Introduction

The ability of students to self-regulate is increasingly important nowadays, especially after the pandemic, which has profoundly changed the modern way of life. The pandemic has highlighted the importance of students gaining autonomy and taking control of their own learning (Trias et al., 2021), especially for students from vulnerable groups who need continuous attention and support to ensure equal access to education (Jeriček Klanšček et al., 2021). SRL is a fundamental process in learning and academic achievement (Dermitzaki & Kallia, 2021). It acts as an indirect factor between individual student characteristics and learning performance (Dent & Koenka, 2015). Interactions with parents and teachers provide an important interpersonal context for the development of SRL (Pino-Pasternak & Whitebread, 2010).

Self-regulated learning and educational inequality

SRL can be defined as the ability of learners to regulate their attention, emotions, and behaviour in order to respond optimally to external and internal demands of the environment (Boekaerts & Corno, 2005; Pintrich, 2005; Schunk & Zimmerman, 2008). It is a complex and multifaceted construct that includes cognitive (e.g., memorization, elaboration, comprehension strategies), metacognitive (e.g., planning, goal setting, monitoring) and motivational (e.g., goal orientation, self-efficacy) aspects of learning (Zimmerman, 2013). The metacognitive aspect refers to the student’s planning, goal setting, organisation, monitoring and self-assessment of the learning process, while the cognitive component involves the use of various learning strategies and tactics to memorise material. The use of cognitive and metacognitive strategies is influenced by the motivational component, which includes beliefs about one’s effectiveness, interest in a task, attributions, and strategies to regulate motivation and emotions (Wolters, 2003). Panadero (2017) asserts that SRL is an umbrella construct within which all variables that influence learning are comprehensively examined (e.g., self-efficacy, metacognition, volition, cognitive learning strategies). It is a dynamic and circular process that involves feedback loops, meaning that students, who are self-regulated learners, set goals and metacognitively monitor their progress toward achieving those goals. Students are active and respond to their monitoring, as well as external feedback, in ways that they believe will help them achieve a goal, such as putting more effort into learning or changing their learning strategy (Schunk & Greene, 2018). In the present study, the theoretical framework of SRL is represented by Ziegler and Stoeger’s (2005) seven-stage cyclical model, which is very similar to Zimmerman’s cyclical model (Zimmerman, 2005). In this model, learning begins with the student’s self-assessment of his or her own learning, followed by goal setting, strategic selection of an effective learning strategy, and strategy implementation; then comes monitoring and possible adjustment of the strategy, and finally, outcome assessment (Ziegler et al., 2012).

Students’ use of SRL learning strategies is context-dependent and varies by subject area (Pintrich, 2005). Although empirical studies have not yet systematically examined the influence of individual subject areas on students’ use of cognitive, metacognitive, and motivational learning strategies, research indicates the domain specificity of SRL (Alexander et al., 2011). Vandevelde et al. (2015) showed that SRL is a dynamic process that differs for students within individual subject areas or tasks, and even between subject areas. Subjects differ in terms of content structure, typical tasks, and appropriateness of use of learning strategies, all of which impact SRL (Schunk & Greene, 2018).

The ability to self-regulate learning is one of the most important protective factors for low socio-economic status (SES) students who have learning difficulties (Azevedo et al., 2023). Research showed that students from low-SES backgrounds have less developed SRL strategies (Pappas et al., 2003; Trias et al., 2021; Vandevelde et al., 2017) and tend to have lower academic achievement than their peers (Boonk et al., 2018). In Slovenia, higher SES students with more books at home showed better reading achievement than their lower SES peers (Klemenčič & Mirazchiyski, 2023). In addition, an international report from PISA showed that the percentage of Slovenian students from the lowest quartile of SES who achieved at least Level 2 in reading was 21% lower than the percentage from the highest quartile, which is slightly lower than the OECD average of 29% (OECD, 2021). However, the paradigm of inclusive education in Slovenia enables different groups of students (with migrant backgrounds, with low-SES backgrounds, with special needs) to receive special individual support and targeted services (e.g., additional special support, assistive devices) in schools (Kavkler et al., 2015; Skubic Ermenc, 2020).

