Elsevier

Comprehensive Psychiatry

Volume 51, Issue 4, July–August 2010, Pages 434-442
Comprehensive Psychiatry

Perceived parental rearing style in childhood: internal structure and concurrent validity on the Egna Minnen Beträffande Uppfostran—Child Version in clinical settings

https://doi.org/10.1016/j.comppsych.2009.08.003Get rights and content

Abstract

Background

We provide the first validation data of the Spanish version of the Egna Minnen Beträffande Uppfostran—Child Version (EMBU-C) in a clinical context. The EMBU-C is a 41-item self-report questionnaire that assesses perceived parental rearing style in children, comprising 4 subscales (rejection, emotional warmth, control attempts/overprotection, and favoring subjects).

Methods

The test was administered to a clinical sample of 174 Spanish psychiatric outpatients aged 8 to 12. Confirmatory factor analyses were performed, analyzing the children's reports about their parents' rearing style.

Results

The results were almost equivalent for father's and mother's ratings. Confirmatory factor analysis yielded an acceptable fit to data of the 3-factor model when removing the items of the favoring subjects scale (root mean squared error of approximation <0.07). Satisfactory internal consistency reliability was obtained for 2 of the 3 scales, rejection and emotional warmth (Cronbach α >.73), whereas control attempts scale showed lower values, as in previous studies. The influence of sex (of children and parents) on scale scores was inappreciable and children tended to perceive their parents as progressively less warm as they grew older. As predicted, the scores for rejection and emotional warmth were related to bad relationships with parents, absence of family support, harsh discipline, and lack of parental supervision.

Conclusions

The Spanish version of EMBU-C can be used with psychometric guarantees to identify rearing style in psychiatric outpatients because evidences of quality in this setting match those obtained in community samples.

Introduction

It is widely recognized that the style of education experienced during growth is correlated with psychosocial development, and the association between dysfunctional parenting and pathology has been extensively studied [1], [2], [3]. Thus, the necessity has arisen to precisely conceptualize and measure this potential risk factor on mental disorders. Since Roe [4] reported her theoretical model as a circular continuum of 6 categories (loving, protecting, demanding, rejecting, neglecting, and casual), empirical research conducted over the past 5 decades has consistently identified 2 main dimensions of parental rearing: love vs hostility and autonomy vs control.

The Egna Minnen Beträffande Uppfostran (EMBU) is one of the most widely applied questionnaires to assess perceived parental rearing behaviors. It was originally created to evaluate the memories of adults during upbringing [5]. The 64-item version and a 22-item short form (S-EMBU) have been successfully adapted in a wide number of countries, including Spain [6], [7]. At the first-order level, the former comprises 4 factors (rejection, emotional warmth, overprotection, and favoring subjects), measured with some 18 items each, whereas the latter removes the favoring subjects scale, which has shown to be specific for some countries, and reduces the number of items per factor to 6 to 9. The rejection scale evaluates whether parents showed hostility, criticism, punishment, or verbal degradation. The Emotional Warmth scale refers to physical and verbal gestures of acceptance, stimulation, support and loving attention, and regard for the subject's point of view. The overprotection scale assesses parents' attempts to control the child's behavior, unreasonable worry about his or her safety and expectancy to know all about what the subject was doing, imposition of strict rules, and high expectations regarding the child's achievement. Finally, the favoring subjects scale refers to a more favorable treatment toward the child than toward other siblings.

To obtain present-day information of relevance to the psychopathology of children and adolescents, 3 new versions of EMBU were developed to assess current practices in children (EMBU-C; [8]), adolescents (EMBU-A; [9]), and parents (EMBU-P; [10]). Several studies have highlighted the importance of child reports because children's perceptions of parenting have been shown to be related to children's psychosocial adjustment [11] and children and parents are considered equally valid reporters [12]. These questionnaires are of special interest in clinical environments (see, eg, in Spain, references [13], [14], [15]), but to date, only incomplete data on its psychometric properties have been reported.

