Parent-Child Relationship, Well-Being and Home-Leaving during the Transition from High School to University

: This prospective study examined changes in parent–child relationship quality and well-being during the transition to university. We also investigated whether living situation (i


Introduction
Transition to university is a critical period during which young individuals go through profound changes, such as making new friends, having more freedom and more responsibilities, and potentially leaving home for the first time [1].This transitional period is both exciting and stressful for emerging adults [2].Particularly, first-year university students compared to older students are vulnerable to stress, given the number of changes that occur within a short period [3].Those who leave home in the first year are at risk for homesickness [4] and potentially increased stress because of adjustment to a new environment and managing new responsibilities.While parents continue to be important sources of support during this transitional and potentially turbulent period, they can also be the main source of conflict, as young individuals are seeking more independence from them [5,6].Therefore, the transition to university requires special attention with respect to parent-child relationships and young individuals' well-being.While ample research has focused on the transition to university, studies investigating familial relationships and well-being of young individuals simultaneously are lacking.As parents continue to play a key role during emerging adulthood, it is vital to investigate how parent-child relationships change and to what extent they are associated with the changes in well-being.Furthermore, most of the research on the transition to university has been conducted primarily in Western countries such as the United States, United Kingdom and Australia [7][8][9].Research conducted in non-Western countries is needed in order to understand the unique experiences of young individuals in culturally different contexts.The present study aimed to address these knowledge gaps in the literature by investigating prospective changes in parent-child relationship quality and well-being in a non-Western country.Furthermore, this study aimed to investigate the association between these changes and the impact of home leaving on these changes in Turkey.

Changes in Parent-Child Relationship Quality during the Transition to University and Leaving Home
Over the course of adolescence, the parent-child relationship becomes more egalitarian [10,11].Emerging adulthood is also a significant time during which parent-child relationships can change profoundly as both young individuals and their parents reexamine it [12].Life Course Theory emphasizes the interconnectedness of family members [13], and its linked lives perspective suggests that individual transitions influence the whole family system and present opportunities for change in relationships.In emerging adulthood, the parent-child relationship shows improvements [14].More specifically, emerging adults have a less conflictual [15] and closer relationship with their parents after the transition to university [16].These patterns are generally consistent for both female and male emerging adults, but relationship improvements tend to be more pronounced for mothers than fathers [14].
During the transition to university, leaving home is another significant event that influences the established dynamics of family relationships [5].Nowadays, young adults continue to live with their parents for a longer period [17], yet high school graduation followed by attending university marks a significant time during which they consider leaving home for the first time [18].First-year students are less likely to live independently compared to older students [8].However, when they leave home, they have less contact with their parents [16], but they also experience less conflict and more warmth in the relationship [19].On the other hand, when emerging adults continue to live at home, they receive higher levels of parental support and have more positive daily encounters (i.e., laughter) than negative ones (i.e., annoying each other) compared to those who moved out [7].However, emerging adults living at home may feel restricted and in need of more independence [9].For parents, recognizing their child's new status (i.e., emerging adult, university student) and allowing them more autonomy may be difficult when the child lives under their roof [6,9].Overall, parent-child relationship quality shows changes during the transition to university, yet previous research either focused only on one aspect of the parent-child relationship, such as conflict [15] or the overall quality of the relationship [20].Therefore, the present study focuses on parental warmth, conflict and autonomy support simultaneously, which are fundamental elements in various theories concerning close relationships [10].
Cultural norms and expectations seem to shape patterns of home-leaving and changes in parent-child relationships.For instance, in more family-oriented cultures (i.e., Portugal, Italy, Turkey), living with parents in emerging adulthood is normative [21].In a study with Italian emerging adults, living with parents was not found to be a restriction from which to break free and become independent [20].A study of Portuguese emerging adults found no difference between those who left home and those who did not regarding conflict with parents [22].Given that connectedness is highly valued in family-oriented cultures, changes in parent-child relationships during the transition to university and after moving out of the parental home cannot be conceived without considering cultural norms and expectations.

