7/25/2019 17-4-6_2

1/10

1

SUMMARY

Objective: The present study aimed to evaluate thepsychometric properties of the Parental Bonding Instrument(PBI) with Turkish university students.

Method: The study included 336 (61.5%) female and211 (38.5%) male students from Ankara and PamukkaleUniversities. The mean age of the participants was 21.55years (SD: 1.65 years; range: 18-26 years). The constructvalidity of PBI was examined by exploratory factoranalysis; the criterion validity was assessed by examiningthe correlation of the PBI with two other instruments, theParental Attitude Research Instrument (PARI) and Parenting

Style Inventory (PSI). The reliability of the PBI was assessedby Cronbachs Alpha, test-retest, and split-half reliabilitystatistics.

Results: The factor analysis yielded twofactor solutionsfor both parents, as did the original PBI. Nevertheless, theitems related to controlling behaviors were loaded on thecare factor instead of the overprotection factor, as found byParker et al. (1979). Thus, these two factors were calledcare/control and overprotection subscales. Additionally,PBI was found to generally demonstrate a high level ofcorrelation with PARI and PSI, suppor ting criterion- validity.Finally, the reliability studies of the PBI demonstrated that itis a highly reliable scale and showed stability over a period

of 3 weeks.Conclusion: The psychometric properties of PBI weresatisfactory and could be used to assess parental attachment.It is suggested that further studies may illuminate theclinical value of PBI and quantify parental contribution topsychopathology.

Key Words:Parental bonding, care, overprotection, validity,reliability, parental bonding instrument

Trk Psikiyatri Dergisi 2006; 17(4)Turkish Journal of Psychiatry

The Parental Bonding Instrument:Evaluation of its Psychometric Propertieswith Turkish University Students

Emine Gl KAPI, Sevgi KKER

INTRODUCTION

There is a general acceptance regarding the ef-fects of the parent-child relationship on a childsdevelopment, relationships with others, and psy-chological adjustment. The initial and most impor-tant steps that resulted in this point of view wereBowlbys research findings, which demonstratedthe importance of the social bond between primarycare giver and child (attachment) to normal childdevelopment (Crockenberg et al., 1993).

According to Bowlby (1969, 1973), the mostimportant factors in the development of attach-

ment between the primary care giver and the childare: a) orientation towards preferring human be-ings and moving objects; b) learning to differenti-ate the most frequently seen people from others;c) approaching acquaintances and avoiding strang-ers; d) differentiating behaviors that lead to desiredresults from those that do not, and to increase those

behaviors.

Based on Bowlby's theory, Ainsworth et al.(1978) defined three attachment styles. Infantswith a caregiver that responds timely to their needs

feel safe to engage in play and exploration. Thiskind of a secure attachment with a caregiver is ben-eficial to the infants general adaptation. In caseswhere the caregivers reactions to an infants needsare inconsistent or when the caregiver cannot re-spond on time, an unstable/anxious attachmentstyle might form. On the other hand, in situationswhere the caregiver is consistently non-responsiveto the infants needs, an anxious/avoidant attach-

Emine Gl Kap MD, e-mail:kapci@education ankara.edu.trSevgi Kker MD, e-mail:skucuker@pamukkale.edu.tr

7/25/2019 17-4-6_2

2/10

2

ment style forms. There are numerous researchstudies regarding the relationship between mentalhealth, both in childhood and adulthood, and theattachment style between the infant and the prima-ry caregiver (Rohner and Britner, 2002).

In parallel with the acceptance of the impor-tance of the parent-child relationship to an indi-viduals life, a number of scales/tools were devel-oped in order to understand the childs/individuals

perception of the parent-child relationship, whichinclude the Own Memories of Childhood RearingExperiences (Perris et al., 1980), Parental Attach-ment Questionnaire (Kenny, 1987), and the Con-tinued Attachment Scale-Parent Version (Bermanet al., 1994).

Tools for assessing the parent-child relation-ship as perceived by the children, adolescents oradultshave also been developed or adapted in Tur-key. The first one of these scales is the ParentalAttitude Inventory developed by Kuzgun (1972).This inventory assesses the perceptions of chil-dren and adolescents about their parents attitudestowards them, i.e. democratic, authoritarian, orcareless. Other tools were also developed basedon Kuzgun's inventory (Bilal, 1984; Polat, 1986).In the Perceived Child Rearing Attitudes Scaledeveloped by Okman (1979), perceived parental

attitudes are evaluated in terms of the followingdimensions: ability to separate from the mother,self-control, autonomy, and personality charater-istics.

