How to improve parenting in Hong Kong by training: the 6As Positive Parenting Program

Purpose – Research suggests that children in Hong Kong are at an elevated risk of emotional problems. Authoritarian parenting, a common parenting style in Hong Kong, is a critical factor associated with childhood mental health problems. The purpose of this paper is to evaluate the effectiveness of the 6As Positive Parenting Program (6As) in modifying parenting attitudes, reduction of parenting stress and increasing self-efficacy in positive parenting, among a sample of 82 Hong Kong parents. 6As focuses on prevention by instilling positive parenting beliefs and principles, thereby reducing reliance on authoritarian and related parenting styles. Design/methodology/approach – The program was evaluated using a controlled pre-post-treatment design. Outcome measures were the Chinese Child-rearing Beliefs Questionnaire, Parental Stress Scale, Parenting Self-efficacy Scale and participant feedback. ANOVA and correlation were utilized to detect treatment effects and relationships between the degree of change among measures and subscales. Findings – Relative to the control group, the 6As Positive Parenting Program significantly changed parents ’ parenting attitudes, reduced parenting stress and increased self-efficacy in positive parenting. In sum, 97.6 percent of the participants agreed that the program is a good fit for the Hong Kong culture. Research limitations/implications – A larger sample would have been desirable for this study. One factor that limited analyzable data was that some of the participating organizations enrolled participants into the program who did not meet research inclusion criteria. Furthermore, the size of groups varied from 5 to 16 participants, which may have produced different group dynamics thatadded variability tooutcomes. Future 6As parenting program research should attempt to standardize group size or to directly compare the effectiveness of smaller vs larger groups. Furthermore, as noted earlier, the research was conducted during the primary school application period, which may have heightened the parenting stress for parents of younger children. Practical implications – The results suggest that an enhanced belief in authoritative, autonomy and training strategiescouldreduceparentalstressandshiftparents ’ attitudetowardamorepositive approachinchild-rearing. Originality/value – Findings suggest that positive parenting can enhance the parent – child relationship and reduce parental stress. The results support governmental, non-governmental organizational and community focus on positive practices for parenting training in Hong Kong. group vs interaction effect was significant for three subscales of the CCRBQ: training, F (1, 80) ¼ 9.882, p ¼ 0.02, Z 2 p ¼ 0.110; shame, F(1, 80) ¼ 5.481, p ¼ 0.022, Z 2 p ¼ 0.064; and authoritative, F(1, 80) ¼ 5.240, p ¼ 0.025, Z 2 p ¼ 0.061. The interaction effect for autonomy was not significant, F(1,80) ¼ 0.128, p ¼ 0.721, Z 2 p ¼ 0.002, and neither were the main effects for Time F(1,80) ¼ 2.259, p ¼ 0.137, Z 2 p ¼ 0.027 and Group F(1, 80) ¼ 0.533, p ¼ 0.468, Z 2 p ¼ 0.007. The interaction effect for PSS was significant, F(1, 80) ¼ 6.739, p ¼ 0.011, Z 2 p ¼ 0.078, as was the interaction effect for positive parenting self-efficacy F(1, 80) ¼ 11.001, p ¼ 0.001, Z 2 p ¼ 0.121. The significant interaction effects show that participants in 6As differed from the control group pre- to post- treatment change in the parenting strategies of training, shame and authoritative parenting, as well as parental stress and parenting self-efficacy. hoc tests using a Bonferroni multiple CCRBQ training, p 0.001; authority, p 0.005; and self-efficacy,


