Effects of Maternal Parenting and Mother-Child Relationship Quality on Short-Term Longitudinal Change in Self-Regulation in Early Adolescence

Abstract

The purpose of the present study was to explore the degree to which short-term longitudinal change in adolescent self-regulation was attributable to maternal parenting and mother-child relationship quality. A total of 821 mother-adolescent dyads provided data in the 1992 and 1994 waves of the Children of the National Longitudinal Survey of Youth-1979 (52.5% male; 24.2% Hispanic, 36.7% African American, 39.1% European American; adolescents’ initial age range = 10-12 years). Consistent with hypotheses, longitudinal improvements in young adolescents’ self-regulation were associated with high levels of mother-child relationship quality and low levels of maternal discipline. The association between self-regulation in 1992 and 1994 was moderated by child sex and maternal discipline. Thus, this study provides further evidence favoring the exploration of the parent-child relational context in addition to discrete parenting behaviors in studies on self-regulation during the early adolescent years.

Keywords

self-regulation NLSY-79 discipline warmth maternal parenting relationship quality

During adolescence, self-regulation involves youth’s abilities to control their emotions, behavior, and attention (Raffaelli & Crockett, 2003). Teenagers who have ample self-regulatory abilities are more likely to demonstrate social-emotional and behavioral adjustment than their less well-regulated peers (e.g., reduced likelihood of delinquency, depression, substance use and sexual risk-taking, and increased likelihood of academic achievement and prosocial behaviors; Brody & Ge, 2001; Finkenauer, Engels, & Baumeister, 2005; Moilanen, 2007; Raffaelli & Crockett, 2003). Although the links with adjustment are well defined, considerably less is known about the forces that may shape the ongoing development of self-regulation during the teen years. Contributions of potential socialization sources have been explored in primarily cross-sectional studies, and longitudinal inquiries have not consistently controlled for prior levels of self-regulation. Thus, insight about the degree to which these forces explain individual change in self-regulation remains limited. Accordingly, the primary goal of the present study was to examine the degree to which two maternal parenting behaviors and one feature of the mother-child relationship explained longitudinal change in adolescents’ self-regulation over 2 years. A secondary goal was to explore whether these effects were moderated by child sex, race/ethnicity, and previous levels of self-regulation. These objectives were met through replicating and extending an existing 1-year longitudinal study conducted with a mid-sized sample of at-risk boys (i.e., Moilanen, Shaw, & Fitzpatrick, 2010), results of which are discussed in the literature review below.

Literature Review

Theoretical support for the present study is provided by the tripartite model of familial influences on emotion regulation (Morris, Silk, Steinberg, Myers, & Robinson, 2007), which stipulates how children’s and adolescents’ emotion regulation is the product of their experiences in the family context (i.e., observation, parenting practices, and the emotional climate of the family). Although not a test of the full model, the present study focuses on two elements of those familial contextual experiences, namely parenting practices (i.e., maternal warmth and discipline) and the mother-child relational context (i.e., mother-child relationship quality). Warm, supportive maternal parenting and close mother-child relationships should promote the development of self-regulation through reducing teens’ negative affect, which should consequently afford youth opportunities to develop and master strategies for self-control (Baumrind, 1991; Brody & Ge, 2001). On the other hand, parental discipline or antagonism likely generates excessive negative affect that teenagers then need to control; focusing on repairing their negative emotions distracts them from learning how to regulate their emotions while experiencing parental discipline (Scaramella & Leve, 2004). Per Morris et al. (2007), these features of the home environment are interrelated and are shaped by the parents’ individual characteristics (e.g., personality traits), an influence not considered in the present investigation. Further, the child’s emotion regulation and adjustment-related behaviors bidirectionally impact each other and the family context. In addition to their direct effects, the child’s other individual characteristics (e.g., sex, age, and temperament) moderate the associations between the familial context and the child’s emotion regulation. Finally, the present study also explored such moderated pathways stipulated by Morris et al. (2007), specifically focusing on child sex, race/ethnicity, and prior self-regulation. Evidence for each of these associations is described below.

Maternal Warmth

In the present study, maternal warmth was conceptualized as the degree to which mothers reported engaging in physical affection and giving praise to their adolescent children. Associations between high levels of maternal warmth and youth’s self-regulation are primarily demonstrated in cross-sectional studies of children and youth. The one longitudinal exception involves the present study’s sample, and revealed that high levels of maternal warmth assessed in early childhood were linked to high levels of self-regulation in middle childhood and rank-order improvements in self-regulation between early and middle childhood (Colman, Hardy, Albert, Raffaelli, & Crockett, 2006). However, in their 1-year longitudinal investigation of an at-risk sample of boys, Moilanen and colleagues (2010) identified only cross-sectional associations between boys’ self-regulation at age 10 and observers’ reports of maternal regulatory supportive parenting during a mother-son discussion task (e.g., the degree to which mothers were responsive, empathetic, and engaged during a conversation about sources of familial conflict). Along these lines, in an analysis of three annual waves of data collected from typically-developing adolescents and their parents, self-regulation and maternal regulatory support were positively correlated at each study wave, but maternal parenting did not predict change in self-regulation at any point across the 2-year study period (Moilanen & Padilla-Walker, 2013). Similar cross-sectional findings were echoed in investigations of other samples of teenagers employing both observational and questionnaire-based measures of positive parenting practices (Crossley & Buckner, 2012; Doan, Fuller-Rowell, & Evans, 2012; Finkenauer et al., 2005; Vazsonyi & Belliston, 2007). In short, while cross-sectional data suggest linkages between positive forms of maternal parenting and youth’s self-regulation, the longitudinal data do not confirm such associations. Consequently, the role of maternal warmth was considered in the present two-wave longitudinal study, in which maternal disciplinary practices and mother-child closeness were also considered.

