Relations between Parental Autonomy Support and Coercion with Children’s Total Difficulties

Child-rearing practices can either negatively or positively influence the psychosocial outcomes of their children. Negative psychological outcomes such as total difficulties can result in peer relationship problem, hyperactivity, emotional symptoms, and conduct problem among children. Thus, this study aims to examine the relations between parental autonomy support and coercion with total difficulties of children in Malaysia. As respondents of this study, a total of 502 children aged between 9 and 12 years were recruited using a multistage proportionate-to-size sampling technique. The descriptive analysis found that 8.4% of respondents experienced an abnormal level of total difficulties score that can result in mental disorder. The results indicated that children who perceived a higher level of autonomy support from parents experienced a lower level of total difficulties. In contrast, children who perceived their parents as coercive reported more total difficulties. The results indicated that autonomy support from parents is vital in protecting their children during middle childhood from behavioral, emotional, and peer relationship difficulties. Parental coercion tends to intensify total difficulties. Thus, to help their children in mitigating the performance of total difficulties, parents should be empowered with positive parenting skills. Hubungan antara Dukungan Otonomi Orangtua (Parental Autonomy Support), Koersif dan Kesulitan Total pada Anak


Introduction
The results from National Health Morbidity Survey stated that the prevalence of psychiatric morbidity among children and adolescents aged between 5 and 15 years have increased from 13. 0% in 1996to 20.3% in 2011(Ministry of Health, Malaysia, 2012. These figures indicate that more Malaysian children and adolescents are facing mental and psychological problem. A recent study in South-East Asia by Saputra, Yunibhand, and Sukratul (2017) draws our attention to the increases in mental health problems among schoolaged children. Mental health problems among schoolaged children are getting common over the years (Centers for Disease Control and Prevention, 2013). Also, parents' low awareness on the psychological distress of children (Schwartz, McFadyen-Ketchum, Dodge, Pettit, & Bates., 1998) may increase the severity of behavioral and psychological problems. A study suggested that a clinical evaluation should be provided for any child aged 4 to 18 years who had some behavioral problems so that early treatment or group parent-training programs can be administered to mitigate the effects of possible neurobehavioral disorder such as hyperactivity (Subcommittee on Attention-Deficit/Hyperactivity Disorder et al., 2011) or any psychological disorders. These psychosocial problems are often referred to the total difficulties that may be experienced by an individual, which include difficulties in handling behavior, emotion, and peer relationships (Goodman, 1997;White & Renk, 2012).
In this study, total difficulties reflect the internalizing and externalizing behaviors of an individual. Children who experience total difficulties have higher tendency to encounter mental health disorder including conduct disorder, hyperactivity disorder, emotional disorder, and peer problems (Goodman, 1997;Goodman, Ford, Simmons, Gatward, & Meltzer, 2000). A study conducted by Wolke, Woods, Bloomfield, and Karstadt (2000) showed that primary school students who involved in direct bullying had higher levels of total difficulties, conduct problems, hyperactivity, and peer problem. Moreover, another study also found that boys who have total difficulties, hyperactivity, and emotional problem are more likely to be the victim of bullying (Johnson et al., 2002). A longitudinal study (Fontaine, McCrory, Boivin, Moffitt, & Viding, 2011) found that conduct problems were correlated with problematic personality and negative psychosocial problems. In one longitudinal research conducted by Taylor, Chadwick, Heptinstall, and Danckaerts (1996) found that conduct problem and hyperactivity of young children contributed to negative developmental outcomes during adolescence such as antisocial behavior, social, and peer problem. Another longitudinal research on the developmental trajectories of conduct problems found that the onset of conduct problem during early childhood caused adjustment problem in adolescence, and hyperactivity, emotional difficulties, and peer problems also reflected conduct problems (Barker, Oliver, & Maughan, 2010). It explained the causal effect between early externalizing problems and later internalizing psychopathology. Children who exhibit total difficulties have a higher tendency to involve in juvenile delinquency and even in adult crimes later (Tanner-Smith et al., 2013). The occurrence of total difficulties in early and middle childhood may increase the vulnerability of children in the future. Thus, it is essential to examine the risks and protective factors for school-aged children to avoid total difficulties during middle childhood, given the potential continuity of total difficulties across the lifespan.