The social context of self-regulated learning

According to the social cognitive theory of SRL, parents and teachers provide an important model for students because they provide them with the support they need to imitate and apply self-regulation skills. At the same time, their presence serves as an external reinforcer for the student’s behaviour (Martinez-Pons, 2002; Zimmerman, 2013). The development of students’ SRL is not innate but the result of an interaction between students’ maturational processes and their education (de Ruig et al., 2023; Moos & Ringdal, 2012; Paris & Newman, 1990). In addition to school experiences, students’ learning experiences in the family context play a very important role in the development of their learning skills and strategies (Dermitzaki & Kallia, 2021). SRL is not a trait. Although individual differences in SRL have been reported (Cadima et al., 2016; Denton et al., 2015; Li et al., 2018; Peeters et al., 2016), research has shown that all students, regardless of intellectual and academic ability, can improve their SRL, which contributes to higher academic achievement (Dent & Koenka, 2015). Despite awareness of the importance of SRL, some students still struggle to effectively regulate their learning (Kron-Sperl et al., 2008; Trias et al., 2021). Therefore, researchers emphasise the need to systematically develop and promote SRL in the initial period of education (van der Stel & Veenman, 2010; Vandevelde et al., 2017).

Learning environments that promote SRL are influenced by several factors: teachers’ beliefs about the importance of SRL, their competence to promote and guide the process of SRL, and their predominant teaching style (Lombaerts et al., 2009). Teachers can promote students’ SRL in the following ways: by engaging students in more complex, open-ended activities in which they have the opportunity to select tasks and determine their level of difficulty; by providing students with appropriate instrumental support; and by encouraging peer support for learning; by creating situations in which students can observe the use of specific learning strategies (Paris & Paris, 2001); by encouraging students to consciously use cognitive, metacognitive and motivational learning strategies, and to reflect on their use; by providing appropriate feedback that encourages students to focus on their own learning progress (Askell-Williams et al., 2012); and by encouraging students to attribute success to the use of appropriate learning strategies rather than to ability or luck (Pintrich, 2005).

The development of SRL is influenced by parental characteristics (e.g., parental capacity for self-regulation, temperament, emotional expression, mental health), parenting (e.g., parental involvement, parenting styles, parental sensitivity), the parent–child relationship (attachment styles), family structure, and the family environment (Baker, 2018), and the frequency and quality of self-regulation that the family environment requires of the child (Dermitzaki & Kallia, 2021). The influence of the family context and the role of parents in the development of a child’s self-regulation varies depending on the age of the child. As children develop, they become more independent of their parents in regulating their own behaviour and emotions (Eisenberg et al., 2010). Co-regulation of learning between parents and children gradually transforms into self-regulation through internalisation processes (Erdmann & Hertel, 2019) .

Parenting styles

Parenting styles represent a relatively stable form of child rearing and include a range of parenting approaches, goals, and patterns of behaviour toward the child in various situations. Parenting styles include the quality of the interaction between the child and the parents, parental attitudes toward the child, and parental behaviour, as well as the emotional atmosphere in the family (Steinberg, 2001). Through emotional attachment to parents, children internalise their parents’ values and learn how to form optimal relationships with other adults (Gregory & Weinstein, 2004). A positive relationship with parents provides a form of social support that enhances students’ ability to self-regulate (Balaguer et al., 2021). Baumrind’s (1971, 2013) classification of parenting styles, which is the most widely used and researched, includes two basic dimensions: demandingness and responsiveness. Demandingness includes parenting behaviours in which parents exhibit control, maturity demands, and monitoring, whereas responsiveness (i.e., affection) includes parenting behaviours in which parents show emotional warmth, acceptance, and involvement. Authoritative parents are characterised by high levels of both demandingness and affection, authoritarian parents by high levels of demandingness and low levels of affection, and permissive parents by too high levels of affection and low levels of demandingness (Maccoby & Martin, 1983).

Previous studies have shown that authoritative parenting is related to high student academic achievement and positive developmental outcomes such as independence, the ability to self-regulate, establishing appropriate social interaction with peers, the tendency to master the environment and learn new things, and flexibility in the school environment, both cross-sectionally (Baumrind, 1991; Darling & Steinberg, 1993; Gagnon et al., 2013; Grolnick & Ryan, 1989; Steinberg, 2001) and longitudinally (Baumrind et al., 2010; Steinberg et al., 1992). In a meta-analytic study, Pinquart (2016) showed that authoritative parenting style, parental responsiveness, emotional warmth, behavioural control, and autonomy-promoting parenting behaviours were associated with better academic achievement in students. Contemporary research has demonstrated associations between parenting practises and various components of SRL (Pino-Pasternak & Whitebread, 2010; Spera, 2005; Thomas et al., 2019). Lee et al. (2012) found that SRL is an important mediator between students’ perceptions of parenting styles and their academic achievement. Similarly, Alnafea and Curtis (2017) reported that authoritative parenting style is positively associated with student self-efficacy, use of cognitive and metacognitive strategies, and effective time management. Similarly, Amani et al. (2020) showed that an authoritative parenting style was positively associated with adolescents’ academic achievement through increased SRL. In addition, emotional support from the father was found to foster students’ self-monitoring and metacognitive talk in the classroom (Stright et al., 2001). Kallia and Dermitzaki (2017) reported that mother’s autonomy-supportive behaviour was positively associated with children’s actual use of cognitive strategies, planning, and monitoring skills during a problem-solving task. Recently, Du et al. (2021) reported that emotional warmth and democratic discipline from parents promote SRL in adolescent students.