Introduction
There is considerable research and popular interest in Hong Kong regarding the qualities of good parenting. Chinese mothers are portrayed as being highly involved in their children's studies, demanding and strict, and focused on training their children for optimal The challenging environment in Hong Kong Considerable research has suggested that traditional Chinese culture has created an environment that encourages and even requires the practice of harsh parenting (Chan et al., 2009;Li et al., 2013;Tang, 1998;Zhou et al., 2008). Despite the rapid integration of new cultural values, filial piety and "guan (管)" are still the core ethical codes of today's Chinese family. Based on these values, children are expected to behave as parents' property and must be obedient to their parents; otherwise, parents are justified in the use of harsh discipline (Tang, 1998). In Hong Kong, it is extremely challenging for children to acquire entry to desirable schools, especially the prestigious primary schools. If their children do not obtain satisfactory academic results, parents are more likely to enact harsh parenting (Li et al., 2013). Moreover, Chinese parents may hesitate to praise their children due to traditional beliefs that excessive praise may cause their children's performance to decline, believing that praise creates children who are too proud and who will exert insufficient effort (Li et al., 2013). Changing family structure may also account for the practice of harsh parenting. In two-wage families, parents may feel that it is imperative to spend their limited family time monitoring their children's academic results, interests and use of time, rather than providing emotional support.
Increasing research suggests that excessive focus on academic performance has led to parent-child relationship problems and children's behavioral problems. Siu (2007) referenced a finding to indicate 75 percent of primary school children are unhappy, primarily due to their academic performance. Research on youth suicide has reported that the top three reasons for young people's (aged 15-29) suicide attempts were poor academic performance, family relationships and romantic relationships (Mak, 2011).
This paper focuses on the potential benefits of positive parenting training for children in Hong Kong. Although parenting is a central life role that profoundly influences both children and parents' lives, Hong Kong couples who are parents or who expect to be parents have very limited access to formal parenting training. Parenting styles have been shown to influence children's problem behaviors (Aunola and Nurmi, 2005). Young (1994) and Bowlby (1988) showed that early caregiver interactions contribute to the development of early-life cognitive schemas, which are stable working models of the self and interactions with others. Young (1994) proposed that negative early childhood interactions can create early maladaptive schemas, which can act as self-fulfilling prophesies and increase vulnerability to numerous forms of later psychopathology (Hardt and Rutter, 2004). In contrast, positive schemas with the themes of self-efficacy, optimism, trust, success and worthiness support resilience and reduce child psychopathology (Keyfitz et al., 2013). These findings are echoed by Martin Seligman (2007), a pioneer in the field of positive psychology, who noted that children's optimism and life success can be enhanced by modifying negative expression of emotions, negative explanatory style and criticism coming from parents, teachers or coaches.
The current research examined the effects of an existing positive parenting training program, 6As Positive Parenting Program in Hong Kong (6As). This program focuses on changing parenting attitudes about parenting goals and strategies. The positive parenting training program was developed based on a book entitled How to be a Hero to your Kids, written by McDowell and Day (2008). Parents participating in the training seminar learn two core beliefs in parenting, which are called limits and love, and the six principles (the 6As) of acceptance, appreciation, affection, availability, accountability and authority. Parents also may have the opportunity to learn behavioral skills in a two half-day workshop involving practice of 6As principles in their interactions with their children.
The 6As program is unique from other parenting training programs in its greater emphasis on attitude change rather than solely behavioral training. 6As program advocates an authoritative parenting style (Baumrind, 1991) which, unlike such an authoritarian style, is both demanding and responsive. This program has been in existence and has been popular since 2006. Hong Kong Character City (2016) provided free 6As seminars for more than sixty schools in Hong Kong and more than 500 organizations and schools have participated in the program and other events promoting 6As Positive Parenting.
Purpose of the research This study represents the first formal controlled evaluation of the 6As Positive Parenting Program. This research aims to evaluate the program's effectiveness in changing parenting attitudes, reduction in parenting stress and self-efficacy in positive parenting in a sample of 82 Hong Kong parents. Considerable published research shows that effective parenting training can better equip parents to utilize sensitive and appropriate child-rearing strategies, which ultimately enhances family relationships and reduces parental stress.
Hong Kong Character City (2016), the agent for this program established in 2006, is registered as a private not-for-profit charity organization, and offers the program

57
The 6As Positive Parenting Program with the overall objective of helping parents to nurture children via the inculcation of positive attitudes. This program was chosen by the investigator for the present study because the focus of 6As is different from other local programs, as it focuses on prevention and working on instilling positive beliefs and principles for parenting roles, rather than emphasizing interventions for emergent problems. The program is open to the general public and is not restricted to parents of children who exhibit challenging behaviors. The program has been well-received in many sectors, including, but not limited to, schools, corporations and the wider community. Until now, there has been no formal evaluation of its effectiveness. Understanding the extent to which participants feel that the 6As program model fits with Hong Kong Chinese cultural models of parenting was an additional goal of this study.