Empirical studies showed that children's perception on the quality of parenting influences their developmental outcomes in aspects of psychological and social behaviors (Respler-Helman, Mowder, Yasik, & Shamah, 2012;Saputra et al., 2017). Parents as primary socializing agents work as nurturer and supporter to mitigate total difficulties among children. Vansteenkiste and Ryan (2013) highlighted the need to investigate both parental autonomy support and coercion because the presence of parental autonomy support cannot be assumed as the absence of parental coercion. Thus, this study aimed to examine the relativistic concept of parental autonomy support and coercion in contributing to total difficulties among children. Autonomy support refers to the presence of encouragement and democracy toward children's preferences and views, whereas coercion is characterized by high demand, restrictions and power assertion by parents (Skinner, Johnson, & Synder, 2005). Children who subjectively feel autonomous, with parental support and guidance, are more likely to resolve the developmental crisis and avoid problematic behavior. In contrast, parenting practices that highly restrict children's preferences and often ignore their opinions can deprive children's opportunities to internalize self-discipline and master problem-solving skill (Soenens, Vansteenkiste & Van Petegem, 2015), which, in turn, may lead to interpersonal, behavioral, and emotional problems among children. Parental autonomy support has been identified as a major contributing factor for minimizing behavioral problem of children by encouraging their interests and appropriate freedom under parental guidance (Inguglia et al., 2016). Recent evidence suggests that maternal power assertion hinders emotional closeness between parents and child attributed to behavioral problems (Kim & Kochanska, 2015). A number of researchers have reported that children who perceived a higher level of autonomy support are less likely to experience total difficulties, whereas coercion as high controlling practices was positively associated with total difficulties in aspects of behaviors and emotions (Griffith & Grolnick, 2014;Joussemet, Mageau, July 2020½Vol. 24 ½ No. 1 & Koestner, 2014Muratori et al., 2017). Parenting practices in Asian context were affected by collectivistic values that emphasize child obedience and parental directiveness (Shwalb & Hossain, 2017). In nurturing their children, parental coercion and controlling are commonly performed by Asian parents than parental autonomy support (Supple, McCoy, & Wang, 2010;Supple & Small, 2006). However, a study on Asian family revealed that parental coercion or controlling are positively related to positive development outcomes such as academic motivation and psychological wellbeing (Supple & Cavanaugh, 2013). Hence, this study is useful in filling the research gap and answering the inconsistent findings on the relations of parental coercion, autonomy support, and total difficulties of children.
A large and growing body of literature (Fousiani, Dimitropoulou, Michaelides & Van Petegem, 2016;Harvey et al., 2016;Soenens et al., 2015) has investigated the effect of autonomy support and coercion on child development with the concept of self-determination theory (SDT). SDT (Deci & Ryan, 1985;Deci & Ryan, 2000) posited that a family structure that provides autonomy supportive and responsive but not controlling practices is ideal for the developmental needs of a child. Parental autonomy support, according to SDT, allows children to make decisions and express their feelings to achieve their developmental goals. Autonomy-supportive parents can recognize children's emotional state and provide a rationale for behavioral requests during the socialization process. Hence, children who received autonomy support tend to experience feelings of autonomy or volition in their behavior (Deci & Ryan, 2000) and are less likely to experience total difficulties (Matte-Gagné, Harvey, Stack, & Serbin, 2015). On the other hand, parental coercion, such as over-control and power assertion, tends to arouse the feeling of being coerced and controlled by parents among children. A coercive family environment that provides extremely obedience and no rationale for behavioral requests is more likely to hinder children's ability in problemsolving, maintenance of peer relationship, and emotional and behavioral regulation (McNamara, Selig, & Hawley, 2010). This study hypothesizes that it is more likely to mitigate children's total difficulties while they are receiving parental autonomy support based on the relativistic perspective on the roles of autonomy support and coercion in SDT. In contrast, perceived maternal coercion among children may associate with a higher level of total difficulties.