Maternal Discipline

Maternal discipline was conceptualized as the degree to which mothers used various forms of disciplinary behaviors during the week prior to being surveyed (e.g., spanking, withdrawing privileges, etc.). Compared with positive parenting practices, evidence for effects of maternal disciplinary practices on adolescent self-regulation is quite robust. For example, Brody and Ge (2001) reported that adolescents’ high levels of short-term self-regulation were contemporaneously related to low levels of both mothers’ and fathers’ harsh or antagonistic parenting methods, which were assessed through parent- and child-report questionnaires. Turning to longitudinal linkages, in their analysis of the present study’s sample during early to middle childhood, high levels of maternal physically punitive discipline in early childhood explained slower change in children’s self-regulation over 4 years (Colman et al., 2006). During adolescence, Moilanen, Rasmussen, and Padilla-Walker (2015) revealed longitudinal effects of adolescents’ questionnaire-based reports of authoritarian mothering, such that high levels of authoritarian parenting practices were linked to slow rank-order change in maternal reports of adolescents’ self-regulation over 1 year. Similarly, Moilanen et al. (2010) revealed that increases in self-regulation between boys’ ages 10 and 11 were linked to low levels of observer-reported maternal antagonism at boys’ age 10 years (i.e., in the context of the same conflict discussion task described above, this was assessed as the degree to which mothers were disengaged, rejecting, sarcastic, emotionally manipulative, etc.). In sum, it is reasonable to expect that low levels of maternal discipline will be linked to improvements in self-regulation over 2 years during early adolescence.

Mother-Child Relationship Quality

Mother-child relationship quality was operationalized as the child’s perception of the degree to which they felt close to their mother and viewed the mother as being involved in their daily life. This potential source of socialization is less well-understood than are parenting behaviors. Two known studies demonstrate cross-sectional associations between poor parent-child relationships and poor self-regulation during the adolescent years. Bynum and Brody (2005) revealed associations between maternal reports of high-quality mother-child relationships to high levels of adolescents’ regulation of behaviors. Loukas and Roalson (2006) also demonstrated that high levels of adolescent-reported negative family relations were associated with low levels of effortful control in a sample of European American and Latino 10-year-olds. Longitudinal studies are rare and reveal inconsistent findings during adolescence. In Moilanen et al. (2010), high-quality mother-son relationships (i.e., sons’ perceptions of high closeness and low conflict with mothers) at age 10 were linked to high levels of boys’ self-regulation at ages 10 and 11 years. Yet, in an 11-year longitudinal study, Meldrum, Young, Hay, and Flexon (2012) revealed that close mother-child relationships predicted positive change in children’s self-control only during early and middle childhood, with no benefits of strong maternal attachments for adolescents’ self-control observed between ages 10 and 15 years. Thus, this pathway was also considered.

Control and Moderating Variables

In addition to evaluating the substantive associations described above, the current study controlled for child sex, race/ethnicity, and children’s prior self-regulation. These control variables have been considered to varying degrees in previous self-regulation research. On an initial basis, the present investigation explored whether the associations between maternal parenting, mother-child relationship quality, and children’s later self-regulation were moderated by any of these control variables. What is known about each of these is discussed in turn below.

Sex/gender

In the tripartite model, Morris et al. (2007) treat child sex/gender as both a direct and moderating influence on the development of self-regulation. Actual sex differences in self-regulation throughout childhood and adolescence are well documented: Girls have consistently higher levels of self-regulation than boys across developmental periods, although self-regulatory abilities appear to grow at similar rates for both sexes (Bowers et al., 2011; Moilanen et al., 2015; Raffaelli, Crockett, & Shen, 2005). This difference in level may have biological bases, or it may also be due to differential parental socialization of emotional expression and regulation by sex. In general, parents tend to have higher expectations for controlling negative emotions for sons than for daughters and utilize different parenting strategies for each sex, ultimately training daughters to express emotions and avoid risks and teaching sons to control their feelings and take risks (Eisenberg, Cumberland, & Spinrad, 1998; Hagan, Simpson, & Gillis, 1988; Morris et al., 2007). In the tripartite model, sex/gender is regarded as a potential moderator of parenting behaviors, but not of the family’s emotional climate (Morris et al., 2007), despite evidence of sex differences in mother-child closeness (i.e., girls tend to view their relationships with their mothers as more highly supportive than do boys; Furman & Buhrmester, 1985). Further, evidence indicates that this pathway may need to be added to the tripartite model, in light of findings demontrating that high levels of maternal non-physical punishment and high-quality mother-child relationships were linked to high levels of later self-regulation for boys but not girls during early adolescence (Chapple, Vaske, & Hope, 2010). Conceptually, boys may be more sensitive than girls to disrupted maternal relationships, which has implications for the development of self-regulation (Hayslett-McCall & Bernard, 2002). These possibilities were considered in the present research.

Race/ethnicity

Although Morris et al. (2007) did not include direct or moderated effects of race/ethnicity in their model, both possibilities were considered in the current study. There is limited evidence of small racial or ethnic differences in self-regulation during childhood (e.g., LeCuyer, Swanson, Cole, & Kitzman, 2011). Moderation of parenting effects is suggested by accumulated evidence demonstrating that, compared with African American and Hispanic peers, European American children are disproportionately likely to experience maternal parenting that is conducive to supporting developments in self-regulation (e.g., Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987). Further, in one investigation on parenting and committed compliance during early childhood, LeCuyer et al. (2011) revealed differential effects for maternal self-reported authoritarian parenting, such that high levels were beneficial for African American children and harmful for European American preschoolers. Yet in a similar subsample drawn from the Children of the National Longitudinal Survey of Youth-1979 (CNLSY-79) data set, the associations between parental supervision and monitoring/discipline and self-regulation were equivalent for White and non-White participants (Pratt, Turner, & Piquero, 2004). Turning to mother-child relationship quality, very little information about how racial/ethnic groups vary in terms of parent-child relationship quality is present in the literature, and what is available suggests that variations in mother-child relationship quality are not attributable to racial or ethnic ancestry (Moilanen & Raffaelli, 2010). Thus on an exploratory basis, race/ethnicity was considered as a potential moderator of the association between mother-child relationship quality and young adolescents’ self-regulation.

Prior self-regulation

Finally, the role of adolescents’ prior levels of self-regulation was also considered in the present longitudinal investigation. Evidence of individual stability in self-regulation during childhood and adolescence is substantial and consistent across various measures of self-regulation (Brody & Ge, 2001; Colman et al., 2006; Crosswhite & Kerpelman, 2012; Li-Grining, 2007; Meldrum et al., 2012; Moilanen et al., 2010). Building on this basis, the current study also explored whether youths’ initial levels of self-regulation moderated the effects of maternal parenting and mother-child relationship quality. Children who struggle to self-regulate appear to be more sensitive to parenting and parent-child relationship qualities than their better-regulated peers: for example, high maternal hostility predicted high levels of externalizing problems for poorly- regulated children, but not their well-regulated peers (Morris et al., 2002). In comparison to their well-regulated peers, self-regulatory abilities may develop more slowly for teens who initially struggle with self-control and experience low maternal warmth, high maternal discipline, and/or have low-quality relations with their mothers. These possibilities were considered in this investigation.

