Departman of Primary Education Malatya, Turkey ============================================================================== INVESTIGATING THE CORRELATION OF PARENTAL ATTITUDES WITH EATING BEHAVIOURS OF CHILDREN IN THE EARLY CHILDHOOD PERIOD

The aim of this study was to determine the effect of parental attitudes on the eating behaviors of children attending kindergartens. The population of the study consisted of the parents of children attending independent kindergartens located in Battalgazi and Yesilyurt districts of Malatya province in 2018-2019 school year. The sample of the study was composed of the parents of 284 children attending independent kindergartens affiliated with Battalgazi and Yesilyurt Municipalities in accordance with the sample calculation made over the population. “Parenting Attitude Scale” and “Children’s Eating Behavior Questionnaire” were used to collect the data. According to the results of the normality test and test of homogeneity conducted in the data analysis, while One-way ANOVA and t-test were used for the normally distributed values, Kruskal Wallis H test and Mann Whitney U test were used for the values which were not normally distributed. As a result of the study, no significant difference (p<.05) was observed in the eating behaviors of children in terms of their gender and age. It was observed that authoritarian attitude behavior increased with increasing number of children but decreased with the higher education level of fathers, parents with a single child exhibited the democratic attitude, overprotective attitude and behavior decreased and permissive attitude and behavior increased as the education level of fathers increased. In addition, a positive difference was observed between the authoritarian and permissive attitudes of parents and the eating behaviors of children.


INTRODUCTION
Children need regular and balanced nutrition in order to maintain their development in a healthy way from the moment they are born. While these needs are met by the adults around them, primarily by their parents, it is important that they become aware of this subject in the following period. The problems experienced in nutrition lead to health problems such as obesity and are a factor affecting the development negatively in general. We are expected to support children about healthy eating behavior from the early period and this issue needs to be a part of the education process (MAGEP, 2007;Kutlu and Civi, 2009;Kobak and Pek, 2015). İt is seen that gains and indicators for children to acquire healthy eating habit are involved in preschool education program (Aral et al., 2011). The studies have shown that eating related problems have increased recently and especially weight gain and obesity have appeared as an important issue in developed and developing countries. Changing lifestyles have also affected eating habits and it is stated that there have been significant increases in the consumption of ready and artificial foods, thus affecting children too much. This change in eating habits has caused weight problems in children and approximately one third of children experiencing this problem in the early childhood period were determined to have this same problem in adulthood (Zametkin, et al., 2004;Uskun et al., 2005;Anonymous, 2010). It is expressed that the transition to supplementary food from the use of breast milk in infants significantly affect the obesity risk in the early childhood period (Okdemir and Esen, 2018). These data show that overweight and obesity have become an increasing issue among children. In addition, it is also known that some parents feed the children inadequately and turn the feeding action into a crisis. Eating behavior which has become a serious problem in children is shaped starting from the early childhood period. This formation process is affected by the family life of the child, attitudes of the parents and the education process if the child is attending a school.
Malnutrition is reported to affect different parts of society, primarily pregnant women and children aged 0-6 years (Arli et al., 2017). The studies on nutrition have reported that nutritional problems have increased 10 times compared to the 1970s and about 10% of children in the world will face these problems in the near future. The World Health Organization states that the number of obese people in the world increased from 400 million to 700 million people from 2008 to 2015 and the number of overweight individuals increased from 1.4 billion to 2.3 billion people. In the USA, it was determined that 16.3% of the children aged between 2-19 years were obese between 2003-2006 and these rates did not change in 2011-2012 period. While the obesity prevalence in the children aged between 0-5 years in Turkey in 2010 was determined as 10.1% in boys and 6.8% in girls, 26.4% of the children were determined to be overweight and obese. According to the OECD's 2017 report, it was determined that one out of every six children living in these countries was obese or overweight and one in two adults living in OECD countries was overweight or obese. In the studies covering Asia, Africa and Europe, it has been reported that the obesity prevalence increased by 30% in the last decade (Demirezen and Cosansu, 2005;Uskun et al., 2005;Ogden et al., 2014;Tedik, 2017). These data show that eating behavior in children has become an important problem. The eating problem experienced by children can be considered not only as eating too much but also as eating less, resisting to eat or having a conflict with the parent in the eating action.
As in many aspects, the development of eating behaviors in children are shaped during the early childhood period. Obesity is reported to be seen most commonly in the first year of life, 5-6 years of age and the puberty period (Koksal and Ozel Gokmen, 2008). In this shaping process, different factors such as physical activity, duration of watching TV, sleep habits, preventive health services, newborn nutrition and eating habits gained in the childhood period are effective and especially the family communication and parental attitudes directly affect children. Parental attitudes defined as a phenomenon involving belief, value, attitude and behaviors of parents about their children are among the most important factors determining the family communication. The effects of parents are seen as the root of many behaviors of children (Kurtulmus and Temel 2003;Palut, 2008;Tedik, 2017;Okdemir and Esen, 2018). Being a parent is a task which implies a lot of responsibilities (Thayer and Zimmerman, 2003). Parents should be aware of this and adjust their attitudes accordingly (Palut, 2008). The family type in which children are raised affects both their personalities and social behaviors (Yavuzer, 1993). Raising children who are socially adaptive to the environment and can answer social expectations is directly associated with parental attitudes and it is emphasized that the communication of children, modelling their parents' attitudes and behaviors, with third parties is directly affected by these attitudes (Aksoy, 2016). Eating behavior style in children is reported to be completed at the end of the preschool period (Ashcroft et al., 2008). Besides, parents are in an important position in the healthy nutrition of children and their awareness of nutrition (Koksal and Ozel Gokmen, 2008). Family structure and family habits are shown to be among the factors affecting the obesity in children. In addition, parents' nutrition attitudes, their income status and mothers' working status are reported to be significant factors in the emergence of nutrition problems in children. Furthermore, the importance of parental attitude and cooperation in the treatment of obesity is emphasized (Pyle et al., 2006;Chan, 2008;Ergul and Kalkim, 2011;Ogden et al., 2014). Based on this information, this study was conducted to examine the correlation between parental attitudes and the eating habits of children.