Because it is believed that mother as the primary caregiver who spends more time with children will bring stronger effect on children development than father, whose traditional role as breadwinner according to social norms, this study aims to study maternal coercion and autonomy support (Qin & Chang, 2013;Lamb, 2010). Mothers as primary socializing agent play a vital role in affecting psychosocial development of their children. Previous studies (Alink, Cicchetti, Kim, & Rogosch, 2009;Verschueren, Dossche, Marcoen, Bakermans-Kranenburg, & Mahieu, 2006) also focused on the maternal factors in predicting externalizing and internalizing behavior. Hence, this study aims to determine the correlation between maternal autonomy support and coercion with total difficulties among school-aged children in Malaysia. Previous studies (Harvey et al., 2016;Sharma & Sandhu, 2006) also found that the practices of parental autonomy support and coercion also varied by age of children. In addition, empirical studies (Currie et al., 2012;Richter et al, 2009;Meins, Centifanti, Fernyhough, & Fishburn, 2013;Van Roy, Groholt, Heyerdahl, & Clench-Aas, 2010) demonstrated that demographic background was significantly contributed to individual's psychosocial outcomes in aspects of behavior, emotion, and interpersonal relationships. Thus, the current study aimed to include age, gender, and family income as socio-economic status as controlled variables in the proposed multivariate model. In summary, this study aims to (1) determine the correlation between maternal autonomy support and coercion with total difficulties among school-aged children and (2) examine the predicting effects of maternal autonomy support and coercion on total difficulties among school-aged children by controlling demographic background (i.e., age, gender, and family income).

Procedure.
A standardized questionnaire was used to collect data from primary school students and their parents. The self-administered questionnaire was translated into Malay, Chinese, and Tamil languages. Permission was obtained before the data collection from the Ministry of Education (Malaysia) and Education Department of each state and respective headmasters of each school. Only one class was selected from each school, and all students completed the questionnaire in class settings. Prior to the data collection session, brief descriptions and explanation about this study were provided. The consent from the respondents was obtained by informing their right to withdraw from this research at any time. The anonymity of respondents and confidentiality of their responses were reassured. Bilingual or trilingual enumerators were also present to provide further assistance to students when necessary. Completed questionnaires were collected on the spot. Also, the respondents were asked to send the consent form, questionnaire, and research explanation sheet to their parents. The research explanation sheet provides the objectives of the research, permanent confidentiality, withdrawal right, and the name and contact number of researchers. Students returned the parental consent form and parent-report questionnaire to their class teachers.

Translation on instrumentation.
For particular instruments, translation job was approved by the corresponding author. Experts in the relevant field were employed to translate English version of the questionnaire into Malay, Chinese, and Tamil languages, and research members and Indian psychologist conducted the back translation. Back translations refer to one translating work from the source to target language and then the sentences being translated from the target language to source language. If the back-translated sentences are considered identical with the source, then the researchers can suggest that the translated version has similar or equivalent meaning to the source language. Besides, to ensure language clarity and maintain consistency between three ethnic groups, three focus groups were conducted to review each translated item. The project members discussed and reviewed the questionnaire prior to distributing the questionnaire.

Measures. Parenting dimensions.
Parental autonomy support and coercion were assessed with subscales from Children version of Parents as Social Context Questionnaire (Skinner et al., 2005). Four items in autonomy support subscales assess democracy and autonomy provided by mothers for children. For instance, "My mother let me do the things I think are important." Four items of coercive were used to measure strict control and power assertion of mothers. An example of the items is, "My mother thinks there is only one right way to do things-her way." Respondents were required to respond on a 4-point Likert scale (1 = not at all true, 2 = not very true, 3 = sort of true, and 4 = very true). Higher scores on particular dimension indicate the greater intensity of certain feature of parenting among mothers. Minimum scores are 4.00, whereas maximum scores are 16.00. The internal consistency of autonomy support and coercion in this study were 0.62 and 0.64, respectively.