The Current Study

The primary purpose of the present study was to replicate the findings of Moilanen et al. (2010) in a national sample of boys and girls. Such a replication is warranted in order to determine if the patterns evidenced in Moilanen and colleagues’ (2010) study of an at-risk community sample of Black and White boys generalize to girls and Hispanic teens in diverse socioeconomic circumstances. Controlling for child race/ethnicity, it was anticipated that high levels of self-regulation in early adolescence (i.e., ages 12-14 years) would be associated with female sex, high levels of prior maternal warmth, mother-child relationship quality and prior self-regulation, and with low levels of maternal discipline. We predicted that levels of self-regulation would be equivalent across ethnic groups (i.e., no main effect for race/ethnicity). A secondary study goal was to explore whether effects of maternal parenting practices and mother-child relationship quality on young adolescents’ self-regulation were moderated by child sex, race/ethnicity, and prior levels of self-regulation. We anticipated that sex would moderate the effects of parenting practices and mother-child relationship quality, such that boys’ self-regulation would be more sensitive to low warmth, low mother-child closeness, and high discipline than would be girls’. We predicted similar moderated associations for youth’s prior self-regulation, such that low warmth, low relationship quality, and high discipline would be more detrimental for teens with low initial levels of self-regulation than for their better-regulated peers. No further a priori hypotheses were advanced.

Study hypotheses were tested using data drawn from two waves of the CNLSY-79 data set, in which study constructs were assessed using different measures and methods than those employed by Moilanen et al. (2010). Thus, the present study is also situated to explore whether these previous findings replicate across samples and measures.

Method

Participants

The current analyses utilized data from a sample of N = 855 children who participated in the CNLSY-79 starting in 1986 (Zagorsky & White, 1999; for complete details of the sample selection procedures, see Raffaelli & Crockett, 2003). Of the original sample, N = 823 families remained in the study through the 1992 or 1994 study waves; attrition analyses revealed no systematic dropout attributable to sex or race between 1986 and 1994. Two cases were dropped from the sample because of late completion of the 1992 study wave. Thus, the final analytic sample included 821 adolescent-mother dyads, with approximately equal representation of boys and girls (52.5% male) from three ethnic backgrounds (24.2% Hispanic, 36.7% African American, 39.1% European American). In 1992, the children’s mean age was 10.79 years (SD = 0.68, range = 10-12), and the mothers’ mean age was 31.53 years (SD = 2.08, range = 27-35). In all, 31.1% of the families lived below the federal poverty line in 1992.

Measures

All covariates were assessed at youths’ ages 10 to 12 years (T1; 1992 survey wave); the dependent variable was assessed at youth’s ages 12 to 14 years (T2; 1994 survey wave). With exceptions noted, for all continuous variables scale scores were created by averaging the responses to items on the scale and high scores represent high levels of that construct. Descriptive statistics are provided in Table 1.

Table 1.

Study Variable Descriptive Statistics and Correlations (N = 821).

Variable		n	$\bar{X}$ (SD)/%	Range	Correlations
Variable		n	$\bar{X}$ (SD)/%	Range	1	2	3	4	5	6	7
1	Male	821	52.5%	0-1
2	Black	821	36.7%	0-1	.01
3	Hispanic	821	24.2%	0-1	.04	−.43***
4	Warmth	704	3.54 (1.37)	0.00-5.00	−.07	−.27***	−.02
5	Discipline	712	0.47 (0.70)	0.00-5.00	.11**	.07	.07	.03
6	Relationship quality	660	0.00 (0.59)	−2.21-0.82	−.05	−.09*	.06	.20***	−.09*
7	Self-regulation 1992	736	2.56 (0.35)	1.00-3.00	−.09*	−.01	−.02	.13**	−.25***	.11**
8	Self-regulation 1994	709	2.55 (0.35)	1.23-3.00	−.08*	.03	−.02	.10*	−.23***	.14**	.61***

p < .05. **p < .01. ***p < .001 (all tests two-tailed).

Demographic variables

Mothers reported on all demographic variables in 1986. Child demographics were biological sex (0 = female, 1 = male) and race/ethnicity, for which two mutually exclusive dummy codes were used, indicating Hispanic heritage (0 = not Hispanic, 1 = Hispanic) and Black race (0 = not Black, 1 = Black).

Maternal warmth

Mothers completed Part D of the Home Observation Measurement of the Environment Short Form (HOME-SF; Caldwell & Bradley, 1979). Mothers reported how many times in the past week they engaged in three warm behaviors (i.e., “praised child for doing something worthwhile?” “shown child physical affection (kiss, hug, stroke hair, etc.)?” and “told another adult [spouse, friend, co-worker, visitor, relative] something positive about the child?”), with possible responses ranging between zero and 99 times. These raw variables were recoded to a response scale ranging from 0 (zero times) to 5 (five or more times) in order to reduce skewness and kurtosis to acceptable levels. Scale scores were computed by averaging for all children who had responses to all three of the items (α = .75). This measure has been employed in previous research involving a separate CNLSY-79 cohort by Crosswhite and Kerpelman (2012), and demonstrated negative associations with concurrent maternal reports of discipline and positive correlations with concurrent maternal reports of autonomy support, mother-child closeness, and interviewer-reported maternal communication.

Maternal discipline

This five-item measure was also based upon Part D of the HOME-SF (Caldwell & Bradley, 1979). Mothers indicated how many times in the past week they engaged in spanking, grounding, taking away privileges, loss of allowance, and sending the child to their room, with possible responses ranging from zero to 99 times. As for maternal warmth, these values were recoded to a 6-point response scale in order to reduce skewness and kurtosis, resulting in a response scale ranging from 0 (zero times) to 5 (five or more times). A scale score was computed by averaging across the items (α = .73). In the same sample, Moilanen and Shen (2014) provided evidence for the concurrent and predictive validity of this measure, in that high levels of maternal discipline in 1992 were associated with high levels of concurrent maternal expectations for children’s completion of household chores and low levels of children’s perceived mother-child closeness, and high levels of children’s mastery in 1998.

