Ⅰ. Introduction and Background
Understanding parenting is perhaps one the most complicated issues, and it is even more complicated in the context of immigration[24]. Interestingly, there has been a growing interest in understanding immigrant parenting with a specific reference to Asian immigrant populations. In New Zealand, Asian immigrants are the fastest growing ethnic group, comprising 12% of the total New Zealand population, and the majority of this expansion stems from ongoing arrival of immigrants[28]. Korean immigrants, who are the focus of this paper, are the fourth largest ethnic Asian group in New Zealand, with 30,200 people[28]. Despite the ever-increasing diversity of Asian immigrants, they remain understudied.
Where Asian immigrant parenting has been studied in Western countries, the existing research has mainly been based on parenting concepts and measures derived from the perspective of Western parenting practices and theories[6, 16, 17]. Measures of parenting are, thus, often criticized for measuring non-Western parents against parenting styles such as authoritarian (e.g., high on control but low on warmth) and authoritative (high on control and warmth) commonly demonstrated by European-American families[13].
Further, researchers argue that, while using Western derived concepts of parenting may be useful for comparative purposes, this has yielded an incomplete picture of Asian parenting as its own entity[19]. For example, Chinese parenting has often been described as controlling or authoritarian, which is usually shown to be negative for children among European Americans[5]. The construct ‘parental control’, however, needs to be interpreted cautiously in non-western cultures because it may have a different meaning for Asians and Asian immigrants than it does for European and European-American parents[5, 28]. Parental control for Chinese parents has been described as guan[5], which means to govern and to care for and is positively evaluated in Chinese culture with children in combination with strict training in proper behavior and high achievement[5, 13].
Alternatively, other researchers argue that Western based conceptualizations of parenting may possibly be generalized to Asian culture[6, 18] and have some significance for children’s development. However, despite these disparate views, both perspectives allow for the idea that the concepts related to parenting such as authoritarian and authoritative behaviors, may have different meanings and implications for parents and children across cultures depending on the sociocultural context in which these practices occur[5, 28]. Research suggests that it is important to assess culture-specific dimensions of Asian parenting. In fact, several parenting measures have been created to capture parenting practices that are specific to Asian and Asian immigrant families. One of the most prominent examples includes measures of five parenting practices emphasized in China developed by Wu et al.[30]. Because the use of Western concepts and measures fails to capture Asian specific aspects of parenting, the purpose of current study is to investigate whether typical parenting practices emphasized in China might be evident in Korean and Korean immigrant settings. In East Asian countries, such as China and Korea, there are certain commonalities within these countries (i.e., similar values around Confucian principles, such as respect for elders and obedience to authority[5]. Hence, confirmatory factor analysis (CFA) was used to verify the original five-factor structure of the parenting constructs emphasized in China as obtained by Wu et al.[30]. More specifically, the current study examines the validity of the measurement model for these five parenting constructs derived from the Chinese parenting literature by ascertaining whether the postulated factors would emerge and to determine whether these five constructs could be used to assess the parenting of a Korean immigrant population.
Ⅱ. Literature review
Although there has been rapid social change in Korea in the past few decades, Korean families have retained traditional values and beliefs based on the traditional teachings of Confucius which shape their child rearing practices and beliefs[26]. Therefore, it is possible that Korean immigrant parents will use traditional methods of Korean parenting practices valued in Korea. In East Asian countries such as China and Korea, Confucian principles of social harmony, modesty, benevolence, cohesiveness of the group, and proper behavior[19, 26] have influenced parenting practices.
Because the parenting concepts formulated in Western cultures may not be adequate for understanding Asian parenting, Wu et al.[30], for example, identified and examined five distinct Chinese-specific parenting constructs stemming from the Chinese socialization literature. These five parenting practices include encouragement of modesty, shaming and love withdrawal, parental protection, maternal involvement, and directiveness.
Encouragement of modesty emphasized behaving in a humble, modest, and socially conforming way when interacting with others. The ability to cooperate with others and prioritize the achievements and successes of in-group members is considered an index of individual social maturity[30]. Modesty is, in fact, a characteristic esteemed by Koreans and modesty appears to be more common in collectivist societies such as Korea. Evidence that collectivist beliefs are still present among Korean immigrant parents comes from a study by Choi et al.[8]. In their study of the family socialization beliefs and practices of Korean-American immigrant families, they found that the ability to attribute one’s success to help from others was considered an ideal attribute that Korean-American parents would like to see in their 11 to 14-year-old adolescents.