In addition to these scales developed in Turkey,there are also various scales that were adapted tothe Turkish culture. The Parental Attitude ResearchInstrument (PARI) developed by Schaefer and Bellin 1958 was adapted by LeCompte et al., 1978 TheParental Acceptance-Rejection Questionnaire-Child Form developed by Rohner et al. (1980)was adapted - by Erdem, 1990. Another scale, the

Parenting Style Inventory, originally developed byLamborn et al., (1991) was adapted to Turkey byYlmaz (2000). One of the most frequently usedscales in the international literature is the ParentalBonding Instrument (PBI) (Parker et al., 1979). Itis also one of the first scales that was based onBowlby's attachment theory (1969, 1973).

Bowlby related deficient/pathological parent-ing with lack of care and excessivecontrol/protec-tion dimensions. The term lack of care includes

providing deficient care, an inability to meet the in-

fants needs, and belittling, criticizing, or rejectingthe child-. On the other hand, the term excessivecontrol was defined as over protection, not sup-

porting the development of the childs autonomy,or excessively controlling the childs behavior. PBIdeveloped by Parker et al., (1979) in order to mea-sure the parent-child attachment pattern, includescare and control/protection dimensions, as Bowlbyhad proposed. Perceived parental behaviors arescored separately according to these dimensions.In addition, the perceived parental behavior pat-tern can be evaluated with a quadrant classifica-tion model by taking these two dimensions intoconsideration. According to this quadrant model,high care-low overprotection is defined as optimalattachment. Low care-low overprotection is classi-

fied as non-attachment or poor attachment. Parentswho display high care overprotection behaviorsare classified in the affectionate constraint group.Lastly, low careoverprotective parental behaviorsare classified as affectionless control.

In the studies conducted with PBI in differentcultures, the issue of whether the scale has two fac-tors (care and overprotection) or three factors (care,overprotection, and control) remains controversial.In addition to the studies supporting that the scalehas two factors (Uehara et al., 1998, Parker et al.,1997), there are also studies that support three fac-

tors in the scale (Cox et al., 2000; Martin et al.,2004; Lizardi and Klein, 2002). Some researchfindings showed that the scale supports both thet-wo factorial and three factorial structures (Qadiret al., 2005). Furthermore, approximately 15-20years after the development of PBI with 25 items,Pederson (1994) proposed that the items may bereduced down to 10 and Kendler (1996) proposedthat items of the inventory might be reduced downto 16.

Fundamentally, the scale assesses a childs per-

ception of the parent-child relationship retrospec-tively. Studies conducted with PBI showed that thescale could evaluate the contribution of the parent-child relationship to mental disorders, and that it istherefore a valuable tool for use in clinical studies.Studies have found a relationship between attach-ment to parents and mental disorders during ado-lescence and adulthood. For example, perceivedlow parental care and excess control were foundto be related to various psychological disorders,such as depression (Uehara et al., 1998), anxietydisorders (Chambers et al, 2004), schizophrenia

7/25/2019 17-4-6_2

3/10

3

(Willinger et al., 2002), adolescent suicide (Vio-lato and Arato, 2004), eating disorders (Meyerand Gillings, 2004), substance abuse (Gerra etal., 2004), and obsessive compulsive disorders(Myhr et al., 2004).

PBI has been shown to be a valid and reli-able tool in Western cultures, and Near and FarEastern cultures, including Pakistan (Quadir et

al., 2005), Japan (Narita et al., 2000), Belgium,Holland, Italy, Spain, and France (Heider et al.,2005).

The primary aim of this study was to evaluatethe psychometric properties of the PBI for theTurkish culture. The PBI assess perceived andactual parenting characteristics retrospectively.(Parker, 1983 can measure the parent-child re-lationship over a long period of time, as muchas 20 years (Wilhelm et al., 2005), has a clinicalvalue andis short, easy to administer and score.

METHOD

Participants

Of the 547 study participants, 265 were (48.4%)from Ankara University (AU) Engineering Fac-ulty, 55 were (10.1%) from AU Science Faculty,121 were (22.1%) from AU Educational SciencesFaculty, and 106 (19.4%) were from Pamukkale

University Education Faculty. The mean age of theparticipants was 21.55 years (SD: 1.65; range: 18-26 years). Sociodemographic characteristics of the

participants are shown in Table I.

