SEARCH

SEARCH BY CITATION

Keywords:

  • Psychopathology;
  • Income;
  • Education;
  • Family stress;
  • Parent–child conflict

Abstract

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

Using a sample of 156 Chinese children aged 2–3 years and their parents, this study examined the effects of socio-economic status, specifically family income and parental education, on the children's internalizing and externalizing psychopathology and whether these effects were mediated by mother–child and father–child conflict. Results indicated that family income, maternal education and paternal education all negatively predicted externalizing symptoms. Income also negatively predicted internalizing symptoms among boys but not girls. Maternal education negatively predicted internalizing symptoms among girls but not boys. The effects of income on psychopathology were fully mediated by mother–child and father–child conflict. In contrast, the effects of education were not mediated or only partially mediated by conflict. Findings are discussed in the framework of the family stress model.

The impact of socio-economic status (SES) on children's development has been intriguing to developmental and clinical psychologists. The extant research in the field, conducted mainly in Euro-American settings, has converged in suggesting that low SES has an adverse influence on children's physical, cognitive and socio-emotional development (Bradley & Corwyn, 2002). During the past decades, economic changes (e.g., increasing income inequality) in many nations have enhanced the ongoing interest in SES as a global construct. With a full-scale move towards a market economy since the early 1980s, China has experienced historically unprecedented growth in economy. This achievement, however, is accompanied with a sharp increase in social and economic inequality. According to the World Bank (2007), China's income inequality as measured by the Gini coefficient rose from 0.30 in 1982 to 0.47 in 2005, overtaking the recognized warning level of 0.40. The present study seeks to add to the existing literature by examining the effects of SES on internalizing and externalizing psychopathology among 2–3-year-old Chinese children and the mechanisms underlying these effects. Such data are crucial because the impact of SES on child psychopathology has rarely been tested in Chinese societies.

THE INFLUENCE OF SES ON PSYCHOPATHOLOGY

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

Socio-economic status is a construct that captures at least three dimensions of social position, namely, prestige, power and economic well-being. There is general consensus among contemporary investigators that income, education and occupation status are three quantitative indicators that provide reasonably good coverage of SES (Bradley & Corwyn, 2002; Conger & Donnellan, 2007).

In the extant literature, it is well established that growing up in low-SES families places children at risk for health problems (e.g., growth retardation, physical illness; Adler et al., 1994) and cognitive disadvantages (e.g., impairment in language, intelligence and academic performance; Lugo-Gil & Tamis-LeMonda, 2008). In comparison, the impact of SES on socio-emotional development (e.g., psychopathology, social competence) is not as consistent as that on physical and cognitive development, especially in young children (Bradley & Corwyn, 2002).

For very young children, earlier studies provide no evidence of a relation between SES and psychopathology (Earls, 1980; Richman, Stevenson, & Graham, 1975). Yet recent large-scale studies suggest that SES has an influence on psychopathology among 2–3-year-olds (Glick, Hanish, Yabiku, & Bradley, 2012; for a review, see Conger, Conger, & Martin, 2010). For instance, using data from the NICHD Study of Early Child Care, a study found that low income was associated with internalizing and externalizing psychopathology at age 3 among children living in poverty (Mistry, Biesanz, Taylor, Burchinal, & Cox, 2004). Similarly, another study, using data from the National Early Head Start Research and Evaluation Project, found that SES, measured by income, welfare receipt and maternal education, had an impact on low-income children's aggression at age 3 (Mistry, Biesanz, Chien, Howes, & Benner, 2008).

The literature on SES has mainly focused on the economic dimension (i.e., income). As noted by Conger et al. (2010), “the two other usual quantitative markers of SES, education and occupational status, have largely been ignored, used as control variables, or combined with income to construct an overall index of SES” (p. 698). Psychologists have called for further research to isolate the unique influences of these two indicators of SES, especially education (Conger et al., 2010; Duncan & Magnuson, 2003). The few studies linking education with child development have paid almost exclusive attention to maternal education, failing to isolate the unique effect of paternal education. Moreover, most of the extant studies have been carried out in Euro-American settings, with little research on SES and child psychopathology in non-Western societies, such as China. Consequently, the present study tests whether family income, maternal education and paternal education are associated uniquely with internalizing and externalizing psychopathology among Chinese young children.

THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

The extant research further suggests that much of the effect of SES on child development can be mediated through processes proximal to the child. This study draws upon tenets of the family stress model (FSM; Conger & Donnellan, 2007) and tests whether conflict between parents and children mediates the effects of income and education on child psychopathology.

The FSM posits that SES disadvantage influences child development indirectly through a series of intervening family processes, including parents' emotional distress, parenting practices, and parent–child and inter-parental relationships (Conger & Donnellan, 2007). The model is initially used to explain how financial difficulties have an adverse effect on family and child well-being. As suggested by the FSM, low-income parents, as compared to high-income parents, may experience more parenting stress, engage in more hostile and aggressive parenting behaviour and have more conflictual relationships with their children. Consequently, these children are more likely to develop psychopathology symptoms (Conger & Donnellan, 2007). Recently, Conger et al. (2010) have proposed that education should have an influence similar to income. Presumably, poorly educated parents are less likely to place a priority on the needs of their child than well-educated parents. They may be less sensitive and responsive to the needs and establish more conflictual relationships with the child, which, in turn, may increase the child's risk for developmental problems. Recent literature suggests that the FSM provides a good model for understanding how income and education influence positive or negative outcomes. For example, Mistry et al. (2004) showed that poor families' income predicted their family processes (e.g., maternal depression, maternal sensitivity), which in turn predicted children's social (i.e., internalizing and externalizing psychopathology, social competence) and cognitive outcomes at age 3. In another study, Huston and Aronson (2005) found that maternal education predicted child outcomes indirectly through family processes (i.e., maternal sensitivity) as well.

In the existing research on the FSM, however, “family stress” or “family process” is nearly always operationalized using measures focused solely on mothers or on the family as a unit, failing to assess processes related to father–child relationships independently. Thus, research has rarely assessed whether the quality of father–child relationships mediates the effect of SES on developmental outcomes, although it has increasingly documented the differential role that fathers play in child development (Lamb, 1997). In this study, we measure both mother–child and father–child conflict in order to enhance our understanding of the mediating role of relationships with both parents.

THE PRESENT STUDY

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

The main goal of this study is to examine the effects of family income and parental education on psychopathology among Chinese young children and whether these effects are mediated by mother–child and father–child conflict. To this end, we used a sample of 2–3-year-old children and their parents in Beijing, China. On the basis of the extensive Western literature, we assume that the increased economic and social inequality in China may contribute to the variations in child psychopathology and family processes. Such inequality has been observed in both rural and urban areas. For example, the Gini coefficient increased from 0.20 in 1993 to 0.32 in 2000 in urban Beijing (Wang, Long, & Zhao, 2003). Hence, there is good reason to believe that the effects of income and education on family and child functioning would be evident in our sample. Indeed, research has documented the substantial impact of income and education on Chinese children's social competence skills and family environments (Zhang, 2012; Zhang, Chen, Zhang, & Sun, 2009). We thus hypothesize that income, maternal education and paternal education would be negatively associated with Chinese children's internalizing and externalizing psychopathology. Furthermore, these associations are expected to be mediated by both mother–child and father–child conflictual relationships. Finally, because the impact of SES on child outcomes has shown similarities (Mistry et al., 2004) as well as differences (Huston et al., 2001) across the child's gender, we test gender differences to evaluate whether gender moderates the effects of income and education on psychopathology.

METHOD

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

Participants

Participants were recruited from four urban preschools (three public and one private) in Beijing, China. Of the 185 2–3-year-old children solicited, 156 children (84.3%; 75 boys, 81 girls; M = 40.9 months old, SD = 4.1 months) participated with their fathers and/or mothers. Most children (95.5%) were only children, and the others had one or two siblings. Most parents (96.8%) were living together, and 3.2% were separated and living alone or divorced.