Given that Confucianism emphasizes harmonious social relations among people, another important aspect of socialization continues to be the use of shaming and love withdrawal techniques. To be aware of another person’s opinions, judgments, and evaluations, children were expected to be sensitive to shame. In addition to the encouragement of modest behavior and shaming/ love withdrawal, parental protection in the Asian cultural context reflects the parental intention of keeping young children safely nearby and fostering dependency on parents meeting the child’s needs[30]. Children are, therefore, taught to depend on their parents from a young age. Maternal involvement places emphasis on mothers sacrificing significant time to help children succeed academically, and the academic success of a child is often described as the success of the mother[4]. Directiveness refers to regulating children’s behavior through scolding and demands. Wu et al.[30] found that these five parenting practices were relatively independent from the parenting constructs (i.e., authoritarian and authoritative parenting) emphasized in North America.
Specific to Korea, devotion is another important traditional parenting practice that persists in modern Korean society[20, 21]. In Korea, parents tend to view unselfish devotion and sacrifice to their children as their basic role and duty[20]. Moreover, devotion encompasses more than a parent’s deep love and generosity; it is associated with a parent’s empathy, care, and unconditional love. For instance, if child makes a mistake or faults are found, a parent tries to accept and even overlook the mistakes made by children.
Furthermore, Park and Cheah[26] and Park et al.[27] reported that even though young Korean parents in contemporary society are more likely to adopt Western individualistic values regarding child development, they devote themselves to nurturing and taking care of their child’s everyday life and expect their child to be dependent on them. The common reasons given for parents’ full responsibility for their child’s everyday life are to establish parent-child interdependent relationships in which both parents and children gain a sense of security[27]. Parents in Korea, therefore, try to maintain close relationships with their children to make sure that their child feels secure, and to make the boundary between themselves and the child minimal by providing for the needs of their children[20]. Korean parents typically are willing to do everything for their children even if it means physical, emotional, and financial sacrifice. Children’s strong emotional and existential dependence needs are satisfied by their parents’ devotion[20].
In some sense. Chinese and Korean cultures are similar and alike in terms of the prevailing cultural and philosophical backgrounds in these societies[26]. Therefore, the purpose of this study was to examine whether a model of parenting constructs emphasized in China are applicable for Korean parents.
To date, however, there has been no parenting or family research conducted with Korean immigrant parents in New Zealand, and no studies could be identified that assessed psychometric properties of the parenting practices emphasized in China and Korea with a Korean immigrant sample. Hence, questions such as, “Can a parenting practice questionnaire titled, ‘The parenting practices emphasized in China,’ developed by Wu et al. [30], be effectively translated, validated, and assessed as a reliable instrument that can be adapted for use with a Korean immigrant parent sample?” remain unanswered.
Ⅲ. Methods
1. Participants
This study was part of a larger survey study of the child rearing practices of Korean immigrant parents in New Zealand[22]. The current study extends and replicates our previous work by comparing the goodness-of-fit of competing models suggested by the extant literature using data for 207 parents.
The total sample consisted of 207 Korean immigrant families, comprising two parent families (68 mothers and 68 fathers) and single parent families (60 mothers and 11 fathers) in New Zealand with a child aged between six and ten. The mean age of mothers was 33.9 years (SD = 12.5) and the mean age of fathers was 34.3 years (SD = 11.0). The average child age was 7.8 years (SD = 1.8). All parents were born in Korea and the average length of residency in New Zealand was 7.4 years (SD = 5.2). The parents came from well-educated middle-class backgrounds. 93% parents had an undergraduate degree or higher. On average, Korean immigrant parents had an annual household income falling between $NZ 60,000 and $NZ 80,000.
2. Procedures
After receiving ethics approval from the University’s Human Participants Ethics Committee, participants were primarily recruited with the cooperation of Korean religious organizations and Korean language schools in New Zealand. The questionnaire package accompanied by a survey questionnaire, a cover letter describing the study, and a stamped self-addressed envelope for the return of their completed survey was distributed by Korean community leaders to eligible parents via post. Other participants were recruited through postings on Korean community websites and newspapers, and in places frequented by Korean parents. Interested parents contacted the researcher by phone or email to obtain questionnaires via mail. Each participant was given both English and Korean versions of the questionnaires and asked to complete them in the language they preferred. The completed questionnaire was mailed to the researcher within a month. The return rate for the survey was 72%.