Research method
In this research, individuals who attended the 6As program were evaluated prior to and following the program and their results were compared to a control group who did not participate, which was also evaluated twice over the same interval of time. Outcome variables were parenting attitudes, parenting stress, parenting self-efficacy and perceived cultural fit of the intervention. By incorporation of a variety of outcome measures, it was intended that the study would both evaluate the effectiveness of the 6As program relative to the control group, and also illuminate some of the mechanisms of change.

Participants
A total of 82 participants were recruited for the 6As intervention (n ¼ 41, 28 women, 13 men) and a control group (n ¼ 41, 30 women, 11 men). The final sample was selected from a larger group of 116 (60 participants in the intervention group and 56 participants in the control group) based on the following inclusion criteria: being a parent and the primary caretaker; being a Hong Kong Chinese local resident; having attained secondary school education level; living with the child(ren); completion of all study measures; having a 100 percent attendance record for those in intervention group; and having not earlier participated in the full 6As (seminar/workshop) program, although those with seminar experience only were accepted. Table I provides participant characteristics. An independent-samples t-test showed that participant age did not differ between the intervention group (M ¼ 43.54, SD ¼ 8.59) and the control group (M ¼ 41.83, SD ¼ 6.64), t(80) ¼ 1.01, p ¼ 0.32. For categorical variables, χ 2 analysis was conducted to evaluate group differences in demographic and parenting training history (Table II). These showed that family income was statistically significantly higher in the control group, and that the control group was statistically significantly more likely to have a child between the ages of three and six.
Measures. All measures were administered and completed in Chinese. Based on a power analysis, the statistical power for analysis in this study for ANOVA with the sample size of 82 in this study is 0.94.
Sociodemographic questionnaire. This questionnaire assessed participants' age, gender, education, the degree of prior parenting training experience over the past two years, monthly family income, age and number of children.
Chinese Child-Rearing Beliefs Questionnaire -Chinese version (CCRBQ; Lieber et al., 2006). The CCRBQ is a 36-item self-report scale divided into four subscales: a training scale (9 items), a shame scale (8 items), an autonomy scale (10 items) and an authoritative scale (9 items). The training scale measures the extent to which the parent values active socialization and moral socialization, by assessing their attitudes toward monitoring, regular reminders, modeling and other social learning strategies. A sample item is "Close monitoring shows children you care." The shame scale measures the use of shame specifically as a training approach for fostering the development of children's social sensitivities. A sample item is "Shaming children helps them learn to behave." The autonomy scale measures parents' need to encourage, protect, and nurture children's exploration, will, self-concept, expression, and autonomy. A sample item is "Parents should stimulate exploring and learning." The authoritative scale measures awareness, respect and encouragement for children's exploration and expression of ideas and emotions, encouraging parent-child communication, and recognizing parents' and children's rights. A sample item is "Parents should seek opportunities to praise children." Each item is rated on a six-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Lieber et al. reported subscale internal consistency (Cronbach's αs) of 0.82, 0.66, 0.67, and 0.71 for the training, shame, autonomy and authoritative subscales, respectively, for a Hong Kong sample. The authors reported that factor analysis of the CCRBQ yielded a four-factor solution consistent with the four subscales (training, shame, autonomy and authoritative). Analysis of internal consistency in the current study yielded Cronbach's αs for the CCRBQ subscales training, shame, autonomy and authoritative belief of 0.87, 0.88, 0.81 and 0.85, respectively.
Parental Stress Scale -Chinese version (PSS; Berry and Jones, 1995). The PSS consists of 17 items rating current parental stress on a six-point Likert type scale (1, strongly disagree to 6, strongly agree). Example items are "I am happy in my role as a parent." and "My child (ren) gives me a more certain and optimistic view for the future." Seven items require reverse scoring. Leung and Tsang (2010) reported that the reliability (Cronbach's α) of the full scale was 0.89. Cronbach's α of the PSS in the present study was 0.94. Berry and Jones (1995) reported that the PSS was significantly and positively correlated with both a common measure of child behavior problems (Eyberg Child Behavior Inventory intensity and problem subscales) (r ¼ 0.44-0.51) and another commonly used measure of parenting stress, the Parenting Stress Index Total score (r ¼ 0.75-0.78).
Parenting Self-Efficacy Scale (CONF). This scale was created for the purpose of this study to evaluate the degree to which participants believed they were able to practice the seven items that reflect the key objectives of 6As Positive Parenting Program. Example items are "I will support and coach my children in building their own sense of security and self-worth" and "I will spend time with my children that is focused on building a positive relationship." Each item was rated on a six-point Likert type scale from 1 (strongly disagree) to 6 (strongly agree). Cronbach's α was 0.72 and 0.85 at pre-treatment and post-treatment, respectively, indicating acceptable internal consistency.
Feedback form. A program feedback form was developed to assess participants' overall satisfaction and their perceptions of the cultural appropriateness of the intervention for Hong Kong parents. The two statements are "6As Positive Parenting Program helps me to be a more positive parent" and "6As Positive Parenting Program could be integrated into Hong Kong Chinese culture." Each item was rated on a six-point Likert type scale from 1 (strongly disagree) to 6 (strongly agree).