Measures

Child psychopathology

Mothers rated their child's internalizing and externalizing symptoms using the Child Behavior Checklist/2–3 (CBCL/2–3; Achenbach, 1992). The checklist is a standardized measure of psychopathology symptoms in children aged 2–3 years and proves to be reliable and valid among Chinese children (Liu, Wu, & Yao, 2003). In this study, 41 items measuring withdrawal, anxious, aggressive and delinquent symptoms were used. Two major groupings of symptoms were identified: “internalizing” (withdrawal and anxiety; 19 items) and “externalizing” (aggression and delinquency; 22 items). Cronbach's α coefficients were .70 and .75 for internalizing and externalizing symptoms, respectively. Higher scores indicate more symptoms.

Parent–child conflict

Fathers and mothers independently rated their own conflictual relationships with their child on a 5-point scale, ranging from 1 (definitely does not apply) to 5 (definitely applies), both using the conflict subscale in the Chinese version of Pianta's (1992) Child–Parent Relationship Scale (CPRS; Zhang & Chen, 2010). The CPRS has been used with children aged 2–5 years and has proven to be reliable and valid in Chinese families (Zhang & Chen, 2010). The conflict subscale contains 12 items and measures a parent's perception of negativity and conflict with the child (e.g., “My child sees me as a source of punishment and criticism”). Cronbach's α coefficients were .76 for mother–child conflict and .77 for father–child conflict. Higher scores indicate more conflictual relationships.

Family SES

Three indicators were used to characterize family SES: family income, maternal education and paternal education. A total of 125 mothers reported monthly income per capita in their family which ranged from ¥500 to ¥15,000 (≤¥1000: 19 families [15.2%]; >¥1000 & ≤¥2000: 57 families [45.6%]; >¥2000 & ≤¥3000: 26 families [20.0%]; >¥3000 & ≤¥5000: 19 families [15.2%]; >¥5000: 4 families [3.2%]). ¥, or RMB, is the primary unit of currency in China. The mean per capita monthly income was ¥2590 (SD = ¥2169), which was approximately equivalent to $320 (SD = $268) at the time of the survey. On the basis of per capita monthly income of urban household in Beijing in 2005 (¥1628; Beijing Statistics Bureau, 2005), we classified the 125 families into high-income (n = 81) and low-income (n = 44) families.

Mothers also provided information on the education of both parents. Maternal education (n = 155) and paternal education (n = 156) were coded as follows: 5 = master's degree or above (26 fathers [16.7%]; 13 mothers [8.4%]), 4 = bachelor's degree (45 fathers [28.8%]; 47 mothers [30.3%]), 3 = vocational college degree (35 fathers [22.4%]; 42 mothers [27.1%]), 2 = high school or vocational school degree (35 fathers [22.4%]; 34 mothers [21.9%]) and 1 = secondary school or below (15 fathers [9.6%]; 19 mothers [12.3%]).

Child demographic characteristics

Two variables were used to represent the child's demographic characteristics: age and gender. Age was measured in terms of months from birth to the time of the survey. It ranged from 27 months to 47 months (M = 40.9 months, SD = 4.1 months). For gender, boys were given a code of “1,” and girls were given a code of “−1.”

Procedure

All data were collected in August 2005. Families were contacted by mail and asked if they would like to participate in a study of child development. After agreeing to participate, mothers and fathers independently filled out several questionnaires, which were delivered to them together with the consent form, and they returned the completed questionnaires to children's teachers. The data collection was finished within a month. The recruitment and data collection procedures were approved by the institutional review board of the author's university.

RESULTS

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

Descriptive statistics and inter-correlations among variables

Table 1 presents the descriptive statistics and inter-correlations among all variables. Income correlated negatively with externalizing symptoms and conflict in both mother–child and father–child relationships. Maternal education correlated negatively with father–child conflict, and internalizing and externalizing symptoms. Paternal education correlated negatively with mother–child conflict and externalizing symptoms. Both mother–child and father–child conflict correlated positively with internalizing and externalizing symptoms.

Table 1. Descriptive statistics and inter-correlations among all variables
VariablesMSD123456789
  1. a

    Effect coded: 1 = high income, −1 = low income.

  2. b

    Effect coded: 1 = boy, −1 = girl.

  3. * p < .05; **p < .01; ***p < .001.