3. Measure
The English version of the questionnaire was translated and back-translated into Korean by a professional bilingual translator. Any discrepancies that were highlighted from back-translation were discussed, evaluated, and reduced through an iterative review process by the translator and the researcher.
Five parenting constructs emphasized in China that were invented by Wu et al.[30] were used in this study to assess Korean parenting practices. This inventory was chosen because the items reflected similar Confucian-influenced parenting practices across China and Korea. The five constructs were: involvement (four items), shaming/ love withdrawal (four items), encouragement of modesty (four items), directiveness (three items), and parental protection (three items). In addition to the five constructs, a new scale with four items, devotion was created to capture both maternal and paternal devotion specific to Korean parents[20, 21]. Parents were asked how strongly they agreed or disagreed with each statement on a 6-point Likert-type scale ranging from 1 = strongly disagree to 6 = strongly agree.
4. Statistical analysis
Statistical analyses were performed with IBM SPSS v19 and AMOS v20. Missing data were minimal in this study with less than 5% of the total number of cases in the data set. Missing data were replaced using the Expectation Maximization (EM) algorithm. The EM algorithm is a method of finding the maximum-likelihood estimate of the parameters from incomplete data or the data that has missing values[10]. CFA is a widely used technique for testing a specific hypothesis and comparing the goodness of fit of competing models. It also enables the researcher to postulate a specific model a priori and then tests whether the data fit the model. A CFA with maximum likelihood procedure was used to assess and verify the theoretical model for the parenting practices emphasized in China proposed by Wu et al.[29]. The quality of fit for CFA models was determined by reference to a number of fit indices. Chi-square per degree of freedom (χ2 /df), a value of less than 5 is an indication of a good fit. The smaller the chi-square, the better the model fit[23]. For the CFI (comparative fit index), a value equal to or greater than .90 signifies an acceptable model fit[23]. For the RMSEA (root mean square error of approximation), a value less than or equal to .05 indicates a good fit, with .08 also suggesting an acceptable-fitting model. In line with current practices [12, 23], acceptable fit for a model was imputed GFI (gamma hat) >.90, and the SRMR (standardized root mean residual) <.08.
The following three criteria were used to identify the best-fitting model for each factor [1] and thus, how many factors to retain for each construct:(1) items had to have a factor loading greater than or equal to .40, (2) items with cross-loadings to other factors >.30 or with strong modification indices (i.e., >20) to other factors were removed, and (3) statistically non-significant paths and items were removed.
The following models were tested, and the fit indices were compared to assess how well each model fit the data. Model 1 is the original five-factor structure of the parenting constructs emphasized in China as obtained by Wu et al.[30]. Model 1, thus, tested and replicated the original 18-item measurement model of the five latent constructs derived from the Chinese parenting literature and the five factor model found in Wu et al. (i.e., encouragement of modesty, parental protection, directiveness, shaming/love withdrawal, and beliefs about maternal involvement). In addition, two more theoretically plausible CFA models were also tested with a sample of 207 Korean immigrant parents. Model 2 is the 24-item six factor model where six items from devotion are added. The researcher developed devotion, a factor derived from extant literature on Korean parenting. The final model tested is the four-factor correlated model. There are two reasons for the decision to run the correlated model. First, the discussion of the relationship among the five parenting practices constructs emphasized in China in Wu et al.[30] indicated that the factors were weakly inter-correlated. Second, examining such a model provides a test of the necessity of incorporating correlated factors by enabling a comparison of the increase in fit between correlated and uncorrelated factors[11].
After examining the goodness-of-fit for alternative (i.e., competing) measurement models for parenting practices emphasized in Korea, the most appropriate model was selected and descriptive statistics were computed. Cronbach’s alpha reliability coefficient was used to calculate the reliability coefficient for the parenting scales.
Ⅳ. Results
1. Confirmatory factor analyses
The model fit indices for the three models (i.e., five-, six-, and four- correlated model) can be found in Table 1.
As seen in Table 1, the five-, six-, and four- correlated models had very poor fit with the data, although comparing the χ2 difference between the five-factor, six-factor and the four-factor correlated models, there is a statistically significant advantage to the four factor correlated model.
