Influence of socioeconomic status on children’s reading abilities: the mediating role of home learning environment and the moderating role of grade level

Abstract

Family socioeconomic status (SES) has been shown to be associated with children’s academic abilities. However, how SES influences children’s reading abilities across grades still warrants investigation. This study recruited 1st - through 6th -grade children (N = 1,534) and examined the role of Home Learning Environment in the relationship between SES and reading abilities. The results revealed that the number of books at home and the reading initiation age mediated the impact of SES on children’s reading abilities, whereas family teaching activities did not have the expected mediating effect. Grade level moderated the influence of SES and reading initiation age on reading; that is, as grade level increased, earlier reading initiation age and higher family SES contributed more to children’s reading performance. These findings illustrate how SES affects children’s reading through Home Learning Environment across grades, which offers practical insights for guiding family activities in promoting children’s reading capacity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03203-z.

Reading ability serves as a cornerstone for academic achievement and lifelong learning [1]. In China, the “National Reading Promotion Plan (2021–2035)” explicitly emphasizes fostering children’s reading literacy as a national strategy [2], while the newly enacted Family Education Promotion Law (2021) [3]underscores the state’s commitment to optimizing family educational environments. Against this backdrop, the disparity in reading abilities among elementary school students has attracted widespread attention. However, the mechanisms underlying these disparities remain poorly understood, particularly regarding how family socioeconomic status (SES) interacts with developmental contexts across grade levels, considering family involvement and school input vary as children grow up. This study addresses this gap by integrating policy imperatives with theoretical frameworks to elucidate how home learning environments mediate SES effects on reading development in Chinese primary schools from lower to higher grades.

Bronfenbrenner’s ecological systems theory posits that children’s development occurs within nested environmental layers [4]. In China’s competitive education ecosystem, SES determines access to material resources (e.g., books, tutoring) and cultural capital (e.g., parent-child reading practices), while the home learning environment acts as a proximal mediator of these effects [5]. Our study operationalizes this framework by examining how SES influences reading outcomes through home learning environment pathways, with grade level moderating these dynamics across developmental stages.

Family SES and children’s reading ability

Family SES represents a critical environmental factor in children’s reading development, with its influence mediated through multiple theoretical lenses. Drawing on Bronfenbrenner’s ecological systems theory [5], SES operates at the macrosystem level by shaping access to educational resources and cultural capital, while simultaneously influencing the microsystem through home learning environments [6]. Moreover, to fully elucidate SES’s role, we integrate three complementary theories: the​resource model [7], the social stratification theory [8], and the cumulative advantage/disadvantage model [9], each providing distinct yet overlapping mechanisms for understanding the influence of SES to academic performance. According to the resource model, SES directly enhances reading outcomes through material resources (e.g., book ownership, educational technology) and cultural capital (e.g., parent-child reading frequency, literary traditions) [7]. Beyond material resources, social stratification theory highlights how SES perpetuates systemic inequities in educational opportunities [8]. Based on cumulative advantage/disadvantage framework, family scientists assume that initial small differences between advantaged and disadvantaged children produce large differences in academic achievements over the life course that have positive or negative consequences of SES [9].

The Program for International Student Assessment (PISA) shows that children with high family SES have greater reading achievement than do those with low family SES [10, 11]. Another international large-scale academic evaluation project, “International Reading Literacy Progress Research” (PIRLS) [12], also highlights the important role of family SES in children’s early reading development. As time passes, SES advantages compound through cumulative processes. Children in high-SES families exhibit accelerating gains in reading fluency. This trajectory aligns with the cumulative advantage hypothesis, where early SES benefits (e.g., preschool literacy interventions) generate multiplicative effects over grades. Notably, Song and Wu [13] reported that children from families with lower monthly per capita income and lower parental literacy levels were more likely to be diagnosed with dyslexia.

The mediating role of home learning environment in the relationship between family SES and reading ability

The family investment model (FIM), which illustrates how families set children’s home learning environments, provides fundamental explanations underlying the influence of family SES on children’s reading ability. According to FIM, high SES enables families to invest in children’s learning environments, thus promoting children’s reading development. Vasilyeva and colleagues [14] tested the FIM in a non-Western setting and found that parental education and income predict children’s literacy skills through family investments (e.g., joint reading activities, access to books). Indeed, family investments may take different forms, known as material investment and emotional investment. Either investment can be manifested in the home learning environment.