SES           
1. Family incomea0.300.96        
2. Maternal education3.011.170.39***       
3. Paternal education3.211.240.26**0.75***      
Child characteristics           
4. Age40.904.12−0.040.010.02     
5. Sexb−0.041.00−0.02−0.08−0.060.00    
Psychopathology           
6. Internalizing10.395.23−0.15−0.17*−0.12−0.070.08   
7. Externalizing12.207.45−0.24**−0.28***−0.28***−0.08−0.18*0.46***  
Parent–child conflict           
8. Mother–child conflict26.616.84−0.19*−0.14−0.16*−0.020.000.39***0.52*** 
9. Father–child conflict28.836.56−0.27**−0.16*−0.120.100.040.32***0.39***0.54***

Effects of income and education on psychopathology

To examine the unique effects of income and education on internalizing and externalizing psychopathology, we conducted a series of multiple regressions that included both income and education in the models. Because the high correlation between maternal and paternal education (r = .75, p < .001) caused a multi-collinearity problem (i.e., a tolerance less than 0.20 and a VIF above 5) when they were incorporated in one model, separate models were tested for maternal and paternal education. The child's age and sex (effect coded: 1 = boy, −1 = girl) were first entered as covariates in the regressions. Then, the main effects of income (effect coded: 1 = high income, –1 = low income) and education (centred) were entered. Finally, the income × sex and education × sex interactions were entered. Table 2 presents the regressions for the models that included maternal education. Table 3 presents the regressions for the models that included paternal education.

Table 2. Effects of family income and maternal education on psychopathology
Independent variableInternalizing (F(6, 117) = 2.16, p = .05)Externalizing (F(6, 117) = 3.27, p = .005)
R2ΔR2B (SE)βR2ΔR2B (SE)β
  1. a

    Effect coded: 1 = boy, −1 = girl.

  2. b

    Effect coded: 1 = high income, −1 = low income.

  3. * p < .05; **p < .01.

(constant)  11.37 (4.87)   20.08 (6.59) 
Step 1.00.00  .03.03  
Child age  −.02 (.12)−.02  −.17 (.16)−.09
Child sexa  .49 (.50)  .09  1.27 (.68).17
Step 2.04.04  .13.10**  
Family incomeb  −.77 (.54)−.14  −1.35 (.73)−.18*
Maternal education  −.48 (.53)−.09  −1.50 (.72)−.20*
Step 3.10.06*  .14.01  
Family income × sex  −1.09 (.55)−.20*  −.86 (.74)−.12
Maternal education × sex  −1.36 (.53).25*    .48 (.72)  .06
Table 3. Effects of family income and paternal education on psychopathology
Independent variableInternalizing (F(6, 117) = 1.42, p = .21)Externalizing (F(6, 117) = 3.41, p = .004)
R2ΔR2B (SE)βR2ΔR2B (SE)β
  1. a

    Effect coded: 1 = boy, −1 = girl.

  2. b

    Effect coded: 1 = high income, −1 = low income.

  3. * p < .05; ** p < .01.

(constant)  12.28 (4.97)   20.35 (6.59) 
Step 1.00.00  .03.03  
Child age  −.04 (.12)−.03  −.18 (.16)−.10
Child sexa  .40 (.51)  .08  1.29 (.67)  .17
Step 2.03.03  .14.11**  
Family incomeb  −.98 (.53)−.18  −1.63 (.71)−.21*
Paternal education  −.20 (.51)−.04  −1.48 (.68)−.20*
Step 3.07.04*  .15.01  
Family income × sex  −.82 (.53)−.15  −.68 (.71)−.09
Paternal education × sex  −1.00 (.51)  .19*    .64 (.68)  .08

As shown in Tables 2 and 3, income, maternal education and paternal education all negatively predicted externalizing symptoms. Children from economically and educationally disadvantaged families had more externalizing symptoms than those from advantaged families. Internalizing symptoms were not predicted by income or education but by the family income × sex, maternal education × sex and paternal education × sex interactions. Further analyses showed that income negatively predicted internalizing among boys (β = −.27, p = .04, ΔR2 = .07) but not girls (β = −.04, ns). Maternal education negatively predicted internalizing among girls (β = −.27, p = .02, ΔR2 = .07) but not boys (β = −.02, ns). Paternal education did not predict internalizing among either boys (β = −.05, ns) or girls (β = −.18, ns).