Total difficulties. Child-report of Strength and Difficulties Questionnaire (Goodman, 1997) was used to assess total difficulties of children with a combination of four different aspects, namely, emotional symptoms, conduct problems, hyperactivity, and peer problems.
Children were asked to respond to five items for each subscale that describe their behavior over the last 6 months with 3-point Likert scale (0 = not true, 1 = somewhat true, 2 = certainly true). To assess emotional problems of children, emotional symptoms subscale was utilized (e.g., I worry a lot). Conduct problems subscale was employed to measure the misbehavior and temper of children (e.g., I get very angry and often lose my temper). Hyperactivity subscale assesses the lack of attention and restless among respondents (e.g., I am restless. I cannot stay till for long). To evaluate the negative relationship of children with their peer, peer problems subscale was used (e.g., other children pick on me or bully me). The total scores reflect total difficulties of children by summing up the scores of emotional symptoms, conduct problems, hyperactivity, and peer problems. In this study, minimum scores are zero, whereas maximum scores are 28.00. Higher scores represent higher tendency to experience total difficulties. The value of Cronbach's alpha for total difficulties is 0.75. Data analysis. Descriptive analysis was used in this study to describe the demographic profile of respondents, levels of children's perception of maternal autonomy support, coercion, and total difficulties among children. The relations between maternal autonomy support, coercion, and total difficulties were measured with bivariate correlation analysis. To determine the unique predictors for total difficulties behavior among children, multivariate regression analysis was used. In the proposed model, demographic background (i.e., age, the gender of children and average family monthly income) was controlled. Maternal autonomy support and coercion as predicting factors were sequentially entered in multiple regression models. The distribution of all of the variables has met the assumptions for regression analysis. Normality, linearity, and homoscedasticity of residuals were implemented prior to conducting multivariate analysis, and all of the variables were found to be normally distributed. There was no multicollinearity detected with the acceptable range of tolerance (>0.20) and variance inflation factor (<5) values.

Results
Descriptive and correlation analysis. Table 1 shows the results of descriptive and correlation analysis. The mean scores of parental autonomy support were higher than the mean scores of coercion scale. The descriptive results implied that children tend to receive a high level of maternal autonomy support and an average level of coercive practices.
Children who scored 0 to 15 in Strength and Difficulties Questionnaire, according to the scoring suggested by Goodman (1997), are considered as the normal group, which is a group of people less likely to experience psychological problem. The range within 16 to 19 marks refers to borderline, whereas children who scored 20 and above are categorized into the abnormal group, which is in contact with mental health problems. The descriptive analysis reported that majority of children in the current study (n = 401, 79.9%) were categorized into the normal group, whereas 59 (11.7%) and 42 (8.4%) children were grouped into borderline and abnormal groups, respectively. The community sample can be considered as normal if there is less than 10% of sample size scores in abnormal band, according to the benchmark from the community-wide deployment of SDQ-based screening (Goodman, 1997;Goodman et al., 2000). In this study, the SDQ-scores distribution thus detected a low risk of mental health problems among the study respondents. The results can also be generalized to Malaysian children aged between 9 and 12 years.
Correlation analysis reported that autonomy support (r = −0.249, p < 0.001) was negatively related to total difficulties. Children who perceived a higher level of maternal autonomy support was associated with a lower level of total difficulties. The results also found that there was a significant negative relationship between maternal coercion (r = 0.328, p < 0.001) and total difficulties among children. The result reveals that children who perceived their mothers as coercive are more likely to experience total difficulties in aspects of emotional symptoms, peer problem, hyperactive, and conduct problems.
