Adolescence is generally acknowledged as a dynamic and critical stage of human development in which individuals rapidly develop capacities for independence, seek to become more autonomous from their parents, and make decisions that can frame their developmental pathways or trajectories (R. M. Ryan et al., 2006; Shlafer et al., 2014). It is also challenging for parents to teach adolescents fundamental values, to guide their regulations, and to support the development of their autonomy (Joussemet et al., 2008; R. M. Ryan et al., 2006). This study focused on the stability of and the changes in autonomy-related parenting profiles and their effects on adolescents’ academic and psychological development in the period from early adolescence to mid-adolescence.
Although various frameworks of parenting practices have been examined (Bornstein, 2015), this study investigated them from the perspective of Self-Determination Theory (Deci & Ryan, 2000). Self-Determination Theory postulates that human behaviors are driven by three universal and innate psychological needs, namely autonomy, competence, and relatedness. Autonomy––the sense of psychological liberty and freedom––weaves throughout the broad framework of Self-Determination Theory (Vansteenkiste et al., 2010) and plays an important role in young people’s internalization of societal norms and rules, the development of motivational orientations, and self-regulation (R. M. Ryan et al., 2006). From the perspective of Self-Determination Theory, adolescents’ socialization contexts can be categorized as autonomy-supportive contexts and controlling contexts (see also Assor et al., 2004). Inspired by this theory, autonomy-supportive parenting, characterized by acknowledging one’s child’s view and encouraging self-initiated activities, has proven to support students’ academic and psychological adjustment (for an overview, see Vansteenkiste & Ryan, 2013). In contrast, controlling parenting, which counteracts young people’s autonomy through, for example, psychological control and conditional regard, can significantly increase adolescents’ risks of developing psychopathology and impair their academic achievements (e.g., Assor et al., 2014; Otterpohl et al., 2019).
Instead of studying parenting as having discrete dimensions, this study used a person-oriented approach and aimed to identify the combinations of parenting dimensions as an undivided whole (Bergman & Trost, 2006; for a classic line of research, see Baumrind, 1995). More importantly, it further aimed to expand our knowledge in this area by examining autonomy-related parenting profiles’ stability and changes and their longitudinal effects on adolescents’ academic achievements, prosocial behavior, and psychopathology from early adolescence (i.e., 10–13 years, also preadolescence) to mid-adolescence (i.e., 14–17 years). To pursue these objectives, five-year longitudinal data were used within a large-scale German project that included both adolescents’ self-reports and parent reports. Latent profile analyses (LPA) and latent transition analyses (LTA) were employed to examine parenting profiles on the basis of dimensions of Self-Determination Theory.
Parenting Dimensions from a Self-determination Perspective and their Effects
To date, the parenting literature has been dominated by the two classic parenting dimensions, namely parental responsiveness and parental demandingness (Maccoby & Martin, 1983). Responsiveness––also warmth––refers to being accepting, sensitive to children’s needs, and emotionally warm, whereas demandingness––also behavioral control––is defined as parenting practices in which parents communicate clear expectations in terms of appropriate behaviors and use rules, instructions, and restrictions to regulate and monitor their children’s behavior (Barber et al., 2005; Baumrind, 1995).
Self-Determination Theory is another theory that has been used to examine the role of parenting in adolescent development. According to this theory, critical parenting practices should support the need for autonomy (i.e., feeling psychological liberty and freedom of internal will), for competence (i.e., feeling able to affect one’s environment), and for relatedness (i.e., feeling bonded and cared for; Deci & Ryan 2000; Vansteenkiste et al., 2020). Although many parenting practices fall into these categories (e.g., provision of structure; Griffith and Grolnick, 2014), the most important practices are autonomy-related supportive practices as they are a prerequisite for the unfolding of positive effects of other supportive parenting practices (Deci & Ryan, 2000; Griffith & Grolnick, 2014). This study investigated four parenting practices––autonomy support, warmth, psychological control, and conditional regard––as they are the most relevant practices that link to the needs for autonomy. Although autonomy-related practices are assumed to be critical for children’s and adolescents’ development, a comprehensive examination of them is still lacking. The present study thus aimed to fill this gap with the following four dimensions.
