SIBLING WARMTH AS A CORRELATE OF YOUTH SELF-ESTEEM

Master’s Thesis, 30 ECTS

Master of Science in Psychology, 60 ECTS Fall 2018

Supervisor: Jeong Jin Yu

SIBLING WARMTH AS A CORRELATE

OF YOUTH SELF-ESTEEM

also appreciate the high standard you have set for me, and all the knowledge I was able to gain throughout this process with your guidance.

Abstract

During adolescence, self-systems and individual perceptions of the self are changing and heavily influenced by family relationships, which are generally the first and longest lasting connections individuals experience. Thus, to understand youth development, it is imperative to investigate the family context, wherein parent-youth relationships and sibling relationships impact one another. The current study examines the effect of a close sibling relationship on youth self-esteem above and beyond the effect of the mother-youth relationship. A cross-sectional study was employed on a sample of 434 families in the US, which all consisted of one mother and at least a pair of siblings. All three members of the family participating filled in an online survey. The target youth were between the 5th to 7th grades. Hierarchical multiple regression analyses find sibling warmth as a significant predictor for youth self-perception, but not for youth self-criticism. The present study shows the importance of considering not only the mother-youth relationship when analyzing youth self-system development, but also the sibling relationship.

Keywords: self-esteem, family dynamics, family context, mother-youth relationship, sibling relationship, maternal warmth, maternal power assertion, maternal favoritism

Abstrakt

Under adolescensen förändras självsystemet, eller individens uppfattning om det egna jaget, och är kraftigt influerat av individens familjerelationer. Dessa relationer är generellt de mest varaktiga relationer som en individ upplever. För att förstå ungdomsutveckling är det därför nödvändigt att undersöka familjekontexten, i vilken relationerna mellan förälder-ungdom och mellan syskon påverkar varandra. Denna studie undersöker effekten av en nära syskonrelation på ungdomars självkänsla utöver och bortom effekten från ungdomars föräldrarelation. Studien är en tvärsnittsstudie av 434 familjer i USA bestående av minst en mor och ett syskonpar. Dessa tre familjemedlemmar deltog alla i en webbaserad undersökning. Denna studies fokus är de av ungdomarna som vid undersökningens tidpunkt gick i årskurs 5 till 7. Med hierarkisk multipel regressionsanalys visas att en varm syskonrelation är en signifikant prediktor för ungdomars självuppfattning, men inte för ungdomars självkritik. Denna studie visar på vikten av att inte endast beakta relationen mellan mor och ungdom när ungdomars självsystem analyseras, utan även syskonrelationen.

Sibling warmth as a correlate of youth self-esteem

Regardless of belief, it is irrefutable that nurture has a hand in the development of children. Taking up a substantial portion of an individual’s socialization, familial relationships heftily contribute to nurture (Macionis, 2006), and as such, parental and sibling relationships are factors that contribute to the fostering of a child. According to Bowen’s (1974) family systems theory, a family is considered as a system that consists of three main relationships: marital, parent-child, and sibling. The familial relationships are generally life-long and one of the first relationships gained (Crocetti, Branje, Rubini, Koot, & Meeus, 2017), and are thus, an important factor in the development of an individual’s life. It is imperative to understand how these relationships interact with each other because of their different inherent structures – the sibling relationship is a dissimilar connection from the mother-youth relationship. For example, whereas there is a hierarchical structure for siblings during early childhood, by adolescence, this relationship generally becomes more egalitarian and equal (Stocker, Lanthier, & Furman, 1997; Buhrmester & Furman, 1990; Davies, 2014). Family systems theory considers that all relationships in the system may potentially impact one another (Bowen, 1974), and that comprehending these relationships are key to comprehending youth functioning (Sameroff, 1994; Thelen & Smith, 1994).

The impact of the mother-youth relationship on children has been well studied (Campione-Barr, Lindell, Greer, & Rose, 2014; Davies, 2014), but sibling relationships are lesser known. Most previous research has focused on the mother-youth relationship’s impact on youth development (McHale, Updegraff, & Whiteman, 2012), understandably, because mothers play a fundamental role in a youth’s life (Ruiz, Roosa, & Gonzales, 2002). However, sibling relationships largely contribute to youth psychosocial development (Noller, 2005). Recently while there is an increase in sibling relationship research (Buist & Vermande, 2014), according to Hakvoort, Bos, van Balen, and Hermmans (2010), most research has only looked at either mother-youth relationships or sibling relationships when attempting to understand family dynamics in relation to youth psychosocial adjustment. However, it is necessary to look at youth growth through the interaction of these two relationships. As sibling relationships impact youths, it is important to consider youth development in the context of these intertwining relationships.

There have been studies investigating the association between mother-youth relationships and sibling relationships, albeit the findings of the exact influence are discordant (Hakvoort et al., 2010). For example, while Boer, Goedhart, and Treffers (1992) suggest that distant mother-youth relationships are compensated by close sibling relationships, positing that the relationships can be conflicting, studies that support the perspective that the relationships build upon each other also exist – the positive aspects of the two relationships can be related (Boer et al., 1992; Brody, Stoneman, & McCoy, 1994; Bryant & Crockenberg, 1980; Pike, Coldwell, & Dunn, 2005), but a negative mother-youth relationship can also be correlated with a negative sibling relationship (Boer et al., 1992; Volling & Bleksy, 1992).

The self-system

Self-systems are an individual’s self-perception, such as self-esteem. This collection arises from caregiver-infant interaction, and helps an individual navigate and make decisions through judgements about the self and balancing his or her place in the world (Sullivan, 1953). Through self-perception, the individual understands how he or she may be perceived by others, and self-criticism allows the individual to pinpoint certain traits or behaviors in him or herself that should be changed. Self-reflection then allows him or her to reevaluate the modified self and compare if the new self is a better match to his or her target self-perception.

Creating a positive self-system necessitates establishing a confident self-esteem – a high self-esteem may serve as a protective factor for youth against negative life results (Ruiz et al., 2002). Furthermore, it helps youth develop confidence in their own abilities and their belief of the world as a positive place, which allows them to grow independent and solve problems on their own (Solomon & Serres, 1999). However, self-esteem may be more elusive to measure because of its emotion- and affect- related subjectiveness (Branden, 1987; McKay & Fanning, 1987). Indeed, the obscurity of self-esteem can be seen in its many definitions.

For instance, Iancu, Bodner, and Ben-Zion (2015, p.115) label self-esteem as “the extent to which one prizes, values, approves of, or likes oneself”. Likewise, Rosenberg (1965, p. 30) defines self-esteem as “a positive or negative attitude toward...the self”, while Cast and Burke (2002) posit that self-esteem is a positive or negative self-evaluation of the worth of an individual, suggesting that a low esteem equates to a low perception and self-evaluation. Indeed, individuals with lower self-esteem focus on self-criticisms and negative traits (Iancu et al., 2015). Thus, Ruiz et al. (2002, p. 70) suggest that self-esteem is visualized as an “innate mechanism” that helps to upkeep a high self-perception. More specifically, a strong self-esteem is key for psychological and emotional development.

