Emotion Regulation and Emotion Understanding in Preschoolers as a Predictor of the Maternal Socialization of Emotion

In this study, the relationship between emotion regulation and emotion understanding skills of preschool children and emotional socialization behaviour of mothers was investigated. This study was designed as a correlation type relational survey model with a total of 180 participants - 90 children between the ages of 3-6 and 90 mothers – who have low, medium, and high socio-economic levels. The data of the study were collected using the Coping with Negative Emotions of Children Scale, Emotion regulation Scale, and Wally Emotion Test. The analysis of data was made with multiple regression analysis. The results of the study showed that emotional socialization behaviour of mothers was a significant predictor of emotion regulation skills of preschool children. Problem based and punitive reactions of mothers that are included in emotional socialization behaviours was identified as significant predictors of emotion regulation skills of children. Accordingly, as problem based, and punitive reactions increased, emotion regulation skills of children also increased. However, the results of the study showed that emotional socialization behaviour of mothers was not a significant predictor of understanding emotional skills of preschool children. Based on the research findings, it is suggested that policy makers, institutions educating teachers or teachers themselves prepare early intervention programs to be implemented in pre-schools to ensure mothers' emotional socialization behaviors recognize the potential effects on their children and to support children's emotional development.


Introduction
Emotion regulation level, which is one of the fundamental development tasks of the preschool period (Macklem, 2008), consists of external and internal processes for monitoring, evaluating, and changing emotional reactions to realise the objectives of an individual (Thompson, 1994). Emotion regulation, which rapidly develops beginning from the first years of life (Eisenberg, Spinrad, & Eggum, 2010), is a process that has starting, sustaining, and changing emotional reactions in both positive and negative events (Bridges & Grolnick, 1995;Grolnick, Bridges, & Connell, 1996). This process covers coping with increasing positive and negative emotions such as joy, pleasure, anxiety, anger, and fear (Kopp, 1989). Adjustment of emotion represents the change in the stimulated emotion. This change could occur in internal processes of individuals as well as interpersonal processes (Cole, Martin, & Dennis, 2004).
Emotion regulation which is a fundamental element of personal functionality, is an important predictor of psychological adjustment and social competence (Thompson & Meyer, 2007). Emotion regulation skills play a critical role on children to form and sustain positive relationships with peers (Denham & Burton, 2003). Accordingly, it is known that children with high emotion regulation skills have high social competency and are liked by their peers (Macklem, 2008). It is known that children who are successful in emotion regulation have higher academic success rates, in-class productivity, early literacy, and mathematic skills (Graziano, Reavis, Keane, & Calkins, 2007). On the contrary, children with inadequate emotion regulation and a tendency for negative emotions could show behavioural problems (Arı & Yaban, 2016;Batum & Yağmurlu, 2007;Eisenberg et al., 1996;Eisenberg et al., 2000).
Emotion regulation is a developmental phenomenon that developed from awareness of children of different emotional impulses and interactions with caregivers (Kopp, 1989). Humour of children, personal characteristics of caregivers, and behavioural interactions are critical for emotion regulation skill development (Calkins, 1994). In other words, emotion regulation skills have a relationship with genetic and environmental factors and affects the interaction of these factors. However, individual differences caused by genetics have constant effects on the adjustment skills of children after one or two years (Eisenberg et al., 2010). Emotion regulation levels significantly affect external interventions and occur by the effects under social context that are important for shaping the emotional impulses of children (Thompson, 1994). From this perspective, emotion regulation mainly develops from the relationship between children and parents or caregivers (Southam-Gerow & Kendall, 2002).
The method of interpreting and evaluating emotions of children that are affected by social interactions taking place around them is important for learning emotional management strategies, gaining self-confidence and self-sufficiency for controlling emotions, and gaining expectation based on gender for cultural and social context. Although these effects occur in various social contexts, since the effect of the family environment is visible in early ages, it is wide spread and multi-directional force on foundational and emotional development (Thompson & Meyer, 2007). Mothers who are actively involved in this process play an important role in emotion regulation strategies and skill development of children (Grolnick, Kurowski, McMenamy, Rivkin, & Bridges, 1998;Kopp, 1989). The ability to understand emotions gains importance in the development of emotion regulation skills. Improvement in children's understanding emotions is nurtured from the acquisition and use of emotion regulation strategies (Eisenberg, Cumberland, & Spinrad, 1998). Understanding emotions, the ability to recognize the emotions reflected in facial and body language, the reasons for the emotions of oneself and others,how to communicate with others, and how to cope with emotions are all part of emotion regulation (Southam-Gerow & Kendall, 2002).