Material investment is associated with family economic capacity, such as the affordability of adequate learning resources and a favourable learning environment [15]. For example, previous studies have shown that families with higher SES may provide their children with more book collections and more enrichment activities that are linked with children’s reading, cognitive, and academic outcomes [16–18]. In other words, children from families with higher SES can receive more and better reading and learning opportunities from the home environment, including material resources such as books at home, educational toys, and digital learning devices.

Emotional investments include interaction and communication with children, encouraging children’s expectations, and investing more time in engaging in children’s reading and writing activities as early as possible [15], which are related to higher parental occupational status and educational level. Parents with high SES share more practical experiences with their children, read more to their children as early as possible, and actively interact more with their children [19, 20]. For instance, parents who begin teaching their children reading earlier (i.e., print exposure) would gain better reading performance at grade 9 than those without print exposure [21]. Second, the Ecological Systems Theory (Bronfenbrenner), reading initiation age emerges from family-level interactions rather than individual child traits. It reflects structured parenting practices (e.g., scheduled shared reading sessions starting at infancy) that create a supportive microsystem for language development. High SES families are more likely to implement such routines, underscoring its role in mediating SES-related disparities in cognitive outcomes. These forms of parental engagement and investment in children’s literacy can provide children with a stimulating home learning environment, helping them develop learning-related abilities to promote further development.

Therefore, in line with FIM, in which SES is associated with children’s performance through material and emotional investments, this study hypothesizes that Home Learning Environment (including the number of books at home, the duration of home teaching activities, and children’s reading initiation age) may be an important mediating variable in the influence of family SES on children’s reading ability.

How SES influences children’s reading ability varies across grades

There remains a hot debate regarding to the impact of SES on reading as children grow older [22]. In early childhood, children are exposed mainly to the nurturing and home environment of their parents. As children transition to primary school (ages 6–12), formal education structures intensify, prompting home-school interactions to diversify into collaborative academic and behavioral support systems [23].

One possibility is that children from low-SES families benefit more from school education because the gap between school and family resources is greater for them [24]. Unified school education can balance the accumulated inequalities from family SES, resulting in a weakened influence of SES after school enrolment [13]. Moreover, another possibility exists. Although some studies suggest that family SES has a greater impact on the reading ability of preschool children [25], it can help ensure that children’s school readiness begets their effective acquisition in the school environment. For example, a study of 177 first-grade children revealed that individual differences in Chinese reading comprehension continue to increase over time, as better reading comprehension ability can lead to greater reading pleasure, higher reading frequency, and more reading practice, creating a virtuous cycle [26]. Moreover, families with higher SES also consistently provide children with more and better learning resources throughout their years of schooling, consolidating the original advantages of preschool. Thus, high SES provides a good start and sufficient support through the journey of learning and reading.

According to the cognitive-developmental theory [27], young children in early grades are in the concrete operational stage, where children are more reply on concrete instruction, such as parent-child shared book reading experience, which is highly related to family SES. For instance, early exposure to reading (i.e., younger reading initiation age) may yield stronger long-term benefits as children advance through school and face increasingly complex reading demands. In contrast, the influence of family teaching activities or the number of books at home may be more stable or subject to external compensatory factors, such as formal schooling [28]. Therefore, it is important to examine whether the strength of the relationship between each HLE component and reading ability changes as children move from lower to higher grades. While, when children progress to the formal operational stage [27], their literacy development becomes increasingly shaped by formal schooling and teacher’s instruction, which may mitigate SES-based disparities. Thus, it is expected that the impact of SES on children’s reading ability may become weak from lower grades to higher grades. Therefore, the present study explores the changes in SES’s impact on reading during primary school. This study examines the moderate effect of grade level on SES’s impact on reading. We are also interested in grade moderation on the mechanism of the SES effect; that is, Home Learning Environment factor may play an increasingly important role in enhancing children’s reading abilities as they progress to higher grades.

The present study

Above all, the present study aims to examine the effect of family SES on children’s reading development through the home learning environment. Moreover, the dynamic influence pattern of family SES on reading, as well as how each aspect of Home Learning Environment influences reading differently across different grades, demands more research. This study selects children in the 1st through 6th grades and uses the family investment model to comprehensively investigate the roles of different home learning environment variables in the relationship between family SES and elementary school children’s reading skills (reading fluency, reading accuracy, and reading comprehension) as well as the corresponding grade-level effects. The findings help clarify the mechanisms by which family SES affects different aspects of children’s reading skill development and provide important implications for how to establish a home learning environment to improve children’s reading ability.

Specifically, this study addresses three interrelated questions: (1) How does family SES influence reading abilities across elementary school grades? (2)To what extent do home learning environment dimensions (number of books at home, family teaching activities, reading initiation age) mediate these effects? (3) Will the influence of material resources and teaching activities in reading abilities remain stable across grades? (4) Does grade level moderate the pathways linking SES, home learning environments, and reading outcomes?