The mediator effect of parent–child conflict

We next conducted a series of regressions and Sobel's z tests to evaluate the mediator effect of parent–child conflict on the relations between SES and child psychopathology. In order for conflict to be such a mediator, the following conditions must be met (Baron & Kenny, 1986): (1) the direct relation from SES to psychopathology (Path C) must be significant; (2) conflict must be significantly associated with both SES (Path A) and psychopathology (Path B); (3) the strength of the relation from SES to psychopathology must be reduced significantly when conflict is added to the model (Path C′). Table 4 shows the unstandardized regression coefficients of Paths A, B, C, and C′ and Sobel's z statistics.

Table 4. Mediator effects
Mediator variablesB (SE)Sobel's z (one-tailed)
Path APath BPath CPath C′
  1. + p < .10; * p < .05; **p < .01; ***p < .001.

Mediating the effect of income on boys' internalizing
Mother–child conflict−1.706* (.862)0.309** (.093)−1.437* (.666)−0.909 (.636)−1.70, p = .045
Father–child conflict−2.059* (.839)0.232* (.102)−1.437* (.666)−1.082 (.686)−1.67, p = .047
Mediating the effect of income on externalizing
Mother–child conflict−1.402* (.665)0.485*** (.082)−1.879** (.681)−1.200 (.614)−1.99, p = .023
Father–child conflict−1.850** (.595)0.429*** (.098)−1.879** (.681)−1.085 (.667)−2.53, p = .006
Mediating the effect of maternal education on girls' internalizing
Mother–child conflict−1.238+ (.690)0.241** (.076)−1.222* (.491)–0.924+ (.475)−1.56, p = .059
Father–child conflict−1.036+ (.612)0.196* (.087)−1.222* (.491)−1.163* (.477)−1.35, p = .089
Mediating the effect of maternal education on externalizing
Mother–child conflict–0.804+ (.479)0.531*** (.074)−1.814*** (.505)−1.386** (.441)−1.63, p = .052
Father–child conflict–0.887* (.449)0.412*** (.086)−1.814*** (.505)−1.546** (.482)−1.83, p = .034
Mediating the effect of paternal education on externalizing
Mother–child conflict–0.895* (.446)0.529*** (.074)−1.705*** (.472)−1.231** (.415)−1.93, p = .027
Father–child conflict–0.640 (.426)0.421*** (.084)−1.705*** (.472)−1.460** (.444)−1.44, p = .075

We first evaluated whether parent–child conflict mediated the effects of income on psychopathology. Table 4 showed that Conditions (1), (2) and (3) were all met for internalizing symptoms among boys (but not girls) and for externalizing symptoms among children of both sexes. Moreover, the effects of income on internalizing and externalizing were significantly reduced and became non-significant when mother–child and father–child conflict were added to the model. Thus, mother–child and father–child conflict fully mediated the effects of income on internalizing (only for boys) and externalizing psychopathology.

Second, we tested whether parent–child conflict mediated the effects of maternal education on psychopathology. Table 4 showed that Condition (1) was met for internalizing among girls (but not boys) and for externalizing among children of both sexes. Neither Condition (2) nor (3) was met for internalizing among girls. Hence, conflict did not mediate the effect of maternal education on internalizing symptoms. For externalizing symptoms, neither Condition (2) nor (3) was met when mother–child conflict was considered a mediator, whereas both conditions were met when father–child conflict was considered a mediator. The effect of maternal education on externalizing was also significantly reduced but remained significant when father–child conflict was added to the model. Therefore, the effect of maternal education on externalizing was not mediated by mother–child conflict but partially mediated by father–child conflict.

Finally, we tested whether parent–child conflict mediated the effect of paternal education on psychopathology. Table 4 showed that Condition (1) was met for externalizing but not for internalizing symptoms. We therefore did not go on with mediation analyses for internalizing. For externalizing, neither Condition (2) nor (3) was met when father–child conflict was considered a mediator, whereas both conditions were met when mother–child conflict was considered a mediator. Moreover, the effect of paternal education on externalizing was significantly reduced but remained significant when mother–child conflict was added to the model. Therefore, the effect of paternal education on externalizing symptoms was not mediated by father–child conflict but partially mediated by mother–child conflict.

