The Role of Self-Esteem in Adolescents' Perception of Parents and Social Anxiety Levels

The purpose of this study is to examine whether the relationship between adolescents' perceptions of parents and their social anxiety levels is mediated by self-esteem. The sample of the study consisted of 694 (324 female and 360 male) volunteer high school students selected with a simple random sampling method from different high school types in Mersin province during the 2017-2018 academic year autumn term. The following tools were used in this study; "Perception of Parents Scale" which was developed by Robbins (1994) and adapted to Turkish by Kocayörük (2009); "Rosenberg SelfEsteem Scale" developed by Rosenberg (1965) and translated into Turkish by Çuhadaroğlu (1986); "Social Anxiety Scale for Adolescents" was developed by La Greca and Lopez (1998) and adapted to Turkish by Aydın and Tekinsav (2007). Multiple regression analysis was used to analyze the data. The results show that there is a negative relationship between adolescents' social anxiety levels and their perception of their parents and self-esteem. In addition to these results, it was concluded that self-esteem had a full mediating effect on the relationship between the perception of both mothers and fathers of their adolescents and their social anxiety levels.


INTRODUCTION
In the process of growth, children begin to form self-perception through messages they receive from their parents through various channels. This concept, called self, is the emotions and thoughts of people about their abilities, behaviors and worthiness (Atkinson et al., 2006). Rogers states that the concept of self is in a certain coherence. This need for consistency affects the way people perceive their surroundings (Gündüz, 2017).
While the individual's thoughts about themselves as a person about their personal characteristics constitute the concept of self (Burger, 2016); self-esteem is the extent to which the individual rewards, values, endorses or loves themselves in the general success of the individual or in areas of life that are particularly meaningful to the individual (Mruk, 2013). Rosenberg treats selfesteem, which is also conceptualized as self-respect, as a positive or a negative attitude of the individual towards themselves. If the self-evaluation of the individual is positive, their self-esteem will be high; and if it contains negativity, it will be low (Torucu, 1990). The common point that humanitarian psychologists have focused on is what people think about themselves, and self-esteem is seen as an important structure that affects their relationship with themselves and others (Burger, 2016). People's well-being is positively affected by self-esteem, which has a close relationship with many human variables (Neff, 2011); otherwise, the level of depression and learned helplessness may increase (Orth and Robins, 2013;Yücel, 2013;Orth, Robins, & Roberts, 2008). Individuals with low self-esteem exhibit overeating behaviors more (Martyn et al., 2009) and there are generally negative relationships between self-esteem and adaptation and behavioral disorders (Zeigler-Hill, 2011).
Teachers and parents contribute to the development of self-esteem by identifying their superiority and distinguishing their children from others (Burger, 2016). In the conducted research, the positive relationships that parents have with their children is reported to be effective on emotional intelligence (Bozdemir, 2015), self-perception (Sezer, 2010), and self-esteem (Erbil, Divan and Önder, 2006). According to research, it is seen that the individuals who grow with democratic parental attitudes have high self-esteem (Sezer, 2010;Ersoy, 2013). According to the research results of Yücel (2013), students' self-esteem is positively related with the perceived democratic parental attitudes and negatively related with the perceived protective and authoritarian parental attitudes.
According to Bowlby, the closeness between the caregiver and the child serves as "a safe base" that the child can use to discover his environment and "a strong shelter" where the child can be protected in case of danger (Sümer and Güngör 1999a). Undoubtedly, these foundations, which were set in the early childhood, also affect the processes of adolescence and after. Parents' attitudes can be vital in this process. Especially in a period such as adolescence when both physical and emotional changes are happening, and academic and professional preferences are shaped, it is important to develop perception of parents and healthy relationships. In other words, the family has an important impact on the development of the child's personality and behavior through being an important institution with its economic, cultural and social dimensions, as well as the successful completion of a critical development process such as adolescence. It can be concluded that parents' support contributes to the basic psychological needs of adolescents, their development of independent sense of self, selfesteem and well-being (Kocayörük, 2012). For a positive sense of self and a high level of self-esteem, the family in which children grow up and the social environment they interact with are important. Adolescents with positive attachment to their parents have high self-esteem (Bayraktar, Sayıl and Kumru, 2009). According to Arslan (2018), there is a positive and significant relationship between perceived acceptance and interest from parents and self-esteem.