Procedure
Before commencing participant recruitment, the approval of the study was obtained from the Institutional Review Board at Alliant International University. To protect confidentiality, each participant was assigned an identification number that was used on all study materials to match pre-and post-data. Participants in the intervention group were recruited from the 6As Positive Parenting Program promotional materials such as leaflets, cooperating organizations' websites, by social workers and through other personal contacts. Most participants in the intervention group were recruited by the cooperating organizations: Fairview Park Alliance Church, The Neighbourhood Advice-Action Council Tung Chung Integrated Services Centre, Life Lutheran Church Yuen Long, Sunshine Lutheran Centre and Elim Full Gospel Church. Program coordinators assigned by each cooperating organization were responsible for the promotional, organizational and screening activities, as well as enrollment within their respective organization. An enrollment form provided to the intervention group included a brief introduction of the study and requirements for participants.
Another enrollment form for the control group was promoted over WhatsApp groups, Facebook and parent groups. Interested potential participants were able to contact the investigator. If the potential participant met the screening requirements and indicated consent to participate, they were enrolled into the control group. After providing informed consent, enrolled participants completed study measures.
The intervention group participated in the 6As Positive Parenting Program which consists of nine hours of direct intervention (Table III) consisting of one seminar and two workshops delivered in a small group format. In total, there were six intervention groups, with between 5 and 16 participants each, conducted at five locations in Hong Kong. The corresponding author of this paper is a certified facilitator for the program who conducted seminars and workshops but was blind to all identifying information associated with questionnaire data.
Study measures were administered to the intervention group, immediately following the two-hour seminar (mid-test, and after completion of the entire program The 6As Positive Parenting Program missed items in the post-treatment measure and this data was otherwise not used in analysis. The interval between pre-and post-questionnaires was two full weeks. Participants in the control group completed study measures at approximately the same time interval.

Statistical analysis
Design. The study utilized a 2 × 2 mixed-model design. Condition (intervention, control) was the between-subjects factor, and time of assessment (pre-and post-intervention) was the within-subjects factor. The p-value to reject the null hypothesis on key study analysis was set at p ¼ 0.01 Preliminary analysis. Skewness and kurtosis statistics were computed to assess the suitability of the data for parametric statistics, with departures from normality defined as values of skewness or kurtosis less than −2 or greater than 2. In total, 12 individuals in the intervention group and 4 individuals in the control group did not complete individual items in some questionnaires. For the intervention group, missing values received corresponding scores for the corresponding mid-test item, and in the control group, missing items received corresponding item scores from the other assessment. Data were checked for outliers.
Data analysis. Group differences in demographic data were evaluated with t-tests (for continuous variables) and chi-square ( for categorical values). Correlations between participant demographic characteristics and study measures were computed. To evaluate the 6As program effectiveness, individual 2 × 2 mixed ANOVAs were conducted to evaluate group (control vs intervention) by time (pre vs post) interaction effects for study measures.

Results
Skewness and kurtosis values were within criteria, and the data were determined to be appropriate for parametric analysis. Table IV provides the means and standard deviations for both the intervention and control groups at pre-and post-treatment intervals.

