Associations between Neighborhood Risk and Protective Factors and Parenting Styles in Chinese American Immigrant Families

Abstract

Despite the rapid increase of the Chinese immigrant population in the U.S., there is limited knowledge on how neighborhood context shapes parenting practices in Chinese immigrant families. Using data from a socioeconomically diverse sample of 239 Chinese American children (aged 7–10 years, 51.9% boys) in immigrant families, the present study examined the unique associations between neighborhood risk and protective factors and parenting styles in Chinese American families. Neighborhood socioeconomic disadvantage, co-ethnic concentration, and ethnic diversity were assessed using the 2010 census tract-level data. Parents rated neighborhood criminal events and reciprocal exchange and parenting styles (authoritative, authoritarian, and intrusive) were rated by parents and children. Path analyses showed that controlling for covariates, neighborhood reciprocal exchange was positively associated with parent-rated authoritative parenting and negatively associated with child-rated authoritarian and intrusive parenting. In contrast, neighborhood Asian concentration was positively associated with child-reported authoritarian and intrusive parenting. Neighborhood criminal events were positively associated with parent-reported intrusive parenting. The findings suggest that risk and protective factors co-exist in neighborhoods and have differential implications for parenting practices in Chinse immigrant families. The results also highlighted parent-child differences in perceptions of parenting styles in immigrant families.

Asian Americans are the fastest growing racial/ethnic group in the United States. With a population of 5.4 million, Chinese Americans are the largest Asian American nationality (Pew Research Center, 2021). Despite the overall higher household median income of Asian Americans compared to other racial/ethnic groups, 10% of Asian Americans live in poverty (Tian & Ruiz, 2024). Thus, vast socioeconomic discrepancies existed among Asian (including Chinese) immigrant families, which are associated with inequality in access to educational resources and social services.

Parenting is a modifiable process shown to play a crucial role in psychological adjustment of children in Chinese immigrant families (Ma, 2020). Although prior research has examined individual, cultural, and family characteristics associated with heterogeneity in parenting among Asian/Chinese immigrant families, there is very limited research on links of neighborhood characteristics to parenting practices in Chinese immigrant families. As a result of the residential racial and socioeconomic segregation prevalent in U.S. metropolitan areas (e.g., Menendian et al., 2021), there is significant heterogeneity in structural, cultural, and social features of neighborhoods where Chinese immigrant families resided (Author, 2014; Juang & Alvarez, 2011; Liu et al., 2009). The present study utilized data from a community-based, socioeconomically diverse sample of Chinese American immigrant families with school-aged children recruited from diverse neighborhoods in a U.S. metropolitan area. Specifically, we examined the unique relations of neighborhood risk and protective factors (socioeconomic disadvantage, criminal events, co-ethnic concentration, ethnic diversity, and reciprocal exchange) to parenting styles controlling for family demographics.

Theoretical Framework Linking Neighborhood to Parenting Styles

According to the Bioecological Theory, neighborhood and parenting are two overlapping contexts that jointly and interactively shape (and are shaped by) child development (Bronfenbrenner & Morris, 2006). The mesosystem connecting neighborhood and parenting captures the mutual influences between the two contexts. There are multiple pathways through which neighborhood characteristics shape parenting (see Cuellar et al., 2015; Shuey & Leventhal, 2019 for reviews). First, the Family Stress Theory (Conger et al., 2010) hypothesizes that socioeconomic disadvantages (including family and neighborhood poverty) increase families’ exposure to stressors, which compromises parent wellbeing and interparental relationships. In turn, elevated parental stress and interparental conflicts can impede parents’ ability to implement supportive and effective disciplines. Although the Family Stress Theory was originally developed to characterize how family socioeconomic disadvantage (such as poverty) shapes child development, it has been applied to inform the research on neighborhood influences on low-income children’s development (Roosa, Jones, Tein, & Cree, 2003). Second, the Family Investment Model theorized that parents in higher socioeconomic status (SES) families or living in affluent neighborhoods have greater access to educational, social, and institutional resources that facilitate or reinforce parental investment in supportive or effective parenting (Duncan et al., 2015). Third, the Social Disorganization Theory hypothesized that parents living in socially disorganized or disengaged neighborhoods lack the social support, role model, and collective efficacy to engage in supportive parenting (Sampson & Groves, 1989). Fourth, the cultural-developmental perspective (Witherspoon et al., 2023) highlights the co-existence of neighborhood socioeconomic disadvantage and cultural assets (e.g., collective efficacy and social capital in ethnically homogeneous neighborhoods). The relations between neighborhood characteristics and parenting can be directional, and individual or family characteristics (e.g., SES, race/ethnicity, immigration status) can shape both families’ selection into neighborhoods and parenting practices (Shuey & Leventhal, 2019).