We posit (see Fig. 1): H1: Higher SES will predict stronger reading abilities through increased home literacy resources and earlier reading initiation. H2: The mediating role of reading initiation age will diminish with grade progression due to diminishing returns of early investments. H3: The influence of material resources and teaching activities in reading abilities would remain relatively stable across grades. H4: Due to the cumulative nature of early reading experiences, grade level would moderate the mediating effects of reading initiation age, number of books at home, and family teaching activities in the relationship between SES and reading ability, with the effects of reading initiation age, number of books at home and family teaching activities expected to become stronger in higher grades.

Fig. 1.

The theoretical model interpreting how family SES influences reading abilities across elementary school grades

Methods

Participants

A total of 1925 primary school children in grades 1–6 from a large provincial capital city were enrolled. Stratified sampling was conducted by gender, semester, and urban/rural area (Table 1). Among them, 1567 students provided complete responses to questions about the home learning environment, and 1534 individuals provided complete responses to questions about socioeconomic status. All analyses were based on a sample of 1534 participants.

Table 1.

The number of participants in each grade (N = 1534)

Grade	Semester	Age M (SD)	n	Gender Proportion
				Female	Male
1	1	6.77 (0.33)	124	40%	60%
	2	7.23 (0.32)	132	54%	46%
2	1	7.73 (0.39)	128	41%	59%
	2	8.31 (0.30)	130	43%	57%
3	1	8.70 (0.33)	130	43%	57%
	2	9.34 (0.36)	130	61%	39%
4	1	9.75 (0.33)	124	43%	57%
	2	10.35 (0.38)	118	49%	51%
5	1	10.71 (0.31)	129	44%	56%
	2	11.29 (0.31)	126	45%	55%
6	1	11.70 (0.29)	128	43%	57%
	2	12.29 (0.35)	135	38%	62%
Total		9.51 (1.77)	1534	45%	55%

Measures

Home learning environment questionnaire

Home Learning Environment questionnaire was revised from previous research [29]. Parents rated eight items on a five-point Likert-type scale. Factor analysis revealed three dimensions: number of books at home (Cronbach’s α = 0.711), home teaching activities (Cronbach’s α = 0.845) and reading initiation age (Cronbach’s α = 0.635). The number of books at home corresponds to material investment from FIM, and both home teaching activities and reading initiation age correspond to emotional investment.

The number of books at home represented home literacy resources and was composed of two items: children’s books and adults’ books (1 = less than 10, 2 = 10–50, 3 = 51–99, 4 = 100 or more than 100). Home teaching activities included three items and were designed to measure the duration of parent‒child teaching and learning activities. The first item was Chinese character teaching time per week, the second item was math teaching time per week, and the third item was shared book reading time per week. All three items were measured on a 5-point Likert scale (1 = 0 h, 2 = 3 or less than 3 h, 3 = 3–6 h, 4 = 6–12 h, 5 = 12 h or more). reading initiation age refers to the age of the children when they first started reading activities. Three items were included: Chinese character teaching age, shared book reading initiation age, and independent book reading initiation age. All three items were measured on a 7-point Likert scale (1 = 0–1 year old, 2 = 1–2 years old, 3 = 2–3 years old, 4 = 3–4 years old, 5 = 4–5 years old, 6 = 5–6 years old, 7 = never). The results of the confirmatory factor analysis indicated that the model fits well:χ²(17) = 110.177, CFI = 0.945, TLI = 0.909, RMSEA = 0.084, SRMR = 0.045.

SES questionnaire

Family SES consisted of parental occupation (PO), parental education (PE), and family income (FI) [30]. Parental occupation (PO) was categorized according to previous studies [31] and was coded into the following 10 categories. 1 = unemployed, 2 = agricultural labourers, 3 = industrial workers, 4 = business services, 5 = individual proprietorship, 6 = clerical workers, 7 = professional and technical personnel, 8 = private entrepreneurs, employees of foreign-funded or joint ventures, 9 = state-owned enterprises and public institutions, 10 = civil servants and military personnel. However, only the father’s occupation was included in the final SES score because factor analysis revealed a negative loading of the mother’s occupation and poor model fit (See Table S2). Parental education (PE) was divided into 5 levels. 1 = primary school or below, 2 = middle school, 3 = high school (regular/vocational), 4 = graduate (normal courses/short-cycle courses), 5 = postgraduate or above. Family income (FI) was assessed by annual household income. 1 = less than 100,000 RMB, 2 = 100,001–200,000 RMB, 3 = 200,001–300,000 RMB, 4 = 300,001–500,000 RMB, 5 = 500,001–1,000,000 RMB, 6 = more than 1,000,000 RMB (100,000 RMB is approximately 14,250 USD. According to the 2021 China Household Finance Survey, the average annual household income in urban China is 71,546 RMB per person, placing the total average household income at a moderate level in the measurement scale. The results of the confirmatory factor analysis indicated that the model fits well: χ²(2) = 26.003, CFI = 0.989, TLI = 0.966, RMSEA = 0.088, SRMR = 0.020.

