Prospective Associations between Mothers’ and Fathers’ Parenting Styles and Adolescents’ Moral Values: Stability and Specificity by Parent Style and Adolescent Gender

Abstract

This study coordinates moral value development in adolescence, parenting style, and gender with issues of stability and specificity. The primary research question asked whether parenting styles of mothers and fathers influence the development of adolescent moral values, and secondary research questions asked whether adolescent moral values were stable and whether gender moderated predictive relations of parenting styles and adolescent moral values. At 14 and 18 years, a sample of 246 adolescents completed the Sociomoral Reflection Objective Measure – Short Form; at 14 years, mothers and fathers self-reported their parenting styles using the Parental Authority Questionnaire. Exploratory and confirmatory factor analyses established a 2-factor model of adolescent moral values across the two ages: Life and Social Contract captured prosocial aspects of morality that are left to individual choice, and Law and Social Order captured acts that are legally or morally obligatory for individuals to perform. Structural equation modeling investigated relations between parental parenting styles and the two adolescent moral value factors, with adolescent age, gender, and family SES as covariates. Both moral values factors had high stabilities across the 4-year period. Mothers’ authoritarian parenting at 14 years, but not their authoritative or permissive parenting, negatively predicted Life and Social Contract moral values, but not Law and Social Order, in adolescents at 18 years, more so for boys. Fathers’ parenting styles did not predict adolescents’ moral values at 18 years. Girls and adolescents from higher SES families had higher Life and Social Contract moral values at 14 years; boys experienced more increases in Life and Social Contract moral values from 14 to 18 years than girls. Stability and parental predictive validity of moral values for adolescence are discussed.

Everyday life is filled with thoughts and acts deemed moral or immoral, and through the ages philosophers, clerics, and psychologists have been drawn to profound questions of human morality, including its ontogenetic origins. Morality is an umbrella the encompasses a plethora of constructs, including values, reasoning, emotions, and behaviors, and in response the study of morality and its attendant developmental science have evolved as complex and nuanced (Glidden et al., 2024; Killen & Smetana, 2022). Moral thought (contra behavior) takes many forms, including judgment, understanding, reasoning, and values. This study is concerned with moral values. Generally speaking, moral values signify our goals and speak to matters of what is desirable to us (Ahn et al., 2020; Schwartz, 1992). In this sense, values compose a part of our identity, shape our attitudes and actions, and serve our judgments of ourselves and others (Maio, 2016; Roccas & Sagiv, 2010; Twito et al., 2019). Moral values are also important because they contribute to successful interactions in social groups (Schwartz, 2017). How children express moral values, how their moral values develop, and the intrinsic and extrinsic forces that shape their moral values are long-standing topics of significant moment in child and adolescent research (Grusec & Goodnow, 1994; Hoffman, 2000).

The present study focused on moral values with several specific aims. The first aim of the study was to provide developmental information that would fill gaps in understanding the structure of moral values across adolescence and the roles of mothers’ and fathers’ parenting styles in the development of adolescents’ moral values. Secondary aims were to investigate gender variation in adolescents’ acquisition of moral values and to chart stability of moral values across adolescence. This study therefore coordinates moral values in adolescence, parenting styles of mothers and fathers, and gender with issues of developmental stability and specificity. As this study touches on these several topics, each is briefly discussed in turn.

Moral Values and Development in Adolescence

Children bring increasingly nuanced thinking and reasoning to moral thought and action as they become competent at respecting multiple competing issues simultaneously and acknowledging their own and others’ moral perspectives as distinct (Møller & Tenenbaum, 2011; Mulvey, 2016; Tenenbaum et al., 2018). In their moral thinking younger children may adhere to simple equality; older children take many factors into consideration. For example, pre-adolescents will allocate resources equally independent of the needs or advantages of recipients, whereas adolescents will allocate resources equitably by accounting for recipients’ needs or advantages (Elenbaas, 2019; Killen & Smetana, 2015; Rizzo et al., 2016).