The relationships between parents and adolescents are associated with many important structures and variables. Many positive and negative characteristics gained in the family can affect the relationships with other people in their social life. The child uses the patterns they have gained from their parents in their relationship with the outside world and others. Adolescence, which includes an important process in the human life, represents an important period in which many changes and developments occur, as well as social anxiety. Adolescence is a period when acceptance by friend groups have the highest importance and when one realizes how important the impressions individuals leave on others truly are (Sübaşı, 2007). Social anxiety is defined as anxiety caused by negative affect, anxiety and negative psychological stimulation in social situations, including the possibility of being evaluated or the possibility of being evaluated by others (Schlenker & Leary, 1982). In this context, adolescents start to avoid social environments with the fear of being evaluated negatively.
Although people, as social beings, pay attention to how they are seen by others at almost every age, this sensitivity can be maximized in adolescents whose identity and personality have not yet been settled. Attitudes, especially when raising children in a family environment, can be seen as determining factors in terms of social anxiety. Erkan (2002) stated that the students with high social anxiety level were exposed to a predominantly authoritarian and protective attitude in their families. On the other hand, it is seen that there is a similar result in the studies that reveal the relationships between the perception of parents and attachment styles of the individuals and their social anxiety levels. Manning et al. (2017) reported a positive relationship between insecure attachment and social anxiety in twenty-eight of the thirty studies they examined in the screening study. In another screening study, Brook and Schmidt (2008) found that over-control in the parent and family interaction and the psychopathology of the parents were the main determinants of social anxiety problems in children.
Studies in literature also show that there is a relationship between self-esteem and social anxiety. While social appearance anxiety was low in those with high self-esteem (Şirin, 2015;Kılıç, 2015), low self-esteem was seen in children who participated less in social events (Çevik Büyükşahin and Atıcı, 2009). According to Sübaşı (2007), the most important predictors of social anxiety are the variables of self-esteem and loneliness. In many studies, self-esteem was negatively and significantly correlated with social anxiety (Kocovski and Endler, 2000;De Jong, 2002;Rasmussen &Pidgeon, 2011;De Jong, Sportel, De Hullu and Nauta, 2011). On the other hand, there are few studies examining the three variables in the present study; self-esteem, parental attitudes and social anxiety. One of them, Eroglu (2018), found a negative correlation between social anxiety and self-esteem, and positive correlation with protective attitudes; and a positive relationship was determined between selfesteem and democratic attitudes.
Self-esteem is an important concept found in direct relationships with many variables as outlined above. In addition to direct relationships, self-esteem was shown to have significant impact as an intermediary variable in some studies. According to this, the mediating role of self-esteem, between social anxiety and awareness (Rasmussen & Pidgeon 2011); problem solving and parental attitudes (Kayaalp and Gündüz, 2018); life satisfaction and social support (Kong, Zhao and You, 2013); automatic thoughts and hopelessness (Çakar, 2014), problem solving skills and loneliness (Karataş, 2014) has been shown.
As seen in the related research, it is seen that there are important relationships between the perception of parents, self-esteem and social anxiety variables developing on the basis of parental relations. When examined, no studies were found that examined these three variables and tested the mediation of self-esteem. It is thought that the findings will provide a scientific basis for school-based psychological counseling for adolescents and their parents. The aim of this study is to examine the role of self-esteem in the relationship between perception of parents and social anxiety.

Models of Research
This descriptive correlational research was conducted with the aim of examining the mediator role of self-esteem in the relationship between adolescents' perception of parents and social anxiety levels.