Reading abilities

Reading ability was measured using three subtests [32]. The first subtest was one-minute character reading, which measured children’s reading fluency when identifying single characters; its Cronbach’s alpha reliability ranged from 0.89 to 0.95 for grades 1–6. The second subtest was 150-character reading. The child read 150 Chinese characters, and the experimenter took note of the number of correct characters. This subtest measured children’s reading accuracy for single characters and had a Cronbach’s alpha reliability coefficient of 0.95–0.99 for grades 1–6. The third subtest was passage reading. The child needed to read an entire passage and answer questions about the passage content afterwards. Three test indicators were obtained. The first one was words correct per minute (WCPM) when they were read aloud. This subtest was similar to one-minute reading but represented reading fluency in a more general manner. The second and third indicators were the performance of reading comprehension tests when read aloud when read aloud and when read silently, both of which represented the reading comprehension ability of the children [33]. The Cronbach’s alpha reliability of the passage reading test ranged from 0.55 to 0.77 in grades 1–6. The results of confirmatory factor analyses indicated an acceptable model fit (χ²(11) = 71.425, CFI = 0.996, TLI = 0.992, RMSEA = 0.053, SRMR = 0.014).

Procedure

Reading tests were conducted one-on-one in a school environment by experimenters who had been strictly trained. The family questionnaire, which included demographic questions, home learning environment questions, and SES questions, was distributed to parents through a Chinese questionnaire platform (https://www.wjx.cn/) and was completed online. This study was approved by the Ethics Committee of the faculty of psychology of Beijing Normal University, with a protocol number of 202,003,160,014. Informed consent form was explained in detail and signed by parents of all participants.

Analytical strategies

Data analysis was conducted in five stages. First, we analyzed missing values. There were missing data on father’s occupation, mother’s occupation, and the age at which children begin to read independently. The proportions of missing cases were 1.6%, 0.5%, and 0.6%, respectively. Little’s MCAR test indicates that the missing data mechanism is completely random (MCAR χ²(8) = 12.94, p =.114). Therefore, listwise deletion is adopted, retaining only the cases with complete data.

Second, because we used parent-rated scales for all home or family variables, we conducted a common method bias test on the questionnaire data using the Harman single-factor test. The results showed that there were 5 factors with eigenvalues greater than 1. The variance explained by the first factor was 28.38%, which is below the critical value of 40%. This indicates that there is no serious common method bias in this study [34]. Third, we conducted descriptive statistics for all variables.

After that, since variables such as socioeconomic status, home learning environment, and reading ability have multiple different observed indicators, exploratory and confirmatory factor analyses were used to explore the internal factor structure of each variable. Principal component analysis was employed to extract factors from the observed indicators of the corresponding variables. The factor scores output by SPSS 25 software were used as the composite indicators of the variables for subsequent path analysis.

Finally, to investigate the effect of family socioeconomic status on the reading ability of children of different grades through the home learning environment, path analysis was conducted using Mplus 8.3 software with the Bootstrap method (5,000 replications). Socioeconomic status was used as the independent variable, home learning environment as the mediating variable, and reading ability as the dependent variable. Student grade (distinguishing between the first and second semesters) was used as a moderating variable, while gender and school location were controlled for. Grade has 12 levels and can was treated as a continuous variable (for example, 1 represents Grade 1 Semester 1, 1.5 represents Grade 1 Semester 2). Gender and school location are categorical variables with two levels each, and they were dummy-coded, that is, the two levels were coded as 0 and 1 (for gender, 0 = Female, 1 = Male; for school location, 0 = suburb, 1 = urban area), respectively. The significance level for statistics was set at α = 0.05.

Results

Descriptive analysis

Descriptive statistics show the means, standard deviations, minimum and maximum values, skewness, and kurtosis of the measures in Table 2. The skewness and kurtosis values were all between − 2 and 2, meaning that the variables followed a normal distribution [35].