First of all, some doubts remain about the internal structure of the EMBU-C. On the one hand, the Spanish child version (41 items) evaluated in a community sample [8] showed a 4-factor structure, with factor loadings more than 0.40 in the expected factor, like the adult memory version adapted into different languages [6]. On the other hand, several modifications of Castro's EMBU-C have been made in Holland and Portugal, suggesting the overlap of the emotional warmth and control attempts dimensions (a new label for the original overprotection factor) [16], removing some items including those of the favoring subjects scale [17], [18], [19] or proposing a bifactorial structure of negative (rejection and control attempts) and positive (emotional warmth) rearing behaviors [18]. Thus, the proposed structures of the EMBU-C vary between 2, 3, and 4 factors, depending on the items that each version is composed of (ranging from 34 to 52 items). This makes it difficult to apply it to cross-cultural studies or to compare it to the adult EMBU and S-EMBU forms in longitudinal studies. From a psychometric point of view, this scenario is undesirable and should be redressed [20], [21].

Another pending matter is the collection of additional evidence on the relationship between emotional warmth and control attempts. The results for infant samples [8], [16], [17] (correlations from 0.20 to 0.46) do not agree with those obtained for adults [6], [7] or adolescents [22] (values from 0 to 0.14), with the exception of a sample of Mexican adolescents [23] (0.43 for fathers' reports and 0.24 for mothers' reports). Taken together, these results suggest that it is necessary to clarify the role of the age of the child and the sex of the parents on EMBU-C internal structure.

Moreover, to facilitate the interpretation of results, some relevant external variables must be related to EMBU-C results. In previous research, none of the scales showed an important relation with the sex of either the children or parents [6], [7], [8], [16], [17], [23], [24], and age was slightly inversely related to emotional warmth and control attempt scores [8]; but to our knowledge, there is a lack of published empirical evidences relating EMBU-C to other external variables in its nomological net. Logical analysis of the constructs led us to begin to fill this gap studying the relation to family adjustment: a negative education style characterized by high rejection and low emotional warmth could be related to problematic management practices as lack of supervision, harsh discipline, and unsatisfactory family environment.

Lastly, validation studies of these present-day versions have focused on psychometric data obtained from community samples [8], [9], [10], [25]. Support for the hypothesis that the questionnaire structure in people with psychological dysfunctions is identical to that found in people without disorders has been obtained for the earlier version of adults' memories of upbringing with a Dutch sample [6]; but to date, no study has been conducted with present-day information using the Spanish versions. Studying the psychometric properties of the child version in disordered patients would meet the American Psychological Association's recommendation to provide empirical evidence of psychometric properties in the particular setting in which the test is to be used [26]. This becomes of special importance as long as the EMBU's measurement model is based on classic test theory, and thus, the relevant psychometric indices are sample-dependent. Moreover, performing this local validation study provides data for validity generalization studies [26].

The purpose of this study is to test whether the Spanish version of EMBU-C can be used with psychometric guarantees to identify rearing style in clinical settings. Thus, the specific objectives are twofold: (a) to evaluate the internal structure and the internal consistency reliability of the Spanish version of the EMBU-C in a clinical sample and (b) to provide validity evidence in relation to the external variables of sex, age, and family adjustment. As mentioned earlier, we expect low influence of sex and age on scale scores and that levels of poorer family adjustment are related positively to rejection and negatively to emotional warmth scores.

Section snippets

Participants

The total sample was composed of 174 psychiatric outpatient children, 110 males and 64 females, recruited from 2 Primary Public Mental Health Care Centres in Barcelona (Spain), between January 1998 and May 2005. Participants were representative of the population that uses mental health services, in terms of age, sex, and socioeconomic status. Age of children ranged from 8 to 12 years (mean, 11.13 years; SD, 1.21 years). Socioeconomic status, which was based on the parents' educational level and

Results

With respect to the endorsement criterion [33], only item 74 did not reach the usual threshold (4.0% of negative responses in father's ratings). Nevertheless, it was maintained, in line with previous studies that only removed an item when the recommended item endorsement was not achieved for both ratings, about fathers and mothers [6], [22]. The influence of the skewed item on the factor solutions was checked by omitting it from analyses. The solutions of the remaining items did not differ

