The Dynamics of the Interrelationships between Conscious Self-regulation, Psychological Well-being and School-related Subjective Well-being in Adolescents: A Three-year Cross-lagged Panel Study

Background Recently, research on psychological well-being and its dynamics and predictors in adolescence, has gained special attention, due to the importance of well-being for mental and physical health, as well as for success in different activities. Self-regulation (SR) is considered a significant resource for maintaining psychological and school-related subjective well-being. Objective The purpose of our study was to identify the role of conscious SR in maintaining pupils’ satisfaction with school life, and to assess the contribution of conscious SR to the development of psychological well-being in adolescence. Design Two three-year longitudinal studies were carried out on samples of young adolescents in Russian schools (N = 148; N = 132; 10–13 years). The studies utilized methods for assessing conscious SR, psychological well-being (PWB), and school-related subjective well-being (SWB), the latter being the cognitive component of life satisfaction. Results Our research revealed differences in the dynamics of PWB and SWB levels in adolescents during their transition from primary to basic secondary school. It also identified the specifics of longitudinal relationships between conscious SR, PWB, and SWB in adolescence. We showed that there was a reciprocal relationship between them. However, the most significant cross-longitudinal effects were established between SR and school-related SWB. These effects changed over time: at the beginning, well-being acted as a significant factor of self-regulation, while later self-regulation acted as a significant resource for maintaining adolescent well-being in the subsequent years. Conclusion School-related SWB is characterized by the most pronounced trajectory of change, while PWB is characterized by greater stability and insignificant growth. Our three-year longitudinal study demonstrated that the link between self-regulation and well-being is consistently reproduced. Conscious self-regulation is a significant resource for both the psychological and school subjective well-being of adolescents.

However, predictors of PWB itself have been studied to a lesser extent. ere is ample evidence that conscious self-regulation of achieving educational goals is a signi cant factor in both academic success and PWB (Gestsdottir & Lerner, 2008;Fomina, Burmistrova-Savenkova, & Morosanova, 2020;Singh, & Sharma, 2018). e purpose of this article is to uncover the dynamics of conscious self-regulation as a signi cant resource for schoolchildren's well-being.

