Mother-son interactions predictive of high parentification self-reported by adolescent sons
Référence bibliographique 
Johnson, Darrell R. 2000. «Mother-son interactions predictive of high parentification self-reported by adolescent sons». Thèse de doctorat, Montréal, Université de Montréal, Département de psychologie.
Questions/Hypothèses « The main hypothesis of this study held that high frequencies of behaviours thought to be congruent with parentification would be predictive of high parentification scores. A second hypothesis predicted the stability of observational ratings of high parentification, and a third predicted that highly-parentified sons would more requently be only children living with single mothers.»
Type de traitement des données : Analyse statistique
« The present study defined parentification as a child’s enmeshment with and overresponsibility for a parent. Sibling and household responsibilities, often considered in the literature to be components of parentification, were judged by the author to better translate a parental than a parentified role. Although informally identified in clinical settings as a problematic parent-child dynamic, parentification has remained empirically vague. The present study represented the first effort to use observations of mother-son interactions in early- and mid-adolescence to predict high parentification near the end of adolescence. Subjects were 82 mothers and their sons who had participated in two laboratory learning tasks in the context of the Montreal Longitudinal Study. Videotapes of mothers helping their sons complete a computer task, which occurred when sons were 13–14 years old and again when they were 15–16 years old, were studied for parentifying and parentifed behaviours in mother and son respectively. The Parentification Scale (Mika, Bergner & Baum, 1987), translated into French for the present study, was administered to the sons when they were 18–19 years old. [...] Four of the mother-son behaviours observed in the lab setting were found to be stable across time, offering support for the second hypothesis. Two behaviours initiated by sons towards their mothers were found to discriminate between the High Parentification (n = 18) and Low Parentification (n = 64) groups, thus offering support for the main hypothesis : higher-than-average frequencies of eye contact, significant both times (967; 2 = 8.17, p inf. .01), and peer-like teasing and flirting behaviours, significant only when the sons were younger. High and Low Parentification groups did not differ significantly on a series of demographic variables and earlier behavioural ratings of the sons. One further discriminator was ’only child’ status; a greater proportion of sons who were only children were in the High Parentification group, 967;2 = 6.66, p inf. .01, although numbers were quite low. Living with a single parent was not found to be associated with higher parentification, and the amount and quality of social and emotional support available to the parent was hypothesized to be a potentially more valid determinant of parentification than would be living in a single-parent or a two-parent family. Finally, in order to explore the relative contribution of each of the three significant discriminators of parentification, they were entered as predictors into a logistic regression equation. The resulting model had a chi-square statistic of 18.6, p inf. .001, and a McFadden’s-R 2 of .27. Odds ratios indicated that sons having peer-like interactions with their mothers at T1 were six times more likely, and that sons who initiated higher-than-average amounts of eye contact with their mothers over the course of both tasks were over three times more likely to be in the High Parentification group. ’Only child’ status predicted an almost seven times greater chance of high parentification. One limitation of the study was a sample size which was too small to permit a validation of the translated Parentification Scale and which contained few examples of what might be considered pathogenic parentification. Methodological limitations included the failure to administer a parallel measure of parentification to the mother and the retrospective nature of mother-son observations. Several possible methodological improvements were discussed and recommendations for future research were suggested. » (Dissertation abstract CD-ROM)