On the other hand, Ju and Lee (2018, p. 354) define self-esteem a bit further as an “inner strength and self-awareness...when faced with danger” and suggest that self-esteem is an important factor for resilience. Indeed, the higher the self-esteem, the more resilient the individual, as he or she has confidence in his or her ability to succeed and overcome. Contrarily, Leary, Tambor, Terdal, and Downs (1995, p. 519) described self-esteem as an “an internal, subjective marker of the degree to which the individual is being included versus excluded by other people”, which fits into Leary’s (2005) proposed sociometer theory, describing how self-esteem is regulated based on the perceived inclusion or exclusion by others. Sociometer theory arises from Cooley (1902) and Mead (1934), who both believed that self-esteem was based on the perceptions an individual has on how they are “regarded and treated by other people” (Leary, 2005, p. 86). Indeed, this idea is supported by other studies (Harter, 1993) – rejection from others can impact youth self-esteem (Kita & Inoue, 2017). Additionally, self-esteem is lowered by experiences such as childhood abuse, wherein a youth is repeatedly reminded of their low value (Ju & Lee, 2018) and subsequent rejection from others, which becomes self-perceptions (Ju & Lee, 2010) thus lowering self-esteem (Leary, 2005).

Campbell (1999) suggests that self-esteem serves as a measure to judge and evaluate the self and is therefore an important scale for mental health. When an individual views him or herself positively, and thus has high self-esteem, then generally mental health is similarly positive. A high self-esteem is correlated with several positive outcomes, such as popularity, academic achievement, and sociability (Kokkinos & Hatzinikolaou, 2011). Additionally, self-esteem is a protective factor against negative results, such as delinquency (Rosenberg, Schooler, & Schoenback, 1989). Thus, the importance of self-esteem is unquestionable.

Hatzinikolaou, 2011; Proctor, Linley, & Maltby, 2009). Gordon (2017, p. 95) defines self-perception as “a collection of beliefs relating to how an individual generally feels about them self and their abilities”, whereas Sanchez-Sandoval (2015) suggests “perception, self-esteem and life satisfaction are concepts referring to the perception or assessment that a person makes of him/herself”, wherein these concepts are subjective and “emotional[ly] charge[d]” (Reina, Oliva, & Parra, 2010, as cited by Sanchez-Sandoval, 2015). Similarly, Kokkinos and Hatzinikolaou (2011, p. 349) suggest that self-perception “refer[s] to all types of self-referent statements about the self that are specific in content”, wherein these perceptions are based on the context and situations surrounding the individual, and their interaction with others (Marsh & Shavelson, 1985). Crocetti et al. (2018) find that high-quality family relationships may influence youth self-perception. Certainly, social relationships are important to the development and adaptation of the individual, which makes it even more relevant during the unstable period of adolescence (Kokkinos & Hatzinikolaou, 2011). This period of life is characterized by both physical changes and fluctuations in the self-systems – self-esteem declines (Sanchez-Sandoval, 2015), however, it generally recovers over time (Parra, Oliva, & Sanchez-Queija, 2004, as cited by Sanchez-Sandoval, 2015) and increases (Kokkinos & Hatzinikolaou, 2011). Furthermore, during this period, different experiences and circumstances lead to different outcomes in adolescent self-perceptions (Harter, 1988, 1999), supporting Leary’s (2005) sociometer theory, wherein adolescents use self-perception and self-evaluations to judge their outward selves, and how they would be perceived and accepted by others, then impacting the youth self-esteem.

Self-perception and self-esteem are also closely interwoven – self-esteem helps the individual regulate his or her self-perception through the cohesive matching of the self-concept with others’ perceived concept of the individual (Luan, Poorthuis, Hutteman, Asendorpf, Denissen, & van Aken, 2018). When youth self-perception matches the perceptions of others and elicits positive responses, the youth feels validated and worthy (Christy, Seto, Schlegel, Vess, & Hicks, 2016; Conger, Conger, & Scaramella, 1997; Crocetti, Moscatelli, Van der Graff, Rubini, Meeus, & Branje, 2016; Plunkett, Henry, Robinson, Behnke, & Falcon 2007), which influences to self-esteem. Moreover, not only is self-esteem based on outside acceptance or rejection, but self-perception also influences the inclusion or exclusion by others (Leary, 2005). For instance, youth self-worth is correlated to feelings of competence in areas that are important to his or her parents (Harter & Marold, 1991).

Self-criticism is defined by Iancu et al. (2015, p. 166) as “the negative inner voice that attacks and judges the individual and his or her actions”. Self-criticism and self-esteem, particularly low self-esteem, create a vicious cycle – a diminished self-esteem leads to more self-criticism which feeds back into an even lower self-esteem (de Jong, Sportel, de Hullu, & Nauta, 2012). Indeed, self-criticism is a risk factor for several mental illnesses (Shahar & Gilboa-Schechtman, 2007), suggesting the importance of a positive self-esteem via decrease in self-criticism and less focus on negative traits. For instance, Noordenbos, Aliakbari, and Campbell (2014) find that victims of eating disorders often develop an inner voice that supports patterns of unhealthy eating behaviors to bolster self-esteem. However, perfectionism, and thus unrelenting self-criticism, allows the eating disorder to persist even upon meeting a targeted weight standard.

unable to achieve success. In order to avoid the expected criticism from others, the individual avoids social interactions. The focus on criticism arises from a mismatch between self-perception and the expectations of others, resulting in a social anxiety that arises from the fear of a negative perception of the individual by others. Moreover, self-esteem is much easier to decrease than increase, and this may be because inclusion is a general default, so inclusion and positive feedback do not lead to as large an increase as exclusion does a decrease (Leary, 2005).

Developing a positive self-system is an especially important developmental task for adolescents, who are evolving their individuality and sense of self (Luan et al., 2018; Santrock, 2011; Sanchez-Sandoval, 2015). Harter (2007) posits that because adolescents have developed more sensitive self-reflection, adolescent self-systems should be considered separately from children. Moreover, adolescents are also comparatively more impacted by an unstable self-system (Harter, 2007), suggesting that adolescents cannot be lumped together with adults, and that there is a need for separation and specific study on the adolescent self-system.

Mother-youth relationships

Numerous studies emphasize the importance of the mother-youth relationship (Driscoll & Pianta, 2011; Garcia, Ren, Esteraich, & Raikes, 2017) – a positive relationship is associated with healthy emotional development (Dawson & Ashman, 2000) and high self-esteem (Coopersmith, 1967; Rosenberg, 1965), and academic achievement (Lopez-Turley, Desmond, & Bruch, 2010) in youths. Additionally, a good relationship with one parent may buffer the effects of marital conflict on the youth (Amato, 1986). For instance, when mothers are perceived to be more open and responsive, youths adapting to a step-family environment will experience more stability (Jensen & Shafer, 2013), indicating the importance and protective effect of a warm mother-youth relationship. Inevitably, the mother-youth relationship is key in youth self-esteem development.

The mother-youth relationship is especially important for the development of adolescent self-perception (Kokkinos & Hatzinikolaou, 2011). Particularly, the parenting style can affect the mother-youth interaction (Brand, Hatzinger, Beck, & Holsboer-Trachsler, 2009; Kokkinos & Hatzinikolaou, 2011) – a positive and effective parenting style is important for contributing to a quality mother-youth relationship (Kokkinos & Hatzinikolaou, 2011). According to Kokkinos and Hatzinikolaou (2011), there are two dimensions to parenting styles: “emotional support or rejection”, and “control or permission”. Emotional support may be measured by parental warmth and acceptance, while control may be measured by parental power assertion (Kokkinos & Hatzinikolaou, 2011). When parenting styles are high in warmth, youth self-perceptions are higher and adolescents view themselves as being more competent (Garber & Flynn, 2001; Heaven & Ciarrochi, 2008), while a neglectful and rejecting parenting style correlates with negative mental health, such as anxiety and depression, and poor self-perception and low self-esteem (Brand et al., 2009; Lamborn, Mounts, Steinberg, & Dornbusch, 1991). When parenting styles are high in control – which may lead to low independence in adolescents – this contributes to a lowered self-esteem (Heaven & Ciarrochi, 2008). Therefore, in the current study, the mother-youth relationship is operationalized by maternal warmth and maternal power assertion.