Mother-child relationship quality

This five-item measure reflected adolescents’ perceived closeness with their mothers. Teens responded to two items (i.e., “How close do you feel to [your mother]?” and “How well do you and your [mother] share ideas or talk about things that really matter?”) on a 4-point response scale, ranging from 1 (not very close/well) to 4 (extremely close/well). The third item (i.e., “How often does your mother know who you are with when you are not home?”) also had a 4-point response scale, ranging from 1 (only rarely) to 4 (all the time). For the fourth item (i.e., “Please think about the time you spend with your mother. Do you think she spends enough time with you, or do you wish she spent more time with you?”), teens answered either 1 (spends enough time with me), 2 (wish she spent more time with me), or 3 (spends too much time with me); this was recoded to a binary variable such that 0 = spends too much/too little time and 1 = spends enough time. The final item (i.e., “How often does [your mother] miss the events or activities that are important to you?”) had a 3-point response scale, ranging from 1 (misses events a lot) to 3 (almost never misses events). Given the three different response scales, all items were standardized before being averaged to form a composite of mother-child relationship quality (α = .54), with high scores corresponding to closer relationships between mothers and teens. This measure has demonstrated concurrent and predictive validity in the same sample, in that high-quality mother-child relationships in 1994 were associated with low levels of children’s autonomous decision making and with low levels of lifetime substance use as of 1998 (Crockett, Moilanen, Raffaelli, & Randall, 2006).

Self-regulation

This measure consisted of 13 conceptually-identified items from the 28-item Behavior Problems Index representing the child’s ability to control their emotions, attention, and behaviors (Raffaelli & Crockett, 2003; Zill, 1990; 1992 and 1994 αs = .85; sample item: “He/she is restless or overly active, cannot sit still”). Mothers reported how well each item described their child’s behavior in the last 3 months on a 3-point scale, from 1 (often true) to 3 (not true). Confirmatory factor analyses of the 1992 items revealed that all items loaded on one latent factor (standardized coefficient range = .41-.62), but also that two- and three-factor solutions each provided comparatively better fits to the data. This same conclusion was reached by Raffaelli et al. (2005) in their similar analysis of the 1994 items in the same sample. This measure has demonstrated criterion-related validity in this sample, evidenced by negative correlations with depression, substance use, and sexual risk-taking (Crockett et al., 2006).

Analysis Plan

Following preliminary analyses, we examined the assumptions of the model. All statistical assumptions were met (i.e., linearity and homoscedasticity) with the exception of normality and the side condition of homogeneity of regression slopes. Statistical tests are robust to violations of normality in the presence of homoscedasticity (Cohen, Cohen, West, & Aiken, 2003). To address homogeneity of regression slopes, we examined whether any of the effects of any continuous variable (i.e., warmth, discipline, relationship quality, and self-regulation in 1992) on self-regulation in 1994 differed for any group (i.e., males/females, Black/non-Black); thus, we identified the statistically significant interactions between sex and race/ethnicity and the continuous IVs, which eliminated the necessity of including an extensive list of interactions in the substantive structural models. Only 1992 self-regulation on 1994 self-regulation differed for males and females and for Black and non-Black children (i.e., the effect of prior self-regulation on post self-regulation was moderated by sex and Black race/ethnicity). Because of this violation, we modeled these differing regression slopes by including two interaction terms, Male × Prior self-regulation and Black × Prior self-regulation, for any model that included the path from 1992 self-regulation to 1994 self-regulation. None of the other regression slopes for the continuous variables (i.e., warmth, discipline, relationship quality, and self-regulation in 1992) on self-regulation in 1994 differed by sex or race/ethnicity (i.e., sex and race/ethnicity did not moderate the effects of the continuous variables on 1994 self-regulation).

The hypotheses were tested via path analysis with observed variables, all of which were conducted in MPlus version 7.2 (Muthén & Muthén, 1998-2012). As missing data analyses suggested that data were missing at random (MAR; i.e., of all study variables, missingness on self-regulation in 1994 was attributable only to child race/ethnicity), all parameters were estimated with all available information using full information maximum likelihood (FIML). All continuous independent variables were grand mean centered. Only exogenous continuous independent variables were permitted to correlate.

A total of five longitudinal models were estimated: the first included sex and race/ethnicity only, with the two parenting variables added in the second model, relationship quality in the third, and prior self-regulation included in the fourth model. The final model included all interactions between prior self-regulation and the two parenting variables and relationship quality (i.e., Self-regulation in 1992 × Warmth, Self-regulation in 1992 × Discipline, and Self-regulation in 1992 × Relationship quality). The primary indicator of acceptable model fit was a non-significant chi-square fit statistic (χ²); however, given that chi-square statistics are sensitive to sampling fluctuation, root mean square errors of approximation (RMSEA) smaller than .06 or a 90% confidence interval that contains .06, and standardized root mean square residuals (SRMR) smaller than .08 were used to indicate acceptable model fit (West, Taylor, & Wu, 2012). However, given that we did not model the correlations among the categorical predictors, poor model fit is to be expected for the models with very few continuous predictors and many categorical predictors (i.e., the parenting, relationship quality, and final models).

Results

Preliminary Analyses

Bivariate correlations between all study variables are provided in Table 1. Mothers of males reported higher levels of discipline and lower levels of self-regulation at both waves. Black youth were more likely to report poorer quality mother-child relationships than their peers, while also experiencing lower levels of maternal warmth. High levels of maternal warmth were linked to high relationship quality and self-regulation at both waves, whereas high levels of discipline were linked to low levels of relationship quality and self-regulation at both waves. High relationship quality was associated with high self-regulation at both time points, and the two self-regulation scores were strongly and positively correlated.

Hypothesis Testing

Parameter estimates and model fit indices for all models are reported in Table 2. In the demographic model, females had higher levels of self-regulation than males and race/ethnicity was not significantly associated with self-regulation. With the addition of the maternal parenting variables in the second model, the effect for sex was mitigated to non-significance; however, in this model high levels of self-regulation were evidenced by Black youth as well as teens who experienced high levels of maternal warmth and low levels of discipline. In the model that added relationship quality, all effects present in the parenting model remained with the addition of relationship quality in this model. High levels of mother-child relationship quality were associated with high levels of self-regulation.