Table 2.

Descriptive statistics (N = 1534)

Variables and Measures	M	SD	Min	Max	skewness	kurtosis
Reading abilities
One-minute reading	91.02	28.56	3	160	−0.37	0.44
150-character reading	92.63	37.91	0	148	−0.82	−0.28
Words correct per minute (read aloud)	149.06	55.65	0	379.19	0.06	0.43
Reading comprehension (read aloud)	0.61	0.26	0	1	−0.37	−0.63
Reading comprehension (read silently)	0.55	0.26	0	1	−0.17	−0.79
Home learning environment
Character teaching age	3.95	1.21	1	7	−0.01	−0.20
Shared book reading initiation age	4.04	1.28	1	7	−0.21	−0.16
Independent reading initiation age	4.42	0.99	1	7	−0.80	0.42
Character teaching time per week	2.58	0.97	1	5	0.75	0.14
Shared book reading time per week	2.72	0.93	1	5	0.66	−0.02
Math teaching time per week	3.01	1.14	1	5	0.27	−0.84
Number of adults’ books at home	2.17	0.99	1	4	0.51	−0.74
Number of children’s books at home	2.49	0.92	1	4	0.28	−0.81
Social economic status
Father’s education level	3.24	1.02	1	5	−0.12	−0.77
Mother’s education level	3.16	1.02	1	5	−0.10	−0.87
Annual household income	2.14	1.34	1	6	1.12	0.32
Father’s occupation	6.99	2.42	1	10	−0.94	−0.12
Mother’s occupation	6.28	3.06	1	10	−0.76	−0.96

The relationship between family SES and reading ability

The next goal was to investigate the associations between SES and reading ability in relation to home learning environment variables and whether grade level moderates the second stage of these mediation paths as well as the direct path from SES to reading ability, controlling for gender and region. Regarding the effects of control variables on outcomes, only school region had a significant association with the number of books at home (see Table S3). Mplus 8.3 software was used, employing the bootstrap method (with 5,000 resamples) for path analysis. Before conducting the path analysis, all variables were standardized, and a saturated model was constructed. Because each variable had multiple observed indicators, composite indices for each variable were constructed. Based on the structural models of the variables derived from the previous factor analysis, principal component analysis (PCA) was used to extract factors from the observed indicators of the variables. The factor scores output by SPSS 25 were used as the composite indices for each variable. The correlations between the composite indices and the observed indicators of each variable were calculated. The correlation coefficients between the composite index of SES and its observed indicators ranged from 0.621 to 0.892, whereas the correlation coefficients between the composite index of reading ability and its observed indicators ranged from 0.405 to 0.934. The correlation coefficients between the three composite indices of Home Learning Environment and their observed indicators ranged from 0.724 to 0.882. These results indicate that the composite indices can represent their corresponding observed indicators.

As shown in Fig. 2 the results indicate that SES had a significant positive association with reading ability. SES also showed significant positive associations with family teaching activities and the number of books at home. Additionally, SES was negatively associated with reading initiation age. The number of books at home correlated positively with reading ability. Both reading initiation age and family teaching activities demonstrated negative associations with reading ability. Family teaching activities, number of books at home, and reading initiation age served as significant mediators between SES and reading ability (β = −0.004, p =.039, 95% CI [− 0.007, − 0.001]; β = 0.055, p <.001, 95% CI [0.036, 0.075]; β = 0.018, p <.001, 95% CI [0.010, 0.026]). After introducing the mediators, the path coefficient of the direct pathway decreased (from β = 0.082 to β = 0.069), indicating that the mediating variables accounted for part of the association. Grade significantly moderated the direct pathway between SES and reading ability (β = 0.064, p =.001, 95% CI [0.026, 0.104]) and the indirect relationship between SES and reading ability through reading initiation age (β = −0.013, p <.001, 95% CI [− 0.021, − 0.007]). The higher the grade level is, the stronger the negative association between reading initiation age and reading ability and the association between of SES and reading ability. In contrast, there was no statistically significant moderating effect on the mediation of family teaching activities (β = 0.002, p =.210, 95% CI [− 0.001, 0.006] and no statistically significant moderating effect on the mediation of the number of books (β = −0.002, p =.835, 95% CI [− 0.025, 0.020]).

Fig. 2.