DISCUSSION

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES

As increasing income inequality has been observed in many countries during the past decades, a growing body of research is seeking to understand the impact of SES on children's development (Bradley & Corwyn, 2002). In line with this research, we set out to examine the effects of family income and parental education on Chinese young children's internalizing and externalizing psychopathology and the mechanisms underlying these effects. Based on the existing literature, we hypothesized that income and education were associated with both internalizing and externalizing psychopathology. In addition, we assumed that these associations were mediated by parent–child conflict.

Our hypothesis that income is negatively associated with psychopathology was supported. Results indicated that young children in low-income families had more externalizing symptoms than their peers in high-income families. In addition, lower income predicted more internalizing symptoms among boys but not girls. One explanation for the differential impacts may be boys' greater behavioural vulnerability to family poverty. A recent investigation also found that boys' internalizing psychopathology was particularly sensitive to the influence of income (Huston et al., 2001). Taken together, these findings underscore the significance that family income may have for Chinese 2–3-year-old children's psychopathology, consistent with recent large-scale studies of very young children conducted in Western societies (Mistry et al., 2004, 2008).

Results also confirmed the negative effect of parental education on child psychopathology. Children whose fathers or mothers had less education exhibited more externalizing symptoms than their counterparts whose parents were well-educated. Also, maternal education negatively predicted internalizing symptoms, but only among girls. One possible explanation for this gender difference is that maternal education may have greater benefits for girls than boys (Hoffman, 1989). Positive effects for girls may result partly from having a well-educated and well-behaved maternal model (Huston et al., 2001). In sum, these findings highlight the role of parental education in psychopathology among Chinese young children, consistent with recent research linking education with child psychopathology (Huston & Aronson, 2005; Mistry et al., 2008).

Our hypothesis that parent–child conflict mediates the effects of income and education on psychopathology was partially supported. In line with the FSM (Conger et al., 2010) and our hypothesis, results suggested that the effects of income were fully mediated by mother–child and father–child conflict. The FSM suggests that economic hardship and poor education result in increased parenting stress, which, in turn, brings about hostile and aggressive parenting behaviour and parent–child conflict (Conger & Donnellan, 2007). Parent–child conflict may reflect children's insecure attachments with their parents. According to Bowbly's (1982) attachment theory, insecure children may develop internal models of the social world as untrustworthy and rejecting and see themselves as unworthy of love. Consequently, they are more likely to develop internalizing and externalizing psychopathology.

However, the effects of education were not mediated or only partially mediated by mother–child or father–child conflict. Specifically, the effect of maternal education on internalizing was not mediated by conflict; the effect of maternal education on externalizing was not mediated by mother–child conflict but partially mediated by father–child conflict; the effect of paternal education on externalizing was not mediated by father–child conflict but partially mediated by mother–child conflict. These findings suggest that, unlike income, education can only sometimes act as a useful resource that helps reduce parent–child conflict.

Indeed, the effect of education on child outcomes can be explained not only by the FSM but also by the family investment model (FIM) in which education increases the investments (standard of living, learning stimulation, etc.) that parents make in the lives of their children and, in turn, leads to their children's academic and social success (Bradley & Corwyn, 2002). For instance, Smith, Brooks-Gunn, and Klebanov (1997) found that the mediation effect of stimulating home environments was particularly strong for maternal education. Presumably, the combination of the FSM and FIM will help understand the impact of education better. Future studies that conduct joint tests of the FSM and FIM are needed.

This study contributed to the literature on the influence of SES in important ways. First, the results suggest that income and education differentially predict children's psychopathology. This pattern of results joins evidence from recent research (Davis-Kean, 2005; Duncan & Magnuson, 2003) showing that both income and maternal education have unique effects on child outcomes. In this sense, the study highlights the drawback of including income as the single indicator of SES and the importance of examining education as an additional indicator (Conger et al., 2010). Moreover, our findings extend past research and suggest that paternal education may also have a unique effect on child outcomes. Finally, the total effect of income and education was substantial in this study, especially for externalizing symptoms. It thus seems that poverty and low education may constitute important family risks for children's development of psychopathology.

