ii.e..
vary with the degree of intactness of the parental rela- mothers. fathers, or teachersichilcl care providers). This
tionship. (The fact that the relatioa: of the PAI to marital finding raises some hope with regard to establishing the
Downloaded by [New York University] at 09:16 18 April 2015

adjustment was of a relatively low order. .20 for moth- predictive vaiidkty of the PM. at Ieast in relation to
ers and .44 for fathers, supports to a degree the concep- fathers.
tual independence of the parenting allrance from that of Overall. aithough some significant relations were
marital adjustment and satisfaction.) That the magrai- found with regard to the PAI's predictive baliditj,
rude of the relation was different for men and women they mere of a relatively io.n order and therefore must
is not surprisrng and seems consistent with our cultural be viewed. at best. as suggestive of predictive valid-
experience in which we find women being more com- ity. Ia is possible that future research on the PA1 will
mitted to thear parenting role despite their m d t a l sar- more clearly establish its predictive validity in rela-
lsfactlon and men reducing &elr parenting role in the tion to chdd ad-iustment. pxticu9arly If children with
face of marital dissatisfaction or dissolution hBelsky. it wider range of adjustment are used in the sample.
Giistrrap,&Rovine, 1984;Belsky. Ysnngbiade, &Rov- The current "sample of opportuni~y"was essentiallg
ine, 19911. a nonclinlcaj sample, and we anticipated that the
When parents' BAI scores were examined in relation predictive validity of the PAI woukd be substantially
to parentrng stress CPSXI. the expected negative relation greater wrth a broader sample, which we hq3e future
was found for hoth mothers and fathers. To the extent research will demonstrate.
that a strong parenting allsance exists. one would expect Our study provides some data that slappon the reli-
:he task of parefiting to go more smoottlIy and for both ability and the inextial construct and concurerat vaiidrty
parents to experience less stress in the parenting roie. of [he PAI. Although data sugpesti\.e of potenkal pre-
These Iow-to-moderate relations between stress and the dictive vairdiry were presented, ir is clear that additional
parenting akliance lends some credibihty to the con- research in thns area rs needed. The furire research wlHk
struct and the predictive validity sf the PAI. One appar- need to examine carefully possible sex differences in
ent noteworthy sex difference was that mothers' PILE relation to hoth the parents and the taget child. The
scores were independent of their feelings of attachment construe! of re parenting alliance znd the validity of the
tom ard their child, whereas for fathers' parenting alli- PA1 needs further exploration in relation to a variety of
ance scores were signlficantiy correlated with their family forms and structures. Parent-grandparent,
attachment scores. Although one mcsr be cautious not same-sex parents. and foster parents systems are but
to overinterpret this Eimd:np. it is consistent with our some of the variants of the parenting nlIiance that need
clinical experience with parents that. regardless sf the to he examined. Yote that. dthough there was some
quality of the marital relationship. most women remain minority represer~tationin our sample. the zdbsoluie
more connected to [hex children. These resuits are wumSer of participants makes generalizations about the
consistent with those of Gsldberg and Easterbrooks PA1 rc minority populations inappropriat~,.Given that
11354). who found that a father's attachment to his the expression of family life IS somewhat t:ultul-ally
children is more related to rtiarntal hzrmony than a specific, the PAI needs to he \.aEidated In relation to the
mother's attachent to her chijdren. minorities groups and cultures to which it is applied.
The separateness of the parenting alliance construct Future research on the PAI in the should dernonstra~e
from rnarltal adjustment was further den~onstratedby that ir makes a unique contribution to the measul-emeaxl
the significant correIations of the PA1 with the PATCS and understanding of family funcf cpning. To accom-
scores arad the absence of a significant relation between plish this goal, it needs to be examined in reIatio~erp
 THE PARENTING ALLIANCE IKVENTORf

other measures of pixrental beliefs, attitudes. and ~ a r e n - ~ s y r ~ m i n g?f.