Psychological Well-being and Subjective well-being
In modern research, the well-being of schoolchildren is assessed by means of various measures. e most theoretically and empirically substantiated are the constructs of psychological well-being and subjective well-being. ese are interrelated but somewhat di erent indicators of positive human functioning (Ryan & Deci, 2001). SWB includes a cognitive component, which is re ected as a self-assessment of overall satisfaction with one's life, and an a ective component, which is measured as a balance between positive and negative emotions (Diener, 1999). e cognitive component is considered as the most stable component in the SWB structure and is analyzed more o en.
It's worth emphasizing that SWB is usually considered in connection with certain speci c areas of life. us, it seemed relevant for us to focus on the study of schoolrelated subjective well-being, which is characterized by students' satisfaction with their success at school, their relationships with teachers and peers, the school climate, etc. (e.g., Steinmayr et al., 2018;Yang et al., 2019).
On the other hand, psychological well-being (PWB) is based on a person's holistic experience, such as feelings of happiness, and satisfaction with oneself and one's life in the broader context of relations between oneself and the world (Huppert, 2009;Ry , 1989;Waterman, 1993, etc.). PWB is a multidimensional phenomenon. Its elements are: 1) self-acceptance; 2) positive relations with others; 3) autonomy; 4) environmental mastery; 5) purpose in life, and, nally, 6) personal growth. ese six constructs de ne PWB both theoretically and operationally, and they specify what promotes emotional and physical health (Ry & Singer 1998).
Although there have been recent attempts to discover a common factor determining SWB and PWB (Garcia, Sagone, De Caroli, & Al Nima, 2017;Heintzelman, 2018), the most promising results have been obtained by di erentiating between them (Ryan and Deci, 2001). Indeed, as a rule, factor analysis con rms a close but still di erent status of SWB and PWB (Compton, Smith, Cornish, & Qualls, 1996;Keyes, Shmotkin, & Ry , 2002). e current scienti c consensus on the relationship between PWB and SWB research can be described as follows: "PWB and SWB are strongly related at the general construct level, but their individual components are distinct once their overlap with the general construct of well-being is partially led out. " (Chen, Jing, Hayes, & Lee, 2013). We adhered to this position in this study. Our main concern was to compare the speci c relationships of conscious self-regulation with both PWB and the cognitive component of SWB in adolescents during their transition from primary to basic secondary school.
Self-regulation (SR), as considered in di erent contexts by empirical studies, is associated with both PWB and SWB. (e.g, Elliot, rash, & Murayama, 2011;Fomina et al., 2020;Hofer, Busch, & Kartner, 2011;Wrosch, Scheier, Miller, Schulz, & Carver, 2003;Tavakolizadeh, 2012). It has been shown that the higher a person's SR, the higher their sense of well-being, and the more e ective their coping strategies (Boekaerts & Corno, 2005;Hofer et al., 2011;Saha, Huebner, Hills, Malone, & Valois, 2014). A study by Saha and colleagues (2014) demonstrated that SR explains a signi cant percentage of the variance across all six PWB measures, with the greatest positive associations found between SR and life goals. It's worth noting that all these data were obtained on samples of university students or adults. ere are very few similar studies on the relationships of SR with PWB and SWB conducted on the samples of adolescents (Steinmayr et al., 2019). Longitudinal studies can make a signi cant contribution to uncovering the relationships between conscious SR, PWB, and SWB, since there are data on di erent trajectories of PWB and SWB in adolescents (Archakova, Veraksa, Zotova, & Perelygina, 2017).
In our approach, the conscious self-regulation of learning activity is understood as a cognitive-personal construct, including cognitive processes (planning goals, modeling signi cant conditions for goals achievement, programming actions, and evaluating results) and regulatory-intrapersonal properties ( exibility, independence, reliability, and responsibility), which serve as tools for initiating and maintaining activity aimed at consciously setting educational goals and managing their achievement (Morosanova, 2004(Morosanova, -2020. Conscious SR is the controlling mechanism for mobilizing all other types of individual resources (cognitive, motivational, and intrapersonal) to achieve a result (Morosanova, 2014). Success, reliability, productivity, and the nal result of actions to achieve the goal depend on the level of development of operational-regulatory processes and regulatory-personal features.
Adolescence has traditionally been associated with the risk of behavioral problems and psychological distress. In this connection we considered it extremely important to study the contribution of SR to PWB and SWB, particularly in early adolescence, when children's sense of well-being is of maximum importance; it gradually decreases later toward high school. Empirical studies show that the period between ages 10 and 12 years is a turning point in the development of individual trajectories of PWB and SWB (Orben, Lucas, Fuhrmann, & Kievit, 2020;Willroth, Atherton, & Robins, 2020). At the same time, according to a number of researchers, personal changes during adolescence create unique opportunities for positive trajectories of development (Lerner et al., 2018). Herewith, SR serves as one of the essential mechanisms contributing to positive youth development (Gestsdottir et al., 2017). e re exivity that takes shape during this period probably facilitates the ability of adolescents to develop such subjective qualities as independence, responsibility, and initiative, which make their signi cant contributions to adolescent well-being. We assumed that conscious self-regulation, being the control level of regulation of educational goals achievement, would determine the level and dynamics of PWB and SWB during this period.
Our research was aimed to answer the following questions: 1. What are the dynamics and specificity of the relationship between conscious self-regulation, psychological well-being, and school-related subjective wellbeing of adolescents during their transition from primary to basic secondary school? 2. Can conscious SR be considered a long-term predictor of PWB and SWB in adolescents during this transition period?
For research purposes, we conducted two studies. e data obtained and results of the analysis are presented and discussed below.

