Family socioeconomic status and young children’s learning behaviors: The mediational role of parental expectation,home environment,and parental involvement

Abstract

Utilizing a national longitudinal sample from China Family Panel Studies (2018), this study examined how family socioeconomic status (SES) is related to young children’s learning behaviors through parental expectation, parental involvement and home learning environment. A total of 1348 children aged 3–5 years were included. Mediation analysis was conducted to test the hypotheses. The results identified a significant relationship between family SES and home learning environment and parental involvement, whereas family SES was only partially related to parental academic expectations. There were no statistically significant relationships between SES and children’s learning behaviors. Moreover, no indirect effects were found between parental expectation, home learning environment, and children’s learning behaviors. However, parental involvement had an indirect effect on children’s learning behaviors. The present study highlights the need to consider the role of parental involvement in supporting the development of young children’s learning behaviors. The findings and discussions raise implications for researchers and practitioners to provide parental involvement support programs for Chinese families from diverse SES backgrounds.

Keywords

socioeconomic status parental expectation home environment parental involvement

To date, many educators and researchers are devoted to investigating how family and parent factors predict children’s development. For example, Bronfenbrenner’s ecological system theory suggests that child development is a complex system affected by multiple levels of their surrounding environment, with family and parent attributes sitting in the hub of this system (Bronfenbrenner & Morris, 2007). Dearing and Tang (2009) further specifically depict the reciprocal relationships among home learning environment, parental expectation, parental involvement, and children’s development and learning. Super and Harkness (1986) claim that children are shaped by three subsystems of developmental niche: the physical and social settings, the customs and practices, and the psychology of the caregivers. As one important predictor, parental belief systems will affect the type of home environment and parenting techniques parents will use, which will ultimately predict children’s development and learning outcomes.

Although abundant studies have been done to examine how family SES is associated with children’s development and learning, studies that focus on preschool children and Chinese families are still inadequate. The purpose of the present study is to use a national-wide household survey dataset, the latest China Family Panel Study (CFPS) 2018, to assess the relationships among family socioeconomic status (SES), parental expectation, home learning environment, parental involvement and Chinese 3–5 years children’s learning behaviors. Studies concerning early childhood education in China received less attention. The CFPS provides invaluable information about Chinese young children and their families regarding early childhood education. However, existing research that is based on CFPS and focuses on the early years is limited. Therefore, the present study will expand the literature on early childhood education by adding more Chinese context. Moreover, family SES is often used as a composite measurement, however, there are studies demonstrating that family income and parental education affect the home learning environment and parental involvement differently (e.g., Son & Morrison, 2010; X. Zhang et al., 2021). Therefore, this study will look into detail if parental education and family income are differently associated with home learning environment, parental involvement and young children’s learning behaviors in Chinese context and how.

Literature Review

Family socioeconomic status (SES), a combined measure of parental education, parental occupation, and family income (Stevens & Featherman, 1981), is considered an important factor in predicting children’s optimal development (Loft & Waldfogel, 2021), academic achievement (Ogg & Anthony, 2020) and children’s learning behaviors (Luo & Gao, 2022). Some studies found that the three components of SES correlate to children’s learning differently. Son and Morrison (2010) claimed that income itself did not result in significant changes in physical home learning environment. X. Zhang and colleagues' (2021) meta-analysis research demonstrated that although both family income and parental education significantly predicted young ESL learners' vocabulary development, parental education is a stronger predictor respectively. Thus, although this study will examine how SES affects the researched dependent variables, I will look at each SES component separately. The influence of SES on children’s learning and achievement is widely believed to be mediated by parental involvement (Tan et al., 2020; F. Zhang et al., 2020). Tan et al. (2020) further summarized six types of parental involvement that are important predictors of children’s learning and achievement, among which, parental expectations, interactional strategies (e.g., shared reading), and parental tutoring are the interest of the present study.

Poon (2020) found that parents from high SES backgrounds tended to set higher academic expectations and involve more in children’s learning both at school and at home. Parental academic expectation positively influences children’s learning because children tend to internalize parents’ expectations and beliefs about learning and achievement (Cheung & Pomerantz, 2012), leading to higher self-regulatory learning behaviors and ultimate school engagement and attainment. Chinese parents were found to place a high value on children’s education, resulting in a prevalent phenomenon that Chinese parents place high educational expectations on their children independent of family SES (Butler, 2014). This study thus aims to test if Bulter’s findings could be found.