? , 349-354
tal functioning. From theoretical and practical perspec- Dadds. M R .&Powell. M B (199l) The relatlonshlp ofinterparen-
tal confi~ctandglobal m t a i a J p t m r r t tcaggrrssion. anxlet).
tives, the establishment of a reliable and valid measure and Imrnatunr! an aggreswe and nonclrn~cchlldren iournaic3f
of the parenting alliance would be useful Theoreti- 4bnnonnt Cil;[dP.r\choloa~ !9,553-567
caHy, it could help clarify our understanding of the Emen R E 11982) Interparental conflsct and the ch11dxn of d~scord
paths of influence of parental attitudes and [he determi- and drvorce Pnih~ic~gacalBuiicim 92 319-330
Enier?. R E . & O'Lear: K D ( 1 9 W ~Manta1 dlscord and child
nants of parenting behavior. Practically, a valid mea- behat nor problem$ ~na noncllnlc sample S c ~ u r noj4bnr~rn1a!
i
sure of the parenting alliance \nould be useful in Child P ~ i c h l ~12g 41 ~ i -420
cI~nicallyassessing and predicting farnilj functioning Emeq R E kelnunob S B Neale J M t i9821 Effectsafmanta
m relation to areas such as child custody. child abuse. d~scoraon the school behm~o:of chidren b lib i c h z o p h ~ n l c .
affects elv d ~ s o r d e ~anJ d . nom~alparents Jouv~~:or.4hnc*~mui:
and special needs children. Future research will need to Ch'ld Pnihoh)g\ 10 225-128
explore the utility of the P.4I in relation to a variet? of Faukr R Forehand R T h o r n . A & lhre5on. t f 'l99i;. A
parenting arrangements. mediat~onalmodel of the Impact of mw.tal confl~cton adoies-
cent adjustment in intact and d~vorcedfamil~es.The role of
disrupted parenting. ChiidD~i~ioprneni, 61, !1 !?-f f 23.
Gafiiand. K. f.. & Day. K. D ; I 9 9 2 P m n d conflict and male
References adolescent probiem behavior. The daurna! q?Genetic P.ycho:-
og?. :r3.201-23.
Abidin, R. R. (1983j. Parenting Stress index (PSI) manuai. Char- Goidberg, \V.A,, & Easterbaooh. M. A. 11984:) Role of marital
iotres\:ille. VA: Pediatric Psychology Press. quality in toddler deveiopment. Bn,eioprncnraiPyc.iioiog,. ?O,
Downloaded by [New York University] at 09:16 18 April 2015

Ahidin, R. R. t 1988). Parenting Allanre Invenro?. Unpublished 5~14-514.