Measures

PBI, The Parental Bonding Instrument (Parkeret al., 1979): PBI assesses the perceived parental- relationship retrospectively as perceived by theindividual. Mainly, the scale includes two factors;care and control/overprotection. Of the 25 items,12 items compose the care dimension, (scores

Table I. Sociodemographic Characteristics of the Participants

n (547) %

Gender

FemaleMale

336 61.5

211 38.5

Classroom Undergraduation level (as years)IIIIIIIV

38 6.9

190 34.7

169 30.9

150 24.4

Mothers education levelIlliteratePrimary schoolHigh-schoolUniversity

42 8.5

242 44.6

135 26.6

88 17.3

Fathers education levelIlliteratePrimary schoolUniversity

11 1.6

190 37.4

132 26.0

175 34.4

Number of siblingsNo sibling1 sibling2 siblings3-8 siblings

n (515) %

40 7.8

163 31.7

168 32.6

144 27.9

7/25/2019 17-4-6_2

4/10

4

range between 0 and 36). High scores reflect theperception of warm, understanding, and accept-ing parents, whereas low scores represent the per-ception of cold and rejecting parents. There are13 items in the control/overprotection dimension

(scores range between 0 and 39) and high scoresindicate the perception of overprotective parentingor not allowing childs autonomous behaviors .The individual is required to rate the parental be-haviors represented in each item regarding the first

Table II.Factor Structure of ABB.

Items

Factors

Mother Form

(n = 419)

Father Form

(n = 431)

Factor I Factor II Factor I Factor I

1. Spoke to me in a warm and friendly voice .0.71 .0.03 .0.75

2 Did not help me as much as I needed .0.55 .0.12 .0.61 .0.08

3. Let me do those things I liked doing .0.61 .0.12 .0.61 .0.32

4. Seemed emotionally cold to me .0.61 .0.08 .0.67

5. Appeared to understand my problems and worries .0.60 .0.66

6. Was affectionate to me .0.62 .0.69

7. Liked me to make my own decisions .0.68 .0.65 .0.26

11. Enjoyed talking things over with me .0.68 .0.11 .0.73

12 Frequently smiled at me .0.71 .0.76

14. Did not seem to understand what I needed or wanted .0.55 .0.23 .0.57 .0.17

15. Let me decide things for myself .0.67 .0.17 .0.64 .0.39

16. Made me feel I wasnt wanted .0.50 .0.08 .0.47 .0.27

17 Could make me feel better when I was upset .0.71 .0.05 .0.78

18 Did not talk with me very much .0.67 .0.13 .0.68

21 Gave me as much freedom as I wanted .0.56 .0.18 .0.51 .0.37

22 Let me go out as often as I wanted .0.56 .0.18 .0.46 .0.32

24 Did not praise me .0.27 .0.14 .0.36

25 Let me dress in any way I pleased .0.54 .0.12 .0.52 .0.34

8. Did not want me to grow up .0.62 .0.54

9. Tried to control everything I did .0.24 .0.61 .0.25 .0.59

10. Invaded my privacy .0.29 .0.44 .0.27 .0.55

13. Tended to baby me .0.59 .0.13 .0.61

19. Tried to make me feel dependent on her/him .0.29 .0.63 .0.25 .0.55

20. Felt I could not look after myself unless she/he was around .0.14 .0.49 .0.51

23. Was overprotective of me .0.16 .0.67 .0.19 .0.55

Eigen value 7.21 2.39 7.80 2.60Variance explained 28.8% 9.6% 31.2% 10.4%

Cronbachs Alpha (Subscales) 0.90 0.70 0.91 0.70

Cronbachs Alpha (Total scale) 0.87 0.89

7/25/2019 17-4-6_2

5/10

5

16 years of his/her life on a 4 point Likert-type rat-ing scale (very like = 3.... very unlike = 0) Theindividual makes two different ratings, one each

for his/her mother and father. In the scoring of thecare subscale, items 2, 3, 4, 7, 14, 15, 16, 18, 21,22, 24, and 25, and items 8, 9, 10, 13, 19, 20, and23 on the control overprotection subscale are re-versely coded. (0-3).