Participants and Procedure
Two separate longitudinal studies were conducted on samples of teenagers (grades 4-6) in Russian state schools which implement the basic education program. e 4 th grade in Russia is the last year of primary school. en children go to the basic secondary school. e design of the studies di ered in the instruments for assessing the schoolchildren's well-being: in the rst study, the methodology for assessing psychological well-being (PWB) was used; in the second, we used the scale for assessing the school-related subjective well-being (SWB). e data were collected in three waves. In Study 1 the sample at T1, T2, and T3 consisted of 148 students. e sample was evenly distributed by sex (50% boys). At T1, the mean age of the participants was 10.2 years (SD = 0.50; range = 10-11 years). Seven months later, at T2 children were on average 10.9 years old (SD = 0.28; range = 10-12 years). One year later, at T3 children were on average 11.9 years (SD = 0.50; range = 12-13 years). In Study 2 the sample at T1, T2, and T3 consisted of 132 pupils (47% boys). At T1, the mean age of the participants was 10.3 years (SD = 0.48; range = 10-11 years). Seven month later, at T2 children were on average 10.8 years old (SD = 0.42; range = 10-12 years). One year later, at T3 children were on average 11.9 years old (SD = 0.32; range = 12-13 years).
Parental and school consent was obtained for all participants. Analyses were carried out on depersonalized data. e study procedure was approved by the relevant institutional review board. Ethical agreement and consent for access to the schools were provided by the Ethics Committee of the Psychological Institute of the Russian Academy of Education (approval number 2017/1-128).

Measures
Psychological Well-Being was accessed using the slightly modi ed Russian adaptation of the 25-item Well-Being Manifestation Measure Scale developed by Masse (Masse, Poulin, Dassa, Lambert, Bélair, & Battaglini, 1998). e questionnaire was previously validated on a sample of 4th-grade pupils in Russian secondary schools . e participants were asked to evaluate to what extent they experienced the described states over the past month on a 5-point Likert scale ranging from 1 (never) to 5 (almost always).
is questionnaire contained the following subscales: 1) Control of Self and Events (e.g., "I was able to face difficult situations in a positive way"); 2) Happiness (e.g., "I found life exciting and I wanted to enjoy every moment of it"); 3) Social Involvement (e.g., "I felt like having fun, doing sports and participating in all my favorite activities and pass-times"); 4) Self-Esteem (e.g., "I had self-confidence"); 5) Mental Balance (e.g., "My life was well-balanced between my family, personal and school activities"); 6) Sociability (e.g., "I got along well with everyone around me"); and the cumulative scale 7) Psychological Well-Being, which summed up the scores on all the scales. e internal reliability coe cients were 0.72-0.78.
School-related Subjective Well-Being was measured by means of the Multidimensional Students Life Satisfaction Scale (MSLSS) (Huebner, 2001, in a Russian adaptation by Sychev, Gordeeva, Lunkina, Osin, & Sidneva, 2018). Its 30 items allow for evaluating schoolchildren's satisfaction in important life domains, including family, school, self, friends, and teachers, on a 5-point Likert scale ranging from 1 (never) to 5 (always). Higher scores indicate higher levels of life satisfaction throughout the scale. All the scales had high reliability (0.82 < α < 0.89).
Self-Regulation was measured by means of Morosanova's Self-Regulation Pro le Questionnaire -Junior (Morosanova & Bondarenko, 2015). It consists of seven selfassessment scales: Planning of goals (e.g., "I know what grades I want to get at the end of the year"); Modeling of signi cant conditions (e.g., "Prior to start solving the task, I always carefully examine its introductory conditions"); Programming of Activity (e.g., "I have no di culty in drawing up a plan of presentation"); Results Evaluation (e.g., "I rarely notice my mistakes"); Flexibility (e.g., "I'm back to studies quickly a er the holidays"); Independence (e.g., "I usually do my homework by myself "); and Responsibility (e.g., "I seek to perform additional tasks").
Each item was scored on a 6-point scale with responses ranging from 1 ("not at all like me") to 6 ("very much like me"). e pupils were to choose to what extent the described behavior was characteristic of them. e general SR level was estimated by adding up the scores on the seven scales. e incentive material was presented in forms accessible for primary school age, such as descriptions of typical situations associated with organization of learning activities and pupils' behavior relative to their training. e coe cients of internal consistency of the items for each scale ranged from 0.62 to 0.79, indicating an overall reasonable homogeneity of the items on each scale.
Statistical Analysis SPSS 26.0 (SPSS Inc.) was used to obtain descriptive statistics for the study variables and bivariate associations. Longitudinal con rmatory factor analyses and bidirectional cross-lagged panel analyses were conducted in AMOS 23. Two separate crosslagged panel models were assessed, one between SR and PWB and the other between SR and school SWB. e models' ts were evaluated using several t indices: a Root Mean Square Error of Approximation (RMSEA); a comparative Fit Index (CFI); and the Tucker-Lewis index (TLI).