Changes in Well-Being during the Transition to University and after Home-Leaving
During the transition to university, the changes young individuals experience may lead to higher levels of stress [23].The World Health Organization (WHO) reported a high prevalence of mental health problems among university students [24].Especially for first-year university students, this is a challenging time as they take on more responsibilities, learn to manage their own lives and adjust to a new environment [23].It is now wellestablished that first-year transitional stress is associated with mental health issues such as depression and anxiety among university students [25,26].
Findings on how young individuals' well-being is influenced by their living situations are inconclusive.For instance, Kins et al. [27] found no difference in well-being between those who moved out and those who still lived at home, yet Seiffge-Krenke [28] reported that those still living at home had better psychological health than those who moved out.While parents provide some sort of protection for their children who live with them, over time, they may feel overburdened by commuting to university and handling daily responsibilities [29].Emerging adults who moved out have more independence and opportunities to further develop their autonomy, which may also increase their life satisfaction and overall well-being than those still living at home.However, moving out clearly adds more challenges to emerging adults' transition to university.Homesickness is common for university students who left home [4], particularly for those who live farthest from parental home [30].
Individual factors may influence the association between living situations and wellbeing.One such factor might be whether young individuals feel ready to leave home or not.Self-Determination Theory (SDT) emphasizes the importance of autonomy, which is defined as the need to be volitional in one's decision and actions, and describes it as one of the key factors influencing well-being [31].Regarding home-leaving, research shows that what matters more for well-being is that emerging adults' living situations reflect their personal and autonomous motivation [27].In high school, adolescents already differ in the extent to which they are motivated to leave home after they graduate [32].However, some students may not have another option but to move out to attend university.If they are not motivated to leave home, they may experience more difficulty in adjusting to their new environment and thus have lower well-being compared to those who want to leave home.Furthermore, nowadays, due to housing shortages and increasing rent in many countries, students cannot find accommodation and are thus forced to stay at home and commute to university.This obstacle may lead to lower levels of life satisfaction and higher levels of stress for university students.

The Transition to University in the Turkish Context: Youth and Family
With the value of higher education increasing over the past few decades, almost half (44.4%) of the young adults (20-24 years old) in Turkey are university students, which is the third highest percentage among European countries [33].Studies show that the characteristics of emerging adulthood (i.e., feeling-in-between, self-focus) mostly exist among university students in Turkey compared to non-university students [34].Close to the prevalence rates in Western societies [24], one-in-fourth of Turkish university students have moderately high levels of clinical depression [35], especially for those coming from low SES than mid-high SES families [3].Furthermore, female university students in Turkey experience higher levels of anxiety and stress than do male university students [3].Therefore, university students in Turkey require special attention, particularly during transitional periods.
Turkey has a family-oriented culture, yet Turkish families in urban settings present a unique context in which both individualistic values, such as autonomy, and collectivistic values, such as family ties, are important [36].In urban areas, Turkish mothers are seen as the primary caregivers, but they are also increasingly participating in the workforce [36].Turkish fathers in urban areas are less disciplinarian and more authoritative than fathers in rural areas [37].Traditional gender roles (i.e., girls more obedient, boys more independent) are less observed compared to rural areas, and female and male adolescents are equally encouraged to pursue higher education [38].With respect to leaving home, Turkey shows similar trends to Southern European countries, characterized by late home-leaving, strong association with marriage, and females moving out at a younger age than males [39].Regardless, Turkey has the highest percentage of university students living in student housing among European countries [40], which suggests that it is culturally acceptable to leave home for university.Thus, the current study will examine how this transitional time is experienced by young Turkish individuals and their parents growing up in urban contexts.