Table 2.

Summary of Longitudinal Models Predicting Self-Regulation at Ages 12-14 Years (N = 821).

	Model
Variable	Demographics	Parenting	Relationship quality	Prior self-regulation	Final
Male	−.16*	−.05	−.05	−.02	−.02
Black	.05	.09*	.09*	.06	.06
Hispanic	−.01	.04	.04	.01	.01
Warmth		.13**	.11**	.04	.04
Discipline		−.23***	−.22***	−.08*	−.09**
Relationship quality			.10*	.08*	.08*
Self-regulation 1992				.54***	.54***
Male × Self-regulation 1992				.10*	.11*
Black × Self-regulation 1992				−.08*	−.06
Self-regulation 1992 × Warmth					.05
Self-regulation 1992 × Discipline					−.07*
Self-regulation 1992 × Relationship quality					−.04
χ² (df)	(0) 0.00***	(6) 97.93***	(9) 101.72***	(18) 108.22***	(27) 112.24***
RMSEA	.00	.14	.12	.08	.06
RMSEA 90% CI	.00-.00	.12-.17	.10-.14	.07-.09	.05-.08
SRMR	.00	.06	.06	.05	.04
R ²	.01	.08***	.09***	.39***	.40***

Note. All coefficients are standardized (βs). Detailed information about the correlations between the predictor variables is available from the first author upon request. RMSEA = root mean square errors of approximation; CI = confidence interval; SRMR = standardized root mean square residual.

p < .05. **p < .01. ***p < .001.

With the addition of prior self-regulation in the fourth model, prior self-regulation was the strongest predictor, such that high levels of self-regulation in 1992 were linked to high levels of self-regulation in 1994. None of the direct effects of the demographic variables and warmth were predictive of self-regulation in 1994 over and above the remaining predictors. Controlling for prior self-regulation, high levels of self-regulation were demonstrated by teens with low levels of discipline and high levels of mother-child relationship quality. The interactions between sex and self-regulation in 1992 and between race and self-regulation in 1992 were statistically significant (see Figures 1 and 2). Simple slope analysis demonstrated that the relationship between self-regulation in 1992 and 1994 was slightly stronger for males (b = .662, SE_b = .041, p < .001) than for females (b = .513, SE_b = .044, p < .001; see Figure 1). A similar pattern was also evidenced by race in the simple slopes analysis of this interaction, such that the relationship between self-regulation in 1992 and 1994 was modestly stronger for non-Black adolescents (b = .644, SE_b = .040, p < .001) than Black adolescents (b = .533, SE_b = .046, p < .001; see Figure 2).

Figure 1.

Depiction of the interaction between self-regulation in 1992 and sex on self-regulation in 1994.

Figure 2.

Depiction of the interaction between self-regulation in 1992 and race on self-regulation in 1994.

The final model provided an acceptable fit to the data. As before, prior self-regulation was still the strongest predictor with high levels of self-regulation in 1992 predicting high levels of self-regulation in 1994. The interaction between Black and self-regulation in 1992 was no longer a statistically significant predictor in the model. All other statistically significant predictors in the previous model continued to persist as statistically significant predictors of self-regulation in 1994. Two interactions were statistically significant. The first interaction was between self-regulation in 1992 and sex on self-regulation in 1994, and was the same as in the preceding model (see Figure 1). The second interaction was between self-regulation in 1992 and discipline on self-regulation in 1994 (see Figure 3). A simple slopes analysis showed that the effect of self-regulation in 1992 on self-regulation in 1994 was strongest for those in lower disciplinary settings (dashed line in Figure 3; b = .629, SE_b = .046, p < .001). However, this relationship was weaker for adolescents in average disciplinary climates (solid line in Figure 3; b = .583, SE_b = .032, p < .001) and weakest for adolescents in higher disciplinary homes (dotted line in Figure 3; b = .537, SE_b = .040, p < .001).

Figure 3.

Illustration of the interaction between self-regulation in 1992 and level of discipline on self-regulation in 1994.

Discussion

Although there is widespread agreement that high levels of self-regulation are valuable for protecting individuals against involvement in risk behaviors and for promoting optimal adjustment, there is limited understanding of the degree to which self-regulation changes during the teen years and which socializing forces may explain its improvement. The goal of the present study was to further address this imbalance by seeking to replicate and extend an existing 1-year longitudinal study conducted with an at-risk sample of young adolescent boys (Moilanen et al., 2010). Armed with a large national sample of young adolescents from the CNLSY-79, these analyses replicated the main effects in Moilanen and colleagues’ (2010) all-male sample, demonstrating that these associations are robust over a 2-year study period and with different measures. Simultaneously, a fruitful exploration of potential moderated effects revealed subtle differences by sex, race, and youth’s prior self-regulation. Overall, these findings provide partial support for two main effects and two moderated effects included in the tripartite model of familial impacts on children’s emotion regulation (Morris et al., 2007), and identify one additional moderated pathway not currently included in this theoretical model.

Main Effects

The effect for maternal discipline was consistent with hypotheses and with prior research. Controlling for prior levels of self-regulation, high levels of maternal discipline were associated with slower change in self-regulation over 2 years. Although the present study focused on relatively mild forms of discipline, these findings are in keeping with theoretical predictions about harms caused by harsh parenting practices (Morris et al., 2007; Scaramella & Leve, 2004), as well as the findings of existing studies including measures of maternal harsh or rejecting parenting (Lengua, 2006; Moilanen et al., 2010). The negative arousal generated through maternal discipline likely prevents children from learning how to regulate their emotions and behaviors, while increasing their need for adequate regulatory skills (Scaramella & Leve, 2004).

Further, results for mother-child relationship quality were also consistent with expectations. Adolescents’ perceptions of high-quality mother-child relationships predicted longitudinal improvements in self-regulation, an effect that replicates findings of Moilanen and colleagues’ (2010) study of at-risk boys between ages 10 and 11 years while contradicting those of Meldrum et al. (2012). Likely these discrepancies are due to study design strategies, in that mothers reported on children’s self-control and attachment in Meldrum et al. (2012), whereas adolescents self-reported on mother-child closeness in the present study as well as in Moilanen et al. (2010). The latter strategy is preferable in these circumstances, as mothers may not be aware of the degree to which youth feel emotionally dysregulated as a consequence of parent-teen relational upheaval (Manczak, McLean, McAdams, & Chen, 2015). Such turmoil may overwhelm young teens’ regulatory capacities and prevent them from learning self-control techniques. Alternately, mothers may provide poor examples of self-regulation during conflicts with their children (Morris et al., 2007).