The ability to understand emotions, which are important components of emotional development, develops at an early age. The critical foundation for children to understand emotions is laid during infancy and early childhood. Thus, children develop the ability to identify the visual and verbal signs of emotions before they begin to talk about emotional experiences and express their emotional experiences (Thompson & Lagattuta, 2006). It is vital for children to understand emotions that has definitions and naming for enabling children to gain social experiences (Denham & Burton, 2003). Children with the ability to understand emotions will form more positive relationships with peers (Cassidy, Parke, Butkovsky, & Braungart, 1992). The ability to understand emotions is at the centre of emotional sufficiency (Denham & Burton, 2003). As children become more able to know their emotion and emotions of other people, they become ready to gain empathy, perspective, and emotion regulation skills (Webster-Stratton & Reid, 2004).
In the development of the child's emotion understanding, the family plays an effective role (Laible & Thompson, 1998). This emotional knowledge of children occurs in a family environment. In this process, some families provide a safe environment for their children in emotional communication and understanding, while others have emotional difficulties that threaten the development of knowledge and skills that transcend coping abilities of children (Thompson & Lagattuta, 2006). Understanding emotions depends on the conversation between parents and children (LaBounty, Wellman, Olson, Lagattuta, & Liu, 2008). In this context, it is known that parents who fulfil their duty as emotional coaches of children successfully have more successful children who understand emotions better (Denham, Mitchell-Copeland, Strandberg, Auerbach, & Blair, 1997).
Understanding emotions is related with cognitive development, however, general information such as structure of the family, number of adults in the house, language spoken in the house, number of children in the house, and education level of parents (Cutting & Dunn, 1999) and family based factors such as parent-child relationships (Kårstad, 2016), reliable connection relationships (Laible & Thompson, 1998) could also be effective. Similarly, emotion regulation skills are affected from family based variables such as cognitive flexibility (Şahin & Arı, 2016), humour of the children (Yağmurlu & Altan, 2010) connecting styles (Cassidy, 1994;Ural, Güven, Sezer, Efe Azkeskin, & Yılmaz, 2015), parentchildren relationship (Ambrose, 2013;Dereli, 2016;Friedlmeier & Trommsdorff, 1999;Kiel & Kalomiris, 2015), and parenting style (Mathis & Bierman, 2015). Another important variable related to the family in the development of emotion regulation and understanding is parental emotional socialization behaviours. Emotional socialization behaviours are defined as the reactions of parents to the emotions of children with negative emotions such as anger, sadness, and fear (Altan Aytun, Yağmurlu, & Yavuz, 2012;Eisenberg et al., 1998). These reactions are evaluated in the context of six different types of reactions which are either positive or negative. Parents helping children who experience negative emotions to feel better (emotion based reaction), helping children to express emotions and encouraging children (reactions take make emotion statement easy), and helping to solve the problems that causes emotional distress on children (problem based reactions) are regarded as positive emotional socialization behaviours. However, becoming anxious for negative emotions of children (problem of parent), underestimating problems and stressful conditions as emotional reactions of children are disregarded (insulting reactions), and oral or physical punishment to control negative emotions (punitive reactions) are considered as negative emotional socialization behaviour (Altan Aytun et al., 2012;Fabes, Eisenberg, & Bernzweig, 1990;Fabes, Poulin, Eisenberg, & Madden-Derdich, 2002;Yağmurlu & Altan, 2010).
Supportive parental characteristics allow children to be prepared to learn the thoughts, feelings, and behaviours of others in an emotional context in addition to attempting to constructively deal with emotions (Eisenberg et al., 1998). It is known that parents who give emotional and emotional reactions to emotions of children's emotions are important predictors of emotional competence and general social competence of pre-school children. In this context, it has been determined that children who are positively supported by their parents have more positive feelings towards their peers, whereas those who are not supported by their parents have a lower social competence (Denham et al., 1997). Positive parental behaviours of mothers have important effect on emotion regulation of children (Altan, 2006;Macklem, 2008) and ability to understand emotions (Denham, Zoller, & Couchoud, 1994). It is known that parents who apply an interactive method in expressing feelings tend to better regulate the emotions of their children's feelings and have no tendency for aggressive behaviour (Macklem, 2008).