The Present Study
In accordance with the Life Course Theory [13], transitions affect the parent-child relationship, and maintaining a positive parent-child relationship is essential for emerging adults' well-being [14].Although ample research has focused on the transition to university, unfortunately, previous studies had some limitations leading to certain knowledge gaps.First, studies often use a cross-sectional design to investigate parent-child relationships and transition to university, which does not allow examining potential changes [22,27].Second, some studies investigating parent-child relationships in emerging adulthood use a wideage range, such as 18 to 30 [7,15], and do not emphasize the specific changes occurring during the transition to university.Third, studies on the transition to university often had participants only from one university [30,35], and just a few are prospective studies investigating changes starting from the time before the actual transitions [41].Finally, research on the transition to university and leaving home has predominantly focused on Western countries.However, it is crucial to pay more attention to the experiences of young individuals in non-Western countries, considering the distinct cultural and familial norms and expectations prevalent in these regions.
Focusing on the transition to university and possibly leaving home for some emerging adults, the current study aims to examine prospective changes in parent-child relationship quality and well-being, and investigates whether university students' living situation and their motivation to leave home prior to the actual transition are associated with these changes in a sample of Turkish emerging adults.We decided to concentrate only on firstyear university students, as new opportunities for autonomy and self-development during this time may lead to greater reexamination of the parent-child relationship and have a greater impact on their well-being compared to those who are not in university.For this, we formulated two research questions: 1.
Do young individuals' motivation to leave home and living situation predict changes in parent-child relationships and well-being?
First, for parent-child relationship quality, we expected warmth, conflict and autonomy support to remain relatively stable in the transition to university (H1).However, we anticipated living situation and motivation to leave home to be associated with changes in parent-adolescent relationship quality.We expected that university students who leave home would show an increase in parental warmth and autonomy support and a decrease in conflict compared to those living at home (H2).We anticipated that the motivation to leave would moderate these associations.For higher levels of motivation, we hypothesized that those who still live at home would show a decrease in warmth and autonomy support and an increase in conflict compared to those who moved out (H3).
Second, for well-being, we expected perceived stress to decline and life satisfaction to increase, given that during the senior year of high school, students in Turkey focus on the approaching (stressful) university exam (H4).We hypothesized that those who leave home would show a weaker decrease in stress compared to those living at home, as adjustment to a new environment and living independently may present challenges (H5).We formulated no specific hypothesis regarding the association between life satisfaction and living situations.Since entering university might be perceived as an achieved goal, this sense of accomplishment may lead to greater satisfaction, regardless of emerging adults' living situation.On the contrary, emerging adults may report less satisfaction because they continue to live at home.Finally, we anticipated an interaction effect between motivation and living situation.For higher levels of motivation to leave home, we anticipated a stronger decrease in stress and a stronger increase in satisfaction for students who moved out than those still living at home (H6).

2.
To what extent are changes in parent-child relationship quality and well-being associated with each other?
We expected a positive association between changes in stress and conflict with parents and a negative association between changes in stress and parental warmth and autonomy support (H7).We expected to find a negative association between changes in satisfaction and conflict with parents and a positive association with parental warmth and parental autonomy support (H8).We did not formulate specific hypotheses regarding differences in the associations between changes in well-being and changes in mother and father relationship qualities separately, but we explored whether the associations between these changes differed based on the present literature indicating differences in parenting styles of mothers and father [14,36].