Family SES is also strongly linked to the quantity and quality of parent-child interactions. According to Adams (1990), by entering primary school, children from middle SES backgrounds received 1000–1700 hours of shared reading at home, outdistancing children from low SES households who only got 25 hours of shared reading. This difference is still existing. Barnes and Puccioni (2017) found that the quantity and quality of parent-child reading significantly vary between different SES families, with more frequent and more high-quality shared reading in higher SES households, as high SES families are academically and financially competent in supporting and involving in their children’s education (Tan et al., 2020). Given parental involvement, a number of studies demonstrate that it contributes to improving children’s social and emotional functioning, including enhancing their self-esteem, self-management, and social awareness (Christenson, 2004). Parental involvement is also associated with children’s high academic performance (Galindo & Sheldon, 2012), high parent-oriented motivation and self-regulated learning (Cheung & Pomerantz, 2012), and reduction in challenging behaviors (Stormshak et al., 2009).

Examining home learning environment, according to Dearing and Tang's (2009) conceptual framework, home learning environment and parent/family characteristics are reciprocally influenced by each other. Parents’ attributes, such as their skills, knowledge, and motivation, will influence their expectations, beliefs and values about children’s learning, which will directly determine what kind of home environment that parents will provide for children. All of these factors play a key role in cultivating children’s motivation and attitudes about learning, and eventually positively predict children’s achievement. Empirical evidence supports this loop. For example, the findings from studies by Peixoto et al. (2022) revealed that mothers’ literacy beliefs predict the quantity and quality of home literacy practices which will eventually contribute to children’s language and literacy development. Although the present study will not attempt to establish the correlation between family/parent characteristics and children’s ultimate learning outcomes, according to Dearing and Tang's (2009) home learning framework, children’s learning behaviors will mediate the effects that how their parents perceive child development and what kind support that parents can provide have on children’s achievement. Therefore, this study is going to explore what kind of family and or parent factors influence children’s learning behaviors and how.

Based on the literature I have reviewed; I proposed a hypothesized parallel mediator model examining the interrelationships among family SES, parental expectation, home learning environment, parental involvement and children’s learning behaviors (see Figure 1). Specifical hypotheses are presented as follows: Hypothesis 1: Family SES positively predicts parental expectation, home learning environment, and parental involvement. Hypothesis 2: Family SES has a significant effect on learning behaviors mediated by parental expectation, home learning environment, and parental involvement.

Figure 1.

A parallel multiple mediator model for the present study.

Method

Data and Sample

The present study is based on the CFPS2018 dataset. CFPS is a nationwide longitudinal survey conducted by Peking University and aims to depict a comprehensive image of contemporary Chinese individuals’ lives and perspectives from a variety of lenses, including family dynamics, education, economic status, health, and child-rearing, etc. (Xie & Hu, 2014). To date, six waves of CFPS have been conducted. This study depended on the CFPS2018 dataset to explore how family and parent factors are related to young children’s learning behaviors. CFPS2018 was conducted by face-to-face or telephone interviews. It employed a three-stage probability proportional to size sampling in the ordinal level: 1. city or county, 2. neighborhood community, and 3. household (Xie & Hu, 2014). Socioeconomic development (i.e., local GDP) and the size of administrative units were used as implicit stratification to ensure a better representation of Chinese society. CFPS2018 consists of five questionnaires: 1. family roster questionnaire; 2. family economic questionnaire; 3. individual self-report; 4. individual proxy questionnaire; 5. child proxy questionnaire (Wu et al., 2021). All five questionnaires can be linked using individual, family, and proxy identifiers. This study integrated the family economic questionnaire and individual self-report into the child proxy questionnaire to add SES and parental information. Listwise deletion approach was used to remove participants if the cases cannot be linked with the other two questionnaires, leading to 25.6% of case deletion. Another 1% of cases were deleted due to over 60% of missing data. The final sample includes 1348 children aged 3–5 years.

Measures

Children’s learning behavior

Young children’s learning behavior was measured by parents’ reports on a 5-point Likert scale from “Totally disagree (1)”, “Disagree (2)”, “Neither agree nor disagree (3)”, “Agree (4)” to “Totally agree (5)”. CFPS measured seven items: studying hard; checking homework; playing only after homework; concentrating on what is doing; respecting rules; always completing what is doing; keeping school things in order. A series of questions featuring display logic are related to this measurement. Parents of children who are enrolled in daycare or kindergarten were first asked if their children had homework last semester. If yes, parents were further asked to evaluate their children’s learning behaviors aforementioned. Face validity was checked according to existing literature. For example, homework checking that seems unusual for preschoolers was also identified as one popular parental involvement interaction in Lau and colleagues’ study that focused on Chinese kindergarten children in Hong Kong and Shenzhen (Lau et al., 2011). This is consistent with the findings from Mao (2020) that found Chinese parents are frequently engaged in helping with homework and homework checking. Pearson’s rs were calculated to check the convergent validity, with all rs=.8 or above (ps<.01, two-tailed), indicating a high level of positive convergent validity according to Terwee et al. (2007). Cronbach’s α was run and the score (α=.74) indicated acceptable internal reliability. Then a composite score on children’s learning behaviors using the mean of each item was computed.

