Developmental status and home environment among children born to immigrant women married to Taiwanese men†
This study was funded by the Taiwan National Science Council (96-2628-B-010-039).
The aims of this cross-sectional study were to examine (a) the developmental status and home environments of children (6–24 months) of immigrant women married to Taiwanese men, and (b) the association of child developmental status with parental socio-demographics, maternal language abilities, and home environment qualities. Participants were 61 children and their mothers from China and Vietnam. Data were collected with interviews, home observations, and developmental testing. The children had lower cognitive and language but higher motor and social development scores compared with native norms. Home environment and maternal perceived language ability were positively associated with child development. The association of home environment and maternal language ability with early childhood development was supported for immigrant populations in Taiwan. © 2011 Wiley Periodicals, Inc. Res Nurs Health 35:121–131, 2012
Awareness of the importance of early life experiences has risen during the past decades, but studies demonstrating the association between the quality of the home environment and early childhood developmental status have been mostly based in the US and other Western countries. There are few studies of the importance of early life experiences focused on non-Western populations. In Taiwan, findings from one early study of native children supported the positive correlation between the home environment and child developmental status (Su, Lu, Chen, Jong, & Chang, 1992). However, the home environment of immigrant women's children in Taiwan and its relationship to early developmental status among these children has not been studied.
On average, children of married immigrant women in Taiwan not only have lower parental educational and income levels, but also older fathers and greater age differences between parents, and they are more likely to live with extended family members than children of native mothers (T. L. Chen et al., 2011; Liang, Chen, Lee, Lung, & Lin, 2010; Social Affairs, 2007). Immigrant mothers usually assume primary responsibility for home making and childcare and their social support is substantially lower than that of native mothers (T. L. Chen et al., 2011). These factors, along with language difficulties and cultural differences could have a differential influence on the home environment and parenting behaviors, and hence the children's developmental outcomes. The purpose of this study was to explore the relationship between the early home environment and the developmental status of children born to immigrant women married to Taiwanese men.
Marriage of immigrant women from countries of origin such as the Philippines, Vietnam, Indonesia, Thailand, and Mongolia to men in countries such as Taiwan, South Korea, and Japan has rapidly increased in recent decades (Asian Regional Exchange for New Alternatives [ARENA], 2007). Many of these immigrant women came from poverty-prevalent countries through marriage brokerages intending to improve their original family's finances. The women were married overseas, alone and separated from their original families (ARENA, 2007). Immigrant women gave birth to 8.68–13.37% of all children born in Taiwan from 2004 to 2009. Many of these women and their husbands have a lower educational level than the native population, in that 65.1% and 49.5% of immigrant women and their husbands received 9 or fewer years of education (junior high school diploma). The corresponding percentages for native women and their husbands are 12.2% and 12.4% (Ministry of the Interior, 2010). In addition, based on the criteria set by the Taiwan government, families of immigrant women are more likely to meet the official cutoff for low-income families compared with native families (2.03% vs. .83%; Liang et al., 2010; Social Affairs, 2007).
Despite these families' lower socioeconomic status, infants born to immigrant women in Taiwan have lower rates of low birth weight and prematurity and fewer health problems than infants born to native women (Liang et al., 2010). However, when these children enter preschool, their cognitive and language abilities lag behind children of native women (Y. H. Chen & Chou, 2008). In elementary and junior high school, their academic performance is lower than that of native children of the same socioeconomic status (C. S. Wang & Tsai, 2008). Their cognitive performance lags behind other children of a similar age as they grow older (R. C. Chen, 2005).
Early childhood is viewed as a critical stage of cognitive development, given that the plasticity of the human brain decreases over the life span (Wallander et al., 2010). Researchers have shown that the quality of experiences in the first 3 years of life, especially caring relationships between infants and parents/primary caregivers, has a significant impact on child development (Kelly, Zuckerman, & Rosenblatt, 2008). Because socioeconomic status is related to cognitive performance, children in families with low socioeconomic status may need more material, social, and emotional resources, including nurturing relationships, a stimulating environment to organize and scaffold children's learning, basic health care, and a safe place to live, for optimal cognitive, social, and emotional development than children of higher income families (Bronfenbrenner, 2005; Fuller et al., 2009; Kelly, Morisset, Barnard, & Patterson, 1996; Ortega, Horwitz, Fang, & Kuo, 2009).