Research Model:
In this study, the aim was to determine the effect of parental attitudes of children attending kindergartens on the children' eating behaviors. The study was designed in a descriptive survey model aiming to define the situation as it was presented.

Study Group:
The population of the study was composed of the parents of children attending independent kindergartens located in Battalgazi and Yesilyurt districts of Malatya province in 2018-2019 school year. The sample of the study included the parents of 284 children attending independent kindergartens affiliated with Battalgazi and Yesilyurt Municipalities in accordance with the sample calculation conducted over the population.

Data Collection Tools
In the study, "General Information Form" for obtaining information about the children and their parents, "Parenting Attitude Scale" (ETO) developed by Karabulut Demir and Sendil (2008) for determining parental attitudes, "Children's Eating Behavior Questionnaire" adapted to Turkish by Yilmaz, Esmeray and Erkorkmaz (2011) for determining the eating behaviors of children were used. All of the data collection tools were filled out by the parents of the children. The General Information Form includes questions about the child's gender, chronic illness, height and weight, gender of the parent filling out the form, child's age, birth order, number of children, family income level, parents' working status, and education level.
"Parenting Attitude Scale (PAS)" which was developed by Demir Karabulut and Sendil (2008) is composed of 46 items and four subscales including democratic attitude (17 items), authoritarian attitude (11 items), overprotective attitude (9 items) and permissive attitude (9 items). Parents are determined to have the attitude of the subscale from which the parents obtain a high score.  (2011) and was used to determine children's appetite. Children's Eating Questionnaire is composed of 35 items and eight subscales including responsiveness to food (5 item), enjoyment of food (5 item), satiety responsiveness (7 item), slowness in eating(4 item), fussiness (3 item), emotional overeating (4 item), emotional undereating (4 item), and desire for drinks (3 items). As a result of the reliability analyses, the Cronbach's alpha values were between 0.61-0.84. Reliability coefficient of the whole scale was calculated 0.69.