Regression analysis. Demographic background (age, the gender of children, and total family monthly income) in this regression model was entered in Step 1 as controlled variables, followed by Step 2 that entered maternal autonomy support and coercion as predictors to total difficulties. In Step 1, demographic background (i.e., age, the gender of children, and total family monthly income) only explained 1.1% of the variability of total difficulties. In Step 2, there was a significant increment in variance amount (R² = 0.155). The inclusion of demographic background and maternal autonomy support and coercion explained 15.5% of the variability of total difficulties. Maternal autonomy support (β = −0.205, p < 0.001) based on the standardized regression coefficients negatively predicted total difficulties, whereas maternal coercion (β = 0.297, p < .001) positively contributed to a higher level of total difficulties among children. The results indicated that both maternal autonomy support and coercion were robust predictors of total difficulties. Children who perceived a higher level of autonomy support from mothers were less likely to experience total difficulties. On the other hand, children who perceived a higher level of maternal coercion experience a greater level of total difficulties (Table 2).

Discussion
The results from Pearson's correlation analysis reported that children's perceived autonomy support from mothers were negatively related to their total difficulties in aspects of emotional symptoms, peer problem, hyperactive, and conduct problems. The findings in this study were consistent with a previous study (Inguglia et al., 2016), which found the negative association between parental autonomy support and total difficulties. Correlation analysis also found that the higher level of parental coercion was correlated with a higher level of total difficulties. Previous studies (Kim & Kochanska, 2015;Verschueren et al., 2006) found that highly directive mothers bring negative impact on children's compliance behavior as compared with autonomous mothers. In this study, parental autonomy support and coercion were correlated with total difficulties that can result in substantial mental health disorders. A study by Griffith and Grolnick (2014) also reached similar findings in which reported that parental autonomy support had a positive correlation with positive developmental outcomes and negative relations with depression, while parental coercion was linked with a higher level of negative outcomes and depression. The first step of hierarchical regression analysis prior to including the effect of parental factors reported that demographic background (age, gender, and family income) failed to predict total difficulties among schoolaged children significantly. Previous study (Soenen et al., 2007) which focused on adolescents found that the effect of age on psychosocial outcomes by explaining adolescents' independence-seeking attribute. The finding in this study differs from that the age of the respondents (aged 9 to 12 years) was no significant predictor of psychosocial outcomes. The non-significant predicting effect may due to the small age range comparison in this study. Previous studies (Currie et al., 2012;Frick, 2016) discussed that boys are more likely to engage in externalizing problem (e.g., conduct and behavioral problems), whereas girls are more likely to experience emotional or social relationship problems. However, difficulties behavior in this study does not vary by the gender of the respondents. As compared with a previous study that measured externalizing and internalizing problems separately, this study assesses total difficulties in a general way by combining the externalizing (i.e., conduct and hyperactivity) and internalizing problems (i.e., emotional symptoms and peer problem) as one scale. Thus, the gender difference in total difficulties was failed to be addressed. The results also showed that total difficulties were not directly caused by family income. The previous study found that family wealth leads to physical health (Currie et al., 2012). It implied that family income might be a predictor of physical health, but it is not a psychosocial development indicator.
Regression analysis found the significant predicting effects of maternal autonomy support and coercion on total difficulties after being controlled for respondents' age, sex, and family income. The results of this study indicated that children who perceived their mothers as autonomy supportive and responsive are more likely to experience fewer total difficulties. Children are less likely to experience total difficulties, with the encouragement and rationale requests from autonomy-supportive caregivers. The results also revealed that children who perceived their caregiver as coercive tend to engage in a higher level of total difficulties. The results of this study further support a local study conducted by Tan (2012), who discussed that parental control is a significant predictor of behavioral and emotional disorders among children in Malaysia. Maternal coercion that offers high control and requests for strict obedience was related to total difficulties of children. Previous research (Mabbe, Soenens, Vansteenkiste, & Van Leeuwen, 2015) has discussed that children who received power assertion and high demand from parents are more likely to engage severe difficulties by performing non-compliance and rebellious behavior. These findings were supported by past studies (Inguglia et al., 2016;Joussemet et al., 2014) that revealed that maternal autonomy support mitigates total difficulties, whereas the perception on parental coercion contributes to the higher level of total difficulties among children. However, this finding is inconsistent with the previous study (Kunz & Grych, 2013), which has reported that maternal autonomy support only negatively predicted externalizing behavior of children but not internalizing behavior. In contrast with the finding of the current study, Kunz and Grych (2013) found that maternal coercion that perceived as high controlling practices failed to predict internalizing and externalizing behaviors among children.