Parenting styles reflect the constellation of parental attitudes communicated to the child that creates the emotional climate of parenting practices (Darling & Steinberg, 1993). Specifically, authoritative parenting is characterized by high parental warmth, responsiveness, and autonomy support and frequent use of reasoning and positive discipline, whereas authoritarian parenting is characterized by low parental warmth and responsiveness and frequent use of harsh and punitive discipline (Baumrind, 1996). In contrast, intrusive parenting is characterized by manipulative and inhibiting parental behaviors that impair children’s healthy development (Barber, 2002). When compared to European-heritage parents, Asian-heritage parents sometimes showed lower levels of authoritative parenting, and higher levels of authoritarian parenting and intrusive parenting (e.g., Shen et al., 2018; Wu et al., 2002). The cultural group variations in parenting have been attributed to Asian cultural emphasis on training, family hierarchy, and high expectations for children’s academic achievement (Ma, 2020). However, despite cultural variations in “normativeness” of specific parenting styles, within-group studies of Chinese immigrants in Western countries found authoritative (or supportive) parenting to be associated with better psychological adjustment, whereas authoritarian and intrusive (or unsupportive) parenting to be associated poorer adjustment in preschool, school-age, and adolescent children (Cheah et al., 2009; Chen et al., 2014; Kho et al., 2019; Kim, Chen, et al., 2009). Importantly, existing research focusing on within-group heterogeneity in Chinese immigrant families has shown that parent, child, or family characteristics such as SES, parent and child cultural orientations (including language proficiency), parental perceived support and stress, and child behavioral challenges accounted for variations in parenting styles (Cheah et al., 2009; Chen et al., 2014; Kho et al., 2019). Yet, there were few quantitative studies on relations between neighborhood characteristics and parenting styles among Chinese immigrant families.

Links of Specific Neighborhood Factors to Parenting Styles in Chinese Immigrant Families

We focused on two neighborhood risk factors (socioeconomic disadvantage and criminal activity) and three putative neighborhood protective factors (co-ethnic concentration, ethnic diversity, and reciprocal exchange).

Socioeconomic disadvantage.

Neighborhood socioeconomic disadvantage refers to circumstances of deprivation in economic, social, and family resources in a residential neighborhood (Ross & Mirowsky, 2001). According to the Family Stress Theory (Conger et al., 2010) and Social Disorganization Theory (Sampson & Groves, 1989), neighborhood disadvantage can lead to elevated family stress and weakened collective efficacy, which in turn undermine parental use of supportive parenting and effective discipline. Consistent with these theories, prior studies of Chinese/Asian American families found that neighborhood disadvantage was associated with lower authoritative parenting, lower parent involvement and monitoring, and lower perceived social support in parents of school-age to adolescent children (Authors, 2014; Liu et al., 2009; Pei et al., 2023; Wei et al., 2021).

Perceived criminal activity.

Residents’ reports of neighborhood criminal activity reflect their perceived neighborhood danger and violence (Cutrona et al., 2000; Sampson et al., 1997). Consistent with the Family Stress Model (Conger et al., 2010) and Social Disorganization Theory (Sampson & Groves, 1989), parental perceived neighborhood danger have been linked to lower positive parenting, warmth, and behavioral control among ethnically and geographically diverse samples (see Cuellar et al., 2015 for a review). However, opposite relations or null findings have been reported. For example, perceived neighborhood danger was associated with higher use of harsh parenting of school-age children among Mexican American mothers who endorsed greater cultural values of familism (White et al., 2019). This suggests that parents may adjust their parenting styles to protect children from the adverse impact of dangerous neighborhood. To our knowledge, no studies examined the links between neighborhood danger and parenting in Chinese immigrant families.

Co-ethnic concentration and ethnic diversity.

Co-ethnic concentration refers to the concentration of same-ethnicity residents, and ethnic diversity refers to the heterogeneity in race/ethnicity of residents in the neighborhood (Budescu & Budescu, 2012; Juang & Alvarez, 2011). According to the Social Disorganization Theory, neighborhood co-ethnic concentration can promote collective efficacy and enhance social capital for ethnic/racial minority residents, whereas ethnic diversity may undermine collective efficacy (Sampson et al., 1997). However, empirical support has been mixed, with some studies showing benefits of co-ethnic concentration for Latinx and Black youths’ socioemotional development, whereases other studies found developmentally inhibiting effects of neighborhood co-ethnic concentration (see Witherspoon et al., 2023 for a review). To our knowledge, only two prior studies examined neighborhood co-ethnic concentration in Chinese American families. In a community-based sample of Chinese American adolescents, Juang and Alverez (2011) found that parent and adolescent perception of co-ethnic density was related to greater adolescent perception of discrimination. Using an earlier wave of data from the present study, we found that neighborhood Asian concentration to be associated with Chinese American parents’ higher use of authoritarian parenting of school-age children (Authors, 2014). This finding is consistent with the theory that the “normativeness” of specific parenting behaviors in the community can shape the frequency, adaptive function, and children’s interpretations of these practices (Lansford et al., 2018). It is possible that living around other Asian residents reinforces Chinese American parents’ value of firm control, which in turn promotes more authoritarian parenting (Lau, 2010). To our knowledge, no studies examined the quantitative relations between neighborhood ethnic diversity and parenting styles.

Reciprocal exchange.