Study 1. Dynamics of the relationship between psychological well-being and conscious self-regulation in adolescents: a cross-lagged panel analysis
e rst study tested the hypotheses about the speci city of the longitudinal relationship between the conscious self-regulation and psychological well-being of adolescents. Figure 1 shows the initial hypothesized model of this relationship. e six PWB variables at each point in time were used as indicators of the latent factor PWB, labeled as PWB_Time1, PWB_Time2, and PWB_Time3. e latent factor SR (labeled as SR_Time1, SR_Time2, and SR_Time3) was represented by seven regulatory indicators.  Table 1. Analysis of signi cant changes in the levels of PWB and SR revealed that the PWB level signi cantly increased in the adolescents when they moved from the 4 th grade (Time 1) to the 5 th grade (Time 2), i.e., from primary to basic secondary school. e same thing happened with SR (p = 0.023). But during their transition from grade 5 (Time 2) to grade 6 (Time 3), no signi cant changes in the levels of PWB and SR were recorded. Correlation analysis data indicated the presence of a moderate to strong relationship between the general level of psychological well-being and the general level of conscious self-regulation. At the next stage, to reveal the cross-longitudinal e ects of PWB and SR, crosslagged panel analyses were carried out using the method of structural modeling. e full model with all cross-lagged paths, auto-regressive paths, and concurrent covariance demonstrated an acceptable t to the data (CFI = .983, TLI = .956, RM-SEA = .038). e path from PWB at Time 1 to SR at Time 2 was statistically signicant. e paths from SR at Time 1 and Time 2 to PWB at Time 2 and Time 3 were all signi cant.
Next, we looked at the individual cross-longitudinal models. e structural models demonstrated a good level of agreement (see Table 2). is allows us to say that there was a signi cant reciprocal relationship between SR and PWB. en we compared the magnitude and signi cance of the standardized PWB and SR coe cients in the obtained models. Results for the auto-and cross-lagged analyses are presented in Table 3. e results suggest that standardized coe cient of the self-regulation score is higher in the cross-longitudinal relationship model of SR (Time 2) and PWB (Time 3), while the standardized regression coe cients for PWB (Time 1) are higher in relation to SR (Time 2). ese results indicated that PWB served as a rm foundation for the SR development at the stage of adolescent adaptation to secondary school (during the transition from primary school), and that a er that, conscious SR (apparently having been formed in the new conditions) acts as a signi cant resource for maintaining PWB in adolescents in later years.