Family socioeconomic status

Family SES is indicated by parental education, parental occupation, and annual family income. Due to a large amount of missing data on parental occupation, in the present study, family SES was indicated by parental education and annual family income. Following Yang et al. (2020), parent’s level of education was transformed into a five-point Likert scale: 1 = below elementary school and elementary school; 2 = junior high school, 3 = senior high school; 4 = bachelor’s degree; 5 = postgraduate and doctoral degrees. Annual household income was measured, and the scores were standardized for later analysis. Previous studies often compute an SES composite score as one independent variable. However, studies that focus on examining the relationships between SES and home learning environment and parental involvement demonstrated that family income and parental education affect home learning environment and parental involvement differently (e.g., Son & Morrison, 2010; X. Zhang et al., 2021). Therefore, this study decided to explore how family income and parental education predict the examined variables separately.

Parental expectation

Child’s future education degree was used to imply parents’ educational expectations for their children. The CFPS asked parents to express their expectations for their children’s highest education degree (1 = elementary school and below; 2 = junior high school, 3 = senior high school; 4 = bachelor’s degree; 5 = postgraduate degree).

Parental involvement

Specifically, four items were used to measure parental involvement: 1. frequency of shared reading; 2. frequency of buying books for children; 3. frequency of outdoor play; 4. frequency of learning games. All the items were measured on a 5-point Likert scale: 1 = several times a year or less; 2 = once a month; 3 = 2–3 times a month; 4 = several times a week; 5 = every day. Face validity was first examined based on existing literature. Previous studies have shown that parents of preschoolers involve in various home involvement discussed in the present study (e.g., Kwok, 2015; Lau et al., 2011; Mao & Pesco, 2022). Convergent validity was further examined. Forty percentage of tested correlations are scoring at .68 or above in Pearson’s r (all ps < .01, two-tailed). The rest is at .3 or above, meaning the parental involvement measurement selected in the present study reaches a positive convergent validity at a moderate to high level according to Terwee et al. (2007). An internal consistency test was run and Cronbach’s α = .68, indicating fairly acceptable internal reliability. A composite score based on mean value was computed.

Home learning environment

To better understand the home environment, the interviewer observed and rated at a 5-level Likert scale from “strongly disagree to strongly agree” regarding if the home environment indicates parents’ care about children’s education, for example, if books and other learning materials that are designed for children were observed by the interviewers.

Covariates

Child, parent, and family characteristics that might affect children’s learning behavior were controlled. These include child age (3, 4, 5) and gender (0 = girls, 1 = boys), child’s school attendance status (0 = not attend school, 1 = attend), residency type (0 = rural areas, 1 = urban areas), and family size (= 2–20 family members).

Analytic Strategies

SPSS 26 version was used to analyze data. Overall, this dataset had minimal missing data (2.9%). Except for the following five items which had more missing data (range = 16%–29%), including “interviewer’s rating in parents’ care about child’s education”, “keeping school things in order”, “respecting rules”, “concentrating on what is doing”, “always completing what is doing”, all other variables had minimal missing data (<5%). Multiple imputation technique that is considered one of the most plausible methods (Prime et al., 2014) was utilized to deal with all missing data. Missing values were substituted with plausible values based 10 times imputations and the result was pooled. Correlation analysis and multiple regression tests were first carried out to interpret the linear relationships among all variables. To explore if children’s learning behaviors are influenced by SES mediated by parental expectation, home learning environment and parental involvement, parallel multiple mediation tests were run using the macro Process for SPSS (Hayes, 2018). In parallel multiple mediator models, the dependent variables are influenced by independent variables directly and by two or more mediators indirectly which are not causally influencing one another though they might be correlated with one another (Hayes, 2018).

Results

Descriptive Statistics

Descriptive statistics are presented in Table 1. The status of residency urbanization was roughly equal. On average, children had a family size of five members. The mean annual income was about 83,000 Chinese Yuan ($16,600). The average age of parents was 32 years, and 80% of them had an education of junior high school and above. Children are roughly equally distributed into three age groups and gender, with a bit more 4 years children and a bit more boys. Regarding school attendance, 70% of children attended school. As for parental expectations, about 90% of them expected their children to have a bachelor’s degree or above.

Table 1.

Descriptive Statistics of Participating Children.