Moral thought emerges in childhood (Abramson et al., 2018; Glidden et al., 2024; Uzefovsky et al., 2016) and develops over the course of the lifespan (Killen & Smetana, 2015; Shachnai & Daniel, 2020), However, moral concerns in childhood are rather circumscribed (often to behaviors such as instrumental helping, comforting, and sharing; Laible & Karahuta, 2014), and prominent theoreticians have pointed to adolescence as a time when moral thought is achieving maturity (Freud, 1930; Kohlberg, 1971; Piaget, 1932). Age-related changes in moral thought in adolescence have been attributed to brain development (Blakemore et al., 2010), to increasing sophistication in sociocognitive abilities (Lagattuta & Weller, 2014; Rizzo & Killen, 2018a,b; Sodian et al., 2020; Wellman & Liu, 2004), and to identity formation (Kroger et al., 2010) at that time. As adolescents prepare for many adult responsibilities (from committed relationships to societal civics), moral values rise in prominence (Barni et al. 2011; Knafo & Schwartz, 2001; Nucci & Ilten-Gee, 2021). Adolescence therefore has special importance in understanding moral thought and action, so the development of moral values across adolescence was studied here.

Parenting Styles and their Effects

Moral values in children reflect their maturing neural, mental, and social abilities, but the inculcation and trajectory of development of moral values is shaped by reciprocal social interactions (Smetana et al., 2014; Turiel, 2015). Notably, social domain theory asserts that morality is one of three domains of social knowledge which are fashioned through social interactions (Nucci, 2001, 2009; Smetana, 2006; Smetana et al., 2014; Turiel, 1983, 2006, 2015). Parenting is one signal domain of interaction for children, and parents shape children’s moral values as well as their moral judgments, emotions, and behaviors via numerous socialization processes, such as reinforcement, modeling, and instruction (Bandura & Walters, 1963; Bornstein et al., 2011; Grusec, 2022; Hoffman, 2000; Kochanska, 1993; Rote & Flak, 2022; Smetana et al., 2019). Parents’ and children’s moral values even tend to correlate (Döring et al., 2016). Therefore, parenting (in particular parenting style) was studied here.

Baumrind (1966) initially described three major parenting styles, which combine the two prominent dimensions of warmth and control in different weights (Kuppens et al., 2009; Larzelere et al., 2013; Maccoby & Martin, 1983). Authoritative parenting combines high levels of warmth with moderate to high levels of control; authoritarian parenting is characterized by low levels of warmth and high levels of control; and permissive parenting places low levels of warmth and control on children. (Later, a fourth style of disengaged parenting, which is close to neglect, was introduced.) For summaries, see Bornstein and Bornstein (2007) and Bornstein and Zlotnik (2008).

To be meaningful in child development parenting must have demonstrated predictive validity. Some past critics have contended that parenting cognitions and practices command comparatively trivial influences over child development; rather, heredity (Plomin, 2012) and children’s peers (Harris, 1998) reign sway. The arguments for genetic endowment and group influences complement, but do not supplant, direct and indirect effects of parenting, and parenting effects have survived criticisms from both quarters demonstrating consistently that parenting effects obtain separate and apart from parent-child shared genetics (Deater-Deckard et al., 2006) and group socialization processes (Wood et al., 2004). Thus, evidence for heritability and peer influences neither negates nor diminishes equally compelling evidence for direct and indirect parenting effects. Germane to the present study, a substantial literature has developed around the effects parenting styles on children’s development (Baumrind, 1967, 1978, 1991). For example, in Western cultural contexts authoritative parenting is consistently associated with children’s achievement of social competence and overall better adaptation, whereas authoritarian parenting is equally consistently associated with poorer mental health and well-being of children (Barber & Harmon, 2002; Cunningham & Boyle, 2002; DeVore & Ginsburg, 2005; Gülay, 2011; Jaffe et al., 2010; Zhou et al., 2002).

The three parenting styles and their effects on adolescents’ moral values in a substantial sample of European American families were assessed here. In addition, this study paid attention to parent and child gender.