Parental autonomy support refers to parents’ active support of their children’s capability to be autonomous and self-initiating by acknowledging their children’s perspectives, allowing and encouraging them to experiment, giving them opportunities to make choices, and providing explanatory rationales for specific expectations (Deci & Ryan, 2012). Meta-analytical (Vasquez et al., 2016) and cross-cultural (Vansteenkiste et al., 2005; Zhang et al., 2017) evidence has shown the positive effect of parental autonomy support on adolescents’ self-regulation, adaptive psychosocial functioning, and academic success. Like Baumrind’s theory, Self-Determination Theory vales parental warmth (a core aspect of involvement; for detail, see Grolnick, 2009). Empirical studies have also found that parental warmth is closely related to autonomy support (for detail, see R. M. Ryan et al., 2006). Self-Determination Theory regards parental warmth as integral to autonomy support. That is, adolescents feel warmly connected to their parents only to the extent that their real selves are accepted by their parents. In contrast, the quality of relatedness suffers if autonomy is perceived as absent (R. M. Ryan et al., 2006). By showing care, support, and compassion, parents can support their children’s needs for relatedness (Grolnick, 2009). Past research has shown parental warmth to have a positive effect on adolescent academic and psychological development as well as on prosocial behaviors (see meta-analysis, Pinquart, 2017).
The opposite of parental autonomy support is parental psychological control, which frustrates adolescents’ needs for autonomy (Joussemet et al., 2008; R. M. Ryan et al., 2006). Psychological control is defined as “parental control that intrudes on the child’s psychological world” (Joussemet et al., 2008, p. 195). Psychological controlling practices include withdrawing love, inducing guilt, shaming, and invalidating the child’s perspective (Barber et al., 2005). Previous studies have consistently shown that parental psychological control is related to lower academic achievement, more externalizing and internalizing symptoms, and fewer prosocial behaviors (e.g., Pinquart, 2017; Wong et al., 2021). Lastly, parental conditional regard (PCR) is the practice through which parents show love and appreciation when children fulfill their parental expectations (Assor et al., 2014). PCR can be divided into conditional negative regard (PCNR) and conditional positive regard (PCPR). In the former, parents withdraw attention and affection when the child fails to comply with their expectations, whereas in the latter, more attention and affection is given when the child acts as expected (Roth et al., 2009). Hence, parents who use PCR strategies engage in a controlling manner in their children’s development by providing relatedness at the expense of autonomy (Vansteenkiste et al., 2010). PCR and psychological control (e.g., love withdrawal) are closely related but also distinct from each other. First, psychological control is a general parenting technique, whereas PCR is domain specific (e.g., academic, affective, and behavioral domains). Second, psychological control contains components of blame that the child cannot change or influence through behavior, whereas PCR refers to showing esteem and attention depending upon the child’s behavior (Assor et al., 2014; Roth et al., 2009). Previous studies have highlighted how PCR has a wide range of detrimental effects on child and adolescent development. Both forms of PCR can cause introjected internalization of parents’ expectations, increase the child’s internal stress, diminish well-being (for an overview, see Assor et al., 2014), and increase the risk of self-regulation failures (e.g., Curran et al., 2017) and the development of internalizing and externalizing problems such as aggressive behaviors, negative emotions, anxiety, and depression (e.g., Otterpohl et al., 2019).
Person-oriented Parenting Approaches
Although parenting dimensions contribute to a vast knowledge on the role of parenting in adolescent development (Bornstein, 2015), the combinations of various parenting dimensions (e.g., parenting styles/profiles) offer a deeper understanding of good parenting (Bornstein, 2015; Darling & Steinberg, 1993; Steinberg et al., 1992). One classic example is the four parenting styles based on responsiveness and demandingness (Maccoby & Martin, 1983): authoritative (high responsiveness and high demandingness), authoritarian (low responsiveness and high demandingness), permissive (high responsiveness and low demandingness), and neglectful (low responsiveness and low demandingness). However, this framework has been criticized for not addressing the important aspect of granting autonomy (e.g., Steinberg et al., 1992). Moreover, the method that creates these parenting styles/profiles is either a scale-mean method (i.e., plus or minus one SD of the mean as high and low levels of a dimension) or a median-split method (i.e., above and below the median as high and low levels of a dimension). These methods are arbitrary in defining cut-off points and are largely inappropriate when multiple dimensions (more than three) are used (Morin & Litalien, 2019). This study adopted a mixture of modeling-backed and person-oriented approaches that can reflect the natural configuration of the combinations of dimensions (Bergman & Trost, 2006). A further advantage of this approach is that its extension to latent transition analysis enables the estimation of the stability of and changes in group membership in profiles (Lanza et al., 2003; Morin & Litalien, 2019).