Neighborhood reciprocal exchange, or the exchange of resources and information among neighbors (Sampson et al., 1999), reflects an aspect of collective efficacy, which combines social cohesion – shared values and trust among neighborhoods – and informal social control – neighborhoods’ willingness to share responsibility for the public good (see Witherspoon et al., 2023 for a review). Consistent with the theory that neighborhood collective efficacy promotes positive parenting by increasing social support, guidance, and modeling for parenting, research has generally shown that collective efficacy (including reciprocal exchange) was related to higher parental warmth, monitoring, and effective discipline in diverse families (see Cuellar et al., 2015 for a review). However, some studies did not find significant relations between social exchange and parenting (Cuellar et al., 2015). Although we found no study on links of neighborhood social relations and parenting in Chinese immigrants, a population-based study of Chinese older immigrants living in the greater Chicago area found that neighborhood cohesion was associated with health benefits such as greater participation in social and cognitive activities, higher sense of mastery, and lower depressive symptoms (Lai et al., 2019; Y. Wang et al., 2022).

When examining the neighborhood-parenting relations, it is valuable to assess parenting from both parents’ and children’s perspectives as each provides unique views about parental behaviors within the household, and a dual-report approach helps identify discrepancy, contributing to our understanding of family dynamics and functioning (see De Los Reyes et al., 2013 and Taber, 2010 for reviews). Particularly, parents may describe their intentions, beliefs, and actions about effective parenting and the overall context of the family, whereas children’s reports are shaped by their lived experiences, developmental stage, and emotional responses to the parenting the receive (Taber, 2010). Moreover, parents typically reported higher levels of positive parenting behaviors (e.g., nurturance and monitoring) compared to children (De Los Reyes et al., 2013). Thus, parent-rated parenting might be more reflective of parents’ perceptions of their own parenting, whereas child-reported parenting might be more reflective of parents’ actual overt parenting behaviors (Taber, 2010). Moreover, in immigrant families, parent-adolescent discrepancies in reports of parenting behaviors may be further contributed by intergenerational acculturation gaps (Fung & Lau, 2010; Hou et al., 2018).

The Present Study

Using a community-based sample of Chinese immigrant families with school-age children, our primary aim was to test the unique relations of neighborhood risk and protective factors to parent- and child-reported parenting styles. Based the literature review, we expected neighborhood disadvantage to be negatively related to authoritative parenting and positively related to authoritarian and intrusive parenting. Based on prior research with Chinese immigrant families (Authors, 2014; Lau, 2010), we expected neighborhood Asian concentration to be negatively related to authoritative parenting and positively related to authoritarian and intrusive parenting. Based on prior research on neighborhood and health outcomes of Chinese older immigrants (Lai et al., 2019; Y. Wang et al., 2022), we expected neighborhood reciprocal exchange to be positively associated with authoritative parenting and negatively associated with authoritarian and intrusive parenting. Due to mixed findings or lack of prior research, we did not have specific hypotheses on links of neighborhood criminal events or ethnic diversity to parenting styles. Given prior research on parent-child discrepancies in reports of parenting, we examined the neighborhood-parenting associations separately for parent- and child-reported parenting styles, although we did not have specific hypotheses regarding the consistency (or incontinency) in the patterns of relations by informants. We considered family SES, child generation, and parents’ length of immigration as covariates (Shuey & Leventhal, 2019).

Methods

Participants

Participants consisted of 239 Chinese American children (51.9% boys, age range = 7 and 10 years, M age = 8.72 years, SD = 0.77), their parents and schoolteachers. Data for the present paper were drawn from the Wave 2 sample (data collected between November 2009 and May 2011) of a longitudinal study of Chinese American immigrant families (Authors, 2014). Detailed sample statistics on parent and family demographics are presented in Table 1. Twenty-three percent of the children were first-generation (i.e., foreign-born) and 77% were second-generation (i.e., U.S.-born and had at least one foreign-born parent). The majority (91.8%) of children lived in two-parent families, and 5.6% were from single-parent (including never married, divorced, or widowed) families. All parents self-identified as Chinese or Chinese American. Most parents (97.9% mothers and 97.5% fathers) were foreign-born: 74.1% of mothers and 72.3% of fathers were born in mainland China, 9.2% and 8.7% in Hong Kong, 2.9% and 2.6% in Taiwan, and 11.7% and 13.9% elsewhere. On average, parents had lived in the U.S. for more than a decade (range = 2 to 43 years, Ms = 13.98 and 22.64 years and SDs = 7.92 and 13.15 for mothers and fathers, respectively). The mean years of parental education were 13.27 (SD = 2.87) years (some education beyond high school) for participating parents. Fifty-seven percent of mothers and 75.8% fathers were employed full-time, 12.9% and 9.8% were employed part-time, and 18.3% and 10.2% were unemployed or homemakers. Families’ annual per capita income was calculated by dividing total family income for the past year by the number of individuals living in the household. The sample’s per capita income ranged from $1,000 to $33,750 (M = $11,909.7, SD = $8,358.8). More than half of the children in the sample (58.5%) were eligible for free or reduced school lunch per parent report. Based on geographic coding of participants’ home addresses at Wave 2, the families lived across 125 census tracts, with a range of 1 to 7 families per tract (the average number of families per tract was 1.91).

Table 1.