Gender of Parent and Adolescent

Historically, parenting science has been primarily occupied with the role of mothers in child development. The reasons for this focus are largely self-evident as, for example, mothers are children’s primary caregivers, mothers are with children more often and for longer periods, mothers assume greater responsibility for children’s health, and so forth (see Bornstein, 2015, for a summary). Mothers still appear as principal participants in developmental science research, including the socialization of morality in children (Grusec, 2022); however, there is persistent call for the inclusion of fathers and the comparison of maternal with paternal approaches to parenting children and adolescents. Mothers and fathers may parent children and adolescents similarly in many respects, but, as recently summarized by Murray et al. (2019) and Parke and Cookson (2019), in many regards mothers and fathers parent children and adolescents differently. Some evidence suggests that mothers may exercise greater sway over children’s moral development (Grusec & Goodnow, 1994). On these bases it is increasingly vital for developmental science studies to include and compare maternal and paternal parenting, as done in the present study.

Similarly, as is well recognized, especially in adolescence, girls and boys develop differently, at different rates, with different interests, and so forth (Soenens, et al., 2019), and parents parent girls and boys differently (Brown & Tam, 2019). With respect to moral values, girls and boys may perceive and respond to parental moral dictates differently (Rote & Flak, 2022). For example, in adolescence boys, but not girls, who perceived stronger parental value ideals increased in their own identification of a moral self (Pratt et al., 2003), and boys, but not girls, perceptions of the appropriateness of their mothers’ parenting predicted more prosocial and less antisocial behavior (Padilla Walker & Nelson, 2010). In consequence, it is equally important to evaluate gender of child in studies of adolescent development, as was done in the present study as well.

Stability and Change in Development

Two central psychometric features of development have to do with stability and change over time. Stability means consistency in the relative ranks of individuals in a group over time, and change points to alteration in the mean level of a group over time; the two are conceptually and statistically independent developmental constructs (Bornstein et al., 2017). Here stability and change in adolescents’ moral values were measured between 14 and 18 years of age. On developmental stability, the fact that individual differences in development may be stable (to some degree) implies that beliefs and behaviors assessed at one point can be assumed to reflect past as well as future beliefs and behaviors. Moral values tend to be stable in maturity (Bardi & Goodwin, 2011; Rokeach, 1973; Schwartz, 1992, 2012) and may be stable in adolescence (Daniel & Benish-Weisman, 2019; Hofmann-Towfigh, 2007; but see Döring et al., 2016). Here, stability of moral values was assessed across 4 years of adolescence. On developmental change, the fact that group mean level changes over time relative to some specific experience gives evidence of the effect of that experience. As maternal and paternal parenting styles were both assessed in the present study, developmental change in adolescents’ moral values could be attributed to one or the other parent.

Specificity

A common assumption in parenting has been that the overall level of parental involvement or stimulation affects the child’s overall level of development, as for example if language acquisition in children were determined by the sheer amount of parent language children hear. In this tradition, classical unidimensional and trait-views derived from psychoanalysis, personality theory, ethology, and attachment theory denoted parenting as “good enough,” “sensitive,” or “adequate” (Ainsworth et al., 1978; Bettleheim, 1987; Brody, 1956; Brody & Axelrad, 1978; Mahler et al., 1975; Rothbaum, 1986; Schaefer, 1959; Symonds, 1939; Winnicott, 1973). Increasing evidence suggests, however, that parenting is multidimensional and modular in character and that nuanced and differentiated pathways govern parenting effects. That is, specific (rather than general) parent cognitions and practices on the part of specific parents appear to relate concurrently and predictively to specific (rather than general) aspects of child competence or performance. It is not the case that the overall level of parenting of mothers and fathers alike directly affects children’s overall level of functioning: Simply providing an adequate financial base, for example, does not guarantee, or even speak to, children’s development of healthy eating habits, an empathic personality, verbal competence, or other valued capacities. Rather, the Specificity Principle states that specific parent cognitions and practices on the part of specific parents at specific times exert specific developmental effects over specific children in specific ways (Bornstein, 2015, 2019). The Specificity Principle is apparently counterintuitive because, according to the Civitas Initiative et al. (2000) national survey, a majority of parents simplistically thinks that the more stimulation a child receives, the better off the child is. However, the Specificity Principle accords with the contemporary relational developmental systems view of parenting (Lerner, 2018).