In human ecology theory, a child's development is a result of complex transactions between the individual and the environment (Bronfenbrenner, 2005). For a young child, home is the most immediate environment shaping development. Numerous investigators have shown the importance of a home environment structured to encourage the child's cognitive and socio-emotional development (Glascoe & Leew, 2010; Pachter, Auinger, Palmer, & Weitzman, 2006). A previous study of children born to immigrant mothers in the US showed that family socioeconomic status and the mother's language ability, as aspects of that environment, which were related to developmental outcomes (Foss, Chantal, & Hendrickson, 2004).
The objectives of the current study were to examine these relationships in families of children aged 6–24 months whose mothers were immigrant women married to Taiwanese men. Parental socio-demographics, maternal language abilities, home environment qualities, and the developmental status of the children were assessed in order to validate the importance of the home environment in child development in a cross-cultural setting, and provide a basis for developing programs to promote the optimal development of these children. The research questions were:
- 1.What is the developmental status relative to national norms of children aged 6–24 months whose mothers are immigrants and are married to Taiwanese men?
- 2.Are parental socio-demographics, maternal language abilities, and home environment qualities associated with the children's developmental status?
In this cross-sectional study of immigrant women and their children aged 6–24 months we used a structured questionnaire administered by a trained interviewer through face-to-face interviews with the women in their homes and a developmental test administered to the child. The quality of the home environment was assessed through direct observation in addition to the interview of the mother.
In a previous study of Taiwanese children the correlation coefficients between the home environment and the child's physical and mental developmental outcomes were .53 and .50, respectively (Su et al., 1992). With the correlation coefficient set at .50 and power at .80, the sample size needed was 32 to achieve an α of .05 (Polit & Beck, 2012). A suggested rule for multiple linear regression is to have at least five observations per predictor (Kleinbaum, Kupper, Nizam, & Muller, 2008). Because we included 10 predictors in this study, the minimum required sample size was 50. We decided to recruit about 60 participants as a conservative approach.
The majority of immigrant women in Taiwan are from China (66.7%) and Vietnam (22.0%; Ministry of the Interior, 2010), therefore we decided to include immigrant women only from China and Vietnam. Because all newborns are registered with the District Health Centers, we cooperated with 2 of the 12 District Health Centers in Taipei to recruit participants. Taipei is the capital of Taiwan and is a metropolitan city with a population of more than 2.6 million people. The inclusion criteria were married immigrant women from China or Vietnam who had healthy children aged 6–24 months and were willing to participate in the study. From October 2007 to May 2008 we contacted 174 eligible immigrant mother–child dyads. Of this group, 30 children were out of the country, 26 mothers could not be reached, and 55 mothers refused home visits. The final sample was 61 mother–child dyads (52%; 61/174), with 30 from China and 31 from Vietnam.
The mean ages of the children and mothers were 11.9 ± 4.7 months and 26.6 ± 4.2 years, respectively. Mothers had resided in Taiwan for a mean duration of 46.7 ± 20.9 months. Half of the mothers had 1 child (n = 31; 50.8%); 24 (39.3%) women had 2 children, and 6 (9.8%) had 3 children. Most women and their husbands had less than a junior high school (9 years) education. The majority of the women were not employed, and most of the husbands had nonprofessional jobs. More than three-fourths of the women answered that their family income was insufficient. More than 80% rated their Chinese ability as good (Table 1). The women who rated their Chinese as poor were all from Vietnam (n = 10).