Descriptive Statistics of Study Variables

Variables	Statistics of the Study Sample						County-Level Census Statistics
	N	Min-Max	Mean	SD	Skewness	Kurtosis	Mean	SD
Demographics
Child age	238	7.49–10.96	9.20	0.73	0.06	−0.70
Parent age (years)	232	29.76–56.36	41.31	5.20	0.31	−0.33
Years of living in US (years)	223	2.08–63.00	21.95	11.29	0.61	0.56
Parent education	231	2.00–20.00	13.27	2.87	−0.52	1.36
Per capita income	227	1000.00–33750.00	11909.65	8358.75	0.93	−0.05
Parenting styles
Authoritative parenting (parent report)	234	1.59–5.00	4.08	0.57	−1.17	2.81
Authoritarian parenting (parent report)	233	1.00–4.15	1.83	0.51	1.49	3.35
Psychological control (parent report)	234	1.00–5.00	2.08	0.64	1.28	3.35
Authoritative parenting (child report)	239	1.06–4.76	2.85	0.82	−0.04	−0.55
Authoritarian parenting (child report)	239	1.00–4.69	1.89	0.69	1.05	1.30
Psychological control (child report)	239	1.00–4.90	2.05	0.75	0.98	1.01
Neighborhood factors
Criminal events (parent report)	230	0.00–3.90	1.23	0.69	0.45	1.07
Reciprocal exchange (parent report)	231	0.22–4.11	2.42	0.78	−0.49	−0.18
Neighborhood disadvantage composite (census)	239	−1.72–2.07	0.00	0.76	0.39	−0.09
Unemployment rate	239	0.00–28.10	9.59	4.33	0.83	1.42	8.72	4.35
Poverty rate	239	0.90–54.00	15.05	11.46	1.12	0.52	10.59	8.69
% Receiving cash public assistance	239	0.00–19.40	4.02	3.78	1.39	2.00	2.79	3.05
% Less than high school education	239	0.00–60.40	21.70	13.77	0.59	−0.21	13.43	11.61
Median household income ($)	239	13,947.00–215,000.00	67254.72	34237.03	1.06	2.06	84080.60	36303.33
% Female householders	239	1.31–12.08	5.77	2.43	0.19	−0.83	5.23	2.52
Neighborhood diversity index (census)	235	0.19–0.78	0.61	0.13	−1.31	1.78	0.55	0.15
% Asian residents (census)	239	3.00–92.00	37.22	19.69	0.46	−0.33	23.56	18.67

Note. The county-level census statistics were calculated for the ethnic Chinese alone population across the seven counties where the sample was recruited.

Study Procedures

Families were recruited from a metropolitan area in western United States. To recruit a socioeconomically diverse sample, a variety of recruitment strategies were employed, including distributing flyers to Asian or Chinese American families in their communities (e.g., Chinatown shopping centers, Asian grocery stores, and Chinese American festival events), establishing partnerships with public and private schools with large Asian student populations to attend their open house events, and seeking referrals from community organizations serving the Chinese American population (e.g., after-school programs, churches, nonprofit organizations). At the recruitment events, research assistants distributed project fliers to parents and asked parents who expressed an interest to fill out a contact sheet that asked for their phone numbers. Trained bilingual research assistants conducted phone screening interviews with interested parents to determine eligibility. To be eligible for the study, the children must meet the following criteria: (a) the child was in third or fourth grade at the time of screening; (b) the child lived with at least one of her/his biological parents; (c) both biological parents identified as ethnic Chinese; (d) the child was either first generation (born outside the U.S.) or second generation (born in the U.S. with at least one foreign-born parent), and (e) both the parent and child were able to understand and speak some English or Chinese (Mandarin or Cantonese).

The family (target child and one parent) visited the university campus and participated in a 2.5-hour lab assessment. Of this sample, 192 children had mothers as the participating parent, whereas 44 children had fathers as the participating parent. Upon parental consent and permission and child’s assent to participate, the parent and the child were interviewed in separate rooms and individually administered the parent or child questionnaires by two trained bilingual assessors. Participants completed the questionnaires in their preferred language (i.e., English, Simplified or Traditional Chinese). Parents were paid $50, and children received a small prize. With parental consent, the child’s classroom teacher was invited to fill out a teacher survey by mail. Teachers were paid $20 per packet. The study procedures were approved by University of California Berkeley Human Research Protection Program (Protocol No. 2010–11-2570).

Measures

Neighborhood Socioeconomic Disadvantage (Census Data)

Following methods widely used in neighborhood research (e.g., Roosa et al., 2005), participants’ addresses were matched with geocodes and census data by using a publicly available application program interface that integrates with R (Macfarlane & Kressner, 2018). The tract-level data from the 2010 U.S. Census and 2010 American Community Survey were pulled to assess on neighborhood disadvantage variables, which included % unemployment, % poverty, % receiving public assistance, % single-parent households, % of individuals with less than high school education, and median household income. These indicators were highly correlated in the present sample (rs > .70). Thus, a composite index of neighborhood disadvantage was computed by averaging the standardized scores of the six indicators (α = .85 in the present sample).