Study 2. Dynamics of the relationship between school-related subjective well-being and conscious self-regulation in adolescents: a cross-lagged panel analysis
In the second study, we examined the speci city of the relationships between subjective school well-being and conscious self-regulation. e means, standard deviations (SD), and correlations among the variables included in the analyses are presented in Table 4. In contrast to PWB, we saw a somewhat di erent picture of the dynamics in relation to the school-related SWB. Younger adolescents, when moving from grade 4 (Time 1) to grade 5 (Time 2), demonstrated a small but signi cant positive trend (p = 0.048), and then a signi cant drop in their SWB level (p = 0.000). e dynamics of self-regulation corresponded to the results of Study 1. e full model with all the cross-lagged paths, auto-regressive paths, and concurrent covariance also demonstrated an acceptable t to the data (CFI = .965; TLI = .958; RMSEA = .029). Further, we created separate cross-longitudinal models of the relationship between SR and SWB ( Table 5). All models turned out to be signi cant, with their conformity indices demonstrating high levels of signi cance. To identify the direction of the cross-longitudinal relationships, we compared the signi cance and magnitude of the standardized regression coe cients in the models ( Table 6). Table 6 Auto-and cross-lagged standardized path coe cients from the cross-lagged panel analyses Notes. *p < .05; **p < .01; ***p < .001.

Cross-longitudinal coe cients are in bold italics
Autoregressive coe cients for the SWB indicators were lower than for SR, which testi ed to a greater variability of the school SWB indicators for adolescents. Evaluation of the cross-longitudinal relationships revealed that when students were moving from grade 4 to grade 5 (Time 1 and Time 2), the relationship between SR and SWB was reciprocal. e cross-longitudinal relationship between SR and SWB during the transition from grade 5 to grade 6 (Time 2 and Time 3) was signi cant only for SR. at is, we can say that a higher level of self-regulation development among students in primary school predicted their well-being in the 5 th grade, and then in the 6 th grade.