Teaching styles

Teachers play an important role in creating learning environments that support students’ basic needs for autonomy, competence, and connectedness (Deci & Ryan, 2008) and allow students to regulate their own learning. Teaching style influences students’ academic achievement, social behaviour, educational beliefs, and aspirations (Ansari et al., 2020; Kiuru et al., 2012; Walker, 2008). Researchers have argued that teaching style in elementary school is similar to parenting style (e.g., Pianta et al., 1997; Walker, 2008; Torff & Kimmons, 2021). Teaching style, like parenting style, consists of attitudes, behaviours, and nonverbal communication that characterise the teacher–child relationship (Kiuru et al., 2012). A considerable amount of research on teaching styles is based on Baumrind’s (1971, 2013) theoretical framework for parenting styles, which includes a mix of teachers’ responsiveness (i.e., warmth, caring) and demandingness (i.e., requirements, control) (e.g., Ertesvag, 2011; Torff & Kimmons, 2021).

Theory and research suggest that an authoritative teaching style, characterised by high levels of responsiveness and demandingness, is associated with higher student academic achievement (Dever & Karabenick, 2011; Walker, 2008). There is some evidence that an autonomy-supportive teaching style (Reeve, 2009; Reeve & Cheon, 2021), which incorporates certain instructional behaviours (e.g., providing explanatory rationales) and supports students’ motivational development and ability to self-regulate autonomously, positively impacts younger students’ self-regulation skills in biology (Whitebread & Grau Cardenas, 2012). Research suggests that teachers’ autonomy support and structure is positively related to students’ SRL learning strategies (de Ruig et al., 2023; Sierens et al., 2009; Vansteenkiste et al., 2012). Moreover, in a longitudinal study, Schuitema et al. (2016) reported that students’ perceptions of teacher autonomy support had positive effects on delay of gratification and the use of metacognitive strategies. Similarly, Zee and de Bree (2017) found that students’ perceptions of the quality of the student–teacher relationship and closeness were directly related to two areas of self-regulation (i.e., task orientation and metacognition). Theory and research suggest that there is a relationship between authoritative teaching and a number of positive outcomes, including academic achievement and student engagement (Kiuru et al., 2012; Vansteenkiste et al., 2012; Yin et al., 2009; Walker, 2008). Teachers who possess SRL skills and have warm and supportive relationships with their students are effective role models for elementary students to regulate their learning (de Ruig et al., 2023).

The present study

SRL is related to student learning competence, positive peer and teacher relationships, and student satisfaction with school (Lee et al., 2012). Students whose self-regulatory skills are less developed often have difficulty connecting with peers and building relationships with teachers and are less successful in school (Piotrowski et al., 2013; Vandevelde et al., 2017). Teachers and parents represent an important factor in the student’s social context (de Ruig et al., 2023; Dermitzaki & Kallia, 2021; Thomas et al., 2019); therefore, careful examination of family and school environmental factors that may contribute to the development of self-regulation is critical. The student characteristics that influence SRL (e.g., Cadima et al., 2016; Denton et al., 2015; Li et al., 2018; Peeters et al., 2016) have been relatively well researched, but much less is known about the influence of parents and teachers on the development of students’ SRL skills, particularly in primary school.

Consistent with the considerations outlined above, the present study had two aims. The first aim was to identify the latent profiles of parenting and teaching styles based on students’ perceived dimensions: emotional warmth, autonomy support, permissive, punitive, and democratic discipline. The second aim was to examine the differences between students’ demographic characteristics (i.e., gender, number of books at home, maternal education, GPA) and students’ SRL approach in the different profiles of parenting and teaching styles according to the results of the latent profile analysis.

Based on the research problem and previous studies (e.g., Amani et al., 2020; de Ruig et al., 2023; Kiuru et al., 2012; Reeve & Cheon, 2021), the following research question and hypotheses were formulated:

  • RQ: What are the differences between the identified parenting and teaching style profiles in terms of student demographic characteristics (i.e., gender, number of books at home, maternal education, GPA)?