Neighborhood Ethnic Density and Neighborhood Diversity (Census Data)

A single variable was drawn from 2010 U.S. Census data to represent neighborhood Asian density: % of residents who identified themselves as being of Asian-only descent. Twenty-six percent of families in the present sample lived in ethnic enclaves with Asian densities above 50%, and 21% lived in neighborhoods with Asian densities lower than 20%. To characterize ethnic diversity in neighborhoods, we computed a general variance (GV) diversity index using the proportion of residents in eight racial/ethnic groups (i.e., white, black or African American, American Indian and Alaska Native, Asian, Native Hawaiian and other Pacific Islander, other race, and two or more races) drawn from 2010 U.S. Census. The GV is computed as $1-\sum _{i=1}^c{P}_i^2$, where C is the number of groups and $P_i$ is the proportion of residents in each ethnic group. A higher value on the diversity index reflects a higher degree of heterogeneity in neighborhood ethnicity (Budescu & Budescu, 2012).

Neighborhood Criminal Events (Parent Report)

Parents reported the frequency of criminal events in their lived neighborhoods using the 10-item Neighborhood Criminal Event Scale (Kim, Nair, et al., 2009). Parents rated how often specific events happened in their neighborhood in the past year (e.g., “violent crimes including stabbings, shootings, violent assaults”, “people taking others’ wallets or purses”, “drug use and dealings in public”) using a 4-point Likert scale ranging from 1(rarely or none of the time) to 4 (most or all of the time). A composite score of neighborhood criminal events was formed by averaging item scores (α = .93 in the present sample).

Neighborhood Reciprocal Exchange (Parent Report)

Parents completed a 9-item Neighborhood Exchange Scale adopted from the Neighboring Activities scale (Unger & Wandersman, 1983) and the Reciprocated Exchange Scale (Sampson et al., 1999). The measure included items assessing how well embedded the family was in their neighborhood (e.g., “How many of the names do you know of the people who live in your neighborhood?”, “How many people in your neighborhood would you consider close friends?”) and items assessing the amount of sharing and interactions among neighbors (e.g., “How many people in your neighborhood would you feel comfortable asking to help you with a home repair?”, “How many people in your neighborhood would you feel comfortable asking for a ride when your car is not working or if you do not have a car?”). For each item, parents used a 5-point Likert scale ranging from 1(None of them) to 5 (All of them). A composite score of neighborhood reciprocal exchange was formed by averaging the item scores (α = 0.89 in the present sample).

Parenting Styles (Parent and Child Report)

Authoritative and authoritarian parenting.

Parent and children completed the Parenting Styles and Dimensions Questionnaire (Robinson et al., 1995). The authoritative scale includes four subscales: warmth/acceptance, reasoning/induction, encouragement of democratic participation, and easy-going/responsiveness. The authoritarian scale included four subscales: non-reasoning/punitive strategies, corporal punishment, directiveness, and verbal hostility. Participants rated each item using a 5-point Likert scale ranging from 1 (never) to 5 (always). The Chinese version of PSDQ has been validated in previous studies of Chinese-heritage families (e.g., Wu et al., 2002; Q. Zhou et al., 2008). We tested the factor structure of parent- and child-reported parenting styles in the present sample (see Supplementary Materials). In the present sample, the αs were .88 for parent- and .90 for child-reported authoritative parenting (17 items), and .81 for parent- and .87 for child-reported authoritarian parenting (14 items).

Intrusive parenting.

Parent and children rated intrusive parenting using the Maternal Psychological Control Scale (Olsen et al., 2002). The scale was suitable for both mothers and fathers due to its gender-neutral wording for the parent scale, and the items were reworded to be appropriate for children. The scale includes items on personal attack, erratic emotional behavior, guilt induction, and love withdrawal. Participants rated the items using a 5-point Likert scale from 1 (never) to 5 (always). We tested the factor structure of parent- and child-reported parenting styles in the present sample (see Supplementary Materials). In the present sample, the αs were .84 for parent- and .84 for child-reported intrusive parenting (10 items).

Data Analytic Plan

First, we examined the descriptive statistics for the neighborhood and parenting variables and compared the objective neighborhood indicators of our sample to the county-level statistics for ethnic Chinese alone population across the seven counties where the sample was recruited using t-tests. Second, to select covariates for the main analyses testing the relations between neighborhood variables and parenting styles, we examined zero-order correlations between family demographic and neighborhood and parenting variables. We also examined the correlations among neighborhood variables. Third, to test the unique relations of neighborhood factors and parenting styles controlling for family demographics, we tested a path analysis model (Figure 1). Based on the literature on parent-child discrepancies in reports of parenting (De Los Reyes & Ohannessian, 2016), the six parenting variables were tested as separate outcomes rather than grouping by latent factors. The model was tested using MPlus 8.6 (Muthén & Muthén, 2017) with full information maximum likelihood (FIML) estimation to handle missing data. Given that participants were clustered in census tracts, the model was estimated using the TYPE = COMPLEX feature in Mplus 8.6, which accounts for the nonindependence of observations when computing standard errors. The sample size was determined based on power analyses for testing the fit of the hypothesized SEM model for the larger longitudinal study. Data and study materials are available upon request. The study was not preregistered.

Figure 1.

Path Analysis Model Testing Unique Relations of Neighborhood Characteristics to Parenting Styles

Notes. In this model, each of the six neighborhood variables is specified to predict each of the six parenting style variables. The effects of covariates (child generation, parent education, parent’s length of stay in the U.S.) on all parenting variables were controlled in the model. Only significant paths were shown. The numbers above parentheses are unstandardized path coefficients, and the numbers inside the parentheses are standardized path coefficients. *p < .05, **p < .01, ***p < .001.