In brief, to detect regular relations between antecedents in parenting on the one hand and outcomes in child characteristics on the other calls for precision in the combinations of independent and dependent variables. The design of the longitudinal present study, which measured moral values of different kinds in adolescent girls and boys and different parenting styles of mothers and fathers, allowed a test of specificity.

Method

Participants

Participants were adolescents and their mothers and fathers recruited through newspaper advertisements and mass mailings in the Washington, DC, metropolitan area. The first wave of data was collected when adolescents were 14 years old (N = 221, M_age = 13.85, SD = 0.27), and the second wave was collected 4 years later when adolescents were 18 years old (N = 190, M_age = 18.22, SD = 0.36). In total, 246 adolescents (56% boys) and their parents provided valid data for at least one wave. At 14 years, parents reported on their education levels and occupations, based on which family socioeconomic status (SES) was calculated using the Hollingshead Four Factor Index of Social Status (Hollingshead, 1975, 2011; M = 54.33, SD = 10.01; see also Adams & Weakliem, 2011).

The sample was composed of ethnically homogenous but socioeconomically heterogeneous European American families. European American families were chosen because, first, a majority of the population of the United States at the time of data collection identified as European American (U.S. Census Bureau, 2019); second, parent–child relationships vary by ethnicity (Halgunseth, 2019; McLoyd et al., 2019; Ng & Wang, 2019); and third, ethnicity and SES are often confounded (Leyendecker et al., 2005). Therefore, we took the initial step of exploring adolescent moral values in the context of parenting styles in European American families so as not to cloud our findings by mixing ethnicities and with the hope that this strategy would stimulate research on moral value development in families of other ethnicities (Bornstein et al., 2013; Jager et al., 2017).

Procedures

Ethical approval for the study was obtained from the Institutional Review Board of the NICHD. Paper and pencil measures were used for 14-year data collection and internet-based surveys for 18-year data collection. Informed consent/assent from all participants was obtained before data collection at each wave.

Measures

Moral values.

At both 14 and 18 years, adolescents completed the Sociomoral Reflection Objective Measure – Short Form (Basinger et al., 2007; Basinger & Gibbs, 1987; Brugman et al., 2007). Eleven moral value evaluation items cover the areas of contract and truth, affiliation, life, property and law, and legal justice (see Table 1). Adolescents rated the importance of each value on a 3-point Likert-type scale (1 = not important, 2 = important, and 3 = very important). Potential factors underlying the items were explored prior to further analyses (see below).

Table 1.

Moral Value Evaluation Items.

Factor	Item Name	Item Content
Life and Social Contract	SAVEFRND	How important is it to save life of a friend?
	SAVEANY	How important is it to save life of a stranger?
	IMPLIVE	How important is it that person lives if that person doesn’t want to?
	PROMFRND	How important is it to keep promises to friends?
	PROMANY	How important is it to keep promises to someone they hardly know?
	PROMCHLD	How important is it that parents keep promises to children?
Law and Social Order	TOJAIL	How important is it that judges send people who break the law to jail?
	OBEYLAW	How important is it to obey the law?
	NOTSTEAL	How important is it not to take things that belong to other people?
	IMPTRUTH	How important is it to tell the truth?
	HELPPRNT	How important is it that children to help their parents?

Note. Items TOJAIL and IMPLIVE were removed from the final factor model.

Parenting styles.

At 14 years, mothers and fathers independently self-reported their parenting styles using the Parental Authority Questionnaire (PAQ; Buri, 1991). The PAQ contains 10 items for authoritative parenting (e.g., “I take my child’s opinions into consideration when making decisions, but I would not decide to do something simply because my child wanted it.”), 10 items for authoritarian parenting (e.g., “Whenever I tell my child to do something, I expect him/her to do it immediately without asking any questions.”), and 10 items for permissive parenting (e.g., “I don’t view myself as responsible for directing and guiding my child’s behavior while s/he was growing up.”). Each item was rated on a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree). Final scores were computed by averaging the 10 items of each parenting style, with internal consistency Cronbach αs ranging from .67 to .81 for mothers and .67 to .84 for fathers.