  • H1: Parenting style profiles with high levels of emotional warmth, autonomy support, and democratic discipline are associated with higher levels of SRL.

  • H2: Teaching style profiles with high levels of emotional warmth, autonomy support, and democratic discipline are associated with higher levels of SRL.

Method

Participants

The sample obtained by quota sampling consisted of 328 ninth graders (aged 13–15 years) from 25 Slovenian primary schools. Girls (n = 171) and boys (n = 157) were almost equally represented in the sample. Schools were distributed across all 12 Slovenian regions and varied in size, with 15 schools located in rural areas and 10 in urban areas. In Slovenia, primary education comprises nine years of compulsory schooling attended by students aged 6 to 15. The Slovenian education system uses a five-point grading scale, with 1 being the lowest possible (negative grade) and all other grades being positive, with 5 the best possible grade (ZOsn, 1996). Students attend a particular school based on geographic criteria and are placed in classes without regard to their gender, ethnicity, achievement level, or SES. Table 1 provides a description of the sample by academic achievement in the previous school year and two SES variables – number of books and mother’s education – as these variables are most used in current research (see Bornstein & Bradley, 2012). Both SES indicators were measured on a six-point scale. In the present study, girls reported a significantly better GPA in the previous school year compared to boys (F(1,326) = 22.74, p < 0.001).

Table 1 Students according to academic performance and SES variables (N = 328)
Full size table

Instruments

Questionnaire for self-regulated learning – 7 (FSL-7)

Designed by Ziegler et al. (2010), FSL-7 measures SRL based on Ziegler and Stoeger’s (2005) seven-step cyclical model of SRL. The questionnaire consists of seven items for each of the three learning scenarios (studying for school, preparing for a knowledge test, missed learning material). For each item, the students choose the one that best describes their approach to learning: external, impulsive and SRL (Ziegler et al., 2012). Each item refers to one of the seven steps of SRL: self-assessment, goal setting, strategic planning, implementation of the planned strategy, monitoring of implementation, adaptation of selected learning strategies, and evaluation of results (Ziegler & Stoeger, 2005). An example of learning for school scenario items (step two: setting goals) is: My teacher or parents tell me what to learn. (external); I don’t set a specific goal when I study. I just start learning. (impulsive); I set a goal and define exactly what I want to learn. (SRL). The questionnaire was used in an adapted version to determine the SRL approach in learning biology. The overall score was calculated by counting the frequency with which the student chose an SRL learning approach and dividing it by the number of items answered. For example, a student who chose the SRL learning approach in 11 of the 28 items would be given a score of 0.52. Therefore, the minimum score was 0 and the maximum score was 1. Sontag and Stoeger (2015) report the following internal consistency coefficient: 0.83 (pre-test), 0.90 (post-test) and 0.94 (follow-up test). Preliminarily, we confirmed an acceptable fit of the model to the predicted factor structure (χ2(1887) = 11332.04; RMSEA = 0.08; SRMR = 0.09; CFI = 0.92; TLI = 0.92) and good reliability of the SRL approach (α = 0.87).

Parenting styles

The Parenting Behaviours and Dimensions Questionnaire (PBDQ, Reid et al., 2015) was used to measure parenting styles. For the purpose of the research, an adapted version of students’ perceptions about parenting behaviours was created. The scale consisted of 27 items on a 5-point scale (1 = never to 5 = always), distributed in five dimensions: emotional warmth (6 items, e.g., My parents tell me how proud they are of me.), punitive discipline (5 items, e.g., My parents lose their patience when I do something to upset them.), autonomy support (5 items, e.g., My parents encourage me to try things for myself before asking for help.), permissive discipline (5 items, e.g., My parents do things for me when I refuse to do them.), and democratic discipline (5 items, e.g., My parents give me reasons for why I am not allowed to do something.). Preliminarily, an adequate fit of the model to the predicted factor structure was confirmed (χ2(289) = 686.68; RMSEA = 0.05; SRMR = 0.06; CFI = 0.97; TLI = 0.97). The reliability of the scales was adequate: αemotional warmth = 0.82, αautonomy support = 0.63, αpermissive discipline = 0.60, αpunitive discipline = 0.70, αdemocratic discipline = 0.75. Maternal and paternal parenting styles were measured together, taking into account the study by Balaguer et al. (2021), which indicated that parenting styles can be validly measured together based on students’ perceptions.