Emotions are not only seen as pre-established, internal processes, but rather are influenced and shaped by the social and cultural environment of the individual (Kitayama & Markus, 1994). Culture plays an influential role in shaping the process of experiencing and expressing emotions (Le, Berenbaum, & Raghavan, 2002;Elfenbein & Ambady, 2002). Emotional socialization behaviours that play an important role in this process are an integral part of culture. Accordingly, the values attributed to emotions can change between cultures. For example, in some cultures, when anger is preferred compared to sadness, in another culture, anger may be attributed to feelings of shame, and parents may be motivated to focus on these emotions that are evident in their culture (Friedlmeier, Corapci, & Cole, 2011). Similarly, emotion regulation skills also emerge in a cultural context, with implications for persons and relationships in the social environment. (Butler, Lee, & Gross, 2007). According to Matsumoto (1989) although the emotions are biologically programmed, controlling the process of expressing and perceiving emotions is highly influenced by cultural factors. Another structure that is influenced by cultural factors is the ability to understand emotions. It is also known that when the emotions are expressed by members own cultural or subcultural group, it is unabled to provide a complete explanation with the confrontation as one of the methods of explaining the possible effect of the intragroup advantage (Elfenbein & Ambady, 2002).
When the related literature was investigated, there were researches on the emotion regulation skills of children and emotional socialising behaviour of parents (Denham et al., 1997;Eisenberg et al., 1998;Gottman, Katz, & Hooven, 1996;Seçer, 2017;Yağmurlu & Altan, 2010) and the ability of children to understand emotions and emotional socialising behaviour of parents (Denham et al., 1994;Garner, Jones, & Miner, 1994;Kılıç, 2012;Perez-Rivera, 2008;Song, 2005). However, there are limited studies in which emotion regulation, understanding emotions, and emotional socialization variables are examined together (Havighurst, Wilson, Harley, Prior, & Kehoe, 2010). When studies conducted in Turkish culture were examined, it was seen that there was no such investigation in this context. There is a need for the research to be done in this context. The findings obtained within the scope of the research will contribute to explaining the role of mothers' emotional socialization behaviors in children's skills to regulate and understand emotions. In this regard, it is thought that it will provide useful information in terms of possible intervention programs for parents to improve the quality of the parental child relationship and to improve the parents' emotional socialization behaviors. At the same time, the results obtained are valuable in terms of deepening the information available and enabling cultural comparability.
In this context, the aim of this study was to investigate the relationship between emotion regulation and understanding emotional skills of preschool children between the ages of 3-6 and emotional socialization behaviour of mothers. For this purpose, the following questions were evaluated: 1-Is emotional socialization behaviour of mothers a predictor of emotion regulation skills of children?
2-Is emotional socialization behaviour of mothers a predictor of the ability to understand emotions of children?

Research Model
This study investigated the predictor relationship between emotion regulation and understanding emotion skills of preschool children and emotional socialization behaviour of mother and adopted correlation type relational survey model to determine if variables changes together, and if there is a change, what is the degree and direction of this relationship (Büyüköztürk, Kılıç Çakmak, Akgün, Karadeniz, & Demirel, 2014;Karasar, 2012,)

Study Group
Since it was possible to observe diverse socio-economic (low-medium-high) levels between variables, the sample of this study was determined using maximum variety sampling method among purposeful sampling types for investigating the sample from a larger perspective. The formation of the working group was carried out in two stages. At the first stage, six independent preschools were set up in three different provinces of Adana to ensure that the participants varied by different socioeconomic levels. In assessing the socioeconomic level, Turkish Statistical Institute's (TÜİK, 2017) per capita monthly average consumption expenditure and parental education level report referenced. In this context, it was determined that all families in the lower socioeconomic level had less than 1642 TL of monthly average consumption expenditure determined by Turkish Statistical Institute and that 60% of their mothers were primary school graduates and 40% of them were junior high school graduates. It was determined that the families in the middle socioeconomic level had monthly income up to 4999 TL and 80% of the mothers were graduated from high school while 20% were graduated from high school. Monthly incomes of all the upper socioeconomic families were over 5000 TL and 83.3% of the mothers received education at the undergraduate level and 16.7% of the mothers were educated at the graduate level.