Maternal warmth. Maternal warmth affects youth development through the amount

symptoms (Heaven, Newbury, & Mak, 2004; Key, 1995), and with social competence, school adjustment, and academic achievement among youths (Chen, Liu, & Li, 2000). Thus, maternal warmth is a key ingredient in the mother-youth relationship and youth development.

Maternal power assertion. According to Grusec and Goodnow (1994), power

assertion is a discipline option that includes physical punishment, withdrawal of privileges or love or resources, or guilt and shame and humiliation, to name a few. Power assertion in this study will be defined as maternal aggression, such as physical, emotional, or verbal aggression on youths. Maternal aggression, especially through psychological, emotional, or verbal aggression, has long-lasting effects on youths (Solomon & Serres, 1999), such as leading to depressive (Briere & Runtz, 1988; Gross & Keller, 1992) and anxiety (Green, 1983) symptoms. Moreover, maternal aggression imbues feelings of incompetence and unworthiness, which impact youth self-confidence in school achievement and peer interaction (Solomon & Serres, 1999), thereby also impacting self-esteem (Khaleque, 2017; Wang, Liu, & Jin, 2015). Not only is maternal psychological aggression a predictor of adolescent depression (Gross & Keller, 1992), but it may be even more detrimental than physical aggression (Claussen & Crittenden 1991; Vissing, Straus, Gelles, & Harrop, 1991). Like maternal warmth, it is the youth’s perception of maternal power assertion that is important, and Lansford et al. (2010) find that youths’ perceptions of high maternal aggression correlate with youth maladjustment. Hence, maternal power assertion via aggression is an important measure when considering mother-youth relationship quality.

Mother-youth relationships and youths’ self-systems

In life, one of the first feedbacks received is the mother’s perception of the youth, leading to a possible lifelong guide for youths and the construction of their self-systems (Luan et al., 2018). Warmth in mother-youth relationships is generally regarded as a good foundation for high youth self-esteem; Chan and Lo (2016) suggest that mother-youth relationships are most important for youth self-esteem, and certainly, a positive mother-youth relationship and high youth self-esteem are correlated (Amato, 1986; Wilkinson, 2004). Maternal warmth and maternal support and acceptance, for instance, tend to help the development of positive self-esteem among youth (Bireda, 2015; Cornell & Grossberg, 1987; Crase, Foss, & Colbert, 1982; Farruggia, Chen, Greenberger, Dmitrieva, & Macek, 2004; Gecas, 1971; Sears, 1970), and correlates with increased motivation in school (Lowe & Dotterer, 2012).

Additionally, Harper, Arias, and House (2003), wherein the maternal warmth was positively correlated with youth self-esteem in victims and witnesses of family violence situations. The results (Bireda, 2015; Farruggia et al., 2004; Lowe & Dotterer, 2012) make maternal warmth and its effect on youth self-esteem a crucial predictor, as high maternal warmth and a close mother-youth relationship results in a healthy self-esteem in youth. However, when paired with maternal power assertion, it might be interesting to see the effects of high maternal warmth combined with high maternal power assertion, or lower maternal warmth and lower power assertion on youth self-esteem.

parental psychological aggression were significant, suggesting that the impact of parental aggression makes a lasting impression (Miller-Perrin, Perrin, & Kocur, 2009; Taillieu & Brownridge, 2013). These findings suggest that even with high parental warmth, high parental aggression may still lower youth self-esteem.

As there are various forms of maternal power assertion, such as psychological or physical, the effects on the youth are different – for instance, yelling and screaming rather than just raising the voice may indicate to the adolescent that the parent in unpredictable, which leads to feelings of insecurity and may threaten the adolescent’s own autonomy, and may actually potentially lead youths to develop counter to their mothers’ wishes (Grusec & Goodnow, 1994). Additionally, low power assertion parenting tactics are more likely to raise youths that become successful in the long run because high power assertion creates a model of anger and hostility, and perhaps high self-criticism, for the youths to learn from (Baumrind, 1971; Hoffman, 1970a, 1970b), therefore rendering the parenting tactic of high power assertion and inflexibility largely ineffective. Certainly, over time, such parenting strategies may segue into adolescent feelings of anxiety, depression, and self-punishment (Grusec & Goodnow, 1994). Moreover, psychoanalysts suggest that youths repress their feelings of anger when confronted with maternal frustration, which leads to self-punishment and self-criticism upon development (Grusec & Goodnow, 1994). The negative effects of parental psychological aggression last even as the child becomes older (Solomon & Serres, 1999), and such controlling, authoritarian parenting styles acts as a risk factor for depression and low self-esteem (Gross & Keller, 1992; Smokowski et al., 2014).

The mother-youth relationship has important outcomes for youth development (Cooper, 2009). A positive mother-youth relationship is indicative of a warm and supportive parenting style (Cooper, 2009), which correlates with positive adolescent outcomes (Aquilino, 2006; Hair, Moore, Garrett, Ling, & Cleveland, 2008), such as better mental health (Hair et al., 2008). Along similar lines, a poor mother-youth relationship relates to lower self-esteem among youth (Wissink, Dekovic, & Meijer, 2006) and lowered mental health (Brand et al., 2009; Lamborn, Mounts, Steinberg, & Dornbusch, 1991). Indeed, a positive self-perception and a positive father-youth relationship correlates with adolescent academic performance (Gordon, 2017). This agrees with social capital theory (Coleman, 1988), wherein parents can motivate youths with social strategies – parents steer adolescents to academic success, which increases youth self-perception, which cycles into determination when facing obstacles (Gordon, 2017). Hence, understanding the role maternal warmth and power assertion plays in the mother-youth relationship is crucial to understanding the effect the relationship has on youth self-systems.

Sibling warmth

found that a good sibling relationship correlates with better youth adjustment (Pike et al., 2005), while a poor sibling relationship correlates with poorer development and behavior (Snyder, Bank, & Burraston, 2005).

According to Crocetti et al. (2017), the mother-youth relationship, the sibling relationship, and the emerging youth self-systems can influence and be influenced by each other, signifying that the family relationships are interrelated. Indeed, Updegraff, Thayer, Whiteman, Denning, and McHale (2005b) suggest that sibling interactions, mother-youth interactions, and maternal stress are linked to children’s psychosocial adjustment. Moreover, a positive sibling relationship may compensate for the negative effects of a poor family context, such as single parenting or marital conflict (East & Khoo, 2005; Jenkins & Smith, 1990). In a similar vein, Brody (2004) suggests that sibling relationships affect both youth development and the amount of impact the mother-youth relationship has on the adolescent. For instance, even after taking mother-youth relationships into account, sibling conflicts exert an influence on children’s antisocial behavior (Bank & Burraston, 2001), and a positive relationship minimalizes the effects of stressful life events (Gass, Jenkins, & Dunn, 2007), suggesting that sibling relationships can affect mother-youth relationships and family dynamics.