scale. llniversity of Virginia. Chariotres~ille. tiolden, G 'JI'.. 6; Eciwards. L. A. (1989). Parrntai attitudes to\vi.ard
Abidin. R. R. (1990). Parenting S:ress Index (PSI] maniiu: (3rd ed ). chiidrearing. Instruments, issues. and implications. Psyhoiog-
Char~ottesuille.V.4: Pediatric Psychology Press. lcai Butlerin, 106. 14-5e.
Achenbach: T.M. (19?8:. The Chiid Behnv~orProfile. I . Boys aged Fiowes. P., & Markman.. H, i !I9891 Marital quaiit? and child
fSi 1. Journal qfConsia!iir;gand Ciinicai P.r?.chologl,46, $78-- functionmg: A longrtudinal investigation. Chiid Dt~..eL.,pme.nt.
48%. 62. 1*%4-1051
Baumrind. D,(1971). Cumnt patterns of parental authorit? [Mono- Hughes. H. M.(1984). Measures of seli-toncep: and self-esteem for
graph;. Deveiopmenrul P.y~rholog,: 4, ? - 103. c5idren ages 3-12 gears- A. revleu and recon~mndadons.
Recker, W. C.ii964). Consequences of different kinds of parental Ciicicat P ~ c h i l i c ! Re>.rrw.
g~ 4. 657-602.
discipline. In M. t.Hoffman & L. W. Hoffman (Eds.). Re~iew Jenkins, I M..6; Smith. M. A. i1990'). Factors protecting chiidrrr,
05fc!zijGideveiopnzenf research [pp. 169-208) New York. Rus- living ic disharmonious honles. Maternal repom. Journil; ~:fthe
sell S a p Foundation. Amenrim Acadprn? of C!i:k' and Adcdescenr Py\,chiarp, 29,
Behar, L., 8; Stnngfield. S, (lY74). A beha\:ior rating scale for the 60-49.
preschool chiid. Developrner~ia/P~viholog?:I S , 631410. Jouriies, E. N.,Murphy, C . h,f., & F m s . A M (1941). h i ~ r i t d
Belsky. J., Giistrap. B., & Rovine. hl. i 19%
1.;! The Pennsylvania ad~uscmen(.parental d~sagreementsabtrur child rearing, and
Infant and Family Development Project. 1. Stability and change behavior problems in boys: Increasing the specsfici~of the
in mother-infant and father-infant intewc:ioa! in a family settnng marital assessment ChiidDe;elopmeni. 62.1424-1433.
at one. three. and nine months. Chii:iDevelnpmen;. 55.692505. Katz. L. F.: 6: Gottman. J. M i :993). Patterns of marital con3tct
Beisky, J., Hertzog. C.. B Rouine, I\?. (1986). Causal analyses of predict chiidren's internal:z~ngand externzilz~ngbehaviors. Dr-
multiple determinnnts of parenting: Ernpixicd m d methodalog- ~ ~ e ~ o p n z e n r a ~ P . ~ ~ cZB.94&450.
halog.
icai advances. In M. L m b , A Brown, 6; B. Rogofi {Eds.), Lmntel, D.. & Van k r Veen. F. (19951. Factors of marital adjust-
Advances 1n del~elopmenraipsych0fo.p (Voi. 4. pp. 153-202). ment in t x k e ^ s marital adjustmnr test Joicrnai r!f hfarrirlge
Millsdale, NJ: Lawrence Erlbaurn Associates, Inc. trnd the firnil? 35 57-63.
Beisky. I., Youngbiacie, L.. & Kovino, M. (1991). Patterns of maiita! Klinc, M.. Johnston. 1.R.. & Tschann. 3. M. :1991). The long shadow
change and parentxhild interaction. Jolountn: ny'hfarriage und of mai:aJ conflict. A m d e l of childrer?'~postd:vorce az-lust-
I ~ Famiiy,
E 33,487498. rnent Jolirnni ofh!urri(lge and the Fm?~i!\..53. 297-33.
Block. J., Block, J.. & Morrison, -4.(198:). Parental a p e m e n t 4 i s - LevIne, S.. El=?, F. F., 6; Lewis, M (1969) Cuiifornia Presc,hooi
agreemen! on chiidrearing onentations and gender-related per- Socta: Cnmpt.rency Sruie. Pale Altc. C.4: Consulting Psychot-
sonaiity correlates in children. CizildDe~~elop~nettf, 52,955-954 ogists Press.
Caldweii, C B., 8; Piant% R. C. (1991). A measure of young Locke. H. .I..B Wailace. K hf. i1959). Shofl marita! adjustment and
children's problems and competence behavior: The Earl! prediction tests. Their rellabilit: and \~aliditg.Marriage and
School Behavior Scale. Jounln~'c~fP.~rnoeducar;omI Assess- Fami+ Living, 21, 25 1-255.
menr, ?. 32-44. OLeq, K., 8 Turkewkz, tf. ( 1 9811. Evaiuating rnariral behavioral
Cl~ngempeei,iV. ( 1 981). Quasi-kin reiations'nips and marital qudity therapy fbr male aicoholics: Effects on marit3 a&ustrnent and
iri step-father famiiies. Jodorsrnai of Personalit). and Soctal Psy- communication from before rc after trearment. B~'lm,iorTher-
c:holeg~.41, 89G-901. CI,~?. 16. 147-167
Coopersmith. S.. 6; Giibeas, R (1982:. Behmvorci! 4cndemic S ~ l f - Reid, W.1.. 8 Crisafulh. A . (i39C)j. Mantal discord and chiid
Esteem. P d o Alto, CA. Consulting Psychoiogisrs Press. hehalrror problems: A mefa-aqalysis. fourrialc~~,4br1o.~malCh:i;;
Cox, M 5.. Owen, M T,,Lewis, J M,, & Henderson, K. ! 19891 Ps?ciroiogv. 18. 105-i 15.
Marriage. adult adjustment. and early parenilng. CniidDei~eiop- Rush. R T t985,. Assessing readability: Formulas and alternatives
3:

menr, 60.I01 5-1 014. Reading Teacher, 39,244283

Crass, C . G., &. Sharpieg, D F. (1981). The Locke-Wailace marital Rutter, M. (1978). Early sources of cecunt? and competeram.. In J . S .
adjustment test reconsidered: Some psychometric findings as B ~ n e& r A. Canon !,Eds 1 Human Growih nnc' I)eveiopitlenr
regards its reliability and factorial valrdiry. Educationa? und (pp 3341).London: Oxford University Press.
Psychologicczt Measurement. 41, 1303- 1306. Spanier, G 8.: 6: Thompson. L. 11982). .4 confirmaton analysis of
Crowne, D. P., & Mxlowe. D. (1960) 4 new scale of soclai desir- the dyadic ae.iustlnen: scale. J~?id,r~ai qfMarriage and Falfz:&.
abiliry independent of psychopatholog) . Journu! ofConsrriiing 44.731-738
 ASIDIN B BRUNNER
Teti, D M., Nakagawa. M., Das, X.,8 Wirtk. 0.(199i). Security of Weissman. S. R., B Cohen, R. S. :'1985).The pmnting ailrance and
attachment between preschoolers and their mothers. Relations adolescence. ddoiescent pehiarry. 12,2-4-45.
amonp social interaction. parenting stress. and mothers' soas of M.'hiE&er. S . , B B q , B. R. j 199; ) Overt and cover( parental cocfiict
the attachment q-set. Developmenra! Fsycizology. 27, 3, &I& and adoiescent problems: Observed mantnl interaction in clinic
447, and nonclinic families. .Adolescence, 26,865-876.
Wallersrein, J.: 6r Kelly. 4. ii9801. Divihd kzorrses: Parents and
children cope with divorce. New York. Basic.
Webster-Stratton, C. 11988) Mothers; and fathers' perceptions of
chnld devsance Roles of parent and chiid behallon and parent
adjus~rnentJournal gfCo~suEr:xgand Chnica: Psjchoioy) 56,
909-9 I5
IZeceiVed December I S 9
Final re~~ueeam
'"'
recerved J m e 20, 1994

Appendix:
The Parenting Allliarlce Inventory

PARENTING ALLIANCE INVEMTOORY

R.R. Abidirt
Curry School erf Education
~ n i v e r s i tsf~ Virginia
DIRECTIONS: The questions listed below concern what happens between you and your child's other
Downloaded by [New York University] at 09:16 18 April 2015

parent, or the other adult mast involved in the care of your child. While you may not find an
answer which exactly describes what you think, please circle the answer that comes closest
Po what you think.
YOUR FlRST REACTION SHOULD BE YOUR ANSWER.

i. My chikl's other parent enjoys beirlg atone with our chilc'. 5 4 3 2 1

-
2. During pregnancy, my child's other parent expressed conf~dencein my ability t o be a good I
parent.
3. WRe(1 thare is a problem wkt! our chlld, we w3rk og? a ~ o o dsolutlor: together. 5 4 3 2 1
4. My child's other gaaent and k communicad~welI about our child. 5 4 3 2 1
8. My &Id's Other garen1 i8 willrng to make personal sacrifices to bclp take care of our child. 5 4 3 2 f

8. Palkina to rnv child% other men: about our child is somethina. B look torward la. 6 4 3 2 1
7. My c W ' s o t b t parent pays a ow: deal o! anerltion t o adr cnilc 6 4 3 2 1
8. My child's nthet parent and L agree on what our child should and shouid not be permitted to
do.
9. Ifee! cbse to my child's o?hcr parent wtler f sea hirn!her play wit%odr child. 5 4 3 2 %
$0. Mv child's other oarent knows how to handie childrec weli. 5 4 5 2 1
I My c W a 6 t h ~aront
~ and I are a good team. 5 4 3 2 1

18. My child" other parent and I would basicalk describe out child in the same way. 5 4 3 2 1
17. If our ctlifd weds tto be punished, n y ch1l3's other parent and I asdally ayree on the type o:
prmimenr.
bout what 1s rqh? for our child. 6 4 3 1 1
5 4 3 2 1
20. My child's other parent an6 I have the same goals for our child. 5 4 3 2 1