Finally, we anticipated that high levels of maternal warmth would be associated with high levels of self-regulation, reflecting the theoretical assumption that positive parenting practices provide opportunities for teens to learn regulatory strategies (Morris et al., 2007). Although high levels of maternal warmth were positively associated with self-regulation in 1994, this effect was mitigated to non-significance when prior levels of self-regulation were controlled. Thus, these null findings were consistent with existing investigations considering such elements of maternal parenting (Eisenberg et al., 2005; Lengua, 2006; Moilanen et al., 2010), further undermining the assumption that maternal warmth facilitates positive changes in self-regulation in adolescence (Kiff, Lengua, & Zalewski, 2011).

Moderated Effects

In addition to the main effect described above, the effect of maternal discipline on self-regulation in 1994 was moderated by teens’ self-regulation in 1992. Specifically, regardless of level of discipline in 1992, young adolescents with low self-regulation scores in 1992 continued to have low self-regulation scores 2 years later (i.e., their levels of self-regulation were stable over time, regardless of the degree to which they experienced discipline in 1992). However, for teens who initially had high levels of self-regulation in 1992, those who experienced low levels of discipline had the highest self-regulation scores in 1994 (i.e., the highest stability in self-regulation between the two waves), whereas those who had high discipline had moderate self-regulation at the 2-year follow-up (i.e., the least stability during the same period). This provides novel support for the theoretical model that was partially tested in this study (Morris et al., 2007) but contradicts previous research suggesting that children with poor self-regulation are particularly vulnerable to the impacts of negative parenting (Morris et al., 2002). Instead, it appears that it is well-regulated youth who are highly sensitive to the impacts of maternal discipline at the transition to adolescence. Given the short-term nature of this two-wave longitudinal study, researchers are cautioned against interpreting these findings as unassailable evidence for a socialization effect (e.g., that highly regulated young adolescents’ positive developments in self-regulation over 2 years may be slowed by the presence of maternal discipline), as they could also reflect a child effect (e.g., that young adolescents who are well regulated elicit less discipline from their mothers, as suggested in Moilanen et al., 2015; Scaramella & Leve, 2004). Both possibilities warrant further exploration.

Two additional instances of moderation were identified. Specifically, the strength of the association between self-regulation in 1992 and 1994 was conditional upon the child’s sex and their race/ethnicity. Both of these interactions were small in size, but their presence alone is noteworthy given the degree to which self-regulation was stable over 2 years. The interaction with sex was present in both of the final two models, and ultimately suggests that boys’ levels of self-regulation improve somewhat faster over 2 years than their female peers. This slight catch-up effect is contrary to the literature, which demonstrates that self-regulation improves at similar rates for both sexes during adolescence (e.g., Bowers et al., 2011; Moilanen et al., 2015). The interaction with race was present only in the model that omitted the interaction terms between prior self-regulation and parenting, and indicated that Black youth demonstrated slightly slower change in self-regulation over 2 years, compared to their non-Black peers. While it appears to be attributable to ethnic variations in parenting, this finding affirms the need for greater consideration of potential ethnic differences in self-regulation during the teen years.

Limitations, Future Directions, and Conclusions

The present study possessed several noteworthy limitations. Though the use of a national data set bolsters confidence about the findings’ generalizability, the analytic subsample was disproportionately comprised of children born to early childbearers. Thus, potential SES effects warrant scrutiny, as children living in low-income households tend to have poorer self-regulation, and are at increased risk of experiencing insecure attachment and parenting behaviors that may undermine self-regulatory development compared with their economically advantaged peers (Lempers, Clark-Lempers, & Simons, 1989; Li-Grining, 2007; Spieker & Booth, 1988). The sample’s predomination by low-SES families and concerns about missing data on the limited available proxy variables for SES precluded such analyses in these data. Further, although the present sample is more diverse than the one used in Moilanen et al. (2010), it is still limited in terms of representation of youth and families from ancestries beyond the three included in the CNLSY-79 sample. Future researchers may also wish to explore simultaneous moderation by sex and race, which was not feasible in the present study due to insufficient subgroup sizes. Finally, some caution is urged about the generalizability of findings to contemporary adolescents given the age of the data. These findings are likely still relevant to present-day teens, as shifts in levels of parenting behaviors over the previous decades do not alter the hypothesized mechanisms of their effects (Trifan, Stattin, & Tilton-Weaver, 2014).

Other limitations stemmed from the use of existing data, including being restricted to the limited available maternal-report measures of maternal parenting, a single child-report measure of mother-child relationship quality, and a somewhat narrow maternal-report questionnaire on adolescents’ self-regulation. The mother-child relationship quality measure evidenced low internal consistency, due to the limited number of possible survey items and having to standardize their various response scales to construct an observed variable. For future research, it would be ideal to assess all study constructs using rich methodologies (e.g., observational measures and multiple informants), and including the elements of the tripartite model omitted from the current inquiry (e.g., maternal depression, marital quality, etc.; Morris et al., 2007). Consideration of paternal parenting behaviors and father-child relationship quality is crucial, as exceptionally little is known about fathers’ impacts on children’s self-regulation.

Finally, the design of the CNLSY-79 limited the present investigation to a two-wave design during early adolescence. Longer term longitudinal investigations remain needed, as little is known about how parenting and relationship quality are linked to changes in self-regulation through late adolescence. Such inquiries cannot be executed using the CNLSY-79 data, as the measure of self-regulation is unavailable after the present study’s second wave, and the measure of relationship quality is not available until its first wave. Along these same lines, little is known about potential bidirectional associations between these three constructs: Youth’s self-regulation may also shape the parenting they receive and the emotional quality of the mother-child relationship (Scaramella & Leve, 2004), as suggested in one recent study about parenting behaviors (Moilanen et al., 2015). Such future investigations should involve tests of models with latent variables in order to demonstrate that such cross-lagged associations are not artifacts of measurement error.