The influence of family socioeconomic status on reading ability: A moderated mediation model controlling for gender and school region. Note. The smaller the value of reading initiation age was, the earlier the parents began teaching their child to read, and the sooner the children started reading independently. Significant paths are represented by solid lines, whereas insignificant paths are indicated by dotted lines. *p <.05, **p <.01, ***p <.001

Discussion

This study examined the relationship between SES and children’s reading ability, and path analysis demonstrated significant associations between family SES and children’s reading development through home learning environment variables. Specifically, SES played a significant role in children’s reading development through the number of books at home and reading initiation age. Grade level moderated both the direct association between SES and reading ability and the mediating pathway through reading initiation age. These findings support family investment theory in terms of the mediating role of family investment in the influence of family SES on children’s development and extend to different aspects of children’s reading development.

Relationship between family SES and children’s reading ability development

This study revealed that family SES had a significant positive correlation with children’s reading skill development. In this study, family SES was measured through the father’s education level, the mother’s education level, annual household income, and the father’s occupation; the mother’s occupation was not included. In the factor analysis of family socioeconomic status, the loading of the mother’s occupation was negative. The observed negative loading of maternal occupation in our analysis may reflect systemic biases inherent in traditional occupational measurement scales. Such scales often inadequately capture the socioeconomic contributions of female-dominated or caregiving roles (e.g., part-time work, informal employment, or homemaking), potentially undervaluing mothers’ occupational status. Additionally, we found a high correlation between mother’s and father’s occupations, suggesting that the influence of maternal occupation might be partially obscured by father’s occupation in our models (See Table S1).

The significant positive association between family SES and children’s reading ability found in this study is consistent with the findings of previous studies [36, 37]. According to an OECD survey of nearly 7,000 children in the United Kingdom, the United States and Estonia in March 2020, family SES has a significant effect on the development of children’s four key areas: early literacy, early arithmetic, self-regulation and social emotion. Children with high family SES have greater levels of development in these areas than do children with low socioeconomic status [37]. Furthermore, the strength of the correlation between family SES and child development varies across countries and regions [38]. Specifically, in areas with good educational equity and educational quality development, school education can reduce the impact of family SES on children’s development. The difference in the correlation strength between SES and academic achievement is affected not only by factors such as the selection and measurement of indicators of SES and academic achievement and the calculation of the correlation coefficient but also by the fact that the samples originate from urban areas, suburbs or rural areas and that the socioeconomic development of the areas from which the samples originate may also have an impact on the correlation strength of both of them to a certain extent [39]. Although schools in different regions were stratified and selected in this study, school region was controlled as a control variable in the data analysis, and the moderating effect of different regions on the mediating role can be further analysed in the future.

Mediating role of the home learning environment

As hypothesized, this study revealed that family SES played a role in children’s reading skill development by enhancing Home Learning Environment through the mediating role of the number of books at home and reading initiation age. These findings align with family investment theory. In this theory, the family investment variables include two aspects: material investment and emotional investment. The former refers to the investment of family reading and writing resources, and the latter refers to the length and start time of parents’ investment in parent‒child reading and writing activities. The mediation model revealed that there were significant relationships between family SES and the number of books at home and reading initiation age. Parents with high family SES invest more literacy resources in children’s reading development, and they also tend to start their material and emotional investment in family education at an earlier age, which in turn promotes children’s literacy development.

However, the path analysis indicated that family teaching activities had a significant negative association with reading ability. This counterintuitive finding may primarily reflect qualitative aspects of parental involvement rather than purely quantitative time investment. Prior research has demonstrated that excessive parental control and negative communication during teaching activities can undermine children’s intrinsic motivation and autonomous learning skills [35–37]. When parental involvement becomes overbearing or overly directive, it may displace children’s self-directed practice and critical thinking. What’s more, we cannot rule out potential measurement ambiguities in the questionnaire (e.g., some parents possibly reporting current rather than preschool activities). Therefore, the current results should be interpreted cautiously and the impact of parental teaching activities still requires further investigation.

This study also emphasized the importance of reading initiation age. Previous studies have investigated the role of the quantity and quality of parental investment in parent‒child activities in children’s reading development [38]; the present study further revealed that the early or late initiation of parent‒child activities by parents was also an important and stabilizing factor that promoted children’s reading development and that parents with a high SES started investing in parent‒child literacy activities at an earlier age. They begin child language development and reading initiation activities early in their development, which can have both immediate and long-term effects on children’s language and reading development. This finding was also reflected in a landmark study by Hart and Risley [40], who estimated that by age 3, children in high-SES families heard 30 million more words than did children from low-SES families, whose reading development may be further promoted through the beneficial cycle of vocabulary and reading. Previous research also revealed that parent-child literacy activities could lay the neural foundation for reading ability, indicating that earlier initiation of literacy activities leads to earlier start of the neural and cognitive aspect of language development. For example, parent‒child conversations promote the development of functions in the linguistic neural region, namely, the left inferior frontal gyrus, and, in turn, increase children’s verbal scores [41].