Data Collection Method:
At the data collection stage, preschool teachers or researchers delivered the Parenting Attitude Scale and Children's Eating Behavior Questionnaire to the parents and they were collected in the same way.

Data Analysis:
According to results of the normality test and homogeneity test conducted in data analysis, One Way ANOVA and t test were used for normally distributed data and Kruskal Wallis H test and Mann Whitney U test were used for data which were not normally distributed. When significant differences were observed between the groups, the cause of the differences was determined by using Tukey test for those showing normal distribution and by using Dunnett's C test for those without normal distribution from Post Hoc Multiple comparison tests. Pearson Correlation Coefficient was used in correlation analysis in examining the correlations between the scales as well as the correlations of the overall scales with their subscales. While interpreting the results obtained from the data analysis, the significance level was used as 0.05. Table 3 and Table 4 show the results indicating the correlation between some demographic characteristics of the children and their total scores of Children's Eating Behavior Questionnaire (CEBQ). According to Table 3, there was no significant correlation between CEBQ total score of children and their birth order, income levels, parental education status and number of children (p>.05). There was a significant correlation between the CEBQ total scores and ages of the children (p<.05). When the cause of the difference was examined, it was seen that the CEBQ scores of 4-year-old children were higher than those of 5-year-olds. According to Table 4, there was no significant correlation between CEBQ total scores of children and chronic illness (p>.05). There was a significant correlation between CEBQ total scores and the children' gender (p<.05). CEBQ scores of girls were higher than CEBQ scores of boys.

RESULTS
Tables 5, 6, 7, 8 and 9 show the results showing the correlation between some demographic characteristics of the parents and the total scores of the subscales of the Parenting Attitude Scale (PAS). When Table 5 was examined, it was determined that there was a significant correlation between the parents' scores of democratic attitude subscale and number of children and birth order (p<.05). When the cause of the difference was examined, it was observed that the scores obtained by parents with a single child from democratic attitude subscale were higher than the scores of the parents with 2 children and the total scores obtained by the parents having the first child from democratic attitude subscale were higher than the scores of parents who have the last child. When Table 6 was examined, it was determined that there was a significant difference between the authoritarian attitude total scores of the parents and the number of children and father's education level (p<.05). When the cause of the difference was examined, it was observed that authoritarian attitude scores of the parents who have 2 children were higher than the parents who have a single child and authoritarian attitude scores of the parents who have 4 and more children were higher than the scores of parents who have a single child. The authoritarian attitude total scores of the fathers who have the education level of primary school were observed to be higher than the scores of fathers who have the education levels of secondary school, high school and undergraduate.  Table 7 was examined, it was determined that there was a significant difference between the overprotective attitude total scores of the parents and the income level and parental education levels (p<.05). When the cause of the difference was examined, it was observed that the overprotective attitude total scores of the families with a low income level were higher than the overprotective attitude scores of the families with a high income level, overprotective attitude scores of the mothers with primary school, secondary school and high school education level were higher than the mothers with undergraduate education level, overprotective attitude scores of the fathers with primary school, secondary school, high school and associate degree education level were higher than the fathers with undergraduate education level, and the scores of fathers with secondary school, high school and associate degree education level were higher compared to the fathers with graduate education level. When Table 8 was examined, it was observed that there was a significant difference between the permissive attitude subscale of parents and the number of children and father's education levels (p<.05). When the cause of the difference was examined, it was observed that the permissive attitude total scores of the parents who have oblz one child were higher than the total scores of the parents who have 3 children and the permissive attitude total scores of fathers with graduate degrees were higher than the fathers with secondary school and high school education levels. When Table 9 was examined, it was determined that there was no significant difference between the authoritarian attitude total scores of the mothers and their working status (p>.05). There was a significant difference between the mothers' working status and their total scores of overprotective attitude and permissive attitude subscales (p<.05). While this difference was in favor of the unemployed mothers for overprotective attitude subscale, it was in favor of the employed mothers in permissive attitude subscale. Table 10 shows correlation analysis results about the correlation between the Children's Eating Behavior Questionnaire scores and the Parenting Attitude Scale scores of the parents. According to Table 10, a positive (r: 0.144) significant (p<.05) but not very strong correlation was observed between the authoritarian attitudes of parents and the eating behaviors of the children. Additionally, a positive (r: 0.118) significant (p<.05) but not very strong correlation was observed between the permissive attitudes of parents and the eating behaviors of the children