Results

Descriptive Statistics and Comparison of Neighborhood Indicators of the Sample with County-Level Census Statistics for the Ethnic Chinese Population

Descriptive statistics of study variables are presented in Table 1. With cutoff values of 2 and 7 for skewness and kurtosis, respectively, all study variables were normally distributed. Families in our sample lived in neighborhoods with an average density of 38.29% Asian residents (SD = 19.43), which was significantly higher than the average of sampling counties (23.56%), t(238)=11.72, p<.001. The average of neighborhood diversity indices of our sample (M=0.61, SD = 0.13) was significantly higher than the average of sampling counties (M=0.55), t(238)=7.72, p<.001. Across all indicators of neighborhood socioeconomic disadvantage, families in our sample resided in more socioeconomically disadvantaged neighborhoods compared to the average of sampling counties. Compared to the county-level averages for ethnic Chinese (alone) population across the 7 counties in which the sample was drawn, our participating families resided in neighborhoods with a higher unemployment rate (9.59% vs. 8.72%), t(238) = 3.12, p = .002, higher poverty rate (15.05% vs. 10.59%), t(238) = 6.02, p<.001, higher percentage of residents receiving public assistance (4.02% vs. 2.79%), t(238) = 5.04, p<.001, higher % of residents with less than a high school education (21.7% vs. 13.43%), t(238) = 9.28, p<.001, lower median household income ($67,254 vs. $84,081), t(238) = −7.60, p<.001, and higher % of female householders (5.77% vs 5.23%), t(238) = −54.96, p<.001.

Correlation Analyses

Correlations with family demographics.

Correlations among family demographic, neighborhood, and parenting variables were reported in Supplementary Table 1. Three demographic variables were correlated with both neighborhood and parenting variables. Compared to 1^st-generation children, 2^nd-generation children lived in neighborhoods with lower co-ethnic concentration, fewer criminal activities, and lower reciprocal exchanges. The 2^nd-generation children also rated their parents lower on intrusive parenting than 1^st-generation children. More educated parents lived in neighborhoods with lower disadvantage, lower co-ethnic concentration, fewer criminal activities, and lower reciprocal exchanges compared to less educated parents. More educated parents also scored higher on parent- and child-rated authoritative parenting than less educated parents. Finally, parents who lived in the U.S. longer lived in neighborhoods with lower disadvantage, lower co-ethnic concentration, fewer criminal activities, and lower reciprocal exchanges compared to those with a shorter length of stay in the U.S. Moreover, parents who lived in the U.S. longer scored lower on child-reported intrusive parenting than parents with a shorter length of stay in the U.S. Thus, based on the recommendations of Steiner et al. (2010), we included child generation, parental education, and parent’s length of stay in the U.S. as covariates for subsequent analyses.

Correlations among neighborhood factors.

Neighborhood disadvantage was positively correlated with ethnic diversity (r = .27, p<.001) and perceived criminal events (r = .40, p < .001), meaning that neighborhoods with higher levels of socioeconomic disadvantage were more ethnically diverse and had more criminal activities. Neighborhood Asian concentration was negatively correlated with ethnic diversity (r = −0.34, p < .01), indicating that neighborhoods with high concentration of Asian residents were less ethnically diverse than neighborhoods with low concentration of Asian residents. Neighborhood reciprocal exchange was positively correlated with criminal events, such that the neighborhoods with higher reciprocal exchanges also had more criminal activities than those with lower reciprocal exchanges.

Path Analysis Testing the Unique Relations of Neighborhood Characteristics to Parenting

Relations between neighborhood risk factors and parenting.

Based on the recommended cutoffs of comparative fit index (CFI) ≥ .95, root-mean-square error of approximation (RMSEA) ≤ .06, and standardized root-mean-square residual (SRMR) ≤ .08 (Hu & Bentler, 1999), the path analysis model fit the data well, χ²(df = 9, N = 223) = 7.30, p = 0.61, CFI = 1.00, RMSEA < 0.001, SRMR = 0.025. The effects of selected covariates (child generation, parental education, and parent’s length of stay in the U.S.) on all parenting style variables were controlled in the model. Among the demographic covariates, parental education was positively related to both parent- and child-reported authoritative parenting (βs = 0.17 and 0.19, ps < .05 and .01), and parents’ length of stay in the U.S. was negatively related to child-reported intrusive parenting (β = −0.17, p < .05). Controlling for the covariates, perceived neighborhood criminal events was positively related to parent-rated intrusive parenting (β = 0.15, p < .05). In addition, we conducted analyses to examine the relations between individual neighborhood factors and parenting styles in separate path analysis models. The results are reported in Supplementary Table 2. Interestingly, different from the full path model, neighborhood disadvantage was positively related to child-reported intrusive parenting. Neighborhood criminal event did not predict any parenting style variables in the individual path model.

Relations between neighborhood protective factors and parenting.