In the second stage, volunteer mothers and children in the lower, middle and upper socioeconomic levels of these six independent preschools were selected. Under this context, the sample group of the study consisted of 180 participants consisting of 90 mothers and children who attends six independent pre-school that are subjected to Adana Provincial Directorate for National Education. 52.2% (47) of the children were male and 47.8% (43) were female. The age distribution of the children was 11.1% three-years-old, 15.6% four-years-old, 32.2% five-years-old, and 41.1% six-years-old. The study environment, Turkey has Eastern and Western culture due to geographical location. In this context, the setting is important since it has rich culture were Eastern and Western culture are mixed.

Data Collection Tools
The data of the study were collected with the Coping with Negative Emotions of Children Scale, which is frequently used in the literature and has the appropriate / high level of psychometric properties to determine the emotional socialization behaviors of the mothers and The Wally Emotional Test was used to assess children's emotional understanding skills and finally Emotion Regulation Scale was used to measure children's emotion regulation skills.
Coping with Negative Emotions of Children Scale: In this study, maternal socialization of emotion was measured with the Coping with Children's Negative Emotions Scale (Fabes et al., 1990). This scale was developed by Fabes et al. (1990) to measure emotional socialising behaviour of parents. The scale consisted of 12 scenarios that indicated negative emotions of children such as anger, fear, and sadness, and six items (e.g; "I say it is natural to cry when you are unhappy to him", "If you do not care, it will be like that") regarding reactions of parents regarding these emotions. These above-mentioned six items corresponded to six sub-dimensions of the scale. The scale was revised by Fabes et al. (2002) and internal consistency coefficients were found as .85 for reactions that make emotion statement easy sub-scale, .80 for emotion based reaction sub-scale, .78 for problem based reactions sub-scale, .69 for punitive reaction sub-scale, .78 for insulting reaction sub-scale, and .70 for problem in parent sub-scale. The scale was adapted to Turkish culture by Altan Aytun et al. (2012). Internal consistency coefficients of Turkish form of the scale was found as . 87, .79, .72, .83, .86, and .65 respectively. Under this context, it is determined that scale could be in the study as reliable and valid measurement tool.
Emotion Regulation Scale: This scale was developed by Shields and Cicchetti (1997) to measure emotion regulation skills of children. This scale consisted of four-point Likert type with 24 items (e.g; "Easily disappointed and angry") and two sub-dimensions. Internal consistency coefficient of the original scale was .83 for emotion regulation sub-scale and .96 for changeability/negativity sub-scale. The scale was adopted to Turkish culture for preschool children by Yağmurlu and Altan (2010). The scale could be applied in two forms; parent and teacher form. The internal consistency coefficient for parent form was .75 and for teacher form was .84. In this research, emotion regulation skills of children were evaluated by their mothers using parent form.
Wally Emotions Test: This scale was developed by Carolyn Webster-Stratton under "Incredible Years" project to evaluate the ability of children to understand emotions. In the test, children were shown eight pictures with positive and negative events and were asked what they felt when the picture was shown (e.g; "imagine that the children in the picture are you and your friend, what do you feel when your friend breaks your special toy? What do your friends feel?") Children would receive 1 point for correct picture definitions and received no point when the pictures were incorrectly defined, or they were unable to answer. The scale was adapted to Turkish culture for children between 3-5 years old by Kayılı and Arı (2015). The reliability of the test was determined Kuder Richardson-20 (KR-20), two semitest and test-retest analysis. Under this context, KR-20 internal consistency was found .76 for three-yearold children, .84 for four year-old children, and .80 for five year-old children. The adaptation study of the test for children aged six years was conducted by Dereli (2008) and according to expert opinion, the scoring reliability coefficient was determined as .79. The test was determined as reliable and valid measurement tool.

Data Collection Process
The data of the research were collected between March-June 2017 in Adana province. Data collection processes had two stages. At the first stage, six independent preschools were identified in three different provinces of Adana to ensure that the participants varied in socioeconomic level. In the second stage, these six independent preschools were identified as volunteer mothers and children in the lower, middle, and higher socioeconomic level who are willing to participate the survey. The administrators and teachers of the six preschools were informed about the purpose of the research and approximately 600 permission forms, Personal Information Form, Coping with Negative Emotions of Children, and Emotion regulation Scales were delivered to the teachers to be sent to the parents. Among the returned forms and scales, 87 were found to be incorrect or incompletely filled, and the remaining 313 were evaluated for socioeconomic level.