Analytic Plan

First, exploratory factor analysis (EFA) on the moral value items evaluated how many factors should be retained using the 14-year data and, next, confirmatory factor analysis (CFA) corroborated the retained factor structure using 18-year data. At a third step, structural equation modeling (SEM) investigated structural relations between mothers’ and fathers’ parenting styles (in separate models) and the moral value factors identified in adolescents. The SEM model was based on measurement-invariant moral value CFA models across the two waves. Adolescent age, gender, and family SES were included as covariates. In a sensitivity analysis, child gender was explored as a moderator of the effects in the mother and father models.

Due to attrition, about 24% of the data could be assumed missing completely at random (MCAR) based on Little’s (1998) MCAR test: χ² (945, N = 246) = 920.39, p = .711. Given the appropriateness of polychoric correlations for categorical or ordinal data (Holgado-Tello et al., 2010), factor analyses were conducted using the weighted least square estimator with mean and variance-adjustment (WLSMV) in Mplus 8 (Muthén & Muthén, 1998–2017). The WLSMV makes use of all of the available data to estimate the factor and path models. Model fit of the factor analysis models was evaluated by χ² statistic, root mean square error of approximation (RMSEA), and Comparative Fit Index (CFI). Adequate model fit is indicated by non-significant χ² test, RMSEA below .08, and CFI equal or above .90 (Brown, 2006; Hu & Bentler, 1999; Yu, 2002).

Results

EFA and CFA Models of Adolescent Moral Values

Table 2 shows descriptive statistics and correlations among the variables in the study. EFA was used to explore how many factors underlay moral value items in the 14-year data. According to the initial eigenvalues greater-than-one rule of thumb (Kaiser, 1960), three factors could possibly be extracted from the 11 moral value items at 14 years. Despite acceptable RMSEA, the EFA model with 1 factor had a significant χ² test. The 2- and 3-factor EFA models showed adequate model fit, with non-significant χ² tests and RMSEAs < .05. Therefore, either the 2- or 3-factor models could be retained based on the EFA results.

Table 2.

Descriptive Statistics and Correlations.

	1	2	3	4	5	6	7	8	9	10	11	12	Mean (SD)
1. Life and Social Contract at 14 y	--												−0.07 (0.65)
2. Law and Social Order at 14 y	0.64	--											−0.02 (0.34)
3. Life and Social Contract at 18 y	0.64	0.53	--										−0.08 (0.67)
4. Law and Social Order at 18 y	0.41	0.70	0.73	--									−0.03 (0.36)
5. Permissive Mothers	−0.16	−0.18	0.01	−0.03	--								20.68 (3.98)
6. Authoritarian Mothers	−0.13	0.06	-0.43	−0.07	-0.23	--							26.70 (5.70)
7. Authoritative Mothers	0.12	0.07	0.19	0.06	-0.12	−0.07	--						40.20 (3.40)
8. Permissive Fathers	−0.04	0.03	−0.08	0.05	0.07	−0.10	0.15	--					21.56 (4.23)
9. Authoritarian Fathers	−0.09	0.15	0.07	0.28	-0.18	0.48	−0.04	-0.18	--				28.26 (6.09)
10. Authoritative Fathers	0.01	−0.06	0.02	−0.05	0.02	-0.15	0.08	−0.08	-0.15	--			39.83 (3.65)
11. Adolescent gender	-0.28	−0.15	0.07	0.02	0.03	−0.02	−0.14	0.14	0.06	−0.06	--		0.56 (0.50)
12. Adolescent age	0.05	0.01	−0.02	−0.03	0.03	0.01	−0.07	0.08	−0.15	0.02	−0.04	--	13.85 (0.28)
13. Family SES	0.26	0.07	0.18	−0.10	0.01	-0.36	0.09	−0.01	-0.42	0.20	−0.11	0.04	54.46 (10.00)

Note. Bold correlations are significant at p ≤ .05.