Procedure and Participants
High schools and university exam tutoring centers in Istanbul, Turkey were invited to participate in the study via email.Schools that were interested in participating and gave permission (five high schools and one tutoring center for university exams) were visited twice during the first measurement.The first wave of data was gathered in the fall of 2017.During the first visit, the study was explained to the senior students during class hours.Those who were interested and older than 18 years of age signed the consent form, and younger students were given an information letter and a consent form to be delivered to their parents.Students were asked to return the signed consent forms to the school counselor within a week.Senior students with signed consent forms (91%) completed a paper-pencil form questionnaire set during one class hour (40 min).The participants did not receive any monetary compensation for their participation.The second wave of data was collected in the winter and spring of 2019.Only the students who had already given consent to participate in the second wave (96.4%) were contacted via email.The students received an online version of the same questionnaire set.Two reminder emails were sent biweekly and those who did not respond were called once if they provided a phone number at first measurement.Participants who completed the questionnaire set received an electronic gift card in the amount of 40 Turkish Liras (TRY) (10 USD).The present study was approved by Koç University Committee of Human Research in Turkey (Protocol no: 2017.156.IRB3.082).The research questions and hypotheses of this study were preregistered ("https://osf.io/t29h4/?view_only=55d19f27262c4a33bebce0235fe93aa8 (accessed on 15 December 2023)").
A total of 562 students participated at Time 1 (T1).All participants were senior high school students (M age =17.74 (0.53), 62.3% female) living with their parents.Most adolescents (84.3%) came from intact family structures, while the rest (16.7%) had either divorced or separated parents.Half of the adolescents came from high SES families (58.1%).Almost two-thirds of the adolescents attended a public high school (74.0%).
Of the 562 adolescents who participated at T1, 342 participated at T2 approximately 18 months later, with a completion rate of 94.4% (n = 323).Overall, the study had an attrition rate of 39.2%, which was moderate and comparable to the studies with a similar sample and design [41,42].Comparisons between those who participated in both waves (n = 323) and those who dropped out (n = 239) on background characteristics and study variables revealed four significant differences.Compared to participants who provided follow-up data, dropouts had higher initial mean scores of paternal autonomy support (t (544) =−2.04, p = 0.04) and paternal warmth (t (544) =−1.98, p = 0.048).The dropout group included significantly more males (χ 2 (1, 562) = 5.94, p = 0.01) and more young individuals from intact family structure (χ 2 (1, 562) = 7.19, p = 0.01) than the remaining group.The two groups did not differ with respect to socioeconomic status (SES), type of high school attended (i.e., private, public), and motivation to leave home.
The final sample of this study consisted of 240 Turkish university students (65.4% female), of which 27.5% had moved out at T2 (M age T1 = 17.74 (0.53), M age T2= 19.03 (0.58)).More than half were graduates of public high schools (69.6%) and came from high SES families (60.8%).The majority (80.4%) had an intact family structure.As the present study only focused on the transition to university, non-university students were not included in the final sample.The university student group differed from the non-university group (n = 83) with respect to a few aspects.A higher percentage of university students were from high SES families (χ 2 (1, 323) = 4.83, p = 0.03), graduates of private high schools (χ 2 (1, 323) = 20.03,p < 0.001), and moved out at T2 (χ 2 (1, 323) = 16.59,p < 0.001) compared to the non-university group.Furthermore, university students had higher levels of motivation to leave home in high school (t (319) = −3.06,p = 0.01), lower levels of stress (t (302) = 2.26, p = 0.02) and higher levels of satisfaction at T2 (t (302 = −3.44,p < 0.001) than the non-university group.The percentage of missing data ranged from 0.4% to 5.8%.Missing value analysis revealed only one significant result.The participants with missing data differed from the participants with complete data only with respect to their parents' marital status (χ 2 (1, 240) = 24.24,p < 0.001), meaning that the participants from divorced or separated families had more missing data on the father relationship qualities.In the main analyses, we utilized Full Information Maximum Likelihood (FIML) estimation in Mplus [43] (to handle the missing data.

Measures
All study variables were measured using self-report questionnaires and answered by the adolescents.

Parent-Adolescent Relationship Quality
Participants self-reported three qualities of their parent-child relationship (warmth, conflict and autonomy support).The Turkish-adapted version [44] of the Network of Relationship Inventory-short form (NRI) [45] was used to measure warmth and conflict in the relationship with both mothers and fathers using a five-point Likert scale ranging from 1 ("hardly at all") to 5 ("extremely much").The Support subscale measured warmth in the relationship with eight items (e.g., "How much does your mother/father really care about you?").The scale exhibited good reliability at both time points (mothers α = 0.79-0.80,fathers α = 0.87-0.90).The Negative Interaction subscale assessed conflict with six items (e.g., "How much do you and your mother/father argue with each other?") and showed excellent reliability across time (mothers α = 0.94-0.94,fathers α = 0.94-0.94).The Balanced Relatedness subscale from Shulman's Intimacy Scale [46] measured parents' autonomy support with seven items.The Turkish-adapted version of this scale [47] used a four-point Likert scale ranging from 1 ("not true") to 4 ("very true") (e.g., "My mother/father respects my ideas").The scale exhibited good reliability at both time points (mothers α = 0.89-0.89,fathers α = 0.90-0.91).The means of the three relationship qualities were calculated separately for mothers and fathers.For all scales, higher scores indicated higher levels of relationship quality.