Participants’ Characteristic	n/mean	%/SD
Residence Status, n (%)
Rural	664	49.3%
Urban	684	50.7%
Family size, mean (SD)	5.48	2.06
Annual income (RMB), mean (SD)	83052	81669
Parent age, mean (SD)	31.53	5.72
Parent education, n (%)
Primary school and below	278	21.1%
Junior high school	501	38.0%
Senior high school	246	18.7%
Bachelor’s degree	283	21.5%
Graduate or postgraduate	10	0.8%
Child age, n (%)
3	448	33.2%
4	487	36.1%
5	413	30.6%
Child gender, n (%)
Girls	624	46.3%
Boys	724	53.7%
School attendance, n (%)
No	404	30.0%
Yes	944	70.0%
Expectation of child’s education, n (%)
Primary school or below	3	0.2%
Junior high school	16	1.2%
Senior high school	126	9.4%
Bachelor’s degree	1009	74.9%
Graduate or postgraduate	193	14.3%

Correlation Analyses

Table 2 presents the Pearson’s correlations among all the variables and the results showed that the majority of correlations were in the expected directions. Inter-correlations among urbanization status, family income, parental education, parent age, school attendance, parental expectation, home learning environment, and parental involvement were positive and statistically significant. Family size was significantly negatively correlated with most of the other variables. Specifically examining the dependent variable, child’s learning behaviors, the correlations were complex. It was not statistically significantly related to most of the variables, whereas it was significantly positively linked to child’s age, school attendance and parental involvement. It was negatively related to urbanization status.

Table 2.

Pearson’s Correlations Among Study Variables.

	1	2	3	4	5	6	7	8	9	10	11
1.Urbanization	1
2.Family size	−.228^**
3.Income	.240^**	.004
4.Parent_Edu	.380^**	−.227^**	.368^**
5.Parent_Age	.060^*	−.004	.052	−.028
6.Child_Age	.045	−.020	−.035	−.043	.133^**
7.Child_Gender	.039	−.024	.014	.008	.011	.046
8.School attendance	.158^**	−.163^**	.099^**	.164^**	.031	.506^**	.004
9.PE	.066^*	−.065^*	.106^**	.189^**	.022	−.016	.038	.080^**
10.HLE	.149^**	−.058^*	.225^**	.228^**	−.011	.001	.047	.061^*	.121^**
11.PI	.263^**	−.205^**	.250^**	.416^**	−.017	.029	−.002	.199^**	.162^**	.182^**
12.LB	−.069^*	−.050	−.040	−.004	.034	.271^**	−.050	.284^**	.013	−.023	.104^**

Note. *p < .05. **p < .01; PE=parental expectation, HLE=Home learning environment, PI=parental involvement, LB=learning behaviors.

Multiple Regression Tests

To explore which variables will predict parental expectation, home learning environment and parental involvement, three hierarchical regressions were separately conducted with the same layers of predictors as presented in Table 3. Based on the data in model three which is our focus, parental expectation was only predicted by school attendance and parent’s education. Examining R², family characteristics and child characteristics in combination explained 1.62% of variance in parental expectation. When SES was added to the model, 2.69% of additional variance in parental expectation was explained, with all ps < .01. As for home learning environment, only two SES factors were significantly positively correlated with it. R² in step 1 was 2.29% with p < .001 and was 2.66% with p = .17 in step 2. In the final model, SES accounted for an additional 5.29% of the variance with p < .001. As for parental involvement, except for child characteristics, all other variables significantly predicted it. Additionally, compared to parental expectation and home learning environment, parental involvement was more explained by the three models, separately with R² = 9.1%, 11.52%, and 21.79% (all ps < .001).

Table 3.

Regression Model of Predictors of parental expectation, home learning environment, and parental involvement.