Forms returning from the mothers are divided into three groups as socioeconomic level. It was determined that 54 of them belonged to the high, 113 belong to the lower and 146 belong to the middle socioeconomic level. It was found that only 32 mothers that in the returned forms approved the interview with their child from the group determined as high socioeconomic level. However, only 30 children could be reached at the high socioeconomic level because one of the identified children did not give verbal consent and the other gave up the interview. This situation has resulted in high socioeconomic level children play a decisive role in determining the number of samples. Accordingly, interviews were conducted with 90 children who were included in the sampling group as sub-upper and middle socio-economic levels to obtain emotion understanding scores. For this purpose, appointments were made for the appropriate day and time from the teachers of the children and interviews were conducted by the researchers. The child and the researcher were seated on a desk and the test materials used during the application were prepared. Wally Emotional Test duration varied between 8 and 12 minutes depending on the responses of the children.
During the current study, ethical principles were followed and maternal written consent was obtained through forms for interviews with children, but verbal consent of the children was requested at the beginning of each interview. Interviews were not held with children who did not want to participate. Parents were informed that the work to be done had only to be used for scientific purposes, explaining that the research only focuses on the process, not focusing on the person.

Data Analysis
SPSS 18.0 package program was used for analysis of research data. Multiple regression analysis was conducted to determine the extent to which the emotional socialization behaviours of mothers predicted emotion regulation and understanding emotions skills of preschool children. In this context, first, the assumptions of multiple regression analysis are tested. In this direction, an initial value analysis has been applied. The Mahalanobis distances were calculated at the extreme values and it was observed that these values changed between .04 and 1.00. Considering that the values accepted as the extreme values were below 0.01, it was clear that research data excluded extreme values. The normal distribution fitness of the research data was controlled for skewness and kurtosis coefficients and these coefficients were found between +2 and -2 in all variables. According to the obtained values, the distribution was considered as normal. Another assumption to perform a multiple regression analysis was that there are no multiple connection problems between predictor variables. Pearson correlation coefficients between the independent variables were examined and it was determined that the correlation coefficients between the variables varied between .16 and .56. According to Büyüköztürk (2012), the correlation over .80 may indicate multiple connections, whereas a correlation of .90 and above indicates a significant multiple connection. Another method for investigating multiple connections was using tolerance value and VIF value obtained during regression analysis. If tolerance value was smaller than .20, and VIF value was equal or higher than .10 indicates multiple connections (Büyüköztürk, 2012). Tolerance value in the research was calculated as minimum .37, and VIF value was calculated as maximum .2.68. Based on these results, it could be said that there was no multiple connection problem between variables.

Results
Regression analysis results for the relationship between emotional socialising behaviour of mothers (problem in parent, punitive reaction, insulting reaction, reactions that make emotion statement easy, emotion based reaction, problem based reaction) and emotion regulation skills of preschool children were indicated in Table 1. In Table 1, when multiple regression results between emotional socialization behaviours of mothers as predictive variables and emotion regulation skills of children as predicted variable were examined, it is clear that there was a low (r = .29) positive relationship between punitive responses and emotion regulation skills of children. When other variable was considered, this correlation was calculated as r=.21. It was seen that there was a low level (r = .20) positive relationship between insulting reactions and emotion regulation skills of children, but when the other variables were considered, it was seen that the relationship between the two variables was low and negative (r = -13). The correlation between reactions that make emotion statement easy and emotion regulation skills of children was positive and low (r = .20) while the relationship was also positive and low (r = .02) when other variables were considered. When the relationship between emotion based reactions and emotion regulation points were considered, the calculated correlation was positive and low (r=.19). However, when other variables were considered, this relationship was low but negative (r=-.04). It was seen that the relationship between the problem-based reactions as other predictor variable, and emotion regulation skills of children were positive and moderate (r = .32), but when the other variables were considered this coefficient changed to r = .23. The scores for problem in parent and emotion regulation skills of children showed a positive and moderate correlation (r = .40), while as the other variables were controlled, the relationship was found to be positive and low (r = .19).