Research Group
In order to create the research group, data was collected from 694 volunteer high school students selected with a simple random sampling method from different high school types in Mersin province during the 2017-2018 academic year autumn term. 324 (46.7%) of the students were female and 360 (53.3%) were male. Of the participants from seven high schools in Mersin, 223 (32.1%) were 9th grade, 178 (25.6%) were 10th grade, 170 (24.5%) were 11th grade and 123 (17.7%) were 12th grade students.

Data Collection Tools
Personal Information Form: In order to define the study group and to determine the independent variable of the study, Personal Information Form prepared by the researchers was used.

Perceptions of Parents Scale (POPS):
In order to measure the perceptions of adolescents towards their parents, the Perceptions of Parents Scale form developed by Robbins (1994) and adapted to Turkish by Kocayörük (2009) was used. With POPS, measurements in three sub-dimensions (participation, support for autonomy and closeness) can be made after making inverse scoring, and the total score of perception towards the parent can be obtained by combining these dimensions. A high score indicates that the parent has a high level of autonomy support and a low score indicates that the parent has more control. The scale consists of 21 items and it is a 7-point Likert type (1 = completely wrong, 7 = completely correct) scale. In the adaptation study, while the Cronbach alpha internal consistency coefficient for the total score of the mother form of the scale was .91, the Cronbach alpha internal consistency coefficients of the sub-dimensions were reported between .58 and .90. In the adaptation study, while the Cronbach alpha internal consistency coefficient for the total score of the father form of the scale was .93, the Cronbach alpha internal consistency coefficients of the subdimensions were reported between .62 and .92. In this study, total scores obtained from the scale were evaluated and Cronbach alpha internal consistency coefficient was found to be .92 according to the total score obtained from the mother form; and the Cronbach alpha internal consistency coefficient was found to be .94 according to the total score obtained from the father form.
Rosenberg Self-Esteem Scale (SES): The Rosenberg Self-Esteem Scale (Short Form), which was developed by Rosenberg (1965) and translated to Turkish by Çuhadaroğlu (1986), was used in order to determine the perceptions of individuals about their self-worth. The scale consists of five positive and five negative items for a total of 10 items and it is a 4-point Likert-type scale. The lowest score that can be obtained from the measurement tool is 10 and the highest score is 40. A high score from the measuring tool indicates a high self-esteem. Çuhadaroğlu (1986), who performed the adaptation study of the scale, reported that the scale had a Cronbach α coefficient of .76. In this study, Cronbach's alpha internal consistency coefficient of the scale was .84. Greca and Lopez (1998) in order to measure the social anxiety of adolescents. The adaptation of the scale to Turkish was performed by Aydın and Tekinsav (2007). The scale is a five-point Likert type (1 = Never, 5 = Always) scale. 18 items of the scale, which consists of 22 items, are taken as scale items and the other 4 items are about other topics and not included in the scoring. In this case, the lowest possible score from the scale is 18 and the highest score is 90. While a total score can be obtained from the scale, the scale can also be taken into consideration in 3 sub-dimensions (Fear of negative evaluation, fear and restlessness in new social situations and fear and restlessness in general social situations). In the adaptation study, the Cronbach alpha internal consistency coefficient for the total score of the scale was .88, whereas the Cronbach alpha internal consistency coefficients of the sub-dimensions were reported between .68 and .83. In this study, total scores obtained from the scale were evaluated and Cronbach alpha internal consistency coefficient was .89.

Procedure
After obtaining the permission from the necessary institutions, the measurement tools were applied during class hours. The participants were informed about the purpose of the study and application was performed with volunteer individuals. The application of measuring tools took approximately 15-20 minutes.

Data Analysis
Data were analyzed using SPSS 20.00 software. Multiple regression analysis was used to analyze the data. The mediation effect was based on the model proposed by Baron and Kenny (1986). The assumptions of multiple regression analysis were tested before data analysis was performed. The test results showed a normal and linear distribution of data.

