Sibling Dynamics in Academic Socialization Within the Family Context

Abstract

When it comes to academic socialization in the family context, many studies have addressed how parents influence their children’s educational outcomes (Jeynes, 2003), and a number of studies have shown that the ways in which parents influence their children’s educational outcomes differ by the macro influences of socioeconomic status or culture (Kim et al., 2020; Yamamoto & Sonnenschein, 2016). In comparison, less is known about the dynamics of academic socialization by individual children within a family.

Eccles and colleagues (Eccles, Arberton, et al., 1993) proposed a model of family influences on children’s motivation and achievement. The model hypothesized the joint influence of family characteristics and child/sibling characteristics on parents’ beliefs and behaviors, which in turn affect children’s educational development. The model highlighted bidirectional influences not only between parents and children but also between siblings. Whereas most relevant studies and reviews of the model (Eccles, 2007; Wigfield et al., 2015) have centered on effects of family characteristics and parents’ beliefs and behaviors on children’s educational outcomes, the current dissertation alternatively focused on the hypothesized effects of a child’s characteristics on parents and on the child’s own academic self-concept as well as siblings’ academic self-concepts. Drawing from the model suggested by Eccles and colleagues (Eccles, Arberton, et al., 1993), I tested four hypotheses in the current dissertation. First, I constructed hypotheses about each child’s individual experiences as well as a sibling’s shared experiences with their parents in the family. Second, I hypothesized effects of each individual child’s achievement on their own academic self-concepts and their sibling’s academic self-concepts. Third, I hypothesized effects of each individual child’s achievement and motivation on their parents’ beliefs and parental support. Fourth, I focused on effects of sibling resemblance on comparison processes between siblings and parents’ similar or differential support for the siblings’ learning. To address the four hypotheses, the current dissertation included three empirical studies.

The first study (Sibling Achievement as an Additional Frame of Reference for Parents’ Beliefs About Each Child’s Academic Ability) focused on parents’ beliefs about each sibling’s academic abilities. To provide empirical evidence for parents’ child-specific ability beliefs, the first study used secondary data (Gladstone et al., 2018) from 95 families collected in two academic-track secondary schools in a rural area of Germany and examined how much variance in parental beliefs could be explained by each parent’s child-specific response compared with the variance present at the child, parent, and family levels. To investigate the formation of parents’ child-specific ability beliefs, the study further addressed effects of siblings’ achievement on parents’ beliefs about each child’s academic ability. We explored whether each parent used the other sibling’s achievement as an additional frame of reference for their beliefs about a child’s abilities within and across domains by applying the internal/external frame of reference (I/E) model (Marsh, 1986b; Möller & Marsh, 2013). The results of a multilevel analysis showed substantial variance in parents’ beliefs at all levels of influence within the family. More specifically, parents’ beliefs about children’s academic abilities varied on the family level, indicating that parents’ beliefs differed from one family to another; varied on the parental level, indicating that parents’ general beliefs about siblings’ academic abilities differed between the mother and father; varied on the sibling level, indicating that both parents’ child-specific beliefs differed between siblings; and varied on the level of each child, indicating that each parent’s child-specific beliefs differed between siblings. These findings newly verified the hierarchical structure of parents’ child-specific beliefs about children’s academic abilities influenced by the multiple levels of influence from different socialization agents within the family. Additionally, the study provided tentative evidence that both the mother and father used each child’s own achievement and their sibling’s achievement as frames of reference for their beliefs about the child’s academic abilities within and across domains, yet the results did not show the clear pattern of the I/E model (Marsh, 1986b; Möller & Marsh, 2013). Overall, the findings of the first study are in line with the hypotheses proposed by Eccles and colleagues (Eccles, Arberton, et al., 1993). The second study (Why Do Siblings Differ in Their Learning Motivation and Perceptions of Parental Support? Reciprocal Relationships Between Parental Support and Each Sibling’s Learning Motivation) investigated whether the reciprocal associations between diverse dimensions of perceived parental support (parents’ expectations, parents’ emotional support, parents’ learning encouragement, and parental control) and children’s learning motivation differed between siblings. Using data from 2,082 school-aged twins collected at age 11 and age 13, we tested reciprocal associations within and between twin pairs across the 2 years, holding individual children’s characteristics of sex, school grades, and personality traits constant. By comparing monozygotic (MZ) and dizygotic (DZ) twins, we further verified whether the resemblance between siblings affected the reciprocal associations in within- and between-pair regressions. The results showed that there were significant reciprocal associations between an individual child’s perceptions of their parents’ learning encouragement and their learning motivation across ages 11 and 13, but only in the within-twin-pair regression for DZ twins. This means that the reciprocal relationships differed within DZ twin pairs, indicating the siblings’ individualized academic socialization with parents across the 2 years. Such effects were not observed for the other dimensions of parental support or for MZ twins. MZ and DZ twins differed in particular in effects of learning motivation at age 11 on their perceptions of parents’ learning encouragement at age 13. The substantial differences between MZ and DZ twins signified the effects of sibling resemblance on how parents supported the siblings in similar or different ways. Overall, the findings point to motivational dynamics and each child’s individualized interactions with their parents within the family.

The third study (What Happens With Comparison Processes When “the Other” is Very Similar? Academic Self-Concept Formation in Twins) investigated effects of individual children’s and their siblings’ achievement on their academic self-concepts, using data from 4,208 twins at age 11 and age 17. Applying the I/E model (Marsh, 1986b), the third study examined whether twin’s and co-twin’s academic achievement affected individual twin’s academic self-concept within and across domains. In addition, the study further compared MZ with DZ twin pairs to determine whether they differed in co-twin’s achievement effects on twin’s academic self-concept to test whether sibling resemblance moderates social comparison processes within twin pairs, in line with social comparison theory (Festinger, 1954). Drawing on the pattern suggested by the I/E model, both MZ and DZ twins used their own achievement as a frame of reference for their academic self-concepts within and across domains (Marsh, 1986b). Yet, only MZ twins further used their sibling’s achievement as an additional frame of reference for their academic self-concept within and across domains, showing the similar I/E pattern in effects of their own achievement on their academic self-concept. This finding provided new empirical evidence for a strong assimilation within perfectly similar sibling pairs, which is referred to and theorized as a mirror effect in the current study. Despite being similar to each other, DZ twins did not show the mirror effect but showed sibling deidentification or nonreferencing (Whiteman et al., 2007; Whiteman, McHale, et al., 2011) because they did not consider their sibling’s achievement as a frame of reference for their academic self-concept within and across domains.

Overall, the findings from the three studies showed the reciprocal effects of individual children’s and sibling’s achievement and learning motivation on their academic socialization from their parents and their academic self-concepts. Accordingly, the studies provided empirical support for the four hypotheses and for the model of family influences on children’s motivation and achievement proposed by Eccles, Arberton, et al. (1993). From a theoretical perspective, the studies produced new insights into the effects of children’s characteristics on the diversity in academic socialization between siblings, which has garnered relatively little attention to date. In addition, by conducting interdisciplinary research, the dissertation showed the relevance and usefulness of applying the I/E model and the transactional model of parenting to address how children’s and siblings’ characteristics reciprocally affect parenting and children’s educational outcomes. From a methodological perspective, multilevel analysis, cross-lagged modeling, and multiple-group analyses within a family are suggested to investigate dynamic academic socialization within the family. The dissertation suggests a new perspective on the model proposed by Eccles and colleagues (1993) and the best method to investigate dynamic academic socialization within the family.