The 18-item original five-factor solution showed a better fit than the six-factor model, though, it too was inadequate. The six-factor model had the poorest fit to the data, with five indices all exceeding conventional cutoff values. A number of items had low factor loadings and strong factor cross-loadings. Specifically, all three items from protection did not load on any factor and did not have a significant relationship with the total scale of the parenting practices emphasized in Korea and hence were excluded from the analysis. Modification indices suggested that the three items from the shaming/ love withdrawal factor loaded onto the encouragement of modesty factor. Hence, the shaming/ love withdrawal factor was merged into the encouragement of modesty factor. The trimmed model consisted of 16 items related to four correlated factors consisting of devotion, involvement, shaming/ love withdrawal and encouragement of modesty and, and directiveness factors. All standardized loading were moderately high (>.40) (Table 2).
Examination of the factor inter-correlations showed that devotion and shaming/ love withdrawal and encouragement of modesty were independent of each other, but that both were positively correlated with involvement and directiveness (Figure 1).
Devotion was more strongly correlated with involvement than shaming/ love withdrawal and encouragement of modesty (r =.80 vs. .28), while both were moderately correlated with directiveness (r =.43 and .63). Therefore, both a positive (i.e., devotion, involvement) and a negative (i.e., directiveness and shaming/ love withdrawal and encouragement of modesty) construct both explained the variance in the caring and strict behaviors. Overall, the CFA fit indices for the four factor correlational model failed to reach the recommended cutoffs. However, the difference in χ2 conditioned by the difference in degrees of freedom between the five-factor model and this trimmed correlational model is statistically significant; hence, this model is preferred (χ2 =373.6, df=100; χ2 /df=3.74; p=.05; CFI=.76; GFI=.86; RMSEA=.115; SRMr=.110). Consequently, given these various conditions, the four factors of devotion, shaming/ love withdrawal and encouragement of modesty, involvement, and directiveness are cautiously considered as the best understanding of Korean parenting.
2. Descriptive statistics
Table 3 shows the means, standard deviations, and internal consistency reliabilities of the four factors (i.e., four-factor correlated model). The mean scores indicated that parents were highly devoted and involved with their children and were moderately directive. They also moderately encouraged children to engage in modest behaviors and used shaming/ love withdrawal.
For the six constructs of Korean parenting practices used in the current study, the Cronbach’s alphas (α ) were .68 for devotion, .75 for involvement, and .55 for directiveness. For scales that yielded poor reliability (α <.70), especially if the scale consisted of fewer than 10 items, Pallant[25] recommended that it may be appropriate to report the mean inter-item correlation for the items. Briggs and Cheek[3] suggested that an optimal range for the inter-item correlation of .20 to .40. The mean inter-item correlations for the current study were in the optimal range of .20 and .40 for most of the scales and the mean of the scores was .30.
Following CFA, two items from encouragement of modesty and three items from shaming/ love withdrawal had to be combined to form a shaming and love withdrawal/ modesty encouragement scale with Cronbach’s alpha of .76. Because all three items from parental protection had factor loadings less than .30, protection was removed; hence, Cronbach’s alphas for parental protection were not analyzed.
V. Discussion
The study attempted to replicate the five-factor structure of the Parenting practices emphasized in China proposed by Wu et al.[30]. Unfortunately, the five-factor model did not fit these data. Instead, a four-factor correlated structure, which consisted of devotion, shaming/ love withdrawal and encouragement of modesty, involvement and directiveness, performed better than the five-factor solution, though neither had good fit. The fact that the four-factor model goodness-of-fit solutions were not as good as the results of Wu et al.[30], may be a function of the relatively small sample size[2]. A sample size of 207 may not be sufficient to perform this analysis, with so many parameters to estimate. Also, possible reasons for the different findings could be contributed to the age differences in children and sample differences. For example, in Wu et al.’s[30] study, the samples comprised mothers of preschool-age children whereas the current study consisted of parents of 6-10 year old, school-aged children. The nature of parenting issues and practices vary across child age groups according to normative developmental changes. Furthermore, the factor structures that this study are based on were validated in both mothers and fathers who may demonstrate different parenting styles and practices when rearing their children. However, Wu et al.’s study included mothers only.