The two-factorial structure of the PBI is evalu-ated as two separate scales and two total scores arecalculated from the subscales. High scores in thecare dimension and low scores in the overprotec-tion/control dimension represent positively per-ceived parental behavior. These two scales can beused independently, as well as together.

In the second type of application, perceived pa-rental behaviors can be quadrant classified as pre-viously described. A number of reliability studieswere conducted for the PBI Test-retest reliabilitywas found to be 0.76 for the care dimension and0.63 for the overprotection/control dimension,split-half reliability was 0.88 and 0.74, and inter-rater reliability coefficients were 0.85 and 0.69 re-spectively. For the concurrent validity study of thePBI, it was shown that scores from the interviewand scores derived from the PBI factors were also

related (care: 0.77; overprotection/control: 0.50).

In the Turkish adaptation study of the PBI, twoqualified academicians translated the scale inde-

pendently into Turkish from English. After makingthe required corrections between these two separatetranslations, the back translation was performed byanother qualified academician. In the comparisonof the original scale and the back-translated form,it was agreed upon that the statement of the itemswere conceptually equal. A group of university stu-dents were asked to evaluate the items in terms oftheir understandability and it was concluded thatthe items were easily understandable. As pointedout above, the original scale has two dimensions,with high scores in the care dimension and low

scores in overprotection/control dimension rep-resenting positively perceived parental behavior.For this study, as the factor structure of the Turk-ish version was not known, all items were scoredin the same direction (increased scores reflecting

positive parental attachment). Only the followingitems were reversely coded: items 2, 4, 8, 9, 10,13, 14, 16, 18, 1, 2, 23, and 24. Increased scoresin both total score and subscale scores would thusrepresent perceived positive attachment towards

parents. The Turkish version of the scale is calledAna Babaya Balanma lei (ABB).

Table III.Correlations between ABB, PSI, and PARI.

Mother Form Father Form

ABBtotal

ABB-care/control

ABB-protection

ABBtotal

ABB-care/control

ABB-protection

ABB-care/control .0.94*** .0.95***

ABB-protection .0.61*** .0.29*** .0.60*** .0.27***

PARI--1. Overprotective motherhing -.0.16** -.0.02 -.0.39*** -.0.06 -.0.03 -.0.26***

PARI-2. Democratic Attitude and Equality .0.41*** .0.42*** .0.17** .0.43*** .0.43*** .0.17***

PARI-3. Rejecting of Homemaking role -.0.32*** -.0.21*** -.0.37*** -.0.18*** -.0.09 -.0.30***

PARI- 4. Marital conflict -.0.27*** -.0.19*** -.0.29*** -.0.41*** -.0.36*** -.0.32***

PARI-5. Strict discipline -.0.35*** -.0.23*** -.0.42*** -.0.29*** -.0.20*** -.0.35***

PSI-1. Aacceaptaence/involvement/autonomy .0.20*** .0.28*** .0.10* .0.23*** .0.31*** .0.11*

PSI-2. Direct control .0.20*** .0.18*** .0.12* .0.28*** .0.25*** .0.21***

PSI-3. Indirect control .0.19*** .0.26*** -.0.07 .0.11* .0.18*** -.0.11*

*P < 0.05; **P < 0.01; ***P < 0.001

7/25/2019 17-4-6_2

6/10

6

Parental Attitude Research Instrument (PARI)(Schaefer and Bell, 1958), adapted by LeCompteet al. (1978): The scale consists of 60 items and5 subscales, which are Overprotective MotheringDemocratic Attitude and Equality, Rejection ofHomemaking Role, Marital Conflict, and StrictDiscipline. Items 2, 29, and 44 are reversely coded.Increased scores of the factors, except DemocraticAttitude and Equality, indicate negative parentalattitude. There is no total score of the scale; onlythe subscale scores are evaluated. Test-retest reli-ability of the factors was between 0.58 and 0.88.In a study conducted by Kk (1987) with highschool students and their parents, construct valid-ity of PARI subscales were also supported. Thistool, which is frequently used in Turkey, was used

with the aim of evaluating the criterion validity ofABB.

Parenting Style Inventory (PSI) (Lamborn etal., 1991), adapted by Ylmaz (2000): The scaleconsists of 26 items and three dimensions of par-enting style: acceptance/involment, psychologicalautonomy and strictness/supervision Items 1-18are scored between 4 and 1, items 19 and 20 arescored between 0 and 6, and items 21-26 are scored

between 1 and 3. Test-retest reliability and internalconsistency of the scale were evaluated with pri-mary school, high school, and university students.