DISCUSSION, CONCLUSION, and RECOMMENDATIONS
In this study examining the correlation between the parenteral attitudes of children in the early childhood period and the children's eating behaviors, it was determined that there was a statistically significant correlation between the parents' authoritarian and permissive attitude behaviors and the children's eating behaviors and as the authoritarian and permissive attitude behaviors of the parents increased, the eating behaviors of the children increased. Parents play an important role in determining the children's diet. Whether or not parents allow to determine which foods are appropriate for children and how much it will be given to the child, how much of which food will be eaten, help children to transition to self-feeding by establishing behavioral boundaries (Cited by Unlu et al., 2006). Eating is considered as an indicator of the correlation between parent and child (Satter, 1986). Besides, it is stated that the eating behaviors and problems in children are generally affected by parents' knowledge and attitudes about nutrition (Unvar and Unisan, 2005;Derman and Basal, 2013).
When the literature is examined, it is stated that authoritarian, repressive and controlling diet styles and behaviors of the adults exhibiting the authoritarian attitude negatively affect the fullness and hunger perception development by the children (Faith and Kerns, 2005;Snoek, et al., 2007). The studies have stated that children's eating behavior may change as a response to authority and oppression, willingness to restricted food may increase and the desire to eat healthy food may decrease (Fisher and Birch, 1999;Birch and Fisher, 2000).
Eating style of the families exhibiting the permissive attitude is also tolerant and permissive. When the literature is examined, it is stated that the children of parents who have tolerant eating style show more eating behaviors and are prone to obesity. The fact that these parents show high level of interest towards the child and consider the emotion of the child a lot and are unable to control their children ensures that they cannot establish a proper bond with their children (Hennessy, et al., 2010;Olvera and Power, 2010). In the studies, it is stated that parents with a tolerant and permissive eating attitude gave less negative feedback during the meal, their children were more appetent and the children had higher body mass indices (Hughes et al., 2011;Olvera and Power, 2010;Erkorkmaz et al., 2013). These studies in the literature support the results of the present study.
When the results were examined, it was determined that there was no significant correlation between the children's birth order, parents' education level, number of children, and income level of the family and the eating behaviors of the children. When the literature was examined, it was determined that the variables such as mothers' age, education level, number of children and working status affected the nutrition styles. It was stated in the studies that young mothers with low education level got higher rates of feeding their children when they were unhappy, sad and restless and higher scores in case of giving the food their children wanted as a reward if they ate the food they did not want (Muslu et al., 2014;Wardle et al., 2002;Vereecken et al., 2009;. Sleddens et al., 2010, Saxton et al., 2009. Some studies reported that the number of children that parents had did not affect their nutrition style (Kroller et al., 2008;Wardle et al., 2002). This was associated with the fact that the number of children that parents have did not chane the eating behavior of children because of not affecting their eating style.
In the study, the eating behaviors of four-year-old children were determined to be higher compared to the five-year-old children. This can be associated with the fact that 5-year-old children had increase awareness in general meaning compared to 4-year-old group and their self-control ability increased. It was determined in the study that the eating behaviors of girls were higher compared to boys. The studies conducted with the children in different age groups at school age and in adolescent period revealed that while no difference was seen between eating behaviors in healthy children in terms of gender, enjoying the food was more in boys diagnosed with obesity but emotional less eating behaviors were more in girls (Ozer et al., 2014). However, no study was found in the literature addressing the eating behaviors of preschool children in terms of age and gender. This was associated with the fact that in the society parents have a more protective approach towards boys causing boys to be less developed than girls in the preschool period. Additionally, the fact that girls had more eating attempts was associated with the fact that their eating attempts were more than boys.
