Le but de l’étude est d’adapter et de valider les sous-échelles de perfectionnisme positif orienté vers soi (PPOS) et de perfectionnisme négatif orienté vers soi (PNOS) de l’Échelle de Perfectionnisme Positif et Négatif (EPPN) (Seidah et al., 2002) pour les enfants de 6 à 9 ans. Quatre cent trente garçons et filles remplissent le questionnaire deux fois à un intervalle de deux semaines, ainsi qu’un questionnaire évaluant les attitudes et les comportements associés aux troubles des conduites alimentaires (TCA), et un autre mesurant le sentiment d’efficacité personnelle spécifique à l’alimentation (SEPA). L’analyse factorielle en composantes principales suggère une structure bifactorielle. La cohérence interne est bonne (PPOS = 0,76 et PNOS = 0,81) et la fiabilité test–retest est modérée (PPOS = 0,67 et PNOS = 0,66). Le PPOS n’est pas associé aux TCA, mais il est associé positivement au SEPA. Le PNOS est positivement lié aux TCA et négativement lié au SEPA. Les énoncés du PPOS et du PNOS possèdent des qualités psychométriques satisfaisantes. Ces sous-échelles représentent un outil important spécifiquement adapté pour les enfants francophones du Québec. Elles peuvent être utilisées en recherche pour améliorer l’état des connaissances quant au perfectionnisme chez les enfants et ainsi favoriser l’élaboration de mesures préventives. Perfectionism traits develop in childhood (Flett et al., 2002). Yet, research on perfectionism has typically focused on adults and adolescents with limited extension to children. The lack of research in children is explained by the fact that, until recently, no questionnaires existed to measure the trait in this population (DiBartolo et al., 2012). A variety of self-report measures have now been developed such as the Adaptive/Maladaptive Perfectionism Scale (Kenneth et Preusser, 2002), the Perfectionistic Self-Presentation Scale-Junior Form (Hewitt et al., 2011), the Child and Adolescent Perfectionism Scale (Flett et al., 1997; see DiBartolo et al., 2012) and the Positive and Negative Perfectionism Scale (PANPS) (Terry-Short et al., 1995). The Échelle de Perfectionnisme Positif et Négatif (EPPN) (Seidah et al., 2002) is the French version of the PANPS for adolescents. This questionnaire is the only one, which distinguishes positive from negative forms of self-oriented perfectionism. The present study aims to adapt and validate the positive self-oriented perfectionism (PPOS) subscale and the negative self-oriented perfectionism (PNOS) subscale from the EPPN for French-speaking children in Quebec aged 6 to 9. The participants were 430 boys and girls at primary school. The EPPN was adapted following Vallerand's validation procedure (Vallerand, 1989). The terms used in the statements were reworded by a committee of experts to correspond to the language skills of 6- to 9-year-old children. Symbols were also added to further facilitate the children's comprehension of the answer choice. A first pre-test was performed on two six-year-old children and a second pre-test was performed on a class of first graders. Then, 430 children filled the questionnaire twice over a two-week interval. They also completed a questionnaire measuring eating disorder attitudes and behaviours and a measure of eating self-efficacy. The principal components analysis shows a two-factor structure. At the first testing, factor 1 (negative factor) accounts for 17.83% of the variance and factor 2 (positive factor) accounts for 15.53% of the variance. At the second testing, factor 1 accounts for 20.53% of the variance and factor 2 accounts for 17.31% of the variance. The average scores were 31.95 (SD = 5.78) for the PPOS subscale, 20.36 (SD = 6.84) for the PNOS subscale, 11.20 (SD = 10.05) for the measure of eating disorders attitudes and behaviours and 3.88 (SD = 0.76) for the measure of eating self-efficacy. Cronbach's alpha indicates good internal consistency (PPOS = 0.76 and PNOS = 0.81) and correlation coefficients for test–retest reliability are moderate (PPOS = 0.67 and PNOS = 0.66). PPOS is not associated with eating disorders attitudes and behaviours but is positively associated with eating self-efficacy [r(398) = 0.19 (P < 0.001)]. On the other hand, PNOS is positively associated with eating disorders attitudes and behaviours [r(398) = 0.33 (P < 0.001)] and negatively associated with eating self-efficacy [r(398) = –0.16 (P < 0.001)]. Items measuring PPOS and PNOS have satisfactory psychometric properties. These two subscales are the first to be adapted and validated for French-speaking children, which take into account the positive and the negative function of self-oriented perfectionism. Perfectionism in children has limited scope in scientific literature and the lack of available instruments has been identified. Furthermore, knowing that perfectionism traits develop in childhood (DiBartolo et al., 2012), it would seem useful to make a measurement tool for children available. Thus, PPOS and PNOS subscales can be used in research to foster the understanding of perfectionism in children. A better understanding of perfectionism in this population could strengthen prevention efforts.