The reliability coefficients for each subscale andfor the 3 study groups ranged between 0.60 and0.93. In the validity study of the scale, a similarfactor structure to the original study was found;however, item total correlation of the third item inall 3 study groups and the factor loading values ofthe 10th and 16th items for the university studentswere low. In conclusion, the scale was reduced toa 23-item scale after excluding these 3 items. Thefactors found for the university students were asfollows: acceptance/involvement/autonomy; di-rect control; indirect control. The scale was used

with the aim of evaluating the criterionvalidity ofABB.

Procedure

The study participants were recruited fromAnkara University Engineering, Science, andEducational Sciences Faculties, and PamukkaleUniversity Education Faculty. The aim of thestudy was briefly explained to the students and 4students from Ankara University Educational Sci-ences Faculty declined to participate because theywere not well or tired; all other students completed

the scales voluntarily. ABB-Mother, ABB-Fa-ther, PSI, and PARI were administered to studentgroups of 25-30, counterbalancing the question-naires in each group. Completion of the scales tookapproximately 40 to 50 minutes.

Data Analysis

To test the stability of ABB scores, correla-tions of the test-retest scores with a 3-week intervalwere examined. For the internal consistency of thescale, Cronbachs alpha reliability coefficient wascomputed, and split-half reliability was examinedusing Guttmann split-half analysis. Exploratoryfactor analysis was conducted in order to deter-mine the factor structure of the scale. Correlationsof PARI and PSI total scores and subscale scoreswere examined in order to evaluate the criterionvalidity of the scale.

FINDINGS

Descriptive findings related to ABB

The mean total scores of the father and motherforms of ABB were 51.44 (SS: 10.93; n= 441)and 49.12 (SS: 12.03; n = 440), respectively, andcorrelation between these 2 forms was 0.59 (P 1 were found, whichexplained 53% of the total variance for the motherform and 57% for the father form. However, whenfactor loadings were evaluated, it was found thatmajority of the items loaded onto two factors withEigen values > 2. The results of the varimax ro-tation showed a 2-factor construct for both forms(Table II). For the control/overprotection factor,items that were related only to the controlling be-haviors in the original scale (3, 7, 15, 21, 22, and25) were found to load on the care dimension inthe present study. As this factor includes control-ling behavior of the parents toward the child, inaddition to care, it was called care/control. Whenitems of the second factor were examined (8, 9, 10,13, 19, 20, 23), it was found that all items included

parental overprotective behavior and was calledoverprotection. Although item 9 loaded onto theoverprotection/control factor in the original scale,this item loaded onto overprotection factor in thisstudy.

Two factors derived from the mother form ex-plained 38.4% of total variance. Corrected itemtotal correlations in this dimension were between0.25-0.68 and were statistically significant (P 0.001). Similarly, correlations between the twosubscales of the mother and father forms were sig-nificant (P < 0.001).

Correlations of PARI and PSI total scores andsubscale scores were examined in order to evalu-ate the criterion- validity. As specified in the meth-ods section, higher scores for the factors of PARI,except the second subscale (Democratic Attitudeand Equality), indicate negative parental attitude.As expected correlations between total scores ofABB mother and father forms, and subscale

scores with PARI were significant,. While the cor-relations between the first subscale of PARI, name-ly, Overprotective Mothering and care/control di-mensions of the mother form were not significant,its correlation with ABBs protection subscaleswere significant. In the father form of ABB, de-spite the significant correlation between the pro-tection subscale and PARI first subscale, there wasno correlation with total ABB scores and thecare/control subscale.

When correlations between ABB total and

subscale scores and PSI subscales are considered,generally, correlations were significant, and as anexpected direction. The correlation between thethird subscale of PSI (indirect control) and over-

protection subscale of ABB-Father form wasnegative and statistically significant while this cor-relation was also negative for the mother form butnot statistically significant.

Reliability Findings

The reliability of ABB was calculated usinginternal consistency, split-half, and test-retest sta-

tistics. The Cronbachs alpha value for the motherform was 0.87 and was 0.89 for the father form.The Cronbachs alpha for the protection subscalesfor both forms was 0.70 and for the care/controlsubscales it was 0.90 for the mother form and 0.91for the father form. In the split-half analysis of theABB, the Guttmann split-half method was used.According to the analysis, the value for the totalmother form was 0.70, for the care/control dimen-sion it was 0.85 and was 0.68 for the protectiondimension. For the father form, these values wereas follows: for the total scale, 0.64; for the care/

7/25/2019 17-4-6_2

8/10

8

control and protection subscales, 0.83 and 0.64,respectively.