	Parental Expectation			Home Environment			Parental Involvement
	b (CI)	β	p	b (CI)	β	p	b (CI)	β	p
Model 1
Constant	3.996 (3.953, 4.038)		<.001	3.53 (3.476, 3.584)		<.001	2.588 (2.518, 2.657)		<.001
Urbanization	.059 (−.001, .119)	.054	.056	.2 (.123, .276)	.144	<.001	.426 (.327, .525)	.229	<.001
Family size	−.029 (−.059, .001)	−.053	.062	−.017 (−.056, .021)	−.025	.375	−.143 (−.193, −.094)	−.152	<.001
Model 2
Constant	4.088 (3.932, 4.244)		<.001	3.565 (3.367, 3.763)		<.001	2.707 (2.454, 2.960)		<.001
Urbanization	.046 (−.014, .107)	.043	.133	.19 (.113, .267)	.137	<.001	.39 (.292, .489)	.21	<.001
Family size	−.022 (−.052, .009)	−.039	.169	−.012 (−.051, .027)	−.017	.537	−.121 (−.171, −.072)	−.129	<.001
Child age	−.049 (−.092, −.007)	−.072	.024	−.03 (−.084, .024)	−.034	.279	−.088 (−.157, −.019)	−.075	.013
Child gender	.042 (−.016, .101)	.039	.158	.059 (−.015, .134)	.043	.119	−.022 (−.117, .074)	−.012	.656
School attendance	.123 (.047, .198)	.103	.002	.08 (−.016, .177)	.053	.103	.369 (.246, .492)	.181	<.001
Model 3
Constant	3.874 (3.701, 4.047)		<.001	3.309 (3.091, 3.526)		<.001	2.018 (1.750, 2.287)		<.001
Urbanization	−.024 (−.088, .040)	−.022	.471	.069 (−.011, .149)	.050	.091	.154 (.055, .253)	.083	.002
Family size	−.012 (−.043, .019)	−.022	.445	−.008 (−.047, .031)	−.011	.688	−.093 (−.141, −.045)	−.099	<.001
Child age	−.03 (−.072, .012)	−.044	.166	.004 (−.049, .057)	.004	.890	−.023 (−.089, .043)	−.02	.493
Child gender	.042 (−.016, .100)	.039	.156	.058 (−.015, .130)	.041	.121	−.022 (−.112, .067)	−.012	.624
School attendance	.084 (.008, .160)	.071	.030	.013 (−.082, .108)	.008	.793	.24 (.123, .357)	.118	<.001
Income	.023 (−.008, .055)	.043	.150	.11 (.070, .149)	.158	<.001	.099 (.050, .148)	.106	<.001
Parent Edu	.083 (.051, .114)	.163	<.001	.096.056, .135	.147	<.001	.265 (.216, .314)	.304	<.001

Mediation Effect Tests

Macro Process Model four was used to test parallel multiple mediation effects (Hayes, 2018). Child age, child gender, school attendance, urbanization, and family size served as covariates were controlled. Table 4 and Table 5 separately presented the summary for the mediator model with SES-income and SES-education as independent variables. After controlling for covariates, except for parental involvement, family income, parental education level, parental expectation and home learning environment did not show significant direct effects on child’s learning behaviors. A total indirect effect was significant though the effect size was small, with b = .008, 95%, BCa CI [.001, .016]. Looking in detail, parental involvement contributed more indirect effects with b=0.011, 95%, BCa CI [.005, .018], whereas the other two were not found significant indirect effects. As for parents’ education, it appeared a similar trend to family income. The total indirect effect was significant: b = .028, 95%, BCa CI [.007, .051], and only parental involvement presented a significant indirect effect with b=.034, 95%, BCa CI [.015, .055].

Table 4.

Mediator Model Summary (Family Income as Independent Variable).

		PE (M₁)				HLE (M₂)				PI(M₃)				LB (Y)
		b	SE	p		b	SE	p		b	SE	p		b	SE	p
Income (X)	a₁	.052	.0155	.001	a₂	.143	.019	<.001	a₃	.183	.025	<.001	c^′	−.023	.014	.096
PE (M₁)													b₁	.001	.024	.969
HLE (M₂)													b₂	−.019	.019	.319
PI(M₃)													b₃	.059	.015	<.001
Constant	i_M1	4.091			i_M2	3.543	.099	<.001	i_M3	2.661	.126	<.001	i_Y	2.708	.134	<.001
		R²=.028				R²=.067				R²=.157			R²=.127
		F (6, 1340) =6.441, p<.001				F (6, 1340) =16.019, p<.001				F (6, 1340) =41.44, p<.001			F (9, 1337) =21.693, p<.001

Table 5.

Mediator Model Summary (Parent Education Level as Independent Variable).

		PE (M₁)				HLE (M₂)				PI(M₃)				LB (Y)
		b	SE	p		b	SE	p		b	SE	p		b	SE	p
Parent Edu (X)	a₁	.09	.015	<.001	a₂	.129	.019	<.001	a₃	.294	.024	<.001	c^′	−.012	.0145	.464
PE (M₁)													b₁	.001	.025	.983
HLE (M₂)													b₂	−.021	.020	.276
PI(M₃)													b₃	.055	.016	.001
Constant	i_M1	3.856			i_M2	3.233	.111	<.001	i_M3	1.949	.136	<.001	i_Y	2.749	.136	<.001
		R²=.042				R²=.059				R²=.208			R²=.127
		F (6, 1310) =9.472, p<.001				F (13.579) =16.019, p<.001				F (6, 1340) =57.382, p<.001			F (9, 1307) =21.151, p<.001

Note. PE=parental expectation, HLE=Home learning environment, PI=parental involvement, LB=learning behaviors.