The results of this study are inconsistent with some studies conducted in Asia (Chuang & Su, 2009;Eisenberg, Liew & Sri Untari, 2001). Given the consideration of cultural and hierarchical norms in Malaysia, parental coercion might be an appropriate parenting behavior to promote compliance and may not necessarily contribute to total difficulties among children. However, the results of this study revealed that parental autonomy support is useful in alleviating total difficulties, and the presence of parental coercion leads to a higher level of total difficulties. Traditional parenting practices such as parental coercion may not be compatible with contemporary families, given the trend of positive and warm parenting.
With the support of SDT, the findings identified that parental autonomy support as positive parenting attenuates the occurrence of total difficulties, whereas parental coercion that undermines children's autonomous ability is more likely to cause severe total difficulties. In SDT, autonomy-supportive mothers offer choices to children and acknowledge their feelings along the development process; in contrast, coercive parents that associated with autonomy need frustration and pressure lead to higher level of reactance among children as well as feelings to compel strict rule setting by parents (Grusec, Danyliuk, Kil, & O'Neill, 2017). In short, children who subjectively feel autonomous tend to internalize parental support and guidance and avoid to engage in total difficulties. However, children who received coercive parenting are more likely to conduct externalizing behavior. On the other hand, extreme compliant children who suppress their feelings and lose their initiatives tend to engage in internalizing behavior such as emotional symptoms and peer problems. The results of this study supported the propositions of SDT that supports from caregivers and primary socialization agents are important to positive child development and prevent children in performing problematic behavior, whereas parental coercion can deteriorate well-being and increase total difficulties of children.

Conclusion
In predicting total difficulties among school-aged children, the current study highlighted the important role of maternal autonomy support and coercion. Children reported having a lower rate of experiencing total difficulties by receiving a higher level of autonomy support from mothers. The higher level of maternal coercion caused a higher risk of undergoing total difficulties among children. As supported by SDT, mother as the main caregiver for school-going children are highly correlated with their development of psychosocial outcomes. Children with autonomous mothers can be intrinsically motivated to mitigate the involvement of total difficulties. Children who received coercive practices tend to have a higher risk of performing externalizing and internalizing difficulties. These findings provide insight into the effect of autonomy support and coercion on negative psychosocial outcomes among children. In a nutshell, supportive parenting mitigates the engagement in total difficulties, whereas coercive behavior from mother leads to a higher level of total difficulties.
The findings of this study can increase the awareness of various parties on the important role of maternal autonomy support and the negative effect of maternal coercion on child development. Thus, parents must empower their positive parenting practices (autonomy support) and avoid negative parenting practices (coercion) to nurture a child with fewer total difficulties. The findings of the current study also contribute to current literature regarding the effect of autonomy support and coercion. Previous studies in Asia context (Ang & Goh, 2006;Chao, 1994;Chuang & Su, 2009) found that authoritarian, which reflects coercion and high control, is effective parenting practices to nurture a compliant and successful child. In contrast, this study revealed that parental coercion leads to a higher level of total difficulties.
There are some limitations to the study. First, this study is a cross-sectional study in which data collection was conducted only at a one-time point. To determine the exact causality direction, future research is suggested to conduct a longitudinal study. Second, this study collected data with the self-administered method. Future study can conduct an interview session or mixed-method to avoid social bias. Although this study provides important insights into the concept of SDT in explaining the direct predicting effect of parental autonomy support and coercion on total difficulties, a full discussion of SDT, however, lies beyond the scope of this study. Future research can include more children's psychological needs variables to provide a more comprehensive chaining and sequential explanation on SDT parenting process.