Patterning after the EFA procedure at 14-years, a 1-factor CFA model was specified in which all 11 items loaded on only one factor in the 18-year data. Two- and 3-factor CFA models were then specified in the 18-year data to confirm the factor structures suggested by the 14-year EFA results. To determine the best model to represent moral values of adolescents, Δχ² tests were conducted to compare the three models. The 2-factor model fit the data better than 1-factor model, Δχ²(1, N = 164) = 6.30, p = .012, but the 3-factor model did not fit the data better than the other two models. Based on these test results and parsimoniousness considerations, the 2-factor model was retained as the final factor structure of adolescent moral values.

The first factor was labeled as “Life and Social Contract” because it mainly captures prosocial aspects of morality that are left to individual choice, such as saving life of a friend or a stranger and keeping promises. The second factor was labeled as “Law and Social Order” because it mainly captures acts that are legally or morally obligatory for individuals to perform, such as obeying the law, not taking things that belong to other people, telling the truth, and helping parents. One item (“TOJAIL”: how important is it that judges send people who break the law to jail) had non-significant factor loading, and a second item (“IMPLIVE”: how important is it that a person lives if that person doesn’t want to) cross-loaded on more than one factor in EFA and had low standardized factor loading in CFA; both items were removed from the factor model.

SEM Models of Parenting Styles and Adolescent Moral Values

To ensure that the moral items were interpreted similarly across waves, SEM models were built based on a measurement-invariant CFA model where the factor loadings from the same latent moral factor to the same indicators were constrained to be equal over time. How mothers’ and fathers’ parenting styles at 14 years predicted adolescents’ moral values at 18 years were examined in separate SEM models, controlling for adolescents’ moral values at 14 years as well as adolescent age, gender, and family SES.

The final SEM mother model showed adequate model fit, χ²(N = 246) = 261.20, p = .012, RMSEA =.03, and CFI = .90. Standardized factor loadings and significant standardized path coefficients are presented in Figure 1. Both moral values factors had high stabilities (.70-.74) across the 4-year period. The moral value factors were also strongly correlated with each other at both 14 years (.63) and 18 years (.67); however, the two factors shared less than half of their common variance at 14 years (40%) and at 18 years (45%). On this statistical basis and because the two factors addressed different theoretical perspectives on moral values, the two factors were treated separately.

Figure 1.

Prospective Associations Between Maternal Parenting at 14 years and Adolescent Moral Values at 18 years.

Note. Adolescent age, gender, and family SES are included as covariates. *p < .05. **p = .01. ***p < .001.

For parenting effects, mothers’ authoritarian parenting style at 14 years (β = −.36, SE = .11, p = .001), but not their authoritative or permissive parenting styles, negatively predicted Life and Social Contract moral values in adolescents at 18 years. For covariates, SES and gender differences in adolescents’ moral values at 14 years emerged, with girls (β = −0.25, SE = .09, p = .004) and adolescents from higher SES families (β = .23, SE = .09, p = .013) having higher Life and Social Contract moral values at 14 years. For 18-year moral values, only a gender difference emerged: Boys showed higher increases in Life and Social Contract moral values from 14 to 18 years than girls (β = .27, SE = .11, p = .010). The final SEM father model showed nearly adequate model fit, χ²(N = 246) = 268.46, p = .005, RMSEA =.03, and CFI = .89. None of the parenting styles of fathers at 14 years predicted adolescents’ moral values at 18 years.

Sensitivity analyses tested the moderating role of adolescent gender. Maternal authoritarianism, but not the other two parenting styles, interacted with gender to predict adolescent Life and Social Contract moral values (β = −.36, SE = .18, p = .047). Mothers’ authoritarian parenting was more negatively associated with moral values in their adolescent boys (β = −.43, p < .001) than girls (β = −.07, p = .728). None of the father parenting styles interacted with gender to predict adolescent moral values at 18 years.

Discussion

Many compelling issues characterize modern perspectives on moral development (Killen & Smetana, 2022). However, some basic developmental and parenting information is still lacking from a rounded account of adolescent moral development and its influences. The present study coordinated the identification and development of moral values in adolescence, parenting styles, and gender with issues of stability and specificity to help bridge gaps in the literature. The primary research questions asked about the structure of moral values in adolescence and whether parenting styles of mothers and fathers at 14 years influenced the development of adolescent moral values at 18 years, and secondary research questions asked about stability in adolescent moral values and whether adolescent gender moderated parent-adolescent predictive relations.