Discussion
Comparative analysis of the longitudinal data made it possible to establish that there are certain di erences in the dynamics of changes in the PWB and SWB levels of adolescents during their transition from primary to basic school. e School-Related Subjective Well-Being was characterized by the most pronounced trajectory of change. It had an increasing tendency in the 5 th grade, and then decreased in the 6 th grade. Psychological Well-Being was characterized by greater stability and slight growth during the period of study from grades 4 to 6. e growth of well-being (both subjective and psychological) from grade 4 to 5, in our opinion, took place due to the changes in educational conditions toward actualizing students' positive expectations, expanding their spheres of communication, and encouraging initiative and independence in self-organization of activities. During this period, adolescents could acquire skills of well-being that would help them build positive conditions of school life (Ng et al., 2015).
A number of researchers have stressed that primary school years represent a critical period when the students' experience provides the foundation for their current and later engagement, achievement, and sense of belonging in school (Suldo et al., 2014;. As we did, these researchers noted that puberty is characterized by decreasing SWB, and that adolescents demonstrate a clear drop in the life satisfaction (Shek & Liu, 2014;Steinmayr et al., 2018;Willroth et al., 2020). Furthermore, a er the end of puberty, their sense of well-being increases (Salmela-Aro & Tuominen-Soini, 2010).
However, we can identify somewhat di erent dynamics for PWB and SWB. ere was evidence that PWB is indeed more sustainable over the long term (Joshanloo, 2018). It is de nitely PWB that determined the level of school-related SWB (the cause-and-e ect relationship mainly goes from PWB to SWB, and not vice versa). In this sense, we have con rmed these conclusions. e trajectory of conscious SR development in both studies was characterized by smooth growth. An increase of the regulatory indicators took place largely due to age characteristics and the social situation for development. In turn, self-regulation became a resource for personal growth in adolescence (Bronson, 2000;Morosanova, Bondarenko, Fomina & Burmistrova-Savenkova, 2018). e relationships between PWB and SR which we uncovered demonstrated the heterogeneity and heterochrony of these properties' development in adolescents. As a whole, the PWB indicators were steadily increasing, and the indicators of conscious SR, rising in the 5th grade, returned to their previous values in the 6th grade. e results of the analysis revealed that high PWB level in the 4th grade predicted a high SR level in the 5th grade.
ese results are consistent with the general conclusions of researchers that PWB a ects a wide range of factors in the lives of children and adolescents. A high level of personality and events management, high self-esteem, well-built relationships with teachers and peers, and mental balance allow 4 th -graders to successfully develop their self-regulation. e 5 th grade is associated with the transition to new learning conditions. ese circumstances challenge the independence of schoolchildren and their ability to regulate their activities.
Cross-longitudinal analysis made it possible to record more signi cant e ects of SR on PWB than of PWB on SR. In other words, the child's PWB level in grade 5 did not signi cantly a ect his/her SR in grade 6. It can be concluded that the PWB level achieved in the 5th grade at the stage of adaptation to new learning conditions served as the foundation for development of conscious SR. Upon completion of the adaptation process, the conscious SR serves as a signi cant resource for maintaining the adolescents' PWB in the basic secondary school. us, during the transition from primary to basic secondary school, PWB can be a signi cant condition for conscious SR development. In turn, SR acts as a resource for the growth of PWB indicators in the future. e longitudinal relationship between SR and school-related SWB was characterized by somewhat di erent features. It should be noted that our study was assessing the cognitive component of school SWB, i.e., life satisfaction. is component is considered more stable, and is most frequently included in studies of youths' perceived quality of life (Suldo et al., 2006). us, during the transition from grade 4 to grade 5, the contributions of SR to school-related SWB and vice versa are commensurate, i.e., there is a reciprocal relationship between them. When the child moves from grade 5 to grade 6, however, the situation changes dramatically; conscious SR makes a more signi cant contribution to SWB, while the contribution of SWB to SR is insignicant. is result is especially interesting because life satisfaction decreases during adolescence, which can in uence many later life outcomes (Blakemore & Mills, 2014;Orben et al., 2020). us, conscious SR acts as a SWB resource throughout the entire schooling period. And its contribution to well-being is especially signi cant during the transition from 5th to 6th grade.
Accordingly, reliance on internal resources becomes important (Goldbeck, Schmitz, Besier, Herschbach, & Henrich, 2007;Steinmayr et al., 2018). During this period, adolescents demonstrate a decline in academic motivation, which inevitably leads to a decrease in academic performance, which cannot but a ect school-related SWB (Martin, & Steinbeck, 2017). e relationship between academic achievement and school-related SWB weakens signi cantly during adolescence (Yang et al., 2019), while conscious SR still remains a reliable predictor of both academic achievement and SWB in adolescents (Fomina & Morosanova, 2019;Gestsdottir & Lerner, 2008).
us, conscious self-regulation, being a foundation for success in educational activity, is a necessary resource for maintaining the SWB of adolescents in this di cult age period.

Conclusion
A three-year longitudinal study demonstrated a stable relationship between conscious self-regulation and both psychological and subjective school-related well-being in adolescence.
Cross-longitudinal analysis then made it possible to establish reciprocal relationships between conscious SR and PWB. It has been shown that during the transition from primary to basic secondary school, PWB can act as a signi cant mechanism for the development of SR. However, in the future, conscious SR can be considered as a signi cant resource for maintaining adolescents' PWB in subsequent years. e longitudinal relationships between SR and school-related SWB were characterized by slightly di erent speci city. e general level of the conscious self-regulation of educational activity in adolescents predicted the level of their subjective wellbeing to a greater extent, and, in this sense, acted as an e ective tool for maintaining well-being in adolescence.

Limitations
e present research did not set the task of studying gender di erences in the dynamics of SWB, PWB, and SR, although, according to previous research, it is necessary to take gender speci city into account in this context (Orben et al., 2020). In addition, certain individual characteristics of adolescents can also play a signi cant role in the dynamics of the studied phenomena. e study of these issues will form the basis for our future research.

Ethics Statement
Parental and school consent was obtained for all participants. Analyses were carried out on the depersonalized data. e study was conducted in accordance with the Helsinki Declaration. Ethical agreement and consent for access to the school were provided by the Ethics Committee of the Psychological Institute of the Russian Academy of Education (approval number 2017/1-128).