Teaching styles

The Teacher’s Interaction Styles scale (Žerak, 2019) was constructed based on a revised Slovenian version of the PBDQ (Reid et al., 2015) questionnaire by changing the context from home to school. The scale consisted of 27 items on a 5-point scale (1 = never to 5 = always), which are distributed into five dimensions: emotional warmth (6 items, e.g., My teacher tells me how proud he/she is of me.), punitive discipline (5 items, e.g., My teacher loses his/her patience when I do something to upset him/her.), autonomy support (5 items, e.g., My teacher encourages me to try things for myself before asking for help.), permissive discipline (5 items, e.g., My teacher does things for me when I refuse to do them.) and democratic discipline (5 items, e.g., My teacher gives me reasons for why I am not allowed to do something.). Participants assessed their respective biology teacher’s teaching style. Preliminarily, an adequate fit of the model to the predicted factor structure was confirmed (χ2(289) = 725.96; RMSEA = 0.06; SRMR = 0.07; CFI = 0.97; TLI = 0.97). The reliability of the scales was adequate: αemotional warmth = 0.84, αautonomy support = 0.73, αpermissive discipline = 0.64, αpunitive discipline = 0.64, αdemocratic discipline = 0.77.

Research design and data analyses

The study was approved by the Ethics Committee of the Faculty of Arts, University of Ljubljana. Before the administration of the questionnaires, parent consent was obtained and only students whose parents provided signed informed consent participated in the study. The questionnaires were executed in paper–pencil form. Participants who did not complete all of the questionnaires were excluded from further analyses.

Statistical analyses were performed in R (R Core Team, 2019) and Mplus version 8.3 (Muthén & Muthén, 1998–2019). The Mplus software was used because it provides an easy-to-use interface and several output options that support the LPA analysis process (Asparouhov & Muthén, 2021), and the R software was used because of its adaptability, flexibility, and free availability. Confirmatory factor analyses were conducted in R with the lavaan package (Rosseel, 2012). Latent profile analysis (LPA) using the MLR estimator was applied to identify unobserved subgroups of participants according to their self-perception of parenting styles and teaching styles. The plausibility of one to four latent profile models was examined. The optimal model was selected based on the conceptual interpretability of the profiles, as well as on a review of several statistical indices: Akaike’s Information Criterion (AIC), Bayesian Information Criterion (BIC), Sample-Size Adjusted Bayesian Information Criterion (SABIC), the Lo-Mendell-Rubin Adjusted Likelihood Ratio Test (aLMRT), and the Bootstrap Likelihood Ratio Test (BLRT). For AIC, BIC and SABIC, lower values indicate better model fit, while significant aLMRT and BLRT results indicate a higher number of subtypes. Additionally, entropy (values > 0.80 and marginal value ≥ 0.70) and average posterior probabilities values (values > 0.70) were also considered (Masyn, 2013). Finally, as a practical criterion, the percentage of individuals in the smallest class was considered (> 5%). The LPAs were carried out separately for parenting styles and teaching styles.

Once the optimal latent profile model was identified, the Bolck-Croon-Hagenaars approach (BCH) was performed to examine the differences in demographic variables (i.e., gender, average grade, number of books, maternal education) and SRL across the latent profiles. The BCH approach is recommended for continuous variables because it uses observation weights that reflect the measurement error of the latent class variable and accounts for individual uncertainty in profile classification (Asparouhov & Muthén, 2021).

Results

The descriptive statistics and correlations between the variables of the study are shown in Table 2. The correlations between SRL and the dimensions of teaching styles, and between SRL and the dimensions of parenting styles, are low to moderate. Table 2 shows that there are strong positive correlations between teacher’s autonomy support, emotional warmth and democratic discipline, and between parents’ autonomy support, emotional warmth and democratic discipline. All correlations are in the expected direction, indicating the validity of the constructs measured.

Table 2 Descriptive statistics and correlations between study variables (N = 328)
Full size table