A separate group of students (n=106) partici-pated in the test re-test reliability study of the scale

with three weeks interval. Seventy-four of thesestudents were female (69.8%) and 32 were male(31.2%); the mean age was 20.82 years (SD: 1.95years). Test-retest reliability of the mother formwas 0.90 and was 0.89 for the father form. For thesubscales, the test-retest reliability of the care/con-trol dimension of the mother form was 0.91 andit was 0.90 for the father form. The reliability forthe protection dimension was 0.64 and 0.78 for themother and father forms, respectively.

When reliability and validity findings are con-sidered, it could be concluded that the psychomet-ric characteristics of ABB seems to be satisfac-tory.

DISCUSSION

The reliability and validity of the PBI, devel-oped by Parker et al. (1979), was evaluated withTurkish university students. In order to evaluatethe validity of the Turkish version of the scale(ABB), a factor analysis was performed forconstruct validity, and ABBs correlations withPARI and PSI were examined for criterion- valid-

ity. Internal consistency, split-half, and test-retestreliability coefficients were performed for theevaluation of ABBs reliability

The original scale consists of two factors; careand overprotection/control. In the present study, al-though ABB displayed a two-factorial construct,it was found that control items that were loaded onthe overprotection/control dimension in the orig-inal scale were found to load on the care factor.Therefore, this factor was called care/control.

As stated in the introduction, discussions re-

garding the factor structure of PBI are ongoing.Some studies showed that the scale has two fac-tors, care and overprotection (Uehara et al., 1998;Parker et al., 1997), and other studies proposedthat the overprotection and control dimensions asseparate factors are more precise and has satis-factory psychometric properties (Cox et al., 2000;Martin et al., 2004; Lizardi and Klein, 2002). Inthis study, control items were neither loaded on theoverprotection factor and nor they appeared as athird factor. As control behaviors of the parents arefound to go together with overprotective parenting

in Western cultures, it appeared as a part of car-ing parenting in Turkish culture. The reason for thecontrol and care items to appear as one factor inthe present study could be related to the fact thatcaring parenting has always been associated withcontrolling and/or to be informed about offsprings

behaviorin Turkish culture

Another finding that could be important is thatwhile the ninth item (tried to control everything Idid) loaded onto the overprotection/control factorin the original study, the same item was found toload onto the overprotection factor in this study.Although this item seems to be related to the pa-rental controlling behavior at first glance, it differsfrom other control items; in fact trying to controleverything could be a denial of the childs psy-

chological autonomy and thus this parental behav-ior might reflect an underlying pattern of overpro-tective parenting There are some studies that foundthis item on the overprotection factor supportingthe above explanation (Narita et al., 2000; Martinet al., 2004).

In the analyses conducted to evaluate the crite-rionvalidity of ABB, it was found that total scaleand subscales of ABB demonstrated acceptablecorrelations with PARI and PSIs subscales. Thisfinding indicates that ABB is a valid instrument

in measuring parenting behavior and thus, has cri-terion-validity. Besides this general finding, somecorrelations were found to be non-significant.Correlations between the care/control subscale ofABB and PARI 1- overprotective motherhingsubscale were not significant. When the items ofPARI 1- overprotective motherhing subscale wereexamined, it was seen that some items were relatedto the pattern of ignoring the psychological auton-omy of the child. Significant correlations with thisdimension of PARI and the ABB overprotectionsubscale support this finding.

In addition to this, the correlation between thetotal score of - ABB Father form and PARI-1overprotective motherhing scale was not signifi-cant A possible reason for this finding could be thatsome items in this scale are related with traditionalmother behavior and some are related to both fa-ther and mother behavior.