Table 1. Descriptives of Participant Characteristics and Comparison of Child Development (CDIIT) and Home Environment (IT-HOME) Scores by Participant Characteristics
|Nationalities of mothers|| || ||t = −0.37|| ||t = −0.27|
| China||30 (49.2)||0.13 (0.74)|| ||33.6 (5.1)|| |
| Vietnam||31 (50.8)||0.20 (0.73)|| ||33.9 (3.4)|| |
|Gender of children|| || ||t = −1.24|| ||t = 0.03|
| Male||35 (57.4)||0.06 (0.73)|| ||33.8 (4.7)|| |
| Female||26 (42.6)||0.30 (0.72)|| ||33.8 (3.9)|| |
|Father's educational level|| || ||F = 0.05|| ||F = 1.16|
| Ninth-grade or lower||34 (55.7)||0.14 (0.78)|| ||33.3 (4.2)|| |
| High school||14 (23.0)||0.21 (0.67)|| ||33.5 (4.2)|| |
| College or Graduate||13 (21.3)||0.16 (0.71)|| ||35.4 (4.7)|| |
|Mother's educational level|| || ||t = 0.95|| ||t = 0.42|
| Ninth-grade or lower||41 (67.2)||0.22 (0.76)|| ||33.6 (3.9)|| |
| High school or above||20 (32.2)||0.04 (0.66)|| ||34.2 (5.1)|| |
|Father's occupation|| || ||F = 1.79|| ||F = 1.14|
| Skills/unskilled/none||47 (77.0)||0.21 (0.72)|| ||33.7 (4.1)|| |
| Semi-professional||7 (11.5)||−0.31 (0.69)|| ||32.3 (4.5)|| |
| Professional||7 (11.5)||0.31 (0.77)|| ||35.7 (5.6)|| |
|Mother's occupation|| || ||t = 0.87|| ||t = 2.55*|
| None||52 (85.2)||0.20 (0.74)|| ||34.4 (4.1)|| |
| Skills/unskilled||9 (14.8)||−0.03 (0.68)|| ||30.6 (4.5)|| |
|Family income|| || ||t = −2.17*|| ||t = −1.73|
| Insufficient||47 (77.0)||0.05 (0.70)|| ||33.3 (4.2)|| |
| Sufficient||14 (23.0)||0.54 (0.74)|| ||35.5 (4.3)|| |
|Mother's language ability|| || ||t = 2.16*|| ||t = 1.62|
| Chinese—good||51 (83.6)||0.25 (0.70)|| ||34.2 (4.3)|| |
| Chinese—poor||10 (16.4)||−0.28 (0.77)|| ||31.8 (4.2)|| |
The study protocol was approved by the institutional review board at National Yang-Ming University. A research assistant used the name list provided by the health centers to contact the immigrant mothers by telephone at the health centers. A description of the study and a consent form were read aloud to the mothers by the research assistant. Those who agreed to participate left their contact information and were called later to arrange a time for a home visit.
During the phone calls to schedule home visits, the interviewer assessed the mother's language ability to determine if an interpreter was needed. Because all the participating women communicated with the interviewer effectively, no interpreter was used during any of the home visits. The interviewers confirmed that the women had immigrated alone without family companionship, married into Taiwanese families, and lived in the same neighborhoods as these native families.
At the home visit, the interviewer introduced herself and explained the purpose and procedures of the study and the participants' rights. We prepared consent forms in simplified Chinese and Vietnamese so that the participants would feel at ease signing their names. After the participants signed the consent form, the interviewer asked to be alone with the child and mother. Family members were asked to stay away at a distance from the interview site in order to avoid distraction. The interviewer completed the socio-demographic and language ability questionnaire with the mothers first, and then performed developmental testing of the child, with the mother nearby. Lastly, the interviewer conducted the Infant/Toddler Home Observation for Measurement of the Environment Inventory (IT-HOME, Caldwell & Bradley, 1984) based on direct observation and information from the interview with the mother. The interviewer walked around the house to assess the home environment. After data collection was complete, the interviewer invited mothers and family members (if present) to share in feedback on the child developmental test results and to discuss childcare issues. Each home visit took from 50 to 60 minutes.