The regression model indicated that emotion regulation skills of children form a moderate and meaningful relationship with punitive reactions, insulting reactions, emotion based reactions, problem based reactions, and problem of parents variables which were emotional socialization behaviours of mother (M = , R2 = 0,248, p & lt; 0.01). All modelled independent variables account for approximately 25% of the total variance of emotion regulation skills of children with dependent variables. This suggests that the 75% change in emotion regulation skills scores may be explained by different variables excluded from regression model. According to the standardized regression coefficients (β), the relative importance of predictive variables on emotion regulation skills of children were problem based reactions, punitive reactions, problem of parent, insulting reactions, emotion based reactions, and reactions that make the emotion statement easy. When the t-test results on the significance of the regression coefficients were examined, among predictive variables, punitive reactions (t=2.025, p<.05) and problem based reactions (t=2.155, p<.05) were important predictors of emotion regulation skills of children. Considering this positive relationship, it can be argued that the increase in punitive reactions and problem based reactions could positively affect the increase in emotion regulation skills of children.
Regression analysis results for the relationship between emotional socialising behaviour of mothers (problem in parent, punitive reaction, insulting reaction, reactions that make emotion statement easy, emotion based reaction, problem based reaction) and preschool children's skills to understand emotions were indicated in Table 2. As seen in Table 2, when multiple regression results between emotional socialization behaviours of mothers as predictive variables and understanding emotions skills of children as predicted variable were examined, it is clear that there was a low (r = -.18) negative relationship between punitive responses and understanding emotions skills of children. When other variable were considered, this correlation was calculated as r=.04. It was seen that there was a low level (r = -.25) negative relationship between insulting reactions and understanding emotions skills of children, but when the other variables were considered, it was seen that the relationship between the two variables was low and negative (r = -.17). The correlation between reactions that make emotion statement easy and understanding emotions skills of children was negative and low (r = -.03) while the relationship was also negative and low (r = -.00) when other variables were considered. When the relationship between emotion based reactions and understanding emotions skills were considered, the calculated correlation was positive and low (r=.01). However, when other variables were considered, this relationship was low but positive (r=.11). When the bilateral and partial correlations of the relationship between problem based reactions, which was another predictor variable, and understanding emotions skills of children were examined, it was calculated that both of the correlation coefficients were negative and low (r = -. 06). The scores for problem in parent and understanding emotions skills of children showed a negative and moderate correlation (r = .18), while as the other variables were controlled, the relationship was found to be negative and low (r = -.02).
Similarly, there was a low and insignificant relationship with the predictive variables such as punitive reaction, insulting reaction, reactions that make emotional statement easy, emotion based reaction, problem based reaction, and problem of parent variables (R = 0,285, R2 = 0,082, p>.01). All independent variables in the regression analysis account for approximately 8% of the total variance of emotion regulation skills of children with dependent variables. This suggests that the 92% change in understanding emotions scores may be explained by different variables excluded from regression model. However, the t-test results on the significance of the regression coefficients showed that the predictive variables that were modelled were insignificant predictors of ability of children to understand emotions.

Discussion and Conclusion
In this study, the objective was to contribute to the literature where there was a limited number of researches in the Turkish context by examining the predictive relationship between the emotional socialization behaviours of mothers and the emotion regulation and understanding emotional skills of preschool children.
The findings of the research indicated that positive (reactions that make emotional statements easy, emotion based reaction, problem based reaction,) and negative (punitive reaction, insulting reaction, problem of parent) reactions of emotional socialization behaviours of mothers explained approximately 25% of variance of emotion regulation skills of children. Among independent variables, problem based and punitive reactions of mothers that are included in emotional socialization behaviours was identified as significant predictors of emotion regulation skills of children. Accordingly, as problem based, and punitive reactions increased, emotion regulation skills of children also increased. Havighurst et al. (2010) stated about positive emotional socialization reactions of parents indicated that positive responses of parents contributed to understanding emotions and emotion regulation skills of children. Similarly, Gottman et al. (1997) indicated that families who can provide the best support for emotional coaching were trying to help and solve the problems of children to understand emotions, helping to express these emotions, and naming the emotions of children. In this sense, it is possible to say that problem based emotion socialising reactions of parents positively affected emotion regulation skills of children were in line with literature.