Discussion

Examining the results from multiple regression tests and mediation analysis, similar trends were identified. SES significantly and positively predicted home learning environment and parental involvement, and partially predicted parental expectation, supporting hypothesis 1. Home learning environment was significantly predicted by two SES factors which is in line with previous studies (Barnes & Puccioni, 2017; Miser & Hupp, 2012). Except for child age and child gender (no significant correlations) and family size (significant negative correlation), all other variables significantly positively correlated to parental involvement. These findings were also suggested by previous studies (Barnes & Puccioni, 2017; Poon, 2020). It is worth noting that family size negatively affects parental involvement which is also identified by (Prime et al., 2014). In 2016, one-child policy was lifted in China, this finding would raise valuable implications for Chinese families. As suggested by many studies, parental resources will be diluted as the children increase, however, the decrease in parental involvement and resources in big size family can be compensated by sibling interactions (Backer-Grøndahl & Nærde, 2017; Prime et al., 2014).

The relationships between SES and parental expectations are mixed. According to the results from multiple regression tests, parental expectation about their children’s future education level was only significantly predicted by school attendance and parents' education level. Also, data showed that 89.2% of parents expected their children to have a bachelor’s degree or above as shown in Table 1. However, mediation analysis demonstrated that parental expectation was linked to both family income and parental education, although with a relatively lower effect size in family income. It is consistent with the mixed results revealed by previous studies. For example, Poon’s (2020) study showed a strong and positive relationship between SES and parental expectations, whereas De Keyser et al. (2020) found an opposite path, reporting significantly higher academic expectations in low-SES households than high SES families.

Examining the studies that were specifically conducted in China, the relationships between SES and parental academic expectations are also mixed. For example, Butler (2014) found no differences between different SES groups regarding parental expectations of students’ English learning. Whereas consistent with the present study, the findings from Luo and Gao (2022) only identified the positive association between parental education level and parental expectations, not family income. This finding could be explained by parents’ cultural belief in “万般皆下品, 惟有读书高” (All jobs are inferior and only study is superior) and expectations of “望子成龙, 望女成凤” (Wishing the children become big shots), indicating that Chinese parents widely hold a high expectation on their children’s education with no relation to SES and other family characteristics.

However, differing from previous studies (Backer-Grøndahl & Nærde, 2017; Luo & Gao, 2022), hypothesis two was not supported, indicating that SES did not present significant direct effects on children’s learning behaviors. Moreover, except for parental involvement, all other two variables were not found significant individual indirect effects. One possible reason for this inconsistency is that the present study only focused on 3–5 years children. As identified by previous studies (Warash et al., 2017), parents of preschoolers tended not to emphasize learning. Additionally, the subscales of child’s learning behaviors measurement could ascribe to these non-significant findings. For example, parents were asked if their children respected rules, concentrated in class, and persisted in what is doing. More than child’s learning behaviors, these subscales reflected a cultural and moral value in respecting orders and perseverance which is highly emphasized in many Chinese families (Li et al., 2014). These cultural values could be less correlated to SES or other family characteristics.

Limitations and Future Directions

This study has some limitations that should be taken into account when interpreting the results. First, future studies could include other longitudinal research methods, for example, experimental and interventional studies, to further investigate the causal conclusions and establish the direction of the associations between parental involvement and children’s learning behaviors. Qualitative research may also be needed to help interpret the associations between family and parental factors and children’s learning. Second, in terms of the CFPS measurements, as the lack of standard, it solely depends on the researchers to interpret which set of questions measures a specific variable. Therefore, slightly different from Luo and Gao (2022), the present study also included the interviewers’ report based on their observations to represent the home learning environment. Additionally, the CFPS measurements were self-reported and might cause the issue of social-desirability bias (Rosenman et al., 2011). To address the above concerns, future studies may employ existing inventories to measure the variables.

Conclusion

In summary, the findings from the current study found that both family income and parental education level were significantly related to home learning environment and parental involvement, whereas only parental education level predicted parental academic expectations. There were no statistically significant relationships between SES and children’s learning behaviors, but parental involvement had an indirect effect on children’s learning behaviors. I would consider this finding a good sign as SES is quite a fixed indicator that educators cannot do much to change. Parental involvement’s positive association with children’s learning behaviors independent of family income and parental education raises implications for educators and practitioners regarding designing parenting programs for families from diverse SES backgrounds. A question remains as to how to design a program that parents favor, can easily implement at home and meanwhile will maintain even after the intervention or experiment is over. Parental involvement as a popular topic has been extensively studied and quite a few specific techniques are proposed and examined, for example, dialogic reading, parent-child coviewing TV programs, reminiscing conversations, etc. Although as discussed above, these parental interaction techniques are tested to be strong predictors of children’s development and learning, few studies were done to explore how parents perceive parental involvement. As a result, a high dropout rate (40%) from parent intervention projects was identified by both Justice et al. (2015) and Lonigan and Whitehurst (1998). I believe that only when learning is reflecting both parents’ and children’s psychological characteristics and is situated in the context of both their social environment do parents and children learn best. Therefore, I proposed that in order to be successful, parental involvement program practitioners must recognize what parents and children truly need and want to learn and collaborate with them to design the programs that best suit both parents’ and children’s needs and interests.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Wei Mao

References

Adams

M. J.

(1990). Beginning to read: Thinking and learning about print. MIT Press.