To date, comparatively fewer studies have examined parenting profiles based on dimensions of Self-Determination Theory, particularly autonomy-related parenting practices. Furthermore, the literature has not attempted to include PCR in the testing of parenting profiles. This study focused on the central idea of Self-Determination Theory, namely an autonomy-supportive vs. a controlling socialization climate at home, by incorporating key parenting dimensions (i.e., autonomy support, warmth, and psychological control) and conditional regard. Thus, unlike previous studies that have examined many parenting dimensions, this study sought to identify parenting profiles associated with adolescents’ needs for autonomy. According to Self-Determination Theory, parental autonomy support facilitates children’s experience of autonomy, whereas parental psychological control undermines children’s internal will and thus frustrates their needs for autonomy. Various researchers have regarded autonomy support and psychological control as two sides of the same coin (e.g., Joussemet et al., 2008; Yotyodying et al., 2020). Therefore, finding parents who are autonomy supportive and psychologically controlling is not very likely. However, from our point of view, low control does not equal autonomy support. The present study hypothesized a supportive parenting profile, characterized by high scores in both supportive dimensions (i.e., autonomy support and warmth) and low scores in both controlling dimensions (i.e., psychological control and conditional regard). Conversely, a highly controlling profile was also expected, characterized by low scores on the supportive dimensions and high scores on the controlling dimensions. Further, it was expected to find a profile in which parents are neither supportive nor controlling. PCR refers to parents’ domain-specific controlling parenting strategies. Following Assor et al. (2014), parents can be autonomy supportive in general and use these strategies in specific domains. Hence, it is reasonable to assume that some parents are generally supportive and regularly use conditional regard strategies.
Parenting Profiles and Adolescent Development
Adolescence is a critical time in the development of academic and psychological functioning. The etiology of the development of academic and psychological maladjustment is multifactorial. Apart from biological, genetic, and situational predispositions, needs-thwarting parenting is seen as the most significant risk factor from the Self-Determination Theory perspective (W. S. Ryan & Ryan, 2019). To understand the role of autonomy-related parenting profiles in adolescent development, this study evaluated multiple domains of adolescent outcomes. These included academic achievement, internalizing and externalizing problems, and prosocial behavior. In the present study, internalizing problems refer to emotional problems, social withdrawal, and associated problems integrating into peer groups. Externalizing problems in turn include hyperactivity, distractions, and delinquent and aggressive behaviors (A. Goodman & Goodman, 2009). In contrast, prosocial behavior is a positive outcome linked to greater academic and social adjustment, and this promoting effect persists into adulthood (Brook et al., 2013). Furthermore, adolescents and their parents often provide divergent assessments of adolescents’ psychological outcomes (Lohaus et al., 2020), and this leads to difficulties in interpretation. In addition to adolescents’ self-reports, the present study also took parent reports of adolescents’ internalizing and externalizing problems and prosocial behavior into consideration.
As previously described, the positive effects of parental autonomy support and warmth, and the detrimental effects of psychological control and conditional regard on adolescents’ academic and psychological adjustment have been well documented. Attempts have been made to examine the effects of different parenting dimension constellations on adolescent outcomes. Several studies have incorporated some autonomy-related parenting dimensions. In a two-wave Portuguese study (Pereira et al., 2009), parenting profiles were created on the basis of parental warmth, rejection (a form of psychological control), and overprotection (low autonomy granting). Four parenting profiles were found and labeled Low Support, Supportive-Controller, Rejecting-Controller, and Supportive. Of these, the Rejecting-Controller profile (i.e., low warmth, high rejection, and high overprotection) was the theoretically most maladaptive profile and was related to the highest level of children’s behavioral problems, whereas the Supportive profile was the most adaptive profile associated with the lowest level of behavioral problems. A more recent study (Shen et al., 2020) focused on the effects of parenting profiles on internalizing problems among Chinese primary school children. The same parenting dimensions were used as those in the Portuguese study (Pereira et al., 2009), and similar profiles were found. The results further showed that the most maladaptive profile was associated with the highest risk of children’s emotional maladjustment. Some longitudinal studies (e.g., Kim et al., 2013; Zhang et al., 2017), which integrated autonomy support or psychological control into the classic two-dimension typology (Baumrind, 1995; Maccoby & Martin, 1983), have consistently shown that autonomy-granting belongs to adaptive parenting profiles and buffers adolescents’ development of psychopathology, whereas psychological control belongs to maladaptive profiles and increases the risk of psychopathology.