A positive sibling relationship correlates with less youth maladjustment (Gamble, Yu, & Kuehn, 2011). While sibling aggression correlates with externalizing problems and maladjustment among youth (Snyder, Bank, & Burraston, 2005), sibling warmth is recognized as a protective factor, with quality sibling relationships associated with decreased youth maladjustment and lower internalizing and externalizing behavioral issues (Branje, van Lieshout, van Aken, & Haselager, 2004). Thus, Piotrowski et al. (2014) posits that sibling relationships may offer a refuge from difficult situations and relationships.

Interestingly, however, Piotrowski et al. (2014) find that while increased sibling aggression correlated with depressive symptoms, increased warmth was not found to be a protective factor. Similarly, Buist and Vermande (2014) also find that sibling warmth is not a protective factor, although this seems to only be relevant for conflictual sibling relationships, with high aggression and low acceptance. This suggests that a poor sibling relationship may result in negative development and youth maladjustment (Buist & Vermande, 2014). Furthermore, it is important to note that warmth and aggression were not on one continuous sliding scale, where the absence of one meant the presence of the other (Piotrowski et al., 2014). Both, however, are important to youth adjustment (Piotrowski et al., 2014).

Sibling warmth and self-esteem during adolescence

While self-esteem can be predicted by high quality relationships denoted by maternal warmth (Amato, 1986; Wilkinson, 2004; Hakvoort et al., 2010), youth development and well-being also hinges on the sibling relationship (Branje, van Lieshout, van Aken, & Haselager, 2004; Brody, 1998; Defoe et al., 2013; Gallagher, Updegraff, Padilla, & McHale, 2018; McHale et al., 2012; Patterson, 1984; Piotrowski, Tailor, & Cormier, 2014). For instance, Stewart (1983), and Teti and Ablard (1989) find that sibling attachment may take the place of an absent mother in the strange situation, suggesting that the sibling relationship is just as important as the mother-youth relationship. Particularly, it is the more egalitarian nature of the sibling relationship that is significant for the development of youth self-systems (Davies, 2014; Piotrowski, Tailor, & Cormier, 2014).

Hakvoort et al. (2010) find that the mother-youth and sibling relationships influence each other, and influence youth adjustment. However, sibling relationships may be even more important than the mother-youth relationship – the sibling relationship, specifically trust and alienation, can influence youth self-worth and is associated with depression, even after controlling for mother-youth trust and alienation, suggesting that the sibling relationship is a more important relationship than previously expected, and that it may influence the mother-youth relationship in family dynamics (Noel, Francis, & Tilley, 2018).

Not only is the sibling relationship important for family dynamics, the direction of the sibling relationship is even more crucial. Sibling relationships, like mother-youth relationships, consist similarly of warmth and conflict (Brody 1998; Buhrmeister & Furman, 1990) – sibling warmth is the closeness in the relationship, while conflict is aggression during interaction (Sanders, 2004). A warm sibling relationship serves as a protective factor for behavior problems in youth, whereas a conflictive sibling relationship is a risk factor in youth development (Buist, Dekovic, & Prinzie, 2013). Along the same lines, a warm sibling relationship relates to higher self-esteem for youths (Crocetti et al., 2017; Hakvoort et al., 2010; Sherman, Landsford, & Volling, 2006). Sibling aggression, on the other hand, influences youth adjustment via hostility, which is correlated with depressive symptoms, risky behavior, perceived self-worth, and perceived romantic competence (Gallagher, Uppdegraff, Padilla, and McHale, 2018). Sibling relationships characterized by conflict and low warmth presents a risk factor for youth development, suggesting that warmth and conflict in sibling relationships are important predictors (Buist & Vermande, 2014 ).

Maternal favoritism

Differential treatment often arises due to the “perceived intelligence, positive behaviors, creativity, and birth order” of the youth than “sex, physical appearance, or personality characteristics”, and is often shown through attention or praise, rather than through physical means such as material items (Zervas & Sherman, 1994, p. 30). It is an important factor in family dynamics – it impacts not only mother-youth relationships, but sibling relationships as well (Boll, Ferring, & Filipp, 2003). For instance, youths who were more heavily disciplined compared to their siblings exhibited more negative behaviors towards those siblings than youths who were treated equally or less disciplined (Boll et al., 2003).

Boll et al. (2013) describe two models that explore favoritism and its effects on family dynamics: the equity theory and the self-esteem maintenance theory. These two models attempt to explain favoritism and its different outcomes. The equity theory suggests that unless a youth is treated equally with the sibling, the sibling relationship will be perceived to be negative. However, whereas the perceived relationship with the mother for the disfavored youth will also be negative, the perceived relationship with the mother for the favored youth will be positive. The favored youth, in this case, may feel a sense of guilt for being favored while perceiving themselves and their sibling as worthy of being treated equally. This may help to understand Zervas and Sherman’s (1994) findings that equally treated siblings have a higher self-esteem than even the favored youth. Thus, the favored youth must discard this guilt by reasoning that differential treatment happens for a reason, leading to a negative perception of the disfavored sibling that lowers sibling relationship quality. This then also leads to a perceived positive relationship with the mother by the favored youth (Boll et al., 2003).

The self-esteem maintenance model suggests that being disfavored leads to a perceived negative relationship with both the sibling and the mother, while being favored leads to a perceived positive relationship with both the sibling and the mother. This hypothesis stems from the idea that being favored increases self-esteem, which thereby decreases the need for the favored youth to dislike or put down the sibling, leading to a perceived positive relationship between the siblings by the favored youth. Understandably, the disfavored youth may have a lower self-esteem due to social comparison with the favored sibling (Tesser, 1980). Because the disfavored youth is in low standing with the mother, the relationship with the mother is also diminished. In this case, both the sibling relationship and the mother-disfavored youth relationship are perceived as negative (Boll et al., 2003). Additionally, Harris and Howard (1985) suggest that favored youths perceive their mother more positively compared to disfavored youths, which strengthens Zervas and Sherman’s (1994) findings that perhaps disfavored youth do not find the mother-youth relationship to be worthwhile, and this relationship and the sibling relationship may be diminished (Boll et al., 2003).

Maternal favoritism and youth self-esteem

Through the equity theory and the self-esteem maintenance model, maternal differential treatment can be seen to cause rifts in family dynamics. Beyond just affecting relationships, favoritism also impacts youth development. According to Luan et al. (2018), a positive youth self-esteem requires a self-perception that matches with the perception of those around the youth. Indeed, symbolic interaction theory (Cooley, 1902; Mead, 1934; Sullivan, 1953, as cited by Zervas & Sherman, 1994) suggests that youth self-view hinges on maternal perception of the youth (Burr, Leigh, Day, & Constantine, 1979; Openshaw, Thomas, & Rollins, 1983; Plunkett et al., 2007). Furthermore, phenomenological theory posits that when the mother favors one youth over another, the self-worth of the disfavored youth may be questioned as the youth attempts to figure out the reason for differential treatment (Combs & Snygg, 1959; Rogers, 1959). When the self-view of the youth and the perception of the youth self-view by the mother do not match, youth self-esteem may be diminished (Conger et al., 1997).