Backer-Grøndahl

Nærde

(2017). Self-regulation in early childhood: The role of siblings, center care and socioeconomic status. Social Development, 26(3), 530-544. https://doi.org/10.1111/sode.12216

Barnes

Puccioni

(2017). Shared book reading and preschool children’s academic achievement: Evidence from the early childhood longitudinal study—birth cohort. Infant and Child Development, 26(6), e2035. https://doi.org/10.1002/icd.2035

Bronfenbrenner

Morris

P. A.

(2007). The bioecological model of human development. In Lerner

R. M.

Damon

(Eds), Handbook of child psychology: Vol 1 Theoretical models of child development (6th ed., pp. 793-828). John Wiley & Sons, Ltd. https://doi.org/10.1002/9780470147658.chpsy0114

Butler

Y. G.

(2014). Parental factors and early English education as a foreign language: A case study in Mainland China. Research Papers in Education, 29(4), 410-437. https://doi.org/10.1080/02671522.2013.776625

Cheung

C. S.-S.

Pomerantz

E. M.

(2012). Why does parents’ involvement enhance children’s achievement? The role of parent-oriented motivation. Journal of Educational Psychology, 140(3), 820-832. https://doi.org/10.1037/a0027183

Christenson

S. L.

(2004). The family-school partnership: An opportunity to promote the learning competence of all students. School Psychology Review, 33(1), 83-104. https://doi.org/10.1521/scpq.18.4.454.26995

Dearing

Tang

(2009). The home learning environment and achievement during childhood. In Christenson

S. L.

Reschly

A. L.

(Eds), Handbook of school-family partnerships (pp. 131-157). Routledge.

De Keyser

Bakker

Rathé

Wijns

Torbeyns

Verschaffel

De Smedt

(2020). No association between the home math environment and numerical and patterning skills in a large and diverse sample of 5- to 6-year-olds. Frontiers in Psychology, 11, 547626. https://doi.org/10.3389/fpsyg.2020.547626

10.

Galindo

Sheldon

S. B.

(2012). School and home connections and children’s kindergarten achievement gains: The mediating role of family involvement. Early Childhood Research Quarterly, 27(1), 90-103. https://doi.org/10.1016/j.ecresq.2011.05.004

11.

Hayes

A. F.

(2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed). Guilford Press.

12.

Justice

L. M.

Logan

J. R.

Damschroder

(2015). Designing caregiver-implemented shared-reading interventions to overcome implementation barriers. Journal of Speech, Language, and Hearing Research: JSLHR, 58(6), S1851-S1863. https://doi.org/10.1044/2015_JSLHR-L-14-0344

13.

Kwok

Y. Y.

(2015). Understanding parental strategies in supporting children’s second language learning from the perspectives of bourdieu’s concept of capital: A study of Hong Kong Chinese parents. [Hong Kong Institute of Education] http://libdr1.ied.edu.hk/dspace/bitstream/2260/4350/1/image_file.html

14.

Lau

E. Y. H.

Rao

(2011). Parental involvement and children’s readiness for school in China. Educational Research, 53(1), 95-113. https://doi.org/10.1080/00131881.2011.552243

15.

Fung

Chen

E. C.-H.

(2014). Taiwanese parent–child conversations for moral guidance: Uncovering the ubiquitous but enigmatic process. In Wainryb

Recchia

H. E.

(Eds), Talking about right and wrong: Parent-child conversations as contexts for moral development (pp. 71-97). Cambridge University Press. https://doi.org/10.1017/CBO9781139207072.006

16.

Loft

Waldfogel

(2021). Socioeconomic status gradients in young children’s well-being at school. Child Development, 92(1), e91-e105. https://doi.org/10.1111/cdev.13453

17.

Lonigan

C. J.

Whitehurst

G. J.

(1998). Relative efficacy of parent and teacher involvement in a shared-reading intervention for preschool children from low-income backgrounds. Early Childhood Research Quarterly, 13(2), 263-290. https://doi.org/10.1016/S0885-2006(99)80038-6

18.

Luo

Gao

(2022). Family SES and self-regulated learning in Chinese preschoolers: A mediation analysis of parental educational expectation and home-based involvement. Early Education and Development, 33(3), 452-468. https://doi.org/10.1080/10409289.2021.1958580

19.

Mao

(2020). Play pedagogy in children's ESL learning: Parents and teachers' perspectives (Publication No. 1286849274) [Master\x{2019}s thesis, Thompson Rivers University]. Theses Canada. https://www.bac-lac.gc.ca/eng/services/theses/Pages/item.aspx?idNumber=1286849274

20.