Well-Being
Perceived stress and life satisfaction were chosen to assess the participants' well-being.We used the Perceived Stress Scale short version (PSS-10) [48], which was adapted and validated for the Turkish sample [49].The scale consists of 10 items rating the frequency of stress occurring in the past month.The scale uses a 5-point Likert scale ranging from 0 ("never") to 4 ("very often"; e.g., "In the last month, how often have you felt nervous and "stressed"?).The scale showed good reliability (α = 0.84-0.89)at both time points.The Turkish version of PSS-10 has satisfactory psychometric properties [50].
For life satisfaction, we used the Satisfaction with Life Scale (SWLS) [51], which was adapted to Turkish [52].The scale measured the overall satisfaction with 5 items using a 7-point Likert scale ranging from 1 ("strongly disagree") to 7 ("strongly agree") (e.g., "In most ways my life is close to my ideal").The scale exhibited good reliability in this sample (α = 0.87-0.84).The Turkish version of the SWLS has satisfactory psychometric properties [53].

Motivation and Living Situation
To measure motivation to leave home at T1, participants answered a single question about whether they would like to leave their parental home after high school, ranging from 0 (definitely no) to 4 (definitely yes).Given that all participants were living with their parents at T1, participants were asked at T2 if they left home or not (0 "living at home", 1 "moved out").

Control Variables
Participants answered single item questions about gender and SES.Gender was coded as male (0) and female (1).Parents' combined highest education level was used as an indicator of SES.Participants' responses were coded as follows: primary/middle school education as low SES (0), high school education as middle SES (1), and university/graduate school as high SES (2).Only one dummy variable was created (0 = low/mid SES, 1 = high SES) since we were mainly interested in seeing the effect of high SES compared to low/mid SES.

Statistical Analyses
As a preliminary step, we tested for longitudinal measurement invariance using Mplus version 8.2 [43] to ensure that the scales used to measure our outcomes measured the same construct over time (see Supplementary Materials).First, we compared configural, metric and scalar models' Chi-square values and checked if they were significantly different from each other (Table S1).All constructs, except for perceived stress, met criteria for full measurement invariance as Chi-square comparisons of these models yielded no significant differences (p > 0.05).Because the configural model of stress was significantly different from the metric and scalar models, we adjusted it by adding error covariances based on similar wordings (i.e., to control, to handle) used in some items to arrive at a well-fitting configural model.Next, we compared the new configural model of stress with metric and scalar models.In comparison to this new configural model of stress (Table S2), both metric and scalar invariance models did not lead to a worse fit beyond the recommended thresholds (≤ 0.010) for changes in RMSEA and TLI values [54].
Next, we used latent change score models (LCS) to assess both intraindividual changes and inter-individual differences in change over time [55] using Mplus version 8.2 [43].We utilized the Steyer et al. [55] parameterization, where the latent variables at T1 were measured using the indicators at T1 and T2, and the latent differences (i.e., change in stress) were identified using solely the indicators at T2.For our first research question, we ran 8 univariate LCS models (Figure 1) in which we regressed the change scores of our outcome variables parent-child relationship quality and well-being on living situation, motivation to leave home and their product in a stepwise fashion.In all models, participants' gender and SES were controlled for, and we ran separate analyses for mother and father relationship qualities.For the second research question, we ran 12 bivariate LCS models to test all correlations between the change scores of stress, satisfaction and the three parentadolescent relationship qualities separate for mothers and fathers.An alpha value of 0.05 was used as a criterion for interpreting the significance of the results.

Preliminary Analyses
Descriptive statistics and bivariate correlations are presented in Table 1.All measures of parentchild relationship quality and well-being were positively associated across both waves.Stress at T1 was only moderately correlated with stress at T2 (0.30).The rest of the variables showed high correlations (0.57-0.73) between T1 and T2.For all relationship qualities, reports on mothers and fathers correlated positively with each other, with a low-to moderate strength (0.15-0.46).Furthermore, warmth was positively associated with autonomy support but negatively associated with conflict for both mothers and fathers.Stress negatively correlated with satisfaction in both waves, with low-to-moderate strength.