Relationship between family SES and children’s reading ability: the moderating role of grade

The results showed that grade level significantly moderated the negative association of reading initiation age and reading ability. Specifically, the higher the grade level, the stronger the negative predictive effect of reading initiation age on reading ability. This finding suggests that the relationship between reading initiation age and reading ability becomes more pronounced as children progress through school.

The moderating effect of grade level may reflect the cumulative advantage of early reading experiences in meeting the escalating cognitive and literacy demands of higher grades.One possible explanation for this moderation effect is that the demand for independent reading skills increases as children advance to higher grades. In early grades, reading instruction is often highly structured and supported by teachers, which may mitigate the impact of individual differences in reading initiation age. However, as children move to higher grades, they are expected to read more autonomously and engage with increasingly complex texts. At this stage, children who started reading at an earlier age may have already developed stronger foundational reading skills and intrinsic motivation, enabling them to adapt more effectively to these demands [42]. In contrast, children who started reading later may struggle to keep pace, leading to a widening gap in reading ability. Additionally, the transition to higher grades often involves a shift from learning to read to reading to learn. This shift places greater emphasis on comprehension, critical thinking, and the ability to extract meaning from texts—skills that are more likely to be developed and reinforced in children with an earlier start in reading.

At the elementary school level, due to China’s school district system, families typically choose to live near their child’s school. To ensure that their children can complete the entire school phase, they generally do not move frequently. This means that the living space of the family remains relatively fixed throughout the school period [43]. Because homes in school districts are often smaller due to high rents, the limited living space further restricts the amount of room available for storing books. As a result, the number of books in the home is usually limited. School resources, such as reading corners in classrooms, can help compensate for the disparities in resources between different households, which is why grade level did not significantly moderate the mediating effect of the number of books at home in this study.

The study also revealed that grade level did not significantly moderate the mediating effect of family teaching activities, which may also be due to measurement issues caused by unclear statements in the questionnaire. Furthermore, during elementary school years, teachers’ guidance may play a more dominant role, and the importance of family teaching activities may be less pronounced. This result is related to the quality of parent‒child interactions. With the rapid development of mobile communication technology and the prevalence of smartphone use, parents are prone to accompanying their children while concentrating on mobile media regardless of the family demographics [44]. Smartphone use was highly correlated with parents’ disengagement from children and parent-child activities, which can lead to a situation where the quality of the accompaniment is low even though the accompaniment time is sufficient. This situation makes it difficult to improve children’s reading ability. Yeo et al. [45] reported that parental engagement in family reading activities has the greatest impact on children’s interest in reading after controlling for the age of the child and the parents’ level of education and that interest in reading also has an effect on children’s reading ability. Another possible explanation is that this study is the individual differences in reading ability. Some high-achieving students in lower grades may demonstrate greater reading ability than low-achieving students in higher grades [46]. In addition, parents may dynamically adjust the time invested according to children’s reading ability. If children’s reading ability is greater, the amount of time parents invest in parent‒child literacy activities will decrease accordingly. Therefore, it is possible that this is one of the reasons why the role of the length of parent‒child literacy activities was not reflected.

The grade level’s moderating role was also found to be significant in the direct relationship between SES and reading ability. The higher the grade level is, the stronger the effect of SES on reading ability. There are several potential reasons for this finding. First, in lower grades, school education may play a stronger compensatory role for children from different SES backgrounds. However, as children progress to higher grades, individual differences in learning ability become more pronounced [26]. Children from lower SES families, due to a lack of extracurricular support, may not be able to take full advantage of the resources provided by schools. This suggests that as grade level increases, the compensatory effect of school education diminishes, and children from higher SES families benefit more from the additional resources they have access to outside of school. A previous study [47] revealed that additional early schooling before elementary school can increase children’s cognitive test scores at the end of the first school year, including literacy tests, and the effect is greater for boys with low family SES. Although this finding showed that school education can reduce the gap between different SES levels, the beneficial effect on test scores disappears at age 11, which is consistent with our study results. Consequently, the influence of SES on reading ability becomes more significant in higher grades. Children from higher SES families are often encouraged from an early age to develop learning habits and a love for reading. Therefore, in higher grades, these abilities become more prominent, amplifying the impact of SES on reading ability.