A check-box format questionnaire was developed to collect data on socio-demographic items and language ability. The family income was rated on a 5-point Likert-type scale—very insufficient (0), insufficient (1), just making a living (2), sufficient (3), or very sufficient (4)—according to the women's perception of the adequacy of family income. Because the relationship between this variable and the developmental score was not linear, we dichotomized the scores of 0–2 as insufficient family income and scores of 3–4 as sufficient family income. The immigrant mothers also self-reported their perceived Mandarin Chinese language abilities using a 5-point self-report scale regarding their comprehension and speaking abilities—very bad (0), bad (1), fair (2), good (3), or very good (4). Summed scores for comprehension and speaking abilities were not normally distributed, therefore for analysis we grouped scores of 0–4 as poor Chinese language abilities and scores of 5–8 as good.
The children's developmental status was measured using the Comprehensive Developmental Inventory for Infants and Toddlers screening test (CDIIT), which includes 87 items scored in a binary (yes/no) fashion (T. M. Wang et al., 1998). The CDIIT has five subscales: Cognitive Development, Language Development, Motor Development, Self-Regulation Development, and Social Development. Scores are age-standardized Z scores, with a mean of 0 and standard deviation of 1. A Z score of <−1 is defined as suspicious for developmental delay. The CDIIT was developed in Taiwan using a sample of 3,703 children (1,832 boys and 1,871 girls) aged 3–72 months who were representative of the age, gender, and geographic distributions of children of the same age group as the population norm. The internal consistency of the CDIIT and its five subscales ranged from .75 to .99 (T. M. Wang et al., 1998). The CDIIT discriminated children with developmental delays with a sensitivity of .65 and specificity of .93. Lower Z scores on the CDIIT at age 5 years predicted the need for special education 1, 4, and 8 years later (T. M. Wang, 2005). The concurrent validity between the CDIIT and the Bayley Scales of Infant Development-II was good (.69–.93, p < .001; Liao, Wang, Yao, & Lee, 2005). In this study, the Kuder–Richardson Formula 20 (KR-20) reliability coefficient was .93.
The quality of the home environment was measured by the IT-HOME (Caldwell & Bradley, 1984). The IT-HOME measures the level of environmental stimulation for children in the home setting. The scale combines behavior-based observations of the mother and child and an interview with mothers to assess the quality of the mother–child interaction. The IT-HOME includes six subscales: (a) Responsivity: parental response to the child's behavior; (b) Acceptance: parental acceptance of less than optimal behavior by the child and avoidance of undue restriction and punishment; (c) Organization: safety of the physical environment and utilization of community services as part of the family support system; (d) Learning materials: presence of appropriate play and learning materials; (e) Involvement: parental involvement in the child's learning and provision of stimulation; (f) Variety: events and persons in daily life that bring variety into the child's life. The IT-HOME includes 45 items with binary (yes/no; 1/0) scoring. A higher score indicates a more positive environment for child development. The Chinese Version of the IT-HOME has good reliability and validity, with a KR-20 reliability coefficient of .86, and is highly correlated with the Bayley Scales of Infant Development (Su et al., 1992). In this study, the KR-20 reliability coefficient was .68.
Data collection was performed by two trained interviewers who were registered nurses with a master's degree. They both completed a 1-day CDIIT training course offered by the CDIIT center. In addition, they received a one half-day training session detailing procedures for home visits and administration of the study instruments. As part of the training, each IT-HOME item was read and discussed carefully. To assess inter-rater reliability, the instrument was pilot tested by conducting a home visit with six immigrant women with young children. One interviewer conducted the IT-HOME and an observer rated the questionnaire at the same time. The inter-rater correlation coefficient was .92 for the IT-HOME and .99 for the CDIIT.
Statistical analyses were performed using SPSS for Windows version 15.0 (SPSS Inc, Chicago, IL). Descriptive statistics used were the mean, SD, frequency, and percentage. The CDIIT age-standardized Z scores were used throughout the data analysis. Bivariate associations between developmental status (CDIIT) and the quality of the home environment (IT-HOME) were examined using Pearson correlations. The mean scores of the CDIIT and IT-HOME by socio-demographics and language ability were examined using Student's t-test and analysis of variance. The CDIIT scores were compared against native norms (M = 0, SD = 1; T. M. Wang et al., 1998). Stepwise multiple linear regression was used to examine the association of independent variables with developmental status.