Preliminary Analyses
Descriptive statistics and bivariate correlations are presented in Table 1.All measures of parent-child relationship quality and well-being were positively associated across both waves.Stress at T1 was only moderately correlated with stress at T2 (0.30).The rest of the variables showed high correlations (0.57-0.73) between T1 and T2.For all relationship qualities, reports on mothers and fathers correlated positively with each other, with a low-to moderate strength (0.15-0.46).Furthermore, warmth was positively associated with autonomy support but negatively associated with conflict for both mothers and fathers.Stress negatively correlated with satisfaction in both waves, with low-to-moderate strength.

Predicting Changes in Parent-Child Relationship Quality and Well-Being
Our first research question concerned if and how parent-child relationship quality and well-being of adolescents changed from last year of high school to first year of university, and whether this change was related to motivation of adolescents to leave home and the actual home-leaving in university.Table 2 presents the mean level changes and corresponding variances of all the univariate LCS models.Only young individuals' perceived autonomy support of the father and stress showed significant change over time.Contrary to our expectation, the autonomy support of fathers decreased from T1 to T2.Regarding stress, compared to the senior year of high school, participants reported lower levels of stress in the first year of university, as hypothesized.When comparing standardized latent change scores, we found that the biggest changes occurred in stress levels.Although average mean level changes were only moderate for most outcome variables, variances were significantly large for all relationship and well-being variables, meaning that there were significant individual differences in the changes that emerging adults experienced over time.Thus, we investigated whether the changes in the parent-child relationship and well-being were associated with university students' living situation and their motivation to leave home while still living with their parents at T1. Table 3 show the results of the hierarchical regression models testing the effect of living situation and motivation to leave home on changes in the parent-child relationship and well-being while controlling for gender and SES.We found a main effect of living situation (b = 0.28 (0.10), p < 0.05) only on change in stress.The positive change estimate indicated less decline for those who moved out compared to those living at home.There were no significant interaction effects of living situation and motivation to leave home on the changes in relationship quality and well-being.Although gender and SES were only control variables, results showed that gender was negatively associated with changes in autonomy support of mother (b = −0.17(0.07), p < 0.05).Females had, on average, a stronger decline in autonomy support of mothers when controlling for SES and living situation.Furthermore, the autonomy support of fathers showed less decline for university students from high SES families compared to low-mid SES families (b = 0.16 (0.07), p < 0.05).

Associations between Changes in Relationship Quality and Well-Being
Our second research question concerned the extent to which changes in parent-child relationship quality were associated with changes in well-being during the transition to university.To answer this question, we tested 12 bivariate LCS models (i.e., 3 maternal and 3 paternal relationship qualities and 2 measures of well-being; see Table 4).In line with our hypotheses, the results showed a significant association between changes in life satisfaction and changes in parent-child relationship quality, except for conflict with father.As satisfaction increased, the warmth and autonomy support of both parents also increased.The negative association between satisfaction and conflict with mother indicated that as satisfaction increased, conflict with mother decreased.Furthermore, except for the significant and positive association between changes in stress and changes in conflict with mother, none of the other changes in parent-child relationship qualities were associated with changes in stress.As stress increased, conflict with the mother also increased.All associations could be qualified as small.Note.b = unstandardized regression coefficients, β = standardized regression coefficients.SE = standard error.Living sit.= living situation, Liv *motiv.= interaction term between living situation and motivation to leave home, * p < 0.05.

Discussion
The current study aimed to investigate changes in parent-child relationship quality and well-being during the transition to university while also examining the effect of homeleaving and motivation to leave home on these changes.Our results showed that both the autonomy support of fathers and young individuals' stress decreased over time, while other relationship qualities and life satisfaction remained relatively stable on average.First-year university students who left home showed less decline in stress than those still living at home.Furthermore, the changes in parent-child relationship quality and well-being (mainly satisfaction) were related to each other.These findings advance our understanding of the multiple transitions, such as entering university and home-leaving, and how they impact emerging adults' parent-child relationships and well-being in a family-oriented culture.