Parenting styles and teaching styles profile identification

Table 3 shows the fit indices of the models with increasing number of profiles. The fit indices of the LPAs for the parenting styles – considering the BLRT index, which was found to outperform other likelihood ratio tests (e.g., Nylund et al., 2007), would suggest a four-profile solution, but considering the theoretical interpretability and informativeness of the different profile solutions and the fact that the smallest class of four-profile solutions contains < 5% of the sample, we opted for the three-profile solution (Table 3). The average individual posterior probabilities for assignment to a particular latent class in the three-profile model were 0.90, 0.87, 0.90, indicating a sufficiently clear classification for class interpretation (Masyn, 2013). The patterns of parenting styles that characterise the three profiles are shown in Fig. 1. The most numerous profile comprises 50% of the participants and can be described as responsive parenting. It is characterised by high levels of parents’ emotional warmth, autonomy support, and democratic discipline, and low levels of punitive and permissive discipline. The second profile includes 42.4% of the participants and can be described as democratic parenting. Compared to responsive parenting, the scores for parents’ emotional warmth, autonomy support and democratic discipline are lower, but still in the upper range of the scale, with slightly higher scores for parents’ punitive discipline and the same score for parents’ permissive discipline. The third profile includes 7.6% of participants and was labelled discipline-oriented parenting. Compared to the other two profiles, it is characterised by higher levels of parents’ punitive discipline and lower levels of permissive discipline, as well as lower levels of democratic discipline, autonomy support, and emotional warmth by parents.

Table 3 Summary of fit statistics for latent profile analysis for parenting styles and teaching styles
Full size table
Fig. 1
figure 1

Profiles of parenting styles

Full size image

For teaching style, the LPA results (Table 3), considering the BLRT index, suggest the four-profiles solution, but further investigation of the fit indices and the theoretical interpretability and meaningfulness of the different profile solutions supported the decision to retain the three-profile solution, which also has an appropriate class membership and entropy value. The average individual posterior probabilities for assignment to a particular latent class in the three-profile model were 0.94, 0.91, 0.89, indicating a sufficiently clear classification for class interpretation (Masyn, 2013). The patterns of teaching styles are shown in Fig. 2. The most numerous profile includes 49% of the participants and can be described as diverse teaching. It is characterised by a medium rating of teacher autonomy support, emotional warmth, democratic and punitive discipline, and a slightly lower rating of teacher permissive discipline. The second profile includes 37% of the participants and can be described as autonomy-supportive teaching. It is characterised by high levels of autonomy support, democratic discipline, and emotional warmth. Compared to diverse teaching, the level of punitive and permissive discipline is lower. The last profile includes 14% of the participants and was described as directive teaching. It is characterised by low levels of permissive discipline and emotional warmth, and low levels of autonomy support and democratic discipline. The level of punitive discipline is highest in this profile, although still slightly lower compared to the diverse teaching profile and slightly higher compared to the autonomy-supportive teaching profile.

Fig. 2
figure 2

Profiles of teaching styles

Full size image

Parenting and teaching styles, student demographic characteristics and SRL

To examine the differences between the latent profiles of parenting and teaching styles, the BCH approach was used (Table 4). Regarding gender and parenting styles, the results showed that there were no statistically significant differences between the profiles, although slightly more girls perceived their parents as responsive. On the other hand, for gender and teaching style profiles, the results showed that there were statistically significant differences between the profiles; there were more boys who perceived the biology teacher’s teaching style as diverse and more girls who perceived the biology teacher’s teaching style as autonomy supportive. In addition, results showed that there are statistically significant differences between parenting profiles and the number of books students have at home. Students who perceive their parents as responsive report having more books at home than students who perceive their parents as discipline-oriented. Moreover, there are statistically significant differences between teaching profiles and maternal education. Students who perceive their biology teacher as autonomy-supportive report higher levels of maternal education compared to students who perceive their biology teacher as directive. In addition, there are statistically significant differences between teaching profiles in terms of student GPA. Students who perceive their biology teacher as autonomy-supportive have a higher GPA than students who perceive their biology teacher’s teaching style as directive or diverse (Table 4).

Table 4 Means and standard errors of auxiliary variables and test of mean differences across parenting and teaching styles
Full size table

Statistically significant differences were found when looking at differences between the SRL and parenting style profiles. Students with a responsive parenting profile reported the highest SRL level, followed by students with a democratic parenting profile and students with a discipline-oriented parenting style. The differences in SRL scores between students in the discipline-oriented parenting profile and those in the responsive parenting profile and between students in the democratic parenting profile and those in the responsive parenting profile were statistically significant (Table 4).

In addition, there were statistically significant differences between the teaching style profiles and the SRL. Students in the autonomy-supportive teaching profile had the highest SRL scores, followed by students in the diverse teaching profile and students in the directive teaching profile. The differences in SRL scores between students in the directive teaching profile and those in the autonomy-supportive teaching profile and between students in the diverse and autonomy-supportive teaching profiles were statistically significant (Table 4).