Second, the results highlight the mediator effect of parent–child conflict on the links between SES and psychopathology. Moreover, because previous research on SES has rarely paid specific attention to the process of father–child relationships, our findings has made an important contribution to the existing literature suggesting that father–child relationships, as well as mother–child relationships, may explain the impact of SES on child outcomes.

Finally, the effects of SES on family relations and child psychopathology received strong support in the context of contemporary China. Indeed, this study represents an initial effort to empirically link SES with psychopathology in Chinese young children. Our findings underscore that economic and educational inequality in contemporary China may increase disadvantaged children's risk for internalizing and externalizing psychopathology in their third and fourth years of life, which are almost the earliest ages at which psychopathology symptoms can be measured.

Even though this study has a few interesting findings to add to the literature, several limitations should be noted. One of the major limitations is the use of cross-sectional data to evaluate causal relationships. It would be useful for future studies to collect longitudinal data and determine whether changes in income and education result in changes in child psychopathology. Another limitation is that our income data were not collected separately for mothers and fathers. Further research that includes both maternal and paternal income would be helpful to determine whether they play differing roles in child psychopathology. Additionally, the sample did not represent the diversity of Chinese society. It would be valuable for future studies to get a national-wide sample that is more representative of the diversity of Chinese society. Finally, it is noteworthy that our Sobel's z tests provided a more conservative analysis of mediation effects than the increasingly popular method of bootstrapping.

Understanding SES and psychopathology is essential for designing interventions and policies to reduce the negative impact of economic and social inequality. This article provides strong evidence for the impact that SES has on Chinese young children's internalizing and externalizing symptoms. Given the increasing social and economic inequality in China, we consider the following two implications for implementing effective family-based intervention and prevention programmes.

First, it is crucial to identify children who have internalizing and externalizing symptoms on the basis of key risk factors and then provide them with specific intervention services aimed at reducing these symptoms. Our results indicate that such risk factors might include financial hardship and educational disadvantage in the family. Thus, evaluating income and education may be an efficient way for identifying children at most risk. Our results also imply that programmes that combine earning supplementation and education improvement should be more effective than those only focusing on one aspect. Interventions conducted in the United States have documented favourable effects of welfare policies and intervention programmes that are dedicated to reduce poverty (Huston et al., 2001) and promote parental education (Magnuson & McGroder, 2001). We suggest that similar policies and programmes be implemented in China to intervene economically and educationally disadvantaged families.

Second, because our findings suggest that parent–child conflict mediates the effect of SES on child psychopathology, parent training should be emphasized to reduce conflict in low-SES families. By doing so, low-SES parents may change the way they interact with their child and, in turn, help their child reduce his or her psychopathology symptoms. Finally, our findings also suggest that programmes that target both mothers and fathers are likely to produce the most pronounced effect and should be adopted.