Discussion

The EMBU-C presented an adequate 3-factor structure (rejection, emotional warmth, and control attempts) in psychiatric settings, after removing the favoring subjects scale, in accordance with the model proposed by 2 previous studies [17], [19], and allowing item 20 of control attempts to also load on the rejection factor. All the items except item 50 of control attempts showed significant factor loadings, which were equivalent across the ratings given by the children for both parents except for

Acknowledgment

This research was partly supported by grant BSO2002-03850 from the Ministerio de Ciencia y Tecnología (Spain) and grant SEJ2005-01786 from the Ministerio de Educación y Ciencia (Spain).

We also thank the Unitat d'Epidemiologia i de Diagnòstic en Psicopatologia del Desenvolupament del Departament de Psicologia Clínica i de la Salut de la UAB for kindly making the database available.

References (51)

  • RoeA.

    Early determinants of vocational choice

    J Couns Psychol

    (1957)
  • PerrisC. et al.

    Development of a new inventory for assessing memories of parental rearing behaviour

    Acta Psychiat Scand

    (1980)
  • ArrindellW.A. et al.

    Parental rearing behaviour from a cross-cultural perspective: a summary of data obtained in 14 nations

  • ArrindellW.A. et al.

    The short-EMBU in Australia, Spain and Venezuela: factorial invariance and associations with sex roles, self-esteem and Eysenckian personality dimensions

    Eur J Psychol Assess

    (2005)
  • CastroJ. et al.

    Exploring the feasibility of assessing perceived parental rearing styles in Spanish children with the EMBU

    The International Int J Soc Psychiatr

    (1993)
  • CastroJ et al.

    Perceived parental rearing style in Spanish adolescents, children and parents: three new forms of the EMBU

  • CastroJ. et al.

    Assessing rearing behaviour from the perspective of the parents: a new form of the EMBU

    Soc Psych Psych Epid

    (1997)
  • GaylordN.K. et al.

    Parents' and children's perceptions of parental behavior: associations with children's psychosocial adjustment in the classroom

    Parent Sci Pract

    (2003)
  • GraneroR. et al.

    What single reports from children and parents aggregate to attention deficit-hyperactivity disorder and oppositional defiant disorder diagnoses in epidemiological studies

    Eur Child Adolesc Psychiatry

    (2008)
  • EzpeletaL. et al.

    Differential contextual factors of comorbid conduct and depressive disorders in Spanish children

    Eur Child Adoles Psy

    (2005)
  • Rojo-MorenoL. et al.

    Dysfunctional rearing in community and clinic based populations with eating problems: prevalence and mediating role of psychiatric morbidity

    Eur Eat Disord Rev

    (2006)
  • CanavarroM.C. et al.

    percepção dos filhos sobre os estilos educativos parentais: a versão Portuguesa do EMBU-C

    Revista Iberoamericana de Diagnóstico y Evaluación/E Aval Psicol

    (2007)
  • HambletonR

    International Test Commission Guidelines for adapting tests

  • van de VijverFJR

    Toward the next generation of instruments in cross-cultural testing: Recent developments in translations and adaptations

  • Márquez-CaraveoM.E. et al.

    Datos psicométricos del EMBU-I “Mis memorias de crianza” como indicador de la percepción de crianza en una muestra de adolescentes de la Ciudad de México

    Salud Ment

    (2007)
  • Cited by (11)

    • Adolescents' perceptions of parental behavior: Psychometric properties of the short Egna Minnen Beträffande Uppfostran-Adolescent version (S-EMBU-A) in a clinical sample

      2012, Comprehensive Psychiatry
      Citation Excerpt :

      Internal consistency was satisfactory, regarding the length of each scale. Mean interitem correlation and Cronbach α values of Overprotection were lower, as in all test validation studies in adults, adolescents, or children [2-8,11,12,14-20], but above the minimum acceptable level [44]. In relation to the mean scores, the ratings given to fathers were slightly smaller than those assigned to mothers for Emotional Warmth and Overprotection.

    View all citing articles on Scopus
    View full text