It is known that punitive or insulting applications of families had negative effects on socialemotional development of children. At the same time, these applications were associated with low emotional knowledge and low emotion regulation skills (Havighurst et al., 2010). Similarly, when the relevant literature was examined, it is indicated that effective and positive emotional socialization processes of parents contribute to the development of emotion regulation skills of children (Denham et al., 1997;Eisenberg et al., 1998;Gottman et al., 1996;Seçer, 2017). However, the findings in this study were different than common literature findings and stated that punitive emotional socialising reactions of mothers had positive affect on emotion regulation skills scores of children. The findings were similar to the findings of Jabeen, Anis-ul-Haque, and Riaz (2013). This study was applied in Pakistan and the results indicated that authoritarian mother attitudes had negative effects on emotion regulation skills of children. Similarly, the study of Rudy and Grusec (2006) indicated that authoritarian parenting is irrelevant from self-esteem of the children. In another study based on Asian culture stated that mothers commonly choose punitive emotional socialising reactions, and this could be appropriate for Indian forms of politeness (Raval, 2001). Based on this sample group with common grounds, positive relationship between punitive emotional socialising reactions of mothers and emotion regulation skills of children could be explained with cultural effect. Turkey is in the middle of individualist and collectivist societies as a position, which causes parenting styles to be influenced by both cultures (Selin, 2013). Göregenli (1997) stated that individualistic elements were observed in Turkey, but mostly values were emphasized for collectivism. It is known that negative and authoritarian parenting styles were more intense in communities where collectivism dominates (Goto et al., 2010;Papps, Walker, Trimboli, & Trimboli, 1995;Rudy & Grusec, 2006). Kağıtçıbaşı (1970) stated that the respect and control elements in the Turkish families were at a high level. Parentsattach importance to parental control because they prefer close relations rather than individualistic separation (Kağıtçıbaşı & Ataca, 2005). In traditional Turkish parents, parents used punitive practices as the most common control method (Sen, Yavuz Muren, & Yağmurlu, 2014;Yağmurlu & Sanson, 2009). In this context, the punitive emotions socialization can be used to motivate respect for authority. This is an important ethical and historical tradition for Turkey (Kağıtçıbaşı, 1970) and as a result, children have problems expressing and showing their emotions. It can be said that this situation is perceived as positive by the mothers and may have been interpreted as children can regulate their emotions. Thus, it is thought that the positive relationship between punitive emotional socialization reactions and emotion regulation can be explained. However, in addition to all that, objective of emotional socialization could be differentiated among families for emotional expression and emotion regulation with same culture based on the provider of socialiser. In this context, it may be difficult to reach consensus on which practices are positive about socialization and what the desired outcomes might be in this process (Eisenberg et al., 1998).
In the context of the research, emotion regulation skills of children were evaluated according to perceptions of their mothers. It is believed that the results would affect methodology of the study. Buck (1984) noted that a frightened child can learn to hide fear if that child is criticized by the parent for showing this feeling, but still have fear in himself/herself (as cited in Root & Denham, 2010). When this information is evaluated under current study, it is thought that mother could evaluate self-adjustment of children, but this evaluations could be influenced parent's positive perception regarding child and they were unable to reflect complete potential of their children.
Positive and negative emotional socialization reactions of mothers contributed to emotion regulation skill scores of children by 25%. This percentage indicates different variable that are effective on emotion regulation skills. When the fact that current study was based on evaluation of mothers, it is believed that fathers who could also affect emotion regulation skills of children could an effective factor. Actually, studies show that the relationship between father and child and the quality of attachment are related to emotion regulation skills of children (Diener, Mangelsdorf, McHale, & Frosch, 2002;McDowell, Kim, O'neil, & Parke, 2002). Accordingly, it is known that there are other factors that are effective on emotion regulation skills in terms of family context (Gross & Thompson, 2007;Morris, Silk, Steinberg, Myers, & Robinson, 2007;Volling, McElwain, Notaro, & Herrera, 2002). Emotion regulation skills of children could be affected from other factors than parents. The studies indicate that humour of children could be one of these variables and significant relationships were shown between humour and emotion regulation skills (Altan, 2006;Mangelsdorf, Shapiro, & Marzolf, 1995;Zimmermann & Stansbury, 2003).
Findings of this study indicated that emotional socialization behaviour of mothers are insignificant predictors of the ability to understand emotions in children. Accordingly, there were no significant relationship between positive and negative emotional socialization reaction of mothers and he ability to understand emotions in children. Findings show similarities with different researches in the literature (Kılıç, 2012;Song, 2005;Garner et al., 1994;Perez-Rivera, 2008). Yağmurlu, Çıtlak, Dost, and Leyendecker (2009) stated that information about parental socialization was based on findings from research conducted mostly with working groups in Western societies. When the cultural influence on socialization process (Brody, 2000;Quintana et al., 2006) was considered, the insignificant relationship between emotional socialization behaviours of mother and the ability of children to understand emotions could emphasise cultural differences. However, the studies that were different than current study in terms of understanding the emotions was often in Western culture (Bennett, Bendersky, & Lewis, 2005;Denham et al., 1994;Denham & Kochanoff, 2002;Dunsmore & Karn, 2001), and studies with similar results was often in Asian culture like Korea and China (Song, 2005;Wang, 2003).