Discussion

The present study attempts to identify the latent profiles of parenting and teaching styles and to examine the differences in student characteristics and SRL between the different profiles. Using the LPA, three subtypes of parenting were identified: responsive parenting, democratic parenting, and discipline-oriented parenting. Results show that students who perceived their parents as responsive (half of the sample) reported about higher number of books at home than students in the discipline-oriented profile, which may indicate that parents of low-SES students are more likely to use discipline-oriented parenting practices. This is consistent with previous research (e.g., Areepattamannil, 2010) suggesting that parents from high-SES backgrounds are more likely to have an authoritative parenting style. In terms of SRL, students who perceived their parents as responsive, with the highest expression of emotional warmth, democratic discipline, and autonomy support, also scored highest on SRL. This is consistent with the findings of previous studies (Alnafea & Curtis, 2017; Amani et al., 2020; Du et al., 2021; Huang & Prochner, 2003; Kallia & Dermitzaki, 2017; Pinquart, 2016; Pino-Pasternak & Whitebread, 2010; Stright et al., 2001), suggesting that positive parenting characterised by democratic discipline, emotional warmth, autonomy-supportive behaviours and understanding of the child’s needs positively impacts students’ SRL and academic achievement. Supporting autonomy allows children to gradually take responsibility for schoolwork and is positively related to the metacognitive aspects of SRL (Pino-Pasternak & Whitebread, 2010). Responsive parents exhibit positive behaviours toward the child (e.g., warmth, responsiveness), consider their child’s perspective, provide contingent support and opportunities for the child to act within certain guidelines, and offer meaningful rationales for why the child needs to do a particular activity (Kallia & Dermitzaki, 2017; Sanders et al., 2019), allowing the child to develop SRL behaviours. On the other hand, students who perceived their parents as discipline-oriented and had the highest punitive discipline scores had the lowest SRL scores, reflecting Pinquart’s (2016) meta-analysis showing that an authoritarian parenting style and parental control are related to lower achievement. An authoritarian parenting style (i.e., verbal hostility, punishment, excessive parental control) negatively impacts students’ SRL (Huang & Prochner, 2003) and mental health (Baumrind et al., 2010).

In addition, LPAs suggest three subtypes of teaching styles: diverse teaching, autonomy-supportive teaching, and directive teaching. The results of this study are consistent with previous research (e.g., Filippello et al., 2020; Reeve & Cheon, 2021; Wang et al., 2016) and suggest that autonomy-supportive teaching is positively associated with student academic achievement. Regarding gender differences in perceived teaching style our findings are consistent with previous research suggesting that female students tend to perceive higher levels of teacher autonomy support than male students (Brandisauskiene et al., 2023; Filippello et al., 2020). It is worth noting that students who perceive autonomy-supportive teaching report the highest educational level of their mothers, which may suggest that teachers behave differently or have lower expectations for students from socioeconomically disadvantaged backgrounds (Auwarter & Arguete, 2008). In addition, students who perceive their teacher as autonomy-supportive have the highest SRL scores (37% of the students). Compared to the other two profiles, teachers in the autonomy-supportive profile have the highest levels of emotional warmth and democratic discipline, which are characteristics of authoritative teaching (Kiuru et al., 2012; Walker, 2008). The findings suggest that authoritative teaching with a particular focus on promoting student autonomy is the optimal context for developing students’ SRL. According to self-determination theory (Deci & Ryan, 2008), the essential conditions for fostering student SRL, and thus high achievement, are relatedness, competence, and autonomy. As Reeve (2009) points out, autonomy-supportive teaching is characterised by teacher interaction with students that recognises, promotes, and develops students’ psychological needs and supports students’ ability to self-regulate. Our findings support the claim that autonomy-supportive teaching (Reeve & Cheon, 2021) is related to students’ ability to regulate their own learning. The results are consistent with previous studies (Brandisauskiene et al., 2023; Schuitema et al., 2016; Sierens et al., 2009; Vansteenkiste et al., 2012; Zee & de Bree, 2017) showing that an autonomy-supportive teaching style is positively associated with students’ SRL. Interestingly, students who perceived their teachers as directive reported the lowest levels of SRL and maternal education. This may be related in part to the teacher’s controlling interaction style (Reeve & Cheon, 2021), which is characterised by authoritarian behaviour with a high expression of controlling instructional behaviour. Therefore, teachers should pay more attention to students from low-SES backgrounds and consciously rethink their behaviour in an autonomy-promoting manner, which could help these students overcome factors that negatively affect their academic progress (Brandisauskiene et al., 2023; Vandevelde et al., 2017).