The full path model showed that neighborhood reciprocal exchange was positively related to parent-rated authoritative parenting (β = 0.18, p < .01), and negatively related to child-rated authoritarian parenting (β = −0.14, p < .05) and intrusive parenting (β = −0.24, p < .01). Moreover, neighborhood Asian concentration was positively related to child-rated authoritarian parenting (β = 0.19, p < .01) and intrusive parenting (β = 0.18, p < .01). Consistent with the full path model, separate path analysis models showed that neighborhood Asian concentration was positively associated with child-rated authoritarian parenting and intrusive parenting. Neighborhood reciprocal exchange was positively associated with parent-reported authoritative parenting, and negatively associated with child-reported intrusive parenting.

Discussion

This study tested the unique relations of neighborhood risk and protective factors to parents’ and school-age children’s perceptions of parenting styles in a sample of Chinese immigrant families living in diverse neighborhoods in a U.S. metropolitan area. We found that family socioeconomic and immigration factors were associated with neighborhood characteristics. Immigrant families with lower socioeconomic status, a shorter immigration history, and 1^st-generation children were more likely to live in neighborhoods with higher socioeconomic disadvantage, more criminal activities, as well as higher co-ethnic concentration and higher reciprocal exchanges compared to families with higher-SES, a longer immigration history, and 2^nd-generation children. Controlling for family demographics, neighborhood factors had divergent relations to parenting styles. Neighborhood reciprocal exchange was positively associated with authoritative parenting (parent report) and negatively associated with authoritarian and intrusive parenting (child report). In contrast, Asian concentration was positively associated with authoritarian and intrusive parenting (child report). Neighborhood criminal activity was positively associated with intrusive parenting (parent report).

The sample was recruited from a metropolitan area with overall higher levels of racial/ethnic diversity and Asian concentration compared to rest of the country (Menendian et al., 2021). This metropolitan area is also marked by high levels of “intra-municipal” racial residential segregation (i.e., the degree of racial residential segregation internal to a city or municipality; Menendian et al., 2021). Indeed, we observed significant variations in neighborhood socioeconomic indicators and racial/ethnic compositions in this sample, with families living in neighborhoods ranging from high co-ethnic concentration (e.g., Chinatown) or low ethnic diversity to low co-ethnic concentration or high ethnic diversity. The significant associations found between family demographic and neighborhood measures are consistent with the literature on neighborhood selection factors (Shuey & Leventhal, 2019). Prior research indicated that Chinese American immigrant families’ reasons for migration are associated with their post-migration SES, such that those families who migrated solely for betterment reasons had higher parental education and income than those who migrated to join family or other reasons (X. Wang et al., 2023). Moreover, because Chinese immigrants with lower SES tend to have limited English proficiency compared to those with higher SES (Chen et al., 2014), they may choose to live in neighborhoods with more Chinese-speaking residents and/or family members (e.g., settling in neighborhoods with higher co-ethnic concentration). Interestingly, in our sample, parents’ perceived neighborhood reciprocal exchange was positively associated with their perceived criminal events, consistent with the cultural-developmental perspective that risk and protective factors co-exist in neighborhoods (Witherspoon et al., 2023).

Relations between Neighborhood Risk Factors and Parenting.

While neighborhood risk and protective factors co-exist and are interrelated, they showed differential relations to parenting styles. Although neighborhood socioeconomic disadvantage did not predict parenting styles in the full model controlling for other neighborhood factors (Figure 1), neighborhood disadvantage was associated with higher child-reported intrusive parenting controlling for family demographics in the individual path model (Supplementary Table 2). This is in line with the Family Stress Theory (Conger et al., 2010) and previous findings on neighborhood disadvantage and unsupportive parenting in Chinese immigrant families (Authors, 2014; Liu et al., 2009; Wei et al., 2021). However, no significant relations were found between neighborhood disadvantage and other parenting style variables. This might be due to the fact that families in our sample resided in neighborhoods with low to moderate disadvantage. Few families (3.4%) in this sample were living in a high poverty-rate neighborhood (defined as ≥ 40%). The association between neighborhood disadvantage and parenting styles may be stronger in samples with a greater proportion of families living in highly disadvantaged neighborhoods. The association between neighborhood disadvantage and parenting styles may also be attenuated by cultural resources unique to Chinese immigrant families (e.g., child-rearing support from grandparents). In Asian American families, grandparents tend to live with their adult children and assume a high level of caregiving (Lin et al., 2015). Future research can investigate whether grandparent involvement in childrearing moderates the associations between neighborhood disadvantage and parenting in Chinese American immigrant families.

Although we did not have specific hypotheses on the relation between neighborhood criminal events and parenting styles due to mixed findings in the literature, we found parent-rated neighborhood criminal events were positively associated with parent-rated intrusive parenting. This is somewhat consistent with the Family Stress Theory that neighborhood danger heightens parental stress and impedes their abilities to engage in supportive parenting (see Cuellar et al., 2015; Shuey & Leventhal, 2019 for reviews). Moreover, it is also possible that Chinese immigrant parents living in unsafe neighborhoods increase intrusive control to protect children from the potential harm of high-crime neighborhoods, similar to the finding from White et al. (2019) with Mexican American families. However, no associations were found between neighborhood criminal events and child-rated parenting styles. This suggests that the impact of neighborhood safety on parenting might manifest more in the domains of parents’ subjective experiences such as parental distress or parental intentions to protect children from dangerous neighborhoods than actual parenting behaviors experienced by children.