FINDINGS
The relationships between the variables in this study where social anxiety is predicted in high school students and the mean and standard deviation values of the variables are given in Table 1. When Table 1 is examined, it is seen that there is a significant positive relationship between the perception of mother and perception of father (r = .53; p <.01) and self-esteem (r = .31; p <.01); and a significant negative relationship with social anxiety (r = -119; p <.01). There was a significant positive correlation between the perception of father and self-esteem (r = .27; p <.01) and a significant negative correlation with social anxiety (r = -. 14; p <.01). However, there was a significant negative correlation between self-esteem and social anxiety (r = -.42; p <.01). The internal consistency coefficients of the variables are between .84 and .94.

Table 2. Regression Analysis Results of the Predictive Role of Self-Esteem among Adolescents' Perceptions of Mothers and Social Anxiety Levels
Steps Step 2 Perception of Mother (P) Self-esteem (C The regression analysis of predicting the mediating role of self-esteem in the relationship between adolescents' perception of their mothers and their social anxiety levels was carried out in three steps as suggested by Baron and Kenny (1986). The results obtained from the analysis are shown in Table 2. In the first step, adolescents' perception that their mothers support their autonomy was seen to reduce their social anxiety levels (B= -.16; β= -.19; P<.001) and explains 19% of the variance. In the second step, adolescents' perception that their mothers support their autonomy was seen to increase their self-esteem (B= .13; β= .31; P<.001) and explains 31% of the variance. In the third step, selfesteem determined as a mediator variable was seen to decrease social anxiety levels (B= -.80; β= -.40; P<.001). Perception of mother and self-esteem together explain 42% of the variance. In addition, when the mediator variable, self-esteem, is taken into consideration, the perception of mother does not significantly predict social anxiety levels (B= -.05; β= -.06; P>.05). When the mediator variable and the predictive variable are analyzed together, the reduction or disappearance of the meaningful relationship between the predictor and the predicted variable is considered to be the last criterion indicating the mediating effect. Therefore, in this study, it can be said that there is a full mediating effect of self-esteem on the relationship between perception of mother and social anxiety.
When the table is examined, perception of mother's support of autonomy increases self-esteem and self-esteem decreases the level of social anxiety (B= .13*-.80= -.10; Sobel z= -6.73; P=.00). This effect can also be interpreted as follows; each unit increase in perception of mother scores, leads to an increase of .13 units in self-esteem scores and -.80 of this increase affects the social anxiety level. The indirect effect of perception of mother on the level of social anxiety is (.13 * -. 80) -.10.

Perceptions of Fathers and Social Anxiety Levels
Steps of the Mediating Variable Test B Sh β t p Step 1 Perception of Father (P) Social anxiety (C) -.09 .03 -.14 -3.73 .00** R= .14, R²= .02, F= 13.907 Step 2 Perception of Father (P) Self-esteem (C The results obtained from the analysis of adolescents' perception of their fathers are shown in Table 3. In the first step, adolescents' perception that their fathers support their autonomy was seen to reduce their social anxiety levels (B= -.09; β= -.14; P<.001) and explains 14% of the variance. In the second step, adolescents' perception that their fathers support their autonomy was seen to increase their self-esteem (B= .09; β= .27; P<.001) and explains 27% of the variance. In the third step, selfesteem determined as a mediator variable was seen to decrease social anxiety levels (B= -.82; β= -.41; P<.001). Perception of father and self-esteem together explain 42% of the variance. In addition, when the mediator variable, self-esteem, is taken into consideration, the perception of father does not significantly predict social anxiety levels (B= -.02; β= -.03; P>.05). When the mediator variable is analyzed together with the predictor variable, with the removal of the significant relationship between the predicting and predicted variable, it can be said that there is a full mediating effect of self-esteem in the relationship between perception of father and social anxiety.
When the table is examined, perception of father's support of autonomy increases self-esteem and self-esteem decreases the level of social (B= .09*-.82= -.07; Sobel z= -6.22; P=.00). This effect can also be interpreted as follows; each unit increase in perception of father scores, leads to an increase of .09 units in self-esteem scores and -.82 of this increase affects the social anxiety level. The indirect effect of perception of father on the level of social anxiety is (.09 * -.82) -.07.