To summarize the main results of this study, EFA and CFA established a 2-factor model of moral values across two ages of adolescence; the two factors were Life and Social Contract and Law and Social Order. Structural equation modeling investigated structural relations between parental parenting styles of mothers and fathers and these two moral value factors controlling for adolescent age, gender, and family SES. The two moral values factors had high stabilities across the 4-year period from 14 to 18 years. In terms of prediction, mothers’ authoritarian parenting at 14 years, but not their authoritative or permissive parenting, negatively predicted Life and Social Contract moral values in adolescents at 18 years. Gender and SES differences in adolescents’ moral values at 14 years emerged with girls and adolescents from higher SES families having higher Life and Social Contract moral values at 14 years, but boys experienced more increases in Life and Social Contract moral values from 14 to 18 years than girls. None of the 14-year parenting styles of fathers predicted adolescent moral values factors at 18 years. Mothers’ authoritarian parenting was more negatively associated with moral values in adolescent boys. Here these main findings are elaborated.

First, as measured in this study adolescent moral values divide into 2 factors. One factor called Life and Social Contract captured prosocial aspects of morality that are left to individual choice, and the second factor called Law and Social Order captured acts that are legally or morally obligatory for individuals to perform. Notably, these two factors roughly match values of caring for others and not harming others that have been identified as prominent across a host of cultures (Schwartz & Bardi, 2001; Schwartz et al., 2012). That said, the composition of moral values could also reflect certain methodological and demographic parameters of a study. Here, for example, the Basinger and Gibbs (1987) Sociomoral Reflection Objective Measure was used when other measures of moral values could result in other factorial compositions of moral values. The 2-factor solution achieved here also comports with current views of domain specificity in children’s mental life and social judgments (Bornstein & Putnick, 2019; Turiel, 2015).

Likewise, the sample consisted of ethnically homogenous European American families and was not representative of the U.S. population. Although the specific findings may or may not generalize to other populations, the findings do generalize to a known population of similar composition to this sample (Bornstein et al., 2013; Jager et al., 2017). Comparative evaluations of the composition of moral values with other instruments as well as in other ethnic and cultural samples of adolescents is called for.

Second, both moral values factors, Life and Social Contract and Law and Social Order, showed high stabilities over 4 years from 14 to 18 years of age. As the study of stability of psychological constructs across development is a principal concern of developmental science (Bornstein et al., 2017), identification of the maintenance of rank order of moral values in a large group of adolescents constitutes a valuable contribution of this study from a developmental science perspective in and of itself. Moreover, stability of individual differences has value for its implications for intervention (Killen & Rutland, 2022; Killen et al., 2022; Lecce et al., 2014). The fact that adolescents maintain their relative positions in a group from 14 to 18 years of age suggests that influences that shape individual variation in moral values likely take place before adolescents reach 14 years. Interventions designed to affect the trajectory of growth of moral values should be implemented before adolescents reach puberty.

Third, of three parenting styles as expressed by the two parents, only mothers’ authoritarian parenting predicted only the adolescent moral value factor of Life and Social Contract. Mothers who were less authoritarian in their parenting with their 14-year-olds had 18-year-olds who increased in valued prosocial aspects of morality. Put another way, mothers who were more authoritarian in their parenting with their 14-year-olds had 18-year-olds who less deeply valued prosocial aspects of morality. These predictive results are relatively conservative in that they controlled for adolescents’ moral values at 14 years as well as adolescent age, gender, and family SES. What might account for them? A next important step in this line of research would be to identify processes underlying this dynamic. Authoritarian parenting entails coercive control that can consist of arbitrary discipline, verbal hostility, and harsh physical punishment. In turn, authoritarian parenting often engenders resentment, evasion, and withdrawal in children (Grusec, 2022). Psychological control, which is a constituent of authoritarian parenting, is characterized by love withdrawal, criticism in the form of shame and disappointment, and possessiveness and overprotection. Psychological control is associated with increased anxiety and social aggression especially during adolescence (Kuppens et al., 2013). Parental resort to psychological control is known to undermine regard for the legitimacy of parental authority (Kuhn & Laird 2011) and enhance undesirable moral values (Gershoff, 2002) in adolescents and thereby alter connections between parenting and adolescents’ moral behavior (Qi, 2019). Notably, in this connection, authoritarian parenting is associated with negative perceptions of children and low levels of warmth (Rudy & Grusec, 2001), and authoritarian parenting (accounting for authoritative parenting) is negatively related to children’s concern for others and positively related to their disregard for others (Hastings et al., 2000). Perhaps due to their gender or temperament, boys perceive and judge authoritarian parenting, which is overcontrolling, hostile, or rejecting) in ways that discourage moral values and promote problem behaviors (Sentse et al., 2009; Soenens & Vansteenkiste, 2020). A last point concerning the role of parenting in adolescents’ acquisition of moral values is that, in accord with the extant literature (Grusec & Goodnow, 1994), mothers seem to play a more compelling part than fathers.