In the study conducted by Bennett et al. (2005), when the positive maternal characteristics were considered completely, 48-month-old children were positively associated with emotional information. Based on this information, the non-significant finding between emotional socialization behaviours of mothers and the ability of children to understand emotions could be explained by that in this study only emotional socialization of mother was considered rather than other properties. In related literature, it is stated that there are significant relationship between attitude, experience, and application of parents' emotional socialization and children's understanding emotions (Edwards, Shipman, & Brown, 2005) and the importance of different emotions regarding beliefs of parents are indicated (Chaplin, Casey, Sinha, & Mayes, 2010;Dunsmore & Karn, 2001;Perez-Rivera, 2008). However, the meta-analysis study conducted to explain the relationship between the parents' styles of expressing emotion and the children's ability to express and understand emotions shows that positive emotional expression strategies of parents and children's emotion understanding were not related to all ages (Halberstadt & Eaton, 2002). According to Denham and Kochanoff (2002), mothers' emotional expression and behavior experiences form a foundation for their attitudes about emotion within childrearing. Accordingly, it has been determined that the attitudes of mothers about emotion education are strong predictors of emotional information of children in 3-year-olds. Within the same study, it was determined that the emotional information of children aged 3 and 4 was predicted by the positive responses and behaviors of the mothers to their emotions. When following research was considered, it is understood that age is an important factor in children's emotion understanding skill. 32.2% of the participant children were five-years-old and 41.1% were six-years-old at the time of the study. The mean age of the research sample can be correlated with the finding that emotional socialization behaviour of mothers do not predict children's emotion understanding.

Suggestions
Under the scope of current study, scores of children regarding emotion regulation skills were obtained by statements of parents. There is limited measurement tool that can be used for obtaining direct data from children in Turkey. Within the scope of the research Emotion Adjustment Scale (Batum & Yağmurlu, 2007) which was used frequently in the field and has high level of psychometric properties. In future researches, it is suggested that using direct measurement tools regarding emotion regulation skills of children could provide impartial evaluation.
In the study, the predictive impact of emotional socialization behaviours of mothers on emotion regulation and the ability to understand emotions of children was investigated and the parents were excluded as a participant in the study. In this context, it may be proposed that further studies with the data obtained from both parents could result with holistic perspective. Another suggestion for further research could be the socioeconomic level of the family and the age of the child as independent variables and to evaluate all the data together with advanced statistical methods. Thus, it is thought that comparative studies between previous studies and cultures can be made and important contributions to international literature can be presented.
Research data about emotional socialization behaviours of mothers and emotion regulation were gathered through forms i.e. directing the measurement tools to the families. Inability to use different data collection tools such as interviewing and observation in this study. In future studies, it is considered important that the collecting of data with multiple methods can achieve more comprehensive results.
In this study, the lack of a relationship between emotional socialization reactions of mothers and the ability of children to understand emotions suggests these two variables could be re-evaluated with different mediating variables could an important topic for further research. At this point, the use of alternative measurement tools in the data collection process is essential for timelessness.
In the context of the research, it was determined that emotional socialization behaviour of mothers was significant predictor of emotion regulation skills of children. In this regard, it is thought that it will provide useful information in terms of mothers to be aware of the potential impacts on their children's emotions and possible intervention programs for parents to improve the parents' emotional socialization behaviors. Through these programs, the mother-child relationship will be strengthened and in a sense it will be provide that the mothers fulfil emotional coaches of children successfully. However, considering that children's emotional development is shaped within the family climate, inclusion of their fathers in the intervention programs is thought to increase the positive effects.

Limitations
Another situation that can be considered as a limitation in this study was the rate of return of data collection tools. Only about 52% of the approximately 600 surveys returned. The surveys after excluding incorrect and incomplete forms resulted decreased the sample size below expected levels. However, mothers did not approve the interview with their child in 59% of useable forms which classified as high socioeconomic level. Accordingly, it is thought as an important situation limiting the sample size.