RESULTS, DISCUSSION AND SUGGESTION
In this study, the mediating role of self-esteem among the parental perception levels of high school students and their social anxiety levels was examined. According to the results obtained from the first correlation analysis, there was a positive correlation between the perception of the mother and the perception of father and self-esteem. According to this, it is observed that the self-esteem of adolescents with healthy parental connections are also higher. According to another result, the social anxiety of the students decreases as their perception of or bond with parents becomes stronger. These obtained results support the study findings that show a positive relationship between positive perception of parents and the self-esteem of the child (Erbil, Divan and Önder, 2006;Yücel, 2013;Arslan, 2018); and a negative relationship with social anxiety (Erkan, 2002;Festa and Ginsburg, 2011). Considering that the child's self-perception is formed primarily by the interactions taken from the parents, it can be said that the self-esteem of the individuals whose basic needs are met in a healthy way, will increase. First the family in which they were born and raised in, and then the teachers, are important people who make efforts to develop children's self-esteem (Burger, 2016). It can be concluded that parents' support contributes to the basic psychological needs of adolescents, their development of independent sense of self, self-esteem and well-being (Kocayörük, 2012). For a positive sense of self and a high level of self-esteem, the family in which children grow up and the social environment they interact with are important. Adolescents with positive attachment to their parents have high self-esteem (Bayraktar, Sayıl and Kumru, 2009).
Similarly, social anxiety levels of children with positive relationships with their parents are also low. According to the attachment theorists, in the normal conditions, the relationship between the mother, who is the child's first caregiver, is the determinant of the future relationship-building behavior (Bowby, 1988). In other words, it is more likely for individuals who have a negative perception of parents to experience problems in their interaction with other people in the following years. According to this, there is a parallelism with the studies reporting that the social anxiety of children who have insecure attachment to their mothers and whose parents are anxious, is also high (Manning et al., 2017;Festa and Ginsburg, 2011;Crosby Budinger, Drazdowski, and Ginsburg, 2013;Knappe, Beesdo, Fehm, Lieb and Wittchen, 2009).
Studies in literature also show that there is a relationship between self-esteem and social anxiety. While social appearance anxiety was low in those with high self-esteem (Şirin, 2015;Kılıç, 2015), low self-esteem was seen in children who participated less in social events (Çevik Büyükşahin and Atıcı, 2009). In many studies, self-esteem was negatively and significantly correlated with social anxiety (Kocovski and Endler, 2000;De Jong, 2002;Rasmussen &Pidgeon, 2011;De Jong, Sportel, De Hullu and Nauta, 2011).
On the other hand, when the results of the analysis on mediation are examined, it is seen that self-esteem has mediating effect on perception of mother and social anxiety. According to this, while the perception of mother's support of autonomy increases self-esteem, this increase in self-esteem also decreases the level of social anxiety. In other words, every increase in perception of mother scores leads to an increase in self-esteem scores and this decreases the level of social anxiety. When the dimension of the study which deals with the father-adolescent relationship is examined, it is seen that similar results have been reached. Again, while the perception of father's support of autonomy increases self-esteem, this increase in self-esteem also seen to decrease the level of social anxiety. In other words, every increase in perception of father scores leads to an increase in self-esteem scores and this decreases the level of social anxiety. In short, it can be said that the self-esteem of adolescents with a positive and healthy perception of their parents increases and consequently their social anxiety decreases significantly. Although there is no study investigating the mediating effect of self-esteem on the parental perception and social anxiety relationship; it has been reported that this concept has a