Relations between Neighborhood Protective Factors and Parenting.

We found that neighborhood reciprocal exchange was uniquely associated with both parents’ and children’s reports of parenting styles. Specifically, parents’ higher perceived neighborhood reciprocal exchange was associated with higher parent-reported authoritative parenting and lower child-report authoritarian and intrusive parenting. These results are consistent with prior research testing the Social Disorganization Theory conducted with other samples (see Cuellar et al., 2015; Shuey & Leventhal, 2019 for reviews), as well as the studies on neighborhood cohesion and health outcomes of Chinese older immigrants living in metropolitan areas (Lai et al., 2019; Y. Wang et al., 2022). The consistency in findings across parent- and child-reported parenting suggests that neighborhood social cohesion is associated with benefits in not only parents’ own psychological experiences but also their actual parenting behaviors. A reliable, trusted neighborhood network can alleviate parental stress and serve as a source of social support and parenting guidance, both of which can promote authoritative/supportive parenting and/or reduce unsupportive (authoritarian and intrusive) parenting in Chinese immigrant families.

We expected neighborhood co-ethnic concentration to be associated with unsupportive parenting and found some support. Specifically, neighborhood Asian concentration was positively associated with child-reported authoritarian and intrusive parenting in the full path model controlling for family demographics and other neighborhood factors. However, neighborhood Asian concentration was unrelated to parent-reported parenting styles. Notably, consistent with findings from other immigrant samples (Fung & Lau, 2010; Hou et al., 2018), parent- and child-rated parenting styles were unrelated with each other, highlighting discrepancies in parents’ and children’s perceptions of parenting behaviors (De Los Reyes & Ohannessian, 2016). It is possible that perceived cultural norm for intrusive parenting is higher in neighborhoods with higher concentration of Asian residents. However, due to intergenerational acculturation gaps in immigrant families (Lui, 2015), children might be more sensitive to the potential aversive psychological effects of intrusive parenting than their parents. Thus, while parents might perceive their own parenting styles to be typical compared to other parents in their social network, children might perceive their parents to be more intrusive compared to the parents of their peers. Therefore, difference in reference group might partly contribute to parent-child discrepancy in perceptions of parenting in immigrant families.

We had no specific hypothesis on relations between neighborhood ethnic diversity and parenting and did not find any significant associations between neighborhood ethnic diversity and parenting styles. In the present sample, neighborhood ethnic diversity was positively associated with neighborhood disadvantage and negatively associated with neighborhood co-ethnic concentration. Thus, the neighborhoods higher on ethnic diversity were low on both socioeconomic/structural resources and cultural resources typically found in ethnic enclaves (e.g., heritage language schools and afterschool programs; M. Zhou & Kim, 2006). Therefore, the co-existence of risk and protective factors in ethnically diverse neighborhoods might have offset the potential beneficial effects of cultural diversity on parenting.

Limitations and Conclusions

This study had several limitations. First, our sample resided in neighborhoods with low to moderate (but not severe) disadvantage, which might have attenuated the strength of associations between neighborhood disadvantage and parenting styles. Second, the cross-sectional design of the study makes it difficult to determine mediating mechanisms and potential bidirectionality in the associations between neighborhood factors and parenting outcomes. Third, parenting style measures used in the study were developed from research conducted with European American families, so they may not capture culturally unique parenting styles among Asian families, such as training (Chao, 1994; Y. C. Wang & Supple, 2010). Future studies should consider integrating measures that evaluates both universal and culturally specific facets of parenting. Fourth, subjective neighborhood measures (criminal events and reciprocal exchanges) were only rated by participating parents (rather than using a random sample of residents living in the neighborhoods). Thus, the shared methods variance might have inflated the link between perceived neighborhood factors and parenting (Cuellar et al., 2015). Future studies should use random sampling to assess residents’ perceptions of neighborhood. Fifth, the study’s sample size limited the statistical power for testing moderation hypotheses (e.g., whether family or child factors moderate the links of neighborhood to parenting). Furthermore, the study used a community-based sample recruited from a metropolitan area with a high concentration of Asian American residents. Thus, the findings might not generate to Chinese American families living in geographic areas with distinct sociocultural contexts.

This study was one of the first to disentangle the relative contributions of risk and protective neighborhood factors to within-group variations in parenting styles among Chinese immigrant families. The findings have revealed nuances in heterogeneity in parenting practices among Chinese immigrants by highlighting the complex relations of neighborhood structural, social, and cultural characteristics to parenting behaviors and parent-child relationship dynamics in Chinese immigrant families. Future studies should explore the potential mediating role of parenting in the relations between neighborhood risk and protective factors and the adjustment of immigrant youth. The findings have implications for school- and community-based programs and services serving low-income Chinese immigrant families. Specifically, greater structural and community resources or supports from schools and local organizations can benefit these families and may buffer the adverse effects of neighborhood danger. Moreover, community-based programs and services that foster and promote social exchanges among neighbors and/or access to culture- and language- competent resources on parenting can build resilience and strengthen parent-child relationships in Chinese immigrant families living in socioeconomically disadvantaged neighborhoods.