In the study, it was determined that the parents with one child displayed a more democratic attitude compared to those with two children and first children faced with more democratic parental attitude compared to the last children and as the number of children increased, permissive attitudes of the parents decreased. As the number of children increased, the parents exhibited a more authoritarian attitude. When the number of children increased, the parents might have exhibited authoritarian attitudes to control them and reduce unwanted behaviors. This may have caused the interest in the first child to be more, the burnout of the parents to increase as the number of children increased, the patient level to decrease and correspondingly the parents' attitudes to be more strict-rigid, thus leading to a decrease in democratic attitude behaviors. Similar to these results, it was stated in the studies by Guneysu (1982), Karadeniz (1994), Ozcan (1996), Er Gazeloglu (2000), Ozyurek, Tezel Sahin (2005), Sak et al., (2015), Ersoy (2013), Inci and Deniz (2015) that as the number of children increased, authoritarian attitude behavior of the parents increased.
It was determined that as the father's education level increased, the authoritarian attitude decreased and permissive attitude increased; as the education level of parents increased, overprotective attitude behaviors decreased. When the literature was examined, it was stated that as the education level of parents increased, overprotective attitude decreased, attitudes towards pressure and discipline decreased, children's academic successes increased, and the rate of democratic attitudes of parents increased (Ari, Bayhan and Artan, 1995;Ozcan, 1996;Er Gazeloglu, 2000;Sendogdu;2000;Ozyurek, 2004;Uzun and Baran, 2019).
It was stated that the unemployed mothers exhibited a more overprotective attitude than employed mothers. It was determined that employed mothers displayed a more permissive attitude behavior than unemployed mothers. When the literature was examined, it was determined that especially housewife mothers had more protective and repressive attitudes than mothers who had occupations (Demiriz and Ogretir, 2007;Sendogdu, 2000;Ozyurek and Tezel Sahin, 2005). The limitation of the study is that the data were collected by parents and teachers.
In accordance with these results; Considering that the parental attitudes are effective on children's eating behaviors in the early period as well as in many other areas, it would be helpful to provide parent training in this regard.
Considering the effect of parental attitudes on children, it can be ensured that parents receive training to improve their related awareness and to regulate their behaviors. The awareness of the preschool teachers about guiding the children and families about children's eating behavior and parental attitudes can be increased. It would be useful for further studies to investigate the correlation between eating behavior and parental attitudes over different variables. Developing different assessment instruments to measure the eating behaviors of children and parental attitudes would also contribute to the field. It is also important to conduct longitudinal studies examining the eating behaviors of children in terms of age and gender. It will be useful for future studies to determine the factors affecting the eating behaviors in order to prevent obesity.

Biographical notes:
Merve Unal completed her bachelor's degree at the Inonu University Faculty of Education in 2003, master degree on preschool education at the Inonu University in 2006 and PhD degree on child development and education at the Ankara University in 2014. She works areas are early childhood education, science and math education in early childhood, problem solving skills in early childhood. She has been working as a lecturer at Inonu University since 2004.
Mehmet Saglam completed his bachelor's degree at the Inonu University Faculty of Education in 2004, master degree on child development and education at the Ankara University in 2011 and PhD degree on child development and education at the Ankara University in 2015. He works areas are developmental psychology, early childhood education, socially disadvantaged children. He has been working as a lecturer at Inonu University since 2015.