In sum, supported by a short-term longitudinal design and reports from two informants, the present findings suggest that the majority of the associations demonstrated in a mid-sized, at-risk community sample of boys ages 10 to 11 years (Moilanen et al., 2010) are also observable in a large-sized, national sample of young adolescent boys and girls. This provides further evidence supporting the consideration of the parent-child relational context in addition to discrete parenting behaviors in studies of the development of self-regulation during the early adolescent years. This investigation reaffirms a subset of the tenets of the tripartite model of familial socialization of children’s emotion regulation (Morris et al., 2007), while identifying the need to slightly modify the model as to include possible ethnic differences in change in self-control.

Footnotes

Authors’ Note

The present study involved analyses of public-use data which can be accessed at

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data preparation was supported in part by National Institutes of Health Grant RO1-MH62977 to Marcela Raffaelli and Lisa J. Crockett. Research assistance was provided by Myesha Albert, Brett Avila, Jennifer Bowers, Rebecca Colman, Lisa Crockett, Jenenne Geske, Sam Hardy, Paul Hernandez, Roya Hossani, Andy Peytchev, Marcela Raffaelli, Brandy Randall, Yuh-Ling Shen, and Devan Starks.

Author Biographies

Kristin L. Moilanen is an associate professor of child development and family studies in the Department of Learning Sciences & Human Development at West Virginia University. Her research interests concern the periods of adolescence and emerging adulthood, and focus on developments in self-regulation, social relationships, and risk behaviors.

Karen E. Rambo-Hernandez is an assistant professor of educational psychology in the Department of Learning Sciences & Human Development at West Virginia University. Her research interests include statistical modeling, particularly growth and multilevel modeling, academic acceleration, STEM education, and educational assessment.

References

Baumrind

(1991). The influence of parenting style on adolescent competence and substance use. Journal of Early Adolescence, 11, 56-95. doi:10.1177/0272431691111004

Bowers

E. P.

Gestsdottir

Geldhof

G. J.

Nikitin

von Eye

Lerner

R. M.

(2011). Developmental trajectories of intentional self-regulation in adolescence: The role of parenting and implications for positive and problematic outcomes among diverse youth. Journal of Adolescence, 34, 1193-1206. doi:10.1016/j.adolescence.2011.07.006

Brody

G. H.

(2001). Linking parenting processes and self-regulation to psychological functioning and alcohol use during early adolescence. Journal of Family Psychology, 15, 82-94. doi:10.1037/0893-3200.15.1.82

Bynum

M. S.

Brody

G. H.

(2005). Coping behaviors, parenting, and perceptions of children’s internalizing and externalizing problems in rural African American mothers. Family Relations, 54, 58-71. doi:10.1111/j.0197-6664.2005.00006.x

Caldwell

B. M.

Bradley

R. H.

(1979). Home observation for measurement of the environment. Little Rock, AR: Center for Child Development and Education, University of Arkansas.

Chapple

C. L.

Vaske

Hope

T. L.

(2010). Sex differences in the causes of self-control: An examination of mediation, moderation, and gendered etiologies. Journal of Criminal Justice, 38, 1122-1131. doi:10.1016/j.jcrimjus.2010.08.004

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.

Colman

R. A.

Hardy

S. A.

Albert

Raffaelli

Crockett

(2006). Early predictors of self-regulation in middle childhood. Infant and Child Development, 15, 421-437. doi:10.1002/icd.469

Crockett

L. J.

Moilanen

K. L.

Raffaelli

Randall

B. A.

(2006). Psychological profiles and adolescent adjustment: A person-centered approach. Development and Psychopathology, 18, 195-214. doi:10.1017/S0954579406060111

10.

Crossley

I. A.

Buckner

J. C.

(2012). Maternal-related predictors of self-regulation among low-income youth. Journal of Child and Family Studies, 21, 217-227. doi:10.1007/s10826-011-9465-0

11.

Crosswhite

J. M.

Kerpelman

J. L.

(2012). Parenting and children’s self-control: Concurrent and longitudinal relations. Deviant Behavior, 33, 715-737. doi:10.1080/01639625.2011.647597

12.

Doan

S. N.

Fuller-Rowell

T. E.

Evans

G. W.

(2012). Cumulative risk and adolescent’s internalizing and externalizing problems: The mediating roles of maternal responsiveness and self-regulation. Developmental Psychology, 48, 1529-1539. doi:10.1037/a0027815

13.

Dornbusch

S. M.

Ritter

Leiderman

Roberts

Fraleigh

(1987). The relation of parenting style to school performance. Child Development, 58, 1244-1257. doi:10.2307/1130618

14.

Eisenberg

Cumberland

Spinrad

T. L.

(1998). Parental socialization of emotion. Psychological Inquiry, 9, 241-273. doi:10.1207/s15327965pli0904_1

15.

Eisenberg

Zhou

Spinrad

T. L.

Valiente

Fabes

R. A.

Liew

(2005). Relations among positive parenting, children’s effortful control, and externalizing problems: A three-wave longitudinal study. Child Development, 76, 1055-1071. doi:10.1111/j.1467-8624.2005.00897.x

16.

Finkenauer

Engels

R. C. M. E.

Baumeister

R. F.

(2005). Parenting behavior and adolescent behavioral and emotional problems: The role of self-control. International Journal of Behavioral Development, 29, 58-69. doi:10.1080/01650250444000333

17.

Furman

Buhrmester

(1985). Children’s perceptions of the personal relationships in their social networks. Developmental Psychology, 21, 1016-1022. doi: 10.1037/0012-1649.21.6.1016

18.

Hagan

Simpson

Gillis

A. R.

(1988). Feminist scholarship, relational and instrumental control, and a power-control theory of gender and delinquency. British Journal of Sociology, 39, 301-336. doi:10.2307/590481

19.

Hayslett-McCall

K. L.

Bernard

T. J.

(2002). Attachment, masculinity, and self-control: A theory of male crime rates. Theoretical Criminology, 6, 5-33. doi:10.1177/136248060200600101

20.

Kiff

C. J.

Lengua

L. J.

Zalewski

(2011). Nature and nurturing: Parenting in the context of child temperament. Clinical Child and Family Psychology Review, 14, 251-301. doi:10.1007/s10567-011-0093-4

21.