Based on these findings, it was expected that the supportive parenting profile would be linked to the lowest scores in adolescent internalizing and externalizing problems, the highest school performance, and the most prosocial behaviors, whereas inverse relationships were expected in the controlling profile. Being neither supportive nor controlling was expected to detrimentally affect adolescents’ development because their needs are not supported. If parents were generally supportive but showed conditional regard in specific domains, negative influences on adolescent outcomes were expected.
When examining the effects of autonomy-related parenting profiles on adolescent outcomes, multiple time-invariant confounders should be taken into consideration. This study was conducted in Germany, where the education system is characterized by high levels of aggregation and social inequality (OECD, 2019). For example, girls and students from high socioeconomic families are overrepresented in the highest track secondary school, whereas boys, students with a migrant background, and students from low socioeconomic families are overrepresented in the lowest track secondary school (Kessels et al., 2014). Moreover, socioeconomic status is associated with adolescents’ mental health and psychological adjustment (Klipker et al., 2018). Parenting styles/profiles vary as a function of social stratum (for an overview, see Hoff & Laursen, 2019) and can be additionally influenced by adolescents’ gender (for detail, see Bornstein, 2013). It has been demonstrated that parents of low socioeconomic status are more controlling in their parenting style (e.g., Benner et al., 2016), and that boys are more likely to experience controlling parenting than girls (e.g., Bornstein, 2013). Therefore, the present study included adolescents’ gender, school type, migration background, and socioeconomic status as covariates.
Parenting Profiles’ Stability and Changes
In addition to the limitations described above, another drawback of the traditional parenting profile studies is that they assume that parenting styles/profiles are largely stable, or put little effort into examining changes in these profiles (e.g., Hoeve et al., 2008; Pereira et al., 2009). Parenting styles/profiles can indeed change, particularly in a period that witnesses drastic changes, such as adolescence (Kuczynski & Parkin, 2007; Sameroff, 2010). In a study with a sample of 2173 Chinese adolescent students, Zhang et al. (2017) identified four parenting profiles (authoritative, authoritarian, average-level undifferentiated, and strict-affectionate), based on six dimensions: warmth, inductive reasoning, encouragement of independence, encouragement of achievement, supervision, and harshness. From childhood to early adolescence, about one-third of authoritative, strict-affectionate, and average-level undifferentiated mothers changed parenting profiles. The profile stability of authoritarian parenting was 50–60%. The authors speculated that authoritarian parents shifted to other profiles due to the modification of their parenting behaviors. In their study of 444 Chinese American parent-adolescent dyads over eight years, Kim et al. (2013) found four parenting profiles (supportive, tiger, easygoing, and harsh) based on eight parenting dimensions (warmth, monitoring, democratic parenting, inductive reasoning, hostility, psychological control, shaming, and punitive parenting). The latent profiles were cross-sectionally examined and compared on the basis of the number of profiles and mean patterns. Their results showed that from early to late adolescence, the proportion of tiger mothers tended to decrease, whereas the proportion of tiger fathers tended to increase. These findings suggest that parents adapt their parenting to a more autonomy-supportive manner in response to their children’s increasing needs for autonomy.
To date, however, longitudinal studies of parenting profiles that highlight autonomy-related practices based on the Self-Determination Theory are still lacking. Consequently, the stability of and change in these profiles are far from well understood. On the other hand, research on the stability of and change in parenting profiles could greatly contribute to the understanding of parenting dynamics and their impact on adolescents’ academic and psychological development.