Generally, correlations between ABB total andsubscale scores, and PSI were significant., How-ever, the correlation between the third subscale ofPSI, indirect control, and the overprotection sub-

7/25/2019 17-4-6_2

9/10

9

scale of ABB was negative for both mother andfather forms, but it was also statistically significant,for the father form. Findings that were opposite toour expectations may have been due to a numberof reasons. Firstly, the present study, demonstrateda difference between perceived parental behavior

patterns but, as the mothers and fathers were notevaluated seperately in PSI, it could not reflect thedifferences among the parents. The findings mightalso have been due to the behaviors measured bythe PSIs indirect control and ABBs overprotec-tion dimensions. For example, a parent showingno effort to find out what a child does after schoolor at night could be accepted as positive parent-ing in PSI. This type of controlling parental behav-iors are loaded on the care/control dimension of

ABB, not in the overprotection dimension, buttheir content are also different (for example, Letme decide things for myself).

The reliability of ABB was performed us-ing internal consistency, split-half, and test-reteststatistics. It was observed that total scores andsubscale scores showed a high degree of consis-tency for the mother and father forms of ABB.In addition, the high Cronbachs alpha values andsplit-half correlation coefficients, and significantitem total correlations support the reliability of thescale.

The care and protective behaviors that Bowlbyproposed to be thetwo primary dimensions of par-enting behavior could be said to be supported inthe Turkish version of the scale, as in the origi-nal. The findings of the present study showed that

ABB was able to assess the universal dimensionsof parent-child attachment, such as care and pro-tection; however, the main difference between PBIand ABB was that while items related to con-

trolling parental behavior go together with over-protective behaviors in PBI, they appear with thecaring parental behaviors in ABB. Controlling

parental behavior which appears with overprotec-tive behaviors in PBI, but with caring behaviorsin the Turkish culture, and yet as a third factor inan other culture (Narita et al., 2000) points out theimportance of cultural variables in the controlling

parental behaviors.

As stated in the introduction, there are manystudies that utilized the PBI to evaluate the pa-

rental behaviors that might contribute to variouspsychiatric disorders, ranging from anxiety dis-orders (Chambers et al., 2004) to schizophrenia(Willinger et al., 2002). A limitation of this studywas that a comparison to a clinical group was not

performed. Studies with different clinical groupsin Turkey might provide information about thediscriminant validity of the scale and also the par-ent-child relationship characteristics contributionto psychiatric disorders unique to Turkey. To the

best of our knowledge, there are no studies that as-sessed the relationship between adults attachmentto their own parents and their offspringin the relat-ed literature. Assessing this pattern using ABB inTurkey could also provide important informationfor interventions aiming to improve parent-childrelationships.

REFERENCES

Ainsworth MDS, Blehar M, Waters E. et. al (1978) Patterns ofAttachment. Hillsdale, NJ:Erlbaum.

Berman WH, Heiss GE Sperling MB (1994) Measuring continued

attachment to parents: The continued attachment scale-Parent version.Psychol Rep, 75: 171-182.

Bilal G (1984) Demokratik ve otoriter olarak alglanan ana-babatutumlarnn ocuklarn uyum dzeylerine etkisi. Unpublished PhDThesis. Ankara: HacettepeUniversity.

Bowlby J (1969). Attachment and Loss: Vol.1. Attachment. NewYork Basic Books.

Bowlby J (1973) Attachment and Loss: Vol. 2. Separation:Anxiety and anger. New York Basic Books.

Chambers JA, Power KG, Durham RC (2004) Parental styles andlong-term outcome following treatment for anxiety disorders. ClinPsychol Psychotherapy, 11: 187-198.

Cox BJ, Enns MW, Clara IP (2000) The parental bondinginstrument: Confirmatory evidence for a three-factor model in a

psychiatric clinical sample and in the national comorbidity survey.Soc Psychiatry Psychiatr Epidemiol, 35: 353-357.

Crockenberg S, Lyons-Ruth K, Dickstein S (1993) The familycontext of infant mental health: II. Infant development in multiple

family relationships. Handbook of nfant mental health, CH Zeanah(Ed), New York Guilford Press, s. 38-55.

Gerra G, Angioni L, Zaimovic A et. al (2004) Substanceuse among high-school students: relationships with temperamentpersonality traits and parental care perception. Substance Use Misuse,39: 345-367.

Heider D, Matschinger H, Bernert S et. al (2005) Empiricalevidence for an invariant three-factor structure of the ParentalBonding Instrument in six European countries. Psychiatry Res, 135:237-47.

Kendler KS (1996) Parenting: A genetic-epidemiologicalperspective. Am J Psychiatry, 153: 11-20.

Kenny ME (1987) The extent and function of parental attachmentamong first year college students. J Youth Adolesc, 16: 17-27.