Parents and teachers can support students’ autonomy by clearly articulating their expectations, providing appropriate supports based on their developmental characteristics and abilities, and providing feedback that promotes students’ perceived competence and SRL while actively involving them in decision making and respecting their opinions (Pino-Pasternak & Whitebread, 2010; Reeve, 2016). However, we must consider reciprocity and interactivity in the relationship between the student and parent, and between the student and teacher. It is possible that the student’s behaviour at home and in the classroom influences the behaviour of parents and teachers. Teachers adapt their behaviour and responses to the student’s behaviour and expression of SRL. They supervise students with less pronounced SRL skills more in the learning process, but when students show higher levels of SRL, teachers support them to a greater extent in terms of autonomy (Schuitema et al., 2016). In addition, students’ perceptions of the family environment affect their perceptions of the learning context, as confirmed by Paulson et al. (1998), who found that students’ perceptions of the teacher’s authoritativeness in the classroom were congruent with their perceptions of their parents’ authoritativeness. Therefore, it is important for educational institutions to build good relationships with students’ parents and actively involve them in their children’s education (Froiland & Davison, 2013; Thomas et al., 2019).

The strength of the present study lies in the examination of both parenting styles and teaching styles and their relationship to student characteristics and SRL. At the same time, we also noted several limitations. First, SRL, parenting styles and teaching styles were measured using student self-assessments. Self-assessment questionnaires assess certain aspects of SRL (e.g., learning strategies), and students’ responses are based on their level of self-awareness of SRL, so they are not entirely accurate, as it is possible that some students are not sufficiently aware of their learning process or use certain learning strategies unconsciously (Vandevelde et al., 2015). Moreover, while researchers (e.g., Kiuru et al., 2012; Paulson et al., 1998; Pinquart, 2016) emphasise the importance of students’ perceptions of the home and school learning environment, these perceptions are not necessarily characteristic of parents’ and teachers’ actual interactions with students. Second, because of the cross-sequential nature of the study, we cannot draw conclusions about the compensatory and cumulative effects of parent and teacher interactions on students’ SRL. Third, the study does not distinguish between parental and maternal parenting styles. Finally, the study was conducted in the Slovenian educational context, and to achieve better generalizability, the results would need to be replicated in other educational contexts. It would be useful to use multiple methods (e.g., classroom observations, questionnaires, interviews, thinking aloud protocols) and data sources (teachers, students, parents) to improve understanding of the relationship between SRL and students’ interactions with parents and teachers.

Conclusion

The results of the present study suggest that parents and teachers are an important aspect of the social context for the development of SRL. Regarding the role of parents and teachers in cultivating students’ SRL skills, emotionally supportive parents and teachers who encourage students’ independent learning and gradually grant them autonomy have been shown to be essential for the SRL development. The findings of the present study contribute to theory and practise in three ways. First, they expand our knowledge of the role of family and school environments in students’ SRL by highlighting autonomy-supportive behaviours (Kallia & Dermitzaki, 2017; Reeve & Cheon, 2021) as well as authoritative parenting (Baumrind, 2013) and authoritative teaching (Kiuru et al., 2012; Walker, 2008). Second, they emphasise the importance of coherence between the family and school environments and parent-teacher collaboration to provide appropriate incentives and meet students’ developmental and learning needs. Third, they suggest that effective interventions to promote SRL in primary school should be developed and targeted not only for the individual student, but also for his or her social environment, particularly for at-risk students and students from low-SES backgrounds (Trias et al., 2021; Vandevelde et al., 2017). Findings could be incorporated into teacher professional development programmes. Teachers can model SRL by teaching explicit instruction of SRL skills in their lessons and explaining the strategies they use to self-regulate their own learning and teaching (Kallia & Dermitzaki, 2017). Therefore, they should be advised on how to create a supportive learning environment that promotes SRL in students (de Ruig et al., 2023; Moos & Ringdal, 2012). On the other hand, special emphasis should be placed on parenting programmes to increase parents’ autonomy, flexibility, and confidence in dealing with difficult situations with their children. Programmes for parents should provide a supportive interpersonal context and empower parents to improve their self-regulation skills (Sanders et al., 2019). It is also important to inform parents about the importance of their active involvement in their children’s education (Boonk et al., 2018) and provide them with specific instructions on how to interact with their child in an authoritative manner that supports the development of SRL skills in the home environment.

Future research that considers the moderating role of student characteristics (e.g., interest, academic goal orientation, personality traits, self-concept) and examines the relative contributions of parental and school practises could deepen our understanding of the factors that influence SRL in students.