Developmental Status of the Child
The overall mean CDIIT scores (.16 ± .73) were not significantly different from the population norm (0 ± 1). Only three children (4.9%) had suspected developmental delay. The group mean subscale scores for cognitive and language development were significantly lower than the norm; the mean scores for motor and social development were significantly higher than the norm (Table 2).
Table 2. t-Tests for Comparison of CDIIT Scores With Native Norms and Correlations of CDIIT With Home Environment (IT-HOME)
The mean CDIIT scores by characteristics of the study participants are presented in Table 1. Insufficiency of family income and mothers' perceived poor Chinese speaking and comprehension were significantly associated with lower scores for child development.
Quality of the Home Environment
The overall mean score for the IT-HOME was 33.8 ± 4.3. For comparison purposes, crude sub-scale scores were divided by full sub-scale scores to yield achievement rates relative to a perfect score. Of the six sub-scales assessed, the average achievement rate was 75.1%, with the lowest achievement rates for Involvement and Variety (Table 3). The items of the IT-HOME with the lowest percentages of positive responses were: “The family has a pet (18.0%)”; “Parents read stories to the child at least 3 times a week (26.2%)”; “Parents invest in new toys that are ahead of the child's developmental age and show/encourage the child to play with them (27.9%)”; “Parents tell the child names of objects or persons (31.1%)”; and “Parents structure the child's play period (36.1%)”. “Parents provide toys for child to play with during the visit (41.0%)”; “Child has three or more books of his/her own (45.9%)”; “At least 10 books are present and visible in the home (47.5%).” These items indicate that the variety of cognitive stimulation at home and parental involvement in the children's learning activities were relatively low.
Table 3. Home Environment (IT-HOME) Scores and Achievement Rates Relative to a Perfect Score
|IT-HOME (45 items)||33.8 (4.3)||75.1|
| Responsivity (11 items)||8.6 (1.5)||78.1|
| Acceptance (8 items)||6.0 (1.2)||74.4|
| Organization (6 items)||5.3 (1.0)||87.4|
| Learning materials (9 items)||7.4 (1.3)||82.2|
| Involvement (6 items)||3.8 (1.4)||63.7|
| Variety (5 items)||2.8 (1.2)||55.4|
The mean IT-HOME scores by study participant characteristics are presented in Table 1. The children of mothers who worked had significantly lower scores than the children of women who did not work. There were no significant differences in the IT-HOME scores by other characteristics.
Correlation Between Home Environment Quality and Child Developmental Status
There were significant positive correlations between the CDIIT and IT-HOME total scores. The IT-HOME was positively associated with all dimensions of development except for social development (Table 2).
Factors Associated With Child Development
Multiple linear regression showed that the quality of the home environment (IT-HOME total score) was the only significant predictor of overall child development (CDIIT). A higher IT-HOME score was associated with a higher CDIIT score (Table 4).
Table 4. Multiple Linear Regression Analysis of Factors Associated With Child Development (CDIIT) Total and Subscales
|Total CDIIT|| || || ||.173||12.34**|
| (Constant)||−2.225||0.685||.416***|| || |
| IT-HOME||0.071||0.020|| || || |
|Cognitive Development|| || || ||.259||10.13***|
| (Constant)||−2.468||0.674||.383**|| || |
| IT-HOME||0.065||0.020||−.264*|| || |
| Mother's Chinese language is poor||−0.515||0.225|| || || |
|Language Development|| || || ||.259||20.58***|
| (Constant)||−3.286||0.663||.509***|| || |
| IT-HOME||0.088||0.019|| || || |
|Motor Development|| || || ||.097||6.36*|
| (Constant)||−1.434||0.690||.312*|| || |
| IT-HOME||0.051||0.020|| || || |
|Self-regulation Development|| || || ||.077||4.93*|
| (Constant)||−1.275||0.613||.278*|| || |
| IT-HOME||0.040||0.018|| || || |
The IT-HOME was significantly associated with all subscales of the CDIIT except Social Development. The mother's perceived Chinese language ability being poor was negatively associated with the child's cognitive development, but not with the other developmental subscales (Table 4). No significant predictors were identified for Social Development and the model for Social Development was not statistically significant (F = 1.99, p = .11). Gender of the children, parents' education, parents' occupation, and sufficiency of family income were not significantly associated with the CDIIT, and thus, were eliminated from the regression models.