7/25/2019 17-4-6_2

10/10

10

Kuzgun Y (1972) Ana-baba tutumlarnn bireyin kendinigerekletirme dzeyine etkisi. Unpublished PhD Thesis. Ankara:Hacettepe University

Kk (1987) The validity of the Turkish form of the PARIsubscales II, III, IV. Unpublished Masters Thesis, Istanbul: BoaziiUniversity.

Lamborn SD, Mounts NS, Steinberg L et. al (1991) Patterns ofcompetence and adjustment among adolescents from authoritative,authoritarian, indulgent, and neglectful families. Child Development,62: 1049-1065.

LeCompte G, LeCompte A, zer S (1978) sosyoekonomikdzeyde Ankaral annelerin ocuk yetitirme tutumlar: Bir lekuyarlamas. Psikoloji Dergisi, 1: 5-8.

Lizardi H ve Klein DN (2002) Evidence of increased sensitivityusing a three-factor version of the parental bonding instrument. JNerv Ment Dis, 190: 619-623.

Martin G, Bergen HA, Roeger L et. al (2004) Depression inyoung adolescents. Investigations using 2 and 3 factor versions of theparental bonding instrument. J Nerv Ment Dis, 192: 650-657.

Meyer C ve Gillings K (2004). Parental bonding and bulimic

psychopathology: the mediating role of mistrust/abuse beliefs. Int JEating Disord, 35: 229-233. (www.interscience.wiley.com).

Myhr G, Sookman D and Pinard G (2004) Attachment securityand parental bonding in adults with obsessive-compulsive disorder:a comparison with depressed out-patients and healthy controls. ActaPsychiatr Scand, 109: 447-456.

Narita T, Sato T, Hirano S et.al (2000) Parental child- rearingbehavior as measured by the parental bonding instrument in aJapanese population: factor structure and relationship to a lifetimehistory of depression. J Affect Disord, 57: 229-234.

Okman G (1979) Bilisel stili belirleyen etkenler: Ergenlerzerinde bir inceleme. Unpublished Assoc. Professorship Thesis,Istanbul: Boazii University.

Parker G (1983) Parental overprotection: A risk factor in

psychosocial development. Grune & Stratton: New York.

Parker G, Tupling H and Brown LB (1979) A parental bondinginstrument. Br J Med Psychol, 52: 1-10.

Parker G, Roussos J, Hadzi-Pavlovic D et.al (1997) Thedevelopment of a refined measure of dysfunctional parenting andassessment of its relevance in patiens with affective disorders. PsycholMed, 27: 1193-1203.

Pederson W (1994) Parental relations, mental health anddelinquency in adolescents. Adolesc, 29: 975-990.

Perris C, Jacobsson L, Lindstrm H et al. ( 1980) Development ofa new inventory for assessing memories of parental rearing behaviour.. Acta Psychiatr Scand, 61: 265-274.

Polat S (1986) Ana-baba tutumlarnn renilmi aresizlikdzeyine etkisi. Unpublished Masters Thesis, Ankara: HacettepeUniversity.

Qadir F, Stewart R, Khan M et al. (2005). The validity of theparental bonding instrument as a measure of maternal bonding amongyoung Pakistani women. Soc Psychiatry Psychiatr Epidemiol, 40:276-282.

Rohner RP, Britner PA (2002) Worldwide mental healthcorrelates of parental acceptance-rejection: Review of cross-cultural

and intracultural evidence. Cross Cult Res, 36: 16-47.Uehara T, Sato T, Sakado K et al. (1998) Parental bonding

instrument and the inventory to diagnose depression lifetime versionin a volunteer sample of Japanese workers. Depr Anxiety, 8: 65-70.

Violato C, Arato J (2004) Childhood attachment and adolescentsuicide: a stepwise discriminant analysis in a case comparison study.Ind Diff Res, 2: 162-168.

Ylmaz A (2000) Anne-baba tututm leinin gvenirlik vegeerlik almas. ocuk ve Genlik Ruh Sal Dergisi, 7: 160-172.

Wilhelm K, Niven H, Parker G et al. (2005) The stability of theParental Bonding Instrument over a 20-year period. Psychol Med, 35:387-93.

Willinger U, Heiden AM, Meszaros K et al. (2002) Schizophrenia

and schizoaffective disorder, considering premorbid personality traits.Austr New Zeal J Psychiatry, 36: 663-668.