Treatment results
Correlations between demographic characteristics and outcomes. Table V provides correlations between demographic and outcome variables. Correlational analysis revealed statistically significant associations between the dependent variables and age, family income and parent's education level. Educational level was negatively and moderately associated with the use of authoritative parenting, and family income was negatively associated with parental stress and parenting strategies of shame and authoritative parenting.
The interaction effect for PSS was significant, F(1, 80) ¼ 6.739, p ¼ 0.011, Z 2 p ¼ 0.078, as was the interaction effect for positive parenting self-efficacy F(1, 80) ¼ 11.001, p ¼ 0.001, Z 2 p ¼ 0.121. The significant interaction effects show that participants in 6As differed from the control group pre-to post-treatment change in the parenting strategies of training, shame and authoritative parenting, as well as parental stress and parenting self-efficacy.
Post hoc tests using a Bonferroni correction for multiple comparisons indicated a statistically significant increase from pre to post for the intervention group, but not the control group for CCRBQ training, p o0.001; authority, p ¼ 0.005; and self-efficacy, p o001. There was also a statistically significant decrease in parental stress for the intervention group, but not the control group, i.e. p o0.001. Interestingly there was also a statistically significant increase in CCRBQ shame, for the control group, but not for the intervention group.

Relationships among change scores on study measures
Pre-to post-intervention change scores on study measures were calculated separately for each group, and correlations between these difference scores were then computed, using a p-value of 0.01 to reject the null hypothesis. In the control group an increase in belief in shame as a child-rearing approach was positively associated with an increase in parental stress, r(41) ¼ 0.433. In contrast, in the intervention group, greater increase in beliefs in training and authoritative parenting approaches, were both associated with a decrease in parental stress, rs(41) ¼ −0.315 and −0.331, respectively, and ps o0.01 Taken together, these results suggest that positive changes in beliefs about training and authoritative parenting are related to a decrease in parental stress. Furthermore, it suggests that in the absence of intervention, parenting stress may tend to increase, as less effective parenting strategies based in shame are continued.
The relationship between specific changes in beliefs. Table VI presents intercorrelations of the change scores for study measures for the intervention group. For the intervention group, the degree of change in the four belief measures were moderately positively intercorrelated, with the exception of CCRBQ training and shame, which were unrelated to degree of change. This finding suggests that parents in the intervention group simultaneously developed a stronger belief in all of the domains. In this context, it is interesting to note that belief in

63
The 6As Positive Parenting Program shame as a strategy significantly increased along with beliefs in all of the other domains except belief in training. The control group did not show this same pattern of correlations, as pre-to post-changes in these measures in the control group were small. Lieber et al.'s (2006) CCRBQ study Independent-samples t-tests were conducted to compare the 2006 Hong Kong sample reported by Lieber et al. (2006) to pre-intervention data for the intervention sample, to contrast the samples and to examine whether there was change in child-rearing beliefs over the last decade. This showed that the current intervention sample scored significantly lower on belief in shaming (M Overall feedback Participants (n ¼ 41) were asked for their level of endorsement of the statement "6As Positive Parenting Program helps me to be a more positive parent." Of the sample, 68.3 percent strongly agreed; 29.3 percent very much agreed and 2.4 percent agreed. No participants endorsed "disagree." Participants were also requested to evaluate the statement "6As Positive Parenting Program could be integrated into Hong Kong Chinese culture," and the result was 36.6 percent strongly agree, 53.7 percent very much agree, 7.3 percent agree and 2.4 percent disagree. The results suggested that parents found the program to be useful and that they felt that its core approaches and principles were culturally relevant.

Discussion
Overall research findings Results of this study reflected that the 6As program changes parenting attitudes, reduces parenting stress, and increases positive parenting self-efficacy. Specific positive change in parenting attitudes were found relative to the control group in use of the strategies  (Lieber et al., 2006). *p o0.05; **po 0.01 Table VI.
Intercorrelations between change scores for study measures for intervention group of training and authoritative parenting. In addition, belief in the use of shame as a parenting strategy statistically significantly increased in the control group and did not increase in the intervention group. Parents strongly endorsed the utility and cultural fit of the program. Taken together, these results provide strong support for the short-term effectiveness of the 6As program.
Correlational results of pre-to post-intervention change scores showed that the degree of change of beliefs in training as a parenting strategy, authoritative parenting style and belief in encouragement of autonomy were associated with decrease in parenting stress, suggesting that success in learning these key components of 6As made parenting more manageable. One critical consideration regarding the changes in parenting stress and use of shaming as a child-rearing technique over the two interventions is that this study was conducted during the primary school application period of March-June, and most parents in the control group had children aged between three and six years. It is possible that this intensified parents' concerns for their children's futures.