We found that although overall developmental status was not significantly different between children of immigrant women married to Taiwanese men and the general child population in Taiwan, children of immigrant women had significantly lower scores for cognitive and language development. We noted that the children did not have low scores across all developmental domains. Although they scored lower on cognitive and language development, they scored significantly higher than the norm for motor and social development, which has not been reported in previous literature. More study is needed to confirm these findings and examine the reasons for the differences. Nonetheless, our results suggested that cognitive and language development are areas that might need intervention. Only three children had suspected developmental delays on CDIIT screening, and they were referred for further diagnostic testing, thus, our results indicate overall healthy development rather than developmental delays among the children of immigrant mothers.
Despite the lack of significant difference in the total CDIIT scores, our findings suggest that more attention should be given to promoting the language and cognitive development of young children of immigrant women in Taiwan, as a lag in cognitive and language development in early childhood can manifest itself in poor academic achievement in school, as demonstrated in Western studies (Glascoe & Leew, 2010; Reynolds, Temple, Robertson, & Mann, 2001; Zigler, Pfannenstiel, & Seitz, 2008). Previous investigators also have established that children with low family socio-economic status are more likely to experience early school failure (Pachter et al., 2006; Tong, Baghurst, Vimpani, & McMichael, 2007).
We found that the quality of the home environment was positively associated with the children's developmental status, a relationship well documented in the literature. When the quality of the home environment was entered into analyses, other factors such as family income, parents' occupations, and educational level were not significantly related to the child's developmental status. In a recent UK study, Violato, Petrou, Gray, and Redshaw (2011) found that parenting practices, a cognitively stimulating home environment, and maternal depression, but not family income, influenced child cognitive, and behavioral development. Our findings concur with these and earlier findings suggesting that home environments during early childhood may be the major factor in unfavorable child developmental outcomes among socially and economically disadvantaged families (Kelly, 1996; Spieker, Nelson, Petras, Jolley, & Barnard, 2003).
The results suggest that in view of the generally low family socioeconomic status for immigrant women, increasing the quality of the home environment could be helpful in improving the children's development. The percentage of families who gave positive responses to IT-HOME items concerning cognitive stimulation at home and parental involvement in the children's learning activities was low. In addition, of the six subscales of the IT-HOME, involvement with the child and providing variety had the lowest achievement rates. In order to support these families, health professionals could refer immigrant mothers to resources such as the public library and parental education classes, and encourage active involvement in the child's learning activities.
Language is a tool that children use to plan their actions, remember information, solve problems, and organize their behavior (Siegler & Alibali, 2005). Through the acquisition of language and social interactions, a child's cognitive processes undergo qualitative developmental change (Rowe & Wertsch, 2004). Language may play a special role in the development of children of immigrant women and perhaps partly accounted for why the children lagged in cognitive and language development. Immigrant women from China speak a Chinese which is similar to the Mandarin Chinese language spoken in Taiwan, but in China they wrote and read the simplified Chinese characters, unlike the traditional Chinese characters used in Taiwan.
Immigrant women from Vietnam had not used the Chinese language in Vietnam. Nevertheless, from our interviews, we found that almost all of the immigrant mothers talked to their children in Mandarin Chinese at home, even the Vietnamese mothers whose Chinese ability was limited. The reasons the Vietnamese mothers gave for not talking to their children in their native language were that the children did not understand Vietnamese and they worried that multiple languages would interfere with the children's ability to speak Chinese. Thus, these immigrant mothers and their family members, as is true for Taiwanese society in general, lacked concern about the children not learning their mothers' native language (Yeh, 2004). In fact, researchers have shown that in the West, bilingual children have more favorable results than their single-language counterparts in terms of attention control, concept formation, analytical reasoning, cognitive flexibility, and complexity (Bialystok, 1999; Carlson & Meltzoff, 2008). Thus, in addition to offering free Chinese language courses for immigrant women, empowering the mothers to use both their native language and mainstream Chinese at home might encourage them to provide rich verbal stimulation, thus benefiting their children's language and cognitive development.