Zervas and Sherman (1994) found that favoritism and self-esteem were indeed correlated, with disfavoritism leading to a lower self-esteem in youth, as was hypothesized by the self-esteem maintenance model (Boll et al., 2003). Interestingly, Neale (1986, as cited by Zervas & Sherman, 1994) finds no correlation between youth self-esteem and maternal favoritism. However, Harris and Howard (1985) posit that maternal favoritism may lead the disfavored youth to feeling upset, and Ross, Dalton, and Milgram (1980, as cited by Zervas & Sherman, 1994) suggest that favoritism may also lead to feelings of unattractiveness and ineptitude (Zervas & Sherman, 1994), which is understandably linked to self-perception and self-esteem (Zervas & Sherman, 1994). Furthermore, and perhaps more importantly, the effects of maternal favoritism on a youth’s self-esteem may be negated when the favoritism is only displayed by one parent, while the other parent shows either no favoritism or favoritism for another youth (Zervas & Sherman, 1994), suggesting that it may be possible that a positive relationship may mitigate the effects of a poor relationship on self-esteem. Maternal favoritism is therefore a necessary component in the goal of shedding some light on not only the mother-youth relationship, but also family dynamics and their relations to mother-youth self-systems.

The present study

There appear to be minimal studies on adolescents in the family context, even though family dynamics play a major role in adolescent life. According to Bowen’s (1974) family systems theory, an individual must be analyzed in the context of the family functioning as a system. Jensen and Shafer (2013) suggest that healthy families attain homeostasis where stability and instability are balanced. Where one relationship may be unstable and negatively affect an individual, a separate positive relationship may help balance the individual. Indeed, it has been seen that mother-youth relationships and sibling relationships can each impact each other (Brody, 2004; Crocetti et al., 2017; East & Khoo, 2005; Gass et al., 2007; Jenkins & Smith, 1990; Uppdegraff et al., 2005b).

The current study aims to investigate the youth self-esteem through the lens of a family context, which is a longstanding relationship system in an individual’s life (Crocetti et al., 2017; Macionis, 2006). Furthermore, most studies use either maternal reports or child reports. However, a more reliable way to obtain a clearer, bigger picture of the interaction and relationships in the family is to receive a report from the mothers on the adolescents and the sibling relationship, and also from the youths on the mother and their sibling relationship. As most individuals report based on his or her own perception (Epkins & Dedmon, 1999), it is important to gather different points of views. For instance, Piotrowski et al. (2014) find that mothers found sibling relationships to be more neutral compared to the siblings’ own perceptions of their relationship. Indeed, Furman, Jones, Buhrmeister, and Adler (1989) find that mothers and siblings each view their relationships differently. This study employs this method of survey incorporating reports of three family members (i.e., target youth, sibling, and mother) in order to gain a more accurate and holistic view of the family dynamics.

Demographic variables and their associations with youth self-esteem

The following demographic variables serving as controls were selected due to their possible influence on youth self-esteem.

Household income level. Household income level was selected because income levels

have an influence on parenting styles (Garcia et al., 2017; Ruiz et al., 2002) and mother-youth relationship quality (Silverberg & Steinberg, 1987), and thus, youth development and youth self-esteem. For instance, worrying about the future due to a low income may lead to maternal stress, which may then impact other areas of the mother’s life, such as relationship management, and so, the mother-youth relationship (Garcia et al., 2017). Wadsworth and Achenbach (2005) propose the social causation hypothesis, wherein a low socioeconomic status may lead to psychological issues and suggest that adversity related to a low income may lead to disorders such as depression (Mollica, McInnes, Poole, & Tor, 1998) or anxiety among children (Levav, Koh, & Schwartz, 1998). Indeed, youths belonging to a lower income class were more likely to develop mental health issues (Wadsworth & Achenbach, 2005), possibly including self-esteem (Heine & Lehman, 1999). Along a similar line, Ruiz et al. (2002) suggest that the environment surrounding families, such as low income, may lead such mothers to parent differently when compared to middle class mothers in order to achieve the same youth development results, such as high self-esteem. For this reason, household income was controlled in this study.

Maternal educational level. Maternal educational level was selected as a control

because education level and income level are often correlated, with more advanced degrees correlated with higher incomes (US Census Bureau, 2017). Similar to the logic of household income, maternal educational level correlates with youth achievement, which may relate to the self-esteem (Senler & Sungur, 2009), warranting its inclusion as a control variable in this study.

Birth order. Birth order was selected because previous studies have shown that it may

be a factor contributing to maternal favoritism (Chalfant, 1994; Zervas & Sherman, 1994), which may influence youth self-perception and self-criticism. Furthermore, Chalfant (1994) posits that the eldest child generally had higher scores in measures such as self-concept, emotional health, and happiness, perhaps suggesting that birth order has an effect on self-esteem.

Youth sex. Youth sex was selected as a control because while birth order seems to be

self-esteem (Amato, 1986). This may be due to the fact that boys externalize conflict behaviors while girls internalize such behaviors (Amato, 1986). The sex of the youth is also a factor in “sibling gender combinations”, which may impact sibling relationships (Buist & Vermande, 2014). According to Buist and Vermande (2014), previous studies have found sister sibling pairs to have higher sibling relationship quality (Aguilar, O’Brien, August, Aoun, & Hektner, 2001; Buist, 2010). However, this finding is inconsistent with other results (Buist & Vermande, 2014), which may find no difference in relationship quality based on sibling gender combinations (McGuire, McHale, & Updegraff, 1996; McHale, Whiteman, Kim, & Crouter, 2007; Sherman, Lansford, & Volling, 2006), and may be a result of having more positive-relationship sister sibling pairs participating in studies (Derkman, 2011; Whiteman & Loken, 2006). In any case, it would be interesting to examine the possible effect of the sex of the youth on self-systems.

Research hypothesis

Figure 1. The conceptual model shows the relationship between the independent variables and

Method Participants

A sample of 434 families from the United States participated previously in an online survey (Yu & Gamble, 2010). The current study was conducted on this database, which investigated the triadic relationship between one parent, specifically the mother, and their youths. Target youths were in the 5th – 7th grades, the adolescent period. Younger siblings had an average age of 11.6 (SD = 1.8), while the closest older siblings had an average age of 14.3 (SD = 2.1). The sample contained 209 older males, 230 younger males, 225 older females, and 204 younger females. Mothers of any age were considered, with the average age being 40 (SD = 7.3).

All ethnicities, economic statuses, education levels, and employment levels were included in the current study, although the sample consisted of 83.1% European American, 5.5% African American, 4% Hispanics, 3.9% Asian American, 1.8% Native American, and 1.7% as identifying from another ethnic group. The income was reported in ranges, such as $9,999 USD or less (= group 1) and $10,000 – 19,999 USD (= group 2), and so on until $100,000 USD or greater (= group 10). The median household income was between $60,000 to $69,999, with a mean of 6.89 (between group 6 and 7; SD = 2.87). Approximately 1% of mothers did not complete high school, 22% obtained a high school level education, 65% obtained a college, university, or vocational level education, and 12% obtained some graduate level education. Approximately 72.5% of mothers were employed either full-time or part-time.

To participate, families must have a mother and more than two children. If the family had multiple siblings, the target youth and the sibling should be those closest in age. The target youth is picked based on their age, which should be in the adolescent period. Only families wherein all three members have completed the survey will be eligible for participation in the current study.

Measures

Two separate surveys – one for youths and one for mothers – were used. Demographic questions were filled out at the beginning of the survey by mothers. Survey questions measure self-esteem via self-perception and self-criticism, the mother-youth relationship via maternal warmth and power assertion, the sibling relationship via sibling warmth, and possible favoritism among the youths by the mother. Survey questions were similar among the two variations, with the only difference being from either the youths’ or the mother’s perspectives. Incomplete surveys will be disregarded from the study.