Changes in Parent-Child Relationship, Well-Being and Leaving Home
In line with previous research [14,56], the overall quality of the parent-child relationship remained relatively stable during the transition to university.Contrary to our expectations, the autonomy support of fathers declined during this period.This was surprising since the literature indicates that emerging adults are in need of more autonomy [1], and parent-child relationships become more egalitarian in emerging adulthood [14,56].This finding may align with one study [19] showing that parental acceptance reported by youth declines throughout adolescence and in the early years of emerging adulthood, yet in this study no difference between maternal and paternal acceptance was detected.Perhaps in our sample, Turkish fathers have yet to acknowledge their emerging adult children's new developmental status and their increased need for autonomy in transition to university.It could be that fathers need more time to readjust their parenting styles, as they are less involved in the relationship with their children compared to mothers [19,54].Another explanation may be that fathers adjust their autonomy support, but young adults simply require more than what their fathers are providing.Our results also indicated a less strong decrease (or stronger increase given that we also found individual differences in change) in autonomy support of fathers in high SES families compared to low-mid SES families.This finding is in line with previous studies showing that autonomy is valued especially among high SES Turkish families living in urban areas [36,38].Furthermore, although we did not detect an average decline in the autonomy support of mothers, female university students indicated a stronger decline (or weaker increase for some) in the autonomy support of their mothers than male university students.From a cultural perspective, girls in Turkish families are expected to be more obedient [38], and thus may receive less autonomy support compared to boys.Moreover, mothers may be more worried during this transitional period and still feel the need to control and monitor their daughters when they start building a new life as university students.
Regarding well-being, only the perceived stress of young individuals declined somewhat over time.It seems that the senior year of high school is a stressful time as students prepare for the university exam in Turkey.After successfully passing the exam and entering university, overall stress declines for many emerging adults.However, we also found individual differences, indicating that not all university students' well-being changed the same way.Some university students showed an increase in perceived stress over time.As anticipated, home-leaving was associated with changes in stress.Those who left home in their first year of university experienced higher levels of stress compared to those living at home.This might be due to home leaving requiring more adjustment to new roles and responsibilities and making young individuals more vulnerable to stress [23].
Contrary to previous research [27] and SDT's perspective [31], we found no evidence for motivation to leave home impacting the changes in parent-child relationship or wellbeing, or for the interaction between living situation and motivation.We did not find lower well-being for students whose earlier motivation in high school did not match their current living situation at university.Perhaps motivation to leave home shaped high school students' university decisions by affecting how far they were willing to move to attend university, but had no role in the changes of well-being once they became university students.Overall, our findings show evidence for the individual differences in parent-child relationship quality and well-being changes, and also leaving home leading to higher levels of stress for first-year university students.

Association between Changes in Parent-Child Relationship and Well-Being
We found evidence for associations between changes in parent-child relationship quality, perceived stress, and life satisfaction.The strongest relation was between changes in the satisfaction and warmth of both mothers and fathers.As emerging adults become more satisfied with their lives, this also reflects on the warmth of their relationships with both parents and vice versa.Also, changes in the autonomy support of both parents were positively related to changes in satisfaction.Overall, the associations between satisfaction and parent-child relationship quality were quite similar for mothers and fathers.Only perceived stress was associated (positively) with changes in conflict with the mother.This suggests that when university students experience stress during the transitional period, it can spill over to their relationship with their mothers, or in the other direction, a conflictual relationship with mothers becomes a stress factor for first-year university students.This may be a temporary increase in conflict, as young individuals adjust to their new environment and manage their responsibilities.Their level of stress may decline further in the following years of university, and this may be linked to a less conflictual and more satisfying relationship with mothers, in line with previous findings [15].
Our findings overall suggest that a negative relationship with fathers (i.e., conflict) may have less impact on emerging adults' well-being compared to a negative relationship with mothers.However, the dynamics between relationships with mothers and well-being need to be further investigated.Perhaps factors such as how frequently first-year university students communicate with their mothers, how often they visit home if they move out, and their topics of conflict (i.e., daily routines, responsibilities, more autonomy) can be explored by future studies to shed more light into these associations.In sum, our results on the associations between parent-child relationships and well-being are in line with previous research showing that more than fathers, mothers influence their children's happiness [57].Together, the present study supports the notion of linked lives, as suggested by the Life Course Theory [13].Parent-child relationships and transitions young individuals go through influence each other, and parents are still key figures in emerging adulthood when it comes to their children's well-being.