We used the IT-HOME and CDIIT to measure the quality of the home environment and child developmental outcomes because these instruments have Chinese versions, have been applied to the Taiwanese population previously, and have demonstrated acceptable reliability and validity (Liao et al., 2005; Su et al., 1992; T. M. Wang, 2005; T. M. Wang et al., 1998). However, there may be concerns that the instruments were not culturally appropriate for immigrant mothers and their children because they had not been used previously in immigrant populations in Taiwan.
The mothers usually had immigrated unaccompanied by other family members and were alone when married into Taiwanese families; as a result, their home environment and the children's development might be more similar to than different from those of native Taiwanese. Indeed, we did not find significant differences in the CDIIT and IT-HOME scores between the Chinese and Vietnamese mothers, suggesting that cultural differences might not have influenced the scores. Nonetheless, the low variability of the IT-HOME (M = 33.7 ± 6.0) suggests that either the home environments of the children born to the two groups of immigrant mothers were similar or the IT-HOME was not sensitive enough to differentiate among these families. Further study is needed to examine the applicability and cultural relevance of these instruments among families with new immigrant mothers in Taiwan.
We applied a cross-sectional design in this research, thus, causal relationships cannot be established. The regression model for social development using factors included in this study was not significant; further study is needed to find factors significantly associated with social development among children of immigrant women. In addition, the response rate was only slightly greater than 50% and the sample size was small (N = 61). Although we did not have information regarding women who did not participate, we believe that the participating women could have been more involved in supporting their children's development than those who did not participate because the benefits of participation described at recruitment included providing feedback on the developmental test results and discussing child care issues with the family. Additionally, the participating women may have had better Mandarin Chinese speaking abilities, given that oral consent was a pre-requisite for participation. However, all the women communicated effectively with the interviewers. This may be because they had moved to Taiwan on their own and needed to use the local language to communicate with their husbands and new extended families. At the time of our interviews, the immigrant women had been in Taiwan for an average of 46.7 months (SD = 20.9), thus, most had had time to learn Chinese.
Both language ability and family income sufficiency were self-reported by the mothers. Although our interviewers believed that their responses were reliable, more objective methods, such as using a standardized language test and determining the actual income, should be considered in future studies. We asked the women about the sufficiency of the family income rather than the actual income because most of them did not work outside the home and may not have known their actual family income. This could limit comparability with previous research using either actual incomes or comparing the income level with national low-income criteria. Only 9 of the 61 immigrant mothers were employed. Further study is needed to examine the association between immigrant mothers' work status and child development. Nonetheless, strengths of the study are that we used home visits rather than relying on self-reporting to assess the home environment and child development was assessed using a standardized instrument, enhancing the validity of our data.
Young children of immigrant mothers married to Taiwanese men in Taiwan were not different from the general population in overall child development, but they lagged behind in cognitive and language development. The quality of the home environment and the mother's perceived Chinese language ability were related to the children's developmental outcome. Although much of the literature on the association between the home environment and early childhood developmental status is US or UK based, our results support those findings, and also concur with previous findings for immigrant populations in a Chinese society. The results suggest the need to assess the children's developmental status and quality of the home environment in order to identify problems and intervene as indicated. Increasing the quality of the home environment through encouraging parental involvement in the children's learning activities such as reading, playing, and talking with the children in both their native language and Chinese could help assure optimal cognitive and language development. Although free Chinese language learning programs for immigrant women currently are offered by the government, more programs are needed to encourage immigrant women to read, play, and talk with their children, and to use libraries and other social resources to enhance their children's cognitive and language development. Nurses who work in the pediatric or community setting can incorporate these components into their practice in order to improve the quality of care for immigrant mothers and their children.