Self-perception. Self-perception is reported by the youth and measured using the

Perceived Competence Scale for Children (Harter, 1982). For instance, youths receive the

statement “I am happy with myself most of the time,” and mark on a 4-point Likert scale how much in agreement with the statement he or she is, either 1 (not true of me), 2 (somewhat true

for me), 3 (mostly true for me), or 4 (very true for me). For this measure, Cronbach’s α was

0.83 for target youths, 0.83 for non-target youths, 0.82 for younger siblings, and 0.82 for older siblings.

Self-criticism. Self-criticism is measured using the self-criticism and efficacy subscales

for me). For this measure, Cronbach’s α was 0.77 for target youths, 0.68 for non-target youths,

0.71 for younger siblings, 0.76 for older siblings, and 0.71 for mothers.

Maternal warmth. Maternal warmth is measured with the personal relationship

subscale of the Parent-Child Relationship Questionnaire (PCRQ; Furman & Giberson, 1995). For instance, youths and parents receive a question such as, “How much do you and your mother go places and do things together?/How much do you and this child go places and do things together?” and mark on a 5-point Likert scale how true the statement is, either 1 (little

or none), 2 (somewhat), 3 (a lot), 4 (very much), or 5 (the most). For this measure, Cronbach’s

α was 0.90 for target youths, 0.92 for non-target youths, 0.79 for the mothers on the target youth, 0.90 for younger siblings on the mother and the older sibling, 0.91 for the older siblings on the mothers and the younger siblings, and 0.80 for the mother on the younger sibling, 0.82 for the mother on the older sibling.

Maternal power assertion. Maternal power assertion is measured using the Power

Assertion Parent-Child Relationship Questionnaire (PCRQ; Furman & Giberson, 1995). For

example, youths and parents receive a question such as, “How much does your mother yell at you for being bad?/How much do you yell at this child for being bad?” and mark on a 5-point Likert scale how true the statement is, either 1 (little or none), 2 (somewhat), 3 (a lot), 4 (very

much), or 5 (the most). For this measure, Cronbach’s α was 0.89 for target youths, 0.89 for

non-target youths, 0.89 for the mothers on the non-target youth, 0.90 for younger siblings, 0.90 for older siblings, and 0.90 and 0.92 for mothers on the younger and older sibling, respectively.

Sibling warmth. The sibling relationship is operationalized with sibling warmth, which

is based on the Sibling Relationship Questionnaire (SRQ; Furman & Buhrmester, 1985). For instance, youths and parents receive a question such as “How much do you and your sister or brother tell each other everything?/How much do these siblings tell each other everything?” or “How much do you admire and respect this sister or brother?/How much do these siblings admire and respect one another?” and mark on a 5-point Likert scale how true the statement is, either 1 (little or none), 2 (somewhat), 3 (a lot), 4 (very much), or 5 (the most). For this measure, Cronbach’s α was 0.96 for target youths, 0.96 for non-target youths, 0.96 for younger siblings, 0.96 for older siblings, and 0.96 for mothers.

Maternal favoritism. Maternal favoritism is measured using the differential parental

treatment subscale of the Sibling Inventory of Differential Experience (SIDE; Daniels, 1986).

For examples, youths and parents receive statements such as “Your mother punishes us for our misbehavior./Do you punish them for their misbehavior?” and mark on two scales, one 5-point Likert scale of how much the statement applies, either 1 (with me much more/with the oldest

much more), 2 (with me a bit more/with the oldest a bit more), 3 (with both of us the same/with both the same), 4 (with brother or sister a bit more/with youngest a bit more), or 5 (with brother or sister much more/with youngest much more). For this measure, Cronbach’s α was 0.79 for

target youths, 0.79 for non-target youths, 0.80 for younger siblings, 0.75 for older siblings, and 0.74 for mothers.

Procedure

The data was collected through an online survey via Global Market Insite (GMI), a platform for researchers to access and recruit interested participants. Before starting the survey, participants read and signed an online consent form, which allowed them to stop at any point in time. During the survey, mothers filled out the same form twice, once for younger sibling and once more for the older sibling. After completion of the survey, participants received compensation also through GMI.

the data does not identify or allow the participants to be tracked, and participants remained anonymous throughout the study. Participants were also informed of the purpose of the study. Because the study involved adolescents, mothers signed consent forms allowing their youths to participate.

Results

Preliminary analyses

The means and standard deviations for the study variables are displayed in Table 1. The sample size decreased from 434 to around 300 because target youths were selected by being within the adolescent range, determined to be between 5th to 7th grade. While siblings could be any age, in some sibling pairs, neither were within the targeted range. For instance, if a sibling pair was in 3rd and 8th grade, this pair was dropped from the sample, thus lowering the sample size. The mean scores for most measures are generally similar among the three respondents. For instance, for parental power assertion scores, the target youth (x_{̅ = 2.38), the sibling report} (x_{̅ = 2.41), the maternal report on target youths (x̅ = 2.35), and the maternal report on siblings} (x_{̅ = 2.40) are relatively close. The only noticeable difference in mean scores is for maternal} report on parental favoritism on siblings (x̅ = 1.95), while the other three parental favoritism mean scores are relatively similar (x_{̅ = 2.98 target youth report; x̅ = 2.88 sibling report; x̅ =} 2.35 maternal report on target youth).

Table 1

Means and standard deviations for study measures

N M SD

Maternal power assertion – TR 323 2.38 .96 Maternal power assertion – SR 321 2.41 .96 Maternal power assertion – MR on T 321 2.35 .89 Maternal power assertion – MR on S 323 2.40 .92

Maternal warmth – TR 323 4.28 .80 Maternal warmth – SR 321 4.26 .85 Maternal warmth – MR on T 321 4.26 .64 Maternal warmth – MR on S 323 4.21 .71 Maternal favoritism – TR 322 2.98 .50 Maternal favoritism – SR 320 2.88 .47 Maternal favoritism – MR on T 303 2.35 .97 Maternal favoritism – MR on S 299 1.95 .97 Sibling warmth – TR 323 3.18 .93 Sibling warmth – SR 321 3.21 .93 Sibling warmth – MR 449 3.28 .89

Note. TR = target youth report; SR = sibling report; MR = mother

report

perspectives and responses are similar. While sibling perception of maternal favoritism and the mother’s perceived favoritism on target youth are also positively and significantly correlated, target youth perception, sibling perception, and maternal perception of favoritism on target youth are negatively and significantly correlated. Similarly, maternal perception of favoritism on target youth and sibling are also negatively and significantly correlated. However, although maternal perception of favoritism on sibling and target youth perception of maternal favoritism are positively correlated, this figure is not significant. The same positive, non-significant correlation is seen for maternal perception of favoritism on sibling and sibling perception of maternal favoritism. Composite measures, except for maternal favoritism, are then created out of the significant correlations by averaging ratings given by each subject to represent the target youth’s family relationships for the hierarchical linear regressions. The subscale scores were transformed into z-scores and then averaged to produce the composite measures.