Finally, four perspectives on the results support a specificity view of extrinsic influences over moral value development in adolescence. One is that maternal, and not paternal, parenting predicted adolescent moral values. The second is that authoritarian, and not authoritative or permissive, parenting was the predictive parenting style. The third is that the Life and Social Contract values factor, and not the Law and Social Order factor, was affected by mothers’ antecedent authoritarian parenting style. Although the results were largely the same for girls and boys, fourth, mothers’ authoritarianism parenting style proved to be particularly detrimental to boys’, but not girls’, Life and Social Contract moral value development.

Limitations and Future Directions

This study deepens our understanding of adolescent moral values, their development, and parenting styles that influence them as well as the roles of gender, stability, and specificity in adolescent moral value development. However, the study also has some limitations. Three have been alluded to already: Two relate to sampling and one to methodology. The sample was ethnically homogeneous European American. European Americans are the largest ethnic population in the United States, and so the results for adolescent moral value structure and parent style influences validly generalize to that majority population. However, the sample is not nationally representative and the composition of ethnicity in the United States is changing. Future studies should examine the factor structure of moral values as well as the predictive validity of parenting styles for adolescent moral values in other ethnic and cultural groups. Parental roles, involvement, and parent-child relationships vary across ethnicities and cultures as well as across socioeconomic statuses (Halgunseth, 2019; Hoff & Laursen, 2019; McLoyd et al., 2019; Ng & Wang, 2019), which means the nature and structure of adolescents’ moral values as well as parenting effects on adolescents’ moral values are also likely to vary. Last, one specific measure of adolescents’ moral values was used. The nature and structure of adolescent moral values likely also reflect the instruments used to measure moral values, and so adolescent performance on other measures of moral values should be explored and compared with the current findings. In addition, other individual difference factors in adolescents were not accounted for here, but should be taken into consideration. For example, adolescents vary in their perceptions of parenting (Rote & Flak, 2022) as well as in their temperaments (Sentse et al. 2009), both of which may moderate their reactions to parenting.

Conclusions

Morality is foundational to human existence and has ramifications in multiple domains of personal and interpersonal life from resource allocation to empathy to conflict resolution. On a personal plane, moral values are vital to wholesome child development as children who lack moral values are at risk for negative relationships well into maturity. On an interpersonal plane, moral values constitute vital antecedents and likely causes of justness, fairness, and social harmony (Rutland & Killen, 2015).

Adolescents’ moral values can be divided into those concerned with prosocial aspects of morality that are left to individual choice, such as saving life of a friend or a stranger and keeping promises, and those concerned with acts that are legally or morally obligatory for individuals to perform, such as obeying the law, not taking things that belong to other people, telling the truth, and helping parents. Across a substantial fraction of adolescence, the two sets of moral values appear to be stable. That said, between mothers and fathers, between different parenting styles, between moral value factors of Life and Social Contract and Law and Social Order, and between girls and boys, a certain specificity of development appears to hold in that mothers may selectively and effectively foster Life and Social Contract moral values by refraining from engaging in an authoritarian parenting style, especially with their adolescent sons.