Mao

Pesco

(2022, May 14-20). Chinese parents' perspectives on supporting young children's language development through dialogic reading and co-viewing [Conference workshop] Canadian Society for the Study of Education, Canada

21.

Miser

T. M.

Hupp

J. M.

(2012). The influence of socioeconomic status, home environment, and childcare on child language abilities. Current Psychology, 31(2), 144-159. https://doi.org/10.1007/s12144-012-9139-0

22.

Ogg

Anthony

C. J.

(2020). Process and context: Longitudinal effects of the interactions between parental involvement, parental warmth, and SES on academic achievement. Journal of School Psychology, 78, 96-114. https://doi.org/10.1016/j.jsp.2019.11.004

23.

Peixoto

Cadima

Leal

(2022). Mothers’ educational level and literacy beliefs: Associations with home literacy experiences. International Journal of Early Years Education, 30(2), 167-183. https://doi.org/10.1080/09669760.2022.2025580

24.

Poon

(2020). The impact of socioeconomic status on parental factors in promoting academic achievement in Chinese children. International Journal of Educational Development, 75, 102175. https://doi.org/10.1016/j.ijedudev.2020.102175

25.

Prime

Pauker

Plamondon

Perlman

Jenkins

(2014). Sibship size, sibling cognitive sensitivity, and children’s receptive vocabulary. Pediatrics, 133(2), e394-e401. https://doi.org/10.1542/peds.2012-2874

26.

Rosenman

Tennekoon

Hill

L. G.

(2011). Measuring bias in self-reported data. International Journal of Behavioural & Healthcare Research, 2(4), 320-332. https://doi.org/10.1504/IJBHR.2011.043414

27.

Son

S.-H.

Morrison

F. J.

(2010). The nature and impact of changes in home learning environment on development of language and academic skills in preschool children. Developmental Psychology, 46(5), 1103-1118. https://doi.org/10.1037/a0020065

28.

Stevens

Featherman

D. L.

(1981). A revised socioeconomic index of occupational status. Social Science Research, 10(4), 364-395. https://doi.org/10.1016/0049-089X(81)90011-9

29.

Stormshak

Dishion

T. J.

Falkenstein

C. A.

(2009). Family-centered, school-based mental health strategies to reduce student behavioral, emotional, and academic risk. In Christenson

S. L.

Reschly

A. L.

(Eds), Handbook of school-family partnerships (pp. 228-245). Routledge.

30.

Super

C. M.

Harkness

(1986). The developmental niche: A conceptualization at the interface of child and culture. International Journal of Behavioral Development, 9(4), 545-569. https://doi.org/10.1177/016502548600900409

31.

Tan

C. Y.

Lyu

Peng

(2020). Academic benefits from parental involvement are stratified by parental socioeconomic status: A meta-analysis. Parenting, 20(4), 241-287. https://doi.org/10.1080/15295192.2019.1694836

32.

Terwee

C. B.

Bot

S. D. M.

Boer

M. R. de

Windt

D. A. W. M. van der

Knol

D. L.

Dekker

Bouter

L. M.

de Vet

H. C. W

(2007). Quality criteria were proposed for measurement properties of health status questionnaires. Journal of Clinical Epidemiology, 60(1), 34-42. https://doi.org/10.1016/j.jclinepi.2006.03.012

33.

Warash

B. G.

Root

A. E.

Doris

M. D.

(2017). Parents’ perceptions of play: A comparative study of spousal perspectives. Early Child Development and Care, 187(5–6), 958-966. https://doi.org/10.1080/03004430.2016.1237511

34.

Dai

Zhen

Wang

(2021). User guide for China family panel studies 2018. Institute of Social Science Survey, Peking University.

35.

Xie

(2014). An introduction to the China family panel studies (CFPS). Chinese Sociological Review, 47(1), 3-29. https://doi.org/10.2753/CSA2162-0555470101.2014.11082908

36.

Yang

Liu

Pang

(2020). Hope and creative self-efficacy as sequential mediators in the relationship between family socioeconomic status and creativity. Frontiers in Psychology, 11, 438. https://doi.org/10.3389/fpsyg.2020.00438

37.

Zhang

Jiang

Ming

Ren

Wang

Huang

(2020). Family socio-economic status and children’s academic achievement: The different roles of parental academic involvement and subjective social mobility. British Journal of Educational Psychology, 90(3), 561-579. https://doi.org/10.1111/bjep.12374

38.

Zhang

Lau

(2021). Home environment and development of English as a second/foreign language for young children in Asian contexts: A systematic review and meta-analysis. Early Education and Development, 0(0), 1-32. https://doi.org/10.1080/10409289.2021.1981065