Table 2 Cross-informant correlations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Maternal power assertion – TR 2. Maternal power assertion – SR .46** 3. Maternal power assertion – MR on T .65** _.40** 4. Maternal power assertion – MR on S .50** _.60** _.73** 5. Maternal warmth – TR -.19** _{-.11 -.17}** _-.17** 6. Maternal warmth – SR -.06 -.25** _-.11* _-.24** _.52** 7. Maternal warmth – MR on T -.12* _{-.06 -.22}** _-.16** _.57** _.42** 8. Maternal warmth – MR on S -.05 -.12* _-.14* _-.21** _.51** _.53** _.76** 9. Maternal favoritism – TR -.30** _.14** _-.12* _.17** _{.02 -.20}** _{.09 -.10} 10. Maternal favoritism – SR .18** _-.22** _.18** _{-.08 -.08 .10 -.05 .03 -.34}** 11. Maternal favoritism – MR on T .06 -.09 .03 -.09 -.01 -.02 -.04 -.03 -.18** _.12* 12. Maternal favoritism – MR on S .01 .10 .04 .05 -.06 -.06 -.07 -.06 .02 .01 -.77** 13. Sibling warmth – TR -.05 -.20** _{-.01 -.08 .44}** _.38** _.32** _.32** _{.02 .09 -.05 -.08} 14. Sibling warmth – SR -.08 -.13* _{-.01 -.06 .33}** _.44** _.32** _.28** _{-.02 .08 -.01 -.11 .75}** 15. Sibling warmth – MR -.05 -.08 .07 .03 .27** _.24** _.39** _.33** _{.01 .05 -.00 -.12}* _.71** _.66** **. Correlation is significant at the 0.01 level (2-tailed).

*. Correlation is significant at the 0.05 level (2-tailed).

A multiple regression analysis was performed to explore the impact of several different independent variables (i.e., mother-youth relationship, sibling relationship, and maternal favoritism) and control variables (i.e., household income level, maternal educational level, birth order, and youth sex) on the dependent variable (target youth perception and self-criticism), using SPSS.

Sibling warmth and youth self-perception

A hierarchical regression was performed to investigate if youth self-perception would be influenced by the sibling relationship, even after controlling for the mother-youth relationship, household income, maternal education level, birth order of the youth, and youth sex. The overall regression model is significant (F(11, 261) = 3.99, p < 0.05), which can be found in Table 3.

Model 1 included the control variables such as household income, maternal education level, target youth birth order, and target youth sex. Model 1 did not find any significant predictors from the control variables. The R2 _{for the first model indicates that this model only} explains 1.7% of the variance of the data from the base model. Model 2 added the maternal power assertion, maternal warmth, and maternal favoritism variables. Model 2 indicates that maternal power assertion (β = -.14, p < .05), maternal warmth (β = .19, p < .01), and target youth report of maternal favoritism (β = .14, p < .05) are significant predictors, with an R2 indicating that the model explains 11.3% of the variance. Finally, sibling warmth was added in model 3, which finds that maternal power assertion (β = -.14, p < .05), target youth report of maternal favoritism (β = .13, p < 0.05), and sibling warmth (β = .20, p < 0.01) were significant predictors. In total, all the predictors accounted for 14.4% of the variance in youth self-perception.

Table 3

Hierarchical regression results for target youth self-perception

Model B SE β R2 _{ΔR F t}

(Constant) 2.73 .39 .14 .03 3.99*** 7.02

Household income .001 .01 .00 .07

Maternal education level .01 .02 .02 .35

Target youth birth order .03 .04 .05 .76

Target youth sex (female) -.09 .06 -.09 -1.49 Maternal power assertion -.09 .04 -.14 -2.35*

Maternal warmth .06 .04 .09 1.35 Maternal favoritism – TR .13 .06 .13 2.10* Maternal favoritism – SR -.08 .07 -.08 -1.24 Maternal favoritism – MR on T .04 .05 .07 .70 Maternal favoritism – MR on S .01 .05 .01 .12 Sibling warmth .12 .04 .20 3.07***

Note. All statistics are taken from the final step in the analysis. TR = target youth report;

SR = sibling report; MR = mother report. * p < .05. ** p < .01. *** p < .001.

low (VIF = 1.35, VIF < 4.00), which excludes multicollinearity. This model shows that a sibling warmth and target youth self-perception are positively correlated, indicating that sibling warmth may have a significant positive effect on target youth self-perception. This finding is consistent with the hypothesis that sibling warmth would impact target youth self-perception even after controlling for maternal power assertion, maternal warmth, maternal favoritism, household income, maternal education level, birth order of target youth, and sex of target youth.

Sibling warmth and youth self-criticism

A second hierarchical regression was performed to investigate if youth self-criticism would be influenced by the sibling relationship, even after controlling for the mother-youth relationship, household income, maternal education level, birth order of the youth, and youth sex. The overall regression model is significant (F(11, 260) = 4.18, p < 0.05), which can be found in Table 4.

Model 1 included the control variables such as household income, maternal education level, target youth birth order, and target youth sex. Model 1 finds that the control variable, target youth sex (β = .12, p < 0.05), was a significant predictor. That is, female adolescents are likely to report higher self-criticism than male adolescents. The R2 _{for the first model indicates} that this model only explains 1.8% of the variance of the data from the base model. Model 2 added the maternal power assertion, maternal warmth, and maternal favoritism variables. Model 2 indicates that maternal power assertion (β = .28, p < 0.001) and target youth report of maternal favoritism (β = -.13, p < 0.05) are significant predictors, with an R2 indicating that the model explains 15.0% of the variance. Finally, sibling warmth was added in model 3, which also finds that maternal power assertion (β = .27, p < 0.001) and target youth report of maternal favoritism (β = -.13, p < 0.05) are significant predictors. On the other hand, sibling warmth was negatively correlated with youth self-criticism and not significant (β = -.03, ns). All the predictors collectively explained 15.0% of variance in youth self-criticism.

Table 4

Hierarchical regression results for target youth self-criticism

Model B SE β R2 _{ΔR F t}

(Constant) 1.98 .51 .15 .00 4.18*** 3.86

Household income .01 .02 .03 .40

Maternal education level -.00 .03 -.01 -.11

Target youth birth order -.01 .05 -.01 -.18

Target youth sex (female) .10 .08 .08 1.28

Maternal power assertion .23 .05 .27 4.53***

Maternal warmth -.04 .06 -.05 -.75 Maternal favoritism – TR -.17 .08 -.13 -2.09* Maternal favoritism – SR .12 .09 .08 1.29 Maternal favoritism – MR on T -.02 .07 -.03 -.27 Maternal favoritism – MR on S .06 .07 .09 .94 Sibling warmth -.02 .05 -.03 -.43

Note. All statistics are taken from the final step in the analysis. TR = target youth report;

The Durbin-Watson score is acceptable (d = 2.05, 1.50 < d < 2.50), suggesting that the significance of sibling warmth can be considered accurate as there is no auto-correlation. The tolerance score for sibling warmth is rather high (= 0.74, > 0.20), and the VIF value is rather low (VIF = 1.35, VIF < 4.00), which excludes multicollinearity. This model shows that a sibling warmth and target youth self-criticism are not significantly associated with each other. This result is not consistent with the hypothesis that sibling warmth would impact target youth self-perception even after controlling for maternal power assertion, maternal warmth, maternal favoritism, household income, maternal education level, birth order of target youth, and sex of target youth.