Strengths, Limitations and Future Directions
The strengths of the present study were the focus on a specific transitional period in a family-oriented culture, the prospective design to investigate factors related to homeleaving before the actual transition, and the examination of several individual and familyrelated factors at once.However, our findings should be interpreted considering some limitations.First, our study had a moderate and acceptable rate of attrition (39.2%) from T1 to T2, which was expected given that between the two time points participants graduated from high school, possibly moved away, and may even have switched e-mail addresses.Second, the present study was conducted in an urban city (Istanbul) and the results may be different in rural areas where a lack of universities within the vicinity may force more young individuals to leave home to attend university.However, it may also be that parents in rural areas have less education and more traditional values, and this may put more pressure on their adolescent child to remain longer at home, which may impact adolescents' motivation to leave home, their university choices, and the overall rate of home-leaving for university.Third, we had a small sample size of those who left home in the first year of university.Perhaps with a larger sample of emerging adults who left home, and thus more power, we might have been able explain more of the individual differences in changes in the parent-child relationship and well-being.Furthermore, we did not differentiate between those who left home completely and those who moved out (i.e., student housing) but went back home on the weekends.As the latter group may not feel as if they left home completely, this may influence both the parent-child relationship quality and their well-being.Fourth, this study examined longitudinal changes over a relatively short period of time (i.e., 18 months).Future research should preferably include additional assessments over a longer period.Finally, only adolescents' perceptions of relationship qualities and their well-being were used in the current study.Including parents, a close friend or a romantic partner as respondents may provide a more complete picture.Future research could benefit from addressing these limitations to gain more insight into the changes in parent-child relationships and well-being as well as the dynamics of home-leaving in transition to university.

Conclusions
To conclude, the present study showed that parent-child relationship quality remained relatively stable during the transition to university, yet there were individual differences.Home-leaving was associated with changes in perceived stress, indicating that first-year university students who left home experienced higher levels of stress than those still living at home.During this transitional period, the changes in parent-child relationship quality and well-being were related to each other.The present study once again highlights the importance of paying attention to first-year university students, particularly to those who leave home and who may need more support from their parents and their universities.Thus, our results may be useful for university staff and mental health professionals working with emerging adults in transitions, as well as parents who continue to be a source of support for their emerging adult's well-being.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/youth4010006/s1,Table S1  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Figure 1 .
Figure 1.Schematic overview of the latent change score models and predictors in the present study.

Figure 1 .
Figure 1.Schematic overview of the latent change score models and predictors in the present study.

Author Contributions:
Conceptualization, R.I.A., L.D.B. and S.B.; formal analysis, R.I.A.; writingoriginal draft preparation, R.I.A.; writing-review and editing, R.I.A., L.D.B. and S.B.All authors have read and agreed to the published version of the manuscript.Funding: This research received no external funding.Institutional Review Board Statement:The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of KOC UNIVERSITY, Istanbul, Turkey on 22 September 2017 (protocol code 2017.156.IRB3.082).

Table 1 .
Means, Standard Deviations, Percentages, and Correlations of All Study Variables.

Table 2 .
Unstandardized and Standardized Latent Change Scores (LCS) and Variances of Parent-Child Relationship Quality and Well-Being.

Table 3 .
Predictors of Latent Change Scores of Mother-Child and Father-Child Relationship Quality, and Well-Being.
: Longitudinal Measurement Invariance: Chi-Square Difference Comparisons for Parent-Adolescent Relationship Qualities and Well-Being; Table S2: Testing Longitudinal Measurement Invariance for Stress.