However, one interesting finding was that the three items from the shaming/ love withdrawal factor loaded onto the encouragement of modesty factor, which may have occurred because Korean parents may consider shaming as a means of reinforcing humility and discouraging self-promotion. Another possible explanation for this may relate to the strong training and socialization children receive to be aware of other people’s opinions, judgments, and evaluations. Korean children are required to acquire sensitivity to shame, be humble, and show modest behavior in order to avoid a bad reputation. According to Ho[15], people in collectivist cultures must learn to inhibit behavioral and emotional expression of their own wants and desires to keep from standing out.
All three items from protection, however, were found to have relatively low factor loadings and, hence, were excluded from the CFA. The reduced salience among these items and their removal relative to other Korean parenting practices items may indicate Korean immigrant parental adaptation to the cultural and social expectations of the New Zealand mainstream society, which includes autonomy and independence in children, much as is found in Korean-American contexts[7]. Another possible reason for the failure to include the protection factor in the present study may be due to the age of the children of the parents in the current sample. The parenting practices emphasized in China were used with mothers of preschool-age children (ages 5 years and under) in Wu et al.’s[30] study that contributed to the understanding and development of a model of parenting constructs emphasized and derived empirically from indigenous Chinese cultural notions. Ages of parents’ children in the present thesis ranged from six to ten years. It is possible that parents of preschoolers may have endorsed items in the protection factor more than parents of slightly older children. For example, mothers of preschoolers tend to be more concerned with their child’s safety, and therefore, may encourage their young child to stay physically close to them and restrict certain activities to keep their children safe from harm[14]. However, as children reach a period of middle childhood and start school, parents are with children less than half as much as before their children started in school. Therefore, interactions between parents and children become less frequent in middle childhood and parental protection is less emphasized[9]. Parenting behavior is affected by these developmental changes. The differences are certainly apparent between preschoolers and school-aged children. Items on the parental protection may or may not apply differently to parents of older children. Therefore, the parental protection factor may not be valid with parents of older children. Hence, child age may also be a factor that affects the factor structure of the parenting practices emphasized in Korea, and these variables should be explored in future research. Other than the protection factor, parenting constructs that were highly valued in China were found to be applicable and relevant for Korean immigrant parents in New Zealand. This suggests that these emphasized parenting practices are highly likely to stem from the same underlying socialization goals that are prevalent in Chinese society.
Though this was the first study attempted to replicate Wu et al.’s[30] five-factor structure and evaluate the reliability of its measure for use in a Korean immigrant sample, the study is not without limitations. First, the study was limited by a small sample size. Because the model did not fit well, this instrument should be used cautiously in future research and requires further study to assess its validity in other samples of Korean immigrant parents. At the item level, further refinement of the parenting practices emphasized in Korea may require removal or revision of items that do not load on any factor (e.g., parental protection). Moreover, a closer inspection may be needed, especially of items that loaded in others factor than previously reported. Modifications may need to be made for Korean parent communities. In future studies, researchers should consider including a larger sample to examine the factor structure of the parenting practices emphasized in Korea and examine them separately for mothers and fathers. A number of studies have consistently shown that both mothers and fathers tend to exhibit different parenting practices[22, 29]. Mother-father differences in parenting may produce a different factor structure. Comparisons in factor structure between parents of older and younger children may also provide differential support for the use of the parenting practices emphasized in Korea in one population than the other. Another limitation regarding Korean immigrant parents’ parenting practices was that the researcher was not able to tell from the data the extent to which they used or endorsed Korean parenting practices before coming to New Zealand. Therefore, future research could investigate Korean parenting practices of Korean immigrant parents before and after they come to New Zealand. Moreover, it would be illuminating to obtain data from Korean parents in Korea and Korean immigrant parents in New Zealand to make comparisons using a model of parenting constructs emphasized in Korea. Comparisons between Korean parents in Korea and Korean immigrant parents in New Zealand with regard to the relations between parenting practices could enhance understanding of the dynamic relations between parenting practices and culture[16].
Although the current analysis did not produce an exact replication of the original model because of a small sample size, the findings have some strengths that have allowed it to contribute to the current parenting literature regarding the parenting practices emphasized in Korea. The sample was Korean immigrant parents of children aged between six and ten. Second, fathers, who are not often represented in parenting research, were included in the sample. Third, the use of the more stringent CFA to test factor structure, compared to the more subjective exploratory factor analysis (EFA), allowed for a deeper understanding of the usefulness of the parenting practices emphasized in Korea. Finally, the current study expanded the study of the model of parenting constructs emphasized from indigenous Chinese cultural notions for Korean immigrant parents of school-aged children.