Implications for public intervention
Several positive parenting training programs have been successfully conducted in Hong Kong. These included Triple-P (Leung et al., 2003;Li et al., 2013) and Parent-Child Interaction Therapy (PCIT; Hembree- Kigin and McNeil, 1995). These programs focus on parent skill training to improve child-parent relationships and to reduce children's problematic behaviors through positive reinforcement and praise. As these parenting training programs were imported from Western countries, whether they were culturally fit to local Hong Kong Chinese was often raised as a concern in the literature. For example, as for PCIT, such concerns included: Chinese parents might not feel comfortable praising their children; it was difficult for participants to exercise the right degree of praise because it was not part of their cultural practice; emphasis on praise might mean ignoring a child's negative behavior; and given that many family members had contact with children, it is challenging to ensure consistent practice among parents and extended family members (PCIT; Hembree- Kigin and McNeil, 1995).
In the current study, 97.6 percent of the participants in the intervention group agreed that 6As had an appropriate cultural fit to the local Hong Kong Chinese. Four possibilities may account for the high acceptability of the program: (1) The program promoted change of parenting attitude emphasizing a balance of love and limits, a belief system well aligned to Chinese tradition. Within traditional Chinese beliefs and values, training children for optimal academic outcomes is a typical expression of concern for children, as is implementing limits through psychological control and punishment (Dewar, 2011), both of which represent a largely negative parenting approach. The 6As program helps parents shift from negative to positive parenting, while maintaining the traditional vision of providing love and limits. Essentially, the program offered parents a parenting alternative that consists of building a loving relationship while continuing to provide structure and protection. Therefore, participants did not feel that the strategies were inconsistent with their essential values; instead, they perceived the new interventions as a change of approach to achieving their goals.
(2) At the beginning of the first workshop, parents received a brief lesson about Erik Erikson's eight stages of development and Kohlberg's stages of moral development. Using age ranges to depict the significance of developmental stages is common in various cultures. In Chinese traditional culture, there is also a common understanding of expectations associated with different life stages.

65
The 6As Positive Parenting Program For example, people at the age of eighteen are considered an adult, and males are expected to be married and have successful careers by age thirty. The core Chinese parental belief "guan (管)," which means to govern, control and restrict, while also implying loving and caring, can be aligned with Kohlberg's stages of moral development which suggest people have different needs for moral development at different ages, so parents are best served to adjust their strategies when implementing rules.
(3) Parents learned skills for building loving relationships and coaching children on problem solving for daily challenges, as well as achieving personal growth at children's different stage of development. The program did not endorse that parents ignore negative child behaviors. Both Lieber et al. (2006) and the current study indicated high correlation between belief in shaming as a child-rearing technique and other strategies, suggesting that correcting negative child behaviors by shaming continues to be utilized regularly by Hong Kong parents. The program encouraged a balanced approach for both praising positive characters and correcting problematic behaviors, and thus does not question the importance of correcting negative child behaviors that is central to traditional Chinese parenting.
(4) Parents learned how to exercise their authority through offering several options to their children, prior to an event. If children do not stay within the provided limits, negative consequences were then implemented, keeping children accountable for their own decisions. The calm implementation of consequences replaced the approach of criticizing children, and from this perspective, parents are still able to experience themselves as authority figures, retaining hierarchy in the family. This honors the Chinese tradition which endorses obedience to authority as a core family value (Lau, 1997).

Changes in parenting beliefs in the past decade
A comparison between the current study and the sample of Lieber et al. (2006) found significantly lower levels of belief in shaming strategy and higher levels of belief in autonomy strategy for both control and intervention groups prior to the intervention.
Although it is possible that these findings reflect different sampling procedures, these results might also suggest that although shaming is still a central Hong Kong parenting strategy, parents are increasingly applying more authoritative strategies that promotes autonomy.