LeCuyer

E. A.

Swanson

D. P.

Cole

Kitzman

(2011). Effect of African- and European-American maternal attitudes and limit-setting strategies on children’s self-regulation. Research in Nursing & Health, 34, 468-482. doi:10.1002/nur.20460

22.

Lempers

J. D.

Clark-Lempers

Simons

R. L.

(1989). Economic hardship, parenting, and distress in adolescence. Child Development, 60, 25-39. doi:10.2307/1131068

23.

Lengua

L. J.

(2006). Growth in temperament and parenting as predictors of adjustment during children’s transition to adolescence. Developmental Psychology, 42, 819-832. doi:10.1037/0012-1649.42.5.819

24.

Li-Grining

C. P.

(2007). Effortful control among low-income preschoolers in three cities: Stability, change, and individual differences. Developmental Psychology, 43, 208-221. doi:10.1037/0012-1649.43.1.208

25.

Loukas

Roalson

L. A.

(2006). Family environment, effortful control, and adjustment among European American and Latino early adolescents. Journal of Early Adolescence, 26, 432-455. doi:10.1177/0272431606291939

26.

Manczak

E. M.

McLean

K. C.

McAdams

D. P.

Chen

(2015). Physiological reactivity during parent-adolescent discussions: Associations with scaffolding behaviors and relationship quality. Annals of Behavioral Medicine, 49, 522-531. doi:10.1007/s12160-014-9680-1

27.

Meldrum

R. C.

Young

J. N.

Hay

Flexon

J. L.

(2012). Does self-control influence maternal attachment? A reciprocal effects analysis from early childhood through middle adolescence. Journal of Quantitative Criminology, 28, 673-699. doi:10.1007/s10940-012-9173-y

28.

Moilanen

K. L.

(2007). The Adolescent Self-Regulatory Inventory: The development and validation of a questionnaire of short-term and long-term self-regulation. Journal of Youth and Adolescence, 36, 835-848. doi:10.1007/s10964-006-9107-9

29.

Moilanen

K. L.

Padilla-Walker

L. M.

(2013, April). Transactional models of adolescents’ self-regulation and maternal and paternal parenting. In Moilanen

K. L.

Trentacosta

C. J.

(Chairs), Bidirectional associations between self-regulation and parenting in childhood and adolescence. Symposium conducted at the Biennial Meeting of the Society for Research in Child Development, Seattle, WA.

30.

Moilanen

K. L.

Raffaelli

(2010). Support and conflict in ethnically diverse young adults’ relationships with parents and friends. International Journal of Behavioral Development, 34, 46-52. doi:10.1177/0165025409348553

31.

Moilanen

K. L.

Rasmussen

Padilla-Walker

L. M.

(2015). Bidirectional associations between self-regulation and parenting styles in early adolescence. Journal of Research on Adolescence, 25, 246-262. doi:10.1111/jora.12125

32.

Moilanen

K. L.

Shaw

D. S.

Fitzpatrick

(2010). Self-regulation in early adolescence: Relations with mother-son relationship quality and maternal regulatory support and antagonism. Journal of Youth and Adolescence, 39, 1357-1367. doi:10.1007/s10964-009-9485-x

33.

Moilanen

K. L.

Shen

(2014). Mastery in middle adolescence: The contributions of socioeconomic status, maternal mastery and supportive-involved mothering. Journal of Youth and Adolescence, 43, 298-310. doi:10.1007/s10964-013-9951-3

34.

Morris

A. S.

Silk

J. S.

Steinberg

Myers

S. S.

Robinson

L. R.

(2007). The role of the family context in the development of emotion regulation. Social Development, 16, 361-388. doi:10.1111/j.1467-9507.2007.00389.x

35.

Morris

A. S.

Silk

J. S.

Steinberg

Sessa

F. M.

Avenevoli

Essex

M. J.

(2002). Temperamental vulnerability and negative parenting as interacting of child adjustment. Journal of Marriage and Family, 64, 461-471. doi:10.1111/j.1741-3737.2002.00461.x

36.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén & Muthén.

37.

Pratt

T. C.

Turner

M. G.

Piquero

A. R.

(2004). Parental socialization and community context: A longitudinal analysis of the structural sources of low self-control. Journal of Research in Crime & Delinquency, 41, 219-243. doi:10.1177/0022427803260270

38.

Raffaelli

Crockett

L. J.

(2003). Sexual risk taking in adolescence: The role of self-regulation and attraction to risk. Developmental Psychology, 39, 1036-1046. doi:10.1037/0012-1649.39.6.1036

39.

Raffaelli

Crockett

L. J.

Shen

Y.-L.

(2005). Developmental stability and change in self-regulation from childhood to adolescence. The Journal of Genetic Psychology, 166, 54-75. doi:10.3200/GNTP.166.1.54-76

40.

Scaramella

L. V.

Leve

L. D.

(2004). Clarifying parent-child reciprocities during early childhood: The early childhood coercion model. Clinical Child and Family Psychology Review, 7, 89-107. doi:10.1023/B:CCFP.0000030287.13160.a3

41.

Spieker

S. J.

Booth

C. L.

(1988). Maternal antecedents of attachment quality. In Belsky

Nezworski

(Eds.), Clinical implications of attachment (pp. 95-135). Hillsdale, NJ: Lawrence Erlbaum.

42.

Trifan

T. A.

Stattin

Tilton-Weaver

(2014). Have authoritarian parenting practices and roles changed during the past 50 years? Journal of Marriage and Family, 76, 744-761. doi:10.1111/jomf.12124

43.

Vazsonyi

A. T.

Belliston

L. M.

(2007). The family → low self-control → deviance: A cross-cultural and cross-national test of self-control theory. Criminal Justice and Behavior, 34, 505-530. doi:10.1177/0093854806292299

44.

West

S. G.

Taylor

A. B.

(2012). Model fit and model selection in structural equation modeling. In Hoyle

R. H.

(Ed.), Handbook of structural equation modeling (pp. 209-231). New York, NY: Guilford Press.

45.

Zagorsky

J. L.

White

(1999). NLSY79 user’s guide: A guide to the 1979–1998 National Longitudinal Survey of Youth data. Washington, DC: U.S. Department of Labor.

46.

Zill

(1990). Behavior Problems Index based on parent report (Publication No. 9103). Washington, DC: Child Trends.