Discussion

It was hypothesized that a warm sibling relationship would positively affect youth self-perception and self-criticism, even after controlling for the mother youth-relationship, maternal favoritism, and control variables including household income level, maternal educational level, target youth birth order, and target youth sex. The results show that while sibling warmth was a significant predictor for youth self-perception, it was not for youth self-criticism. In other words, while sibling warmth maintained an influence on youth self-perception above and beyond the mother-youth relationship, the same effect was not found for youth self-criticism. These results suggest that perhaps a warm sibling relationship may compensate for a poor mother-youth relationship in youth perception development, but not for youth self-criticism.

The result for youth self-perception is significant and consistent with previous studies suggesting the importance of the sibling relationship. For instance, Piotrowski et al. (2014) finds that low self-perception and self-esteem is correlated with a lack of sibling warmth. More importantly, these results indicate that a warm sibling relationship may serve as a protective factor against a poor mother-youth relationship. On the other hand, the result for youth self-criticism was not found to be significant. It is curious that sibling warmth did not impact the mother-youth relationship and youth criticism, since esteem, criticism, and perception overlap and intermix in the system. A low perception and high self-criticism are often cyclic, with low self-perception leading to self-self-criticism, which further drive self-perception under (Iancu et al., 2015).

The preliminary results find that the reports among the three respondents – the target child, the sibling, and the mother – were fairly similar, indicating that the cross-informants were relatively accurate in perceiving and describing their family dynamics (Yu & Gamble, 2010). Individual perceptions of family relationships are subjective and can be biased, thus, the parallel answers allow for not only a clearer picture of the family dynamics from separate perspectives but makes sure that the perceptions are accurate and correspond with the experience of other family members (Yu & Gamble, 2009). Maternal favoritism was the only predictor that was not matched among the cross-informants. Understandably, the youth can perceive the mother as being unfair and favoring the sibling when in fact it may not be true (Harris & Howard, 1985; Salmon, 2003; Salmon, Shackelford, & Michalski, 2012). Additionally, it may be difficult for a mother to admit differential treatment of her children (Coldwell & Dunn, 2008; Daniels, Dunn, Furstenberg, & Plomin, 1985; Gaylord, Kitzmann, & Coleman, 2003;Pike, Reiss, Hetherington, & Plomin, 1996; Salmon et al., 2012).

maternal power assertion via psychological aggression often leaves negative consequences, such as low self-esteem and depression (Briere & Runtz, 1988; Gross & Keller, 1992; Smokowski et al., 2014; Solomon & Serres, 1999), and is often long lasting (Solomon & Serres, 1999). Indeed, Downs and Miller (1998) find that paterntal verbal and physical aggression led to women’s later lowered self-esteem. According to the symbolic interaction perspective (Burr et al., 1979; Openshaw & Thomas, 1990) and Cooley’s (1902) “looking-glass self”, adolescents develop a view both of him or herself and his or her environment through interaction with his or her surroundings and with others (Mead, 1934; Sullivan, 1953). As a result, the mother and sibling perceptions of the youth and the resulting actions and responses are internalized by the adolescent, and help to develop the youth’s own sense of him or herself (Mead, 1934; Plunkett et al., 2007). The youth takes the perceptions of significant others in combination with his or her own sense of self and creates a holistic picture of him or herself, forming a subjective evaluation of his or her self-worth (Openshaw & Thomas, 1990; Thomas, 1928). Accordingly, youth perception of high maternal power assertion send the message that the mother perceives the youth as being incompetent and unworthy, who will only become competent through obedience (Barber & Harmon, 2002; Barber, Stolz, & Olsen, 2005, as cited by Plunkett et al., 2007), and is therefore a risk factor for the adolescent development of low perception, self-evaluation, or self-worth (Christy et al., 2016; Conger et al., 1997; Crocetti et al., 2016; Plunkett et al., 2007), and self-esteem (Garber, Robinson, Valentiner, 1997). Indeed, Barber (1996) and Plunkett et al. (2007) find that maternal power assertion is a risk factor for depression among youths through lowered self-esteem. For boys, this may be due to a perceived lack of independence and autonomy, while for girls, this may be due to an inability to develop self-competence and the loss of the caregiver relationship (Barber, 1996). Through low self-esteem, girls may perceive power assertion and control as the parent viewing them as being incompetent, and unworthy of affection and proper responses to their needs. Being constantly reminded of incompetence and rejection by the mothers, understandably, leads to low self-esteem (Leary & Baumeister, 2000). Thus, the significant role of maternal power assertion in shaping youth self-systems found in the current study suggests that the effect of maternal power assertion be especially considered when investigating youth self-systems.

into question in an attempt to figure out why the parent favors one child over another (Combs & Snygg, 1959; Rogers, 1959).

The present study suggests that sibling warmth may be a significant predictor for youth self-systems – a warm sibling relationship may counteract a poor mother-youth relationship’s impact on youth self-perception. Certainly, self-esteem and self-perception and self-criticism relies on the expectation of others (Luan et al., 2018) – when the youth’s sense of self does not match the perception by the mother, self-worth is decreased (Burr et al., 1979; Cooley, 1902; Mead, 1934; Openshaw et al., 1983; Sullivan, 1953). However, if the youth’s perception of him or herself is matched by a sibling, adolescent self-perception may be maintained despite the dissimilarity in maternal perception of the adolescent. Indeed, family systems theory suggests that familial relationships have an effect on one another (Bowen, 1974), thus highlighting the necessity of viewing youths through the family dynamic lens. On the other hand, sibling warmth was not found to be a significant predictor for youth self-criticism. The cognitive model of social anxiety disorder has suggested that self-criticism is a consequence of the divergence between the adolescent self-perception and the perception of the youth by significant others – when such a mismatch occurs, the youth may self-criticize in attempt to correct the discrepancy in perceptions (Hofmann, 2007; Morrison & Heimberg, 2013). However, similarly to Miller-Perrin et al.’s (2009), and Taillieu and Brownridge’s (2013) findings on the persistence of the impact of maternal power assertion on youth self-esteem despite maternal warmth, sibling warmth may not be enough to mitigate this discrepancy and thus, the self-criticism.

Demographic variables such as household income, maternal educational level, birth order, and youth sex may have an effect on youth self-systems, and as such, were included in the present study, albeit the results indicate that these factors were not significant. While youth birth order can contribute to maternal differential treatment (Chalfant, 1994; Zervas & Sherman, 1994) – shown to be a significant predictor in both regressions – the impact of birth order on youth self-perception and self-criticism was not significant. Similarly, although youth sex was included because previous studies found mothers favoring sons (Chalfant, 1994), which is associated with an impact on youth self-esteem, and that youth sex may also affect sibling relationships in the case of sisters having higher relationship quality (Aguilar, O’Brien, August, Aoun, & Hektner, 2001; Buist, 2010), the sex of the youth also had a non-significant influence on youth self-esteem. Household income and maternal educational level are often correlated (US Census Bureau, 2017), with household income having an impact on parenting styles (Garcia et al., 2017; Ruiz et al., 2002) and thus the mother-youth relationship (Silverberg & Steinberg, 1987) and youth self-systems. Moreover, the social causation hypothesis proposes that a poor socioeconomic status can bring on stress that leads to psychological issues such as depression (Mollica et al., 1998) and anxiety (Levav et al., 1998) for the mother and youth (Wadsworth & Achenbach, 2005). However, the current study did not find significant results for household income and maternal educational level on youth esteem. Hence, youth self-perception and self-criticism may be less affected by these demographic variables, at least during adolescence.

Limitations