Limitations
A larger sample would have been desirable for this study. One factor that limited analyzable data was that some of the participating organizations enrolled participants into the program who did not meet research inclusion criteria. Furthermore, the size of groups varied from 5 to 16 participants, which may have produced different group dynamics that added variability to outcomes. Future 6As parenting program research should attempt to standardize group size or to directly compare the effectiveness of smaller vs larger groups. Furthermore, as noted earlier, the research was conducted during the primary school application period, which may have heightened parenting stress for parents of younger children.

Implications for future Hong Kong parenting training
(1) There is considerable need for a holistically designed parenting training that is culturally credible and sensitive. The teaching in the 6As Positive Parenting Program was developed by the founder of Hong Kong Character City Movement, and the founder is a local Hong Kong woman. Being a leading facilitator, I (i.e. the corresponding author of this paper) added a brief psycho-education unit on human developmental stages. Therefore, the current training design is conducted with consideration of evidence-based parenting approaches (authoritative parenting); integration with Hong Kong Chinese beliefs and values; and psycho-education regarding developmental psychology. Collectively, these factors likely produce training that best addresses parents' needs.
(2) The results suggest that enhanced belief in authoritative, autonomy and training strategies could reduce parental stress and shift parents' attitude toward a more positive approach in child-rearing. It is hoped that future positive parenting training programs might continue to enhance family relationships and well-being and reduce the incidence of mental health difficulties faced by children in Hong Kong.

Implications for future research
Future research which replicates this study and recruits a sizable sample, utilizes standardized group size and directly compares the effect of group size will further elucidate factors influencing the effectiveness of the 6As program for promotion of healthy parenting in Hong Kong. A larger sample size would also permit the evaluation of parent and family characteristics that mediate successful outcomes.
In the training, the importance of a consistent parenting approach between fathers and mothers was emphasized. However, mothers bear the primary child-rearing role, based on existing literature and as reflected by the small number of fathers in the current study. Future research comparing mothers and fathers using the present study measures will provide further information on the value of parent training for each parental role.

Policy recommendations for Hong Kong
Evaluation of parenting training programs in Hong Kong has shown that successful programs help parents change their parenting style, reducing their children's problematic behaviors and enhancing the parent-child relationship. However, since the Hong Kong governmental strategic initiatives of 2000 and 2001, no further activities have been announced on the website of the Education Bureau, Department of Health or NGOs. In view of that, certain recommendations for Hong Kong SAR Government are proposed as follows: (1) To continue building up a positive parenting culture in Hong Kong, extensively promote positive parenting training through broad community channels, such as

67
The 6As Positive Parenting Program within Department of Health of Hong Kong SAR Government and NGOs, through schools via parents' associations, and The Family Planning Association of Hong Kong for pre-marriage couples.
(2) To ensure positive parenting programs promoted in the community are culturally fit to local Hong Kong Chinese. For example, Hong Kong SAR Government can tailor-make positive parenting training curricula or can modify imported parenting training programs to address the target audience's parenting beliefs, current parenting skills and understanding of children's stages of development.
(3) To promote importance of the father's role in parenting, and encourage further collaborative efforts and consistent parenting styles between parents or caretakers, such as grandparents and/or domestic helpers.
(4) To nurture younger generations to be positive parents by introducing more positive parenting concepts in sexuality education.

Conclusion
This paper showed that the 6As Positive Parenting Program can effectively change parents' parenting attitudes, reduce parenting stress, and increase self-efficacy in parenting. Furthermore, results indicated that positive change in beliefs in autonomy, authoritative and training strategies of child-rearing are associated with reduced parental stress. However, a belief in the strategy of shaming in child-rearing is still common among Hong Kong Chinese parents, which may reflect that Hong Kong Chinese parents are still strongly influenced by the traditional collectivistic culture which focuses on instilling social sensitivities. It is believed that the 6As Positive Parenting Program strikes a balance between evidence-based strategies and accountability to the traditions of Chinese culture, while endorsing an authoritative parenting approach. Although outcome measures were not entirely utilized to assess the degree of psycho-educational insight gained during the training intervention, based on the participants' verbal feedback and the investigator's observation, this component helped participants better understand children's developmental stages so that their communication style and content would correspond to the developmental stage of their children. Therefore, providing education on basic developmental psychology could enhance the effectiveness of positive parenting training.