REFERENCES

  1. Top of page
  2. Abstract
  3. THE INFLUENCE OF SES ON PSYCHOPATHOLOGY
  4. THE MEDIATOR EFFECT OF PARENT–CHILD CONFLICT
  5. THE PRESENT STUDY
  6. METHOD
  7. RESULTS
  8. DISCUSSION
  9. REFERENCES
  • Achenbach, T. (1992). Manual for the Child Behavior Checklist/2-3 and 1992 profile. Burlington, VT: University of Vermont Department of Psychiatry.
  • Adler, N. E., Boyce, T., Chesney, M. A., Cohen, S., Folkman, S., Kahn, R. L., & Syme, S. . L. (1994). Socioeconomic status and health: The challenge of the gradient. American Psychologist, 49, 1524.
  • Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 11731182.
  • Beijng Statistics Bureau (2005). Beijing statistics yearbook 2005. Beijing: China Statistics Press.
  • Bowlby, J. (1982). Attachment and loss: Attachment (2nd ed.). New York, NY: Basic Books.
  • Bradley, R. H., & Corwyn, R. F. (2002). Socioeconomic status and child development. Annual Review of Psychology, 53, 371399.
  • Conger, R. D., Conger, K. J., & Martin, M. J. (2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72, 685704.
  • Conger, R. D., & Donnellan, M. B. (2007). An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology, 58, 175199.
  • Davis-Kean, P. E. (2005). The influence of parent education and family income on child achievement: The indirect role of parental expectations and the home environment. Journal of Family Psychology, 19, 294304.
  • Duncan, G., & Magnuson, K. A. (2003). Off with Hollingshead: Socioeconomic resources, parenting, and child development. In M. H. Bornstein & R. H. Bradley (Eds.), Socioeconomic status, parenting, and child development (pp. 83106). Mahwah, NJ: Erlbaum.
  • Earls, F. (1980). Prevalence of behavior problems in 3-year-old children: A cross-national replication. Archives of General Psychiatry, 37, 11531157.
  • Glick, J., Hanish, L., Yabiku, S., & Bradley, R. (2012). Migration timing and parenting practices: Contributions to social development in preschoolers with foreign-born and native-born mothers. Child Development, 83, 15271542.
  • Hoffman, L. W. (1989). Effects of maternal employment in the two-parent family. American Psychologist, 44, 283292.
  • Huston, A. C., & Aronson, S. R. (2005). Mothers' time with infant and time in employment as predictors of mother-child relationships and children's early development. Child Development, 76, 467482.
  • Huston, A. C., Duncan, G., Granger, R., Bos, H., McLoyd, V., Mistry, R., … Ventura, A. (2001). Work-based anti-poverty programs for parents can enhance the school performance and social behavior of children. Child Development, 72, 318337.
  • Lamb, M. E. (1997). The role of the father in child development (3rd ed.). New York, NY: John Wiley.
  • Liu, L., Wu, L., & Yao, K. (2003). The development of the Child Behavior Checklist/2–3 test norms in urban Chinese children. Chinese Journal of Child Health Care, 11, 377379.
  • Lugo-Gil, J., & Tamis-LeMonda, C. (2008). Family resources and parenting quality: Links to children's cognitive development across the first 3 years. Child Development, 79, 10651085.
  • Magnuson, K. A., & McGroder, S. M. (2001). Intergenerational benefits: The effect of maternal education on young children's academic problems and school readiness. Unpublished manuscript. Northwestern University.
  • Mistry, R. S., Biesanz, J. C., Chien, N., Howes, C., & Benner, A. D. (2008). Socioeconomic status, parental investments, and the cognitive and behavioral outcomes of low-income children from immigrant and native households. Early Childhood Research Quarterly, 23, 193212.
  • Mistry, R. S., Biesanz, J. C., Taylor, L. C., Burchinal, M., & Cox, M. J. (2004). Family income and its relation to preschool children's adjustment for families in the NICHD-Study of Early Child Care. Developmental Psychology, 40, 727745.
  • Pianta, R. C. (1992). Child–Parent Relationship Scale (CPRS). Charlottesville, VA: University of Virginia.
  • Richman, N., Stevenson, J., & Graham, P. (1975). Prevalence of behavior problems in 3 year old children: an epidemiological study in a London borough. Journal of Child Psychology and Psychiatry, 16, 277287.
  • Smith, J. R., Brooks-Gunn, J., & Klebanov, P. K. (1997). The consequences of living in poverty for young children's cognitive and verbal ability and early school achievement. In G. J. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing up poor (pp. 132189). New York, NY: Russell Sage.
  • Wang, H., Long, W., & Zhao, J. (2003). Predictive modeling method of Gini coefficient and its application in Beijing economic development analysis. Systems Engineering, 21, 6165.
  • World Bank (2007). World development report. New York, NY: Oxford University Press.
  • Zhang, X. (2012). The effects of parental education and family income on mother-child relationships, father-child relationships, and family environments in the People's Republic of China. Family Process, 51, 483497.
  • Zhang, X., & Chen, H. (2010). Reciprocal influences between parents' perceptions of mother-child and father-child relationships: A short-term longitudinal study in Chinese preschoolers. The Journal of Genetic Psychology, 171, 2234.
  • Zhang, X., Chen, H., Zhang, Y., & Sun, B. (2009). Family income and social competence in early childhood: Examining mediation and moderation effects. Acta Psychologica Sinica, 41, 613623.