Limitations and future directions

This study revealed the direct association between SES and reading ability. Specifically, within Home Learning Environment, the number of books at home and the earliest reading initiation age serve as mediating factors in the relationship between SES and reading ability.We also found that grade level played a significant moderating role in the mediation of reading initiation age. These results supported the family investment model. However, there are still shortcomings in this study, which are reflected mainly in the following aspects. First, this study used a cross-sectional design. A longitudinal approach could be used to examine the critical period when family SES is more likely to influence children’s reading ability. Second, we used composite indices instead of latent variables in our study. Initially, we attempted to construct a model using latent variables but the model failed to converge. These issues may have arisen because of the wide age range of our sample, which likely led to significant variability in the observed indicators and made it difficult to achieve model convergence. While this approach allowed us to address our research questions effectively, future studies with more homogeneous age groups or larger sample sizes may be better suited to employ latent variable modeling and further validate our findings. Third, the development of children’s reading ability is also affected by their schools, especially because children from different families are likely to attend schools at different levels, and differences in educational resources, teacher strength, educational styles, and learning atmospheres in schools are likely to have an impact on children’s reading interests and habits, which in turn affects their reading ability. In this study, the school factor was not taken into account when examining the impact of family SES on children’s reading ability. Family and school factors can be further considered together to obtain more accurate conclusions. Fourth, as mentioned before, the measure of Family teaching activities in our study could be inaccurate due to ambiguous wording in the questionnaire. Future study would benefit from more nuanced assessments of Family teaching activities. Finally, although the study focused on specific components of family investment, it did not account for other potentially important factors, such as parents’ interest in reading [48], parents’ stress [49], and parental self-efficacy [50], all of which have demonstrated associations with children’s literacy development.

Conclusions and educational implications

This study revealed that family SES affects children’s reading ability through the mediation of the number of books at home and reading initiation age. Moreover, grade level moderated the direct effect of SES on reading ability and the mediating effect of reading initiation age. These findings theoretically support the family investment theory that family SES influences children’s development and may provide practical guidance for family investment activities in children’s reading development. Specifically, regardless of family SES, parents should consciously provide reading initiation for their children before school entry. In the process of parent‒child reading with children, they not only read books to their children but also encourage communication and discussion to promote the development of children’s reading ability. Moreover, as children grow, parents should emphasize the construction of the reading environment; this is reflected not only in the purchase of appropriate reading books for children but also in parents emphasizing the cultivation of their own reading habits and setting an example for children.

Schools should also strengthen the joint education work of home and school. Schools can provide students with rich and interesting reading materials, regularly carry out activities such as reading cards and good book-sharing clubs to create a reading environment for children, try to compensate for the differences in children’s reading abilities caused by the differences in the economic status of their families, and encourage children to complete the transformation from “learning to read” to “learning in reading”.

At the community level, educator-led parent training workshops could teach adaptive scaffolding techniques while book-sharing initiatives may help mitigate resource disparities. Policy efforts should simultaneously focus on expanding access to early childhood programs that provide literacy-rich environments and public library partnerships offering family literacy support. Such multi-level approaches may help address the material and behavioral dimensions of reading development while acknowledging systemic constraints. Future interventions might particularly benefit from combining these home-based strategies with institutional supports to create more equitable opportunities for literacy acquisition across SES groups.

Overall, this study advances existing literature in three critical ways. First, it bridges the micro-level (family SES) and meso-level (school transition) perspectives by integrating grade-level moderation into the analytical framework. Second, it operationalizes home learning environment through a culturally tailored instrument validated for the Chinese context, incorporating both material resources (e.g., number of books) and cultural practices (e.g., early reading initiation). Third, by adopting a cross-grade design (Grades 1–6), this study captures developmental nuances often obscured in cross-sectional research, thereby enhancing ecological validity for policy implementation.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Yageng Li and Yingqi Yu conceptualized and designed the study. Yingqi Yu collected and organized the data. Yageng Li performed data analysis. Yageng Li drafted the manuscript. Siqi Zhang, Minqi Gao and Xiujie Yang reviewed and edited the manuscript. All authors read and approved the final version of the manuscript.

Funding

This research was supported by the Beijing Social Science Foundation (23JYC016) to Xiujie Yang.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Human ethics and consent to participate declarations

This study was conducted in accordance with the ethical principles stated in the Declaration of Helsinki and was approved by the Ethics Committee of the faculty of psychology of Beijing Normal University (Approval number: 202003160014). Prior to participation, all participants provided written informed consent after receiving detailed explanations of the study’s objectives, procedures, risks, and benefits. Participants were informed that their involvement was voluntary, with the option to withdraw at any time.

Consent for publication

All authors read and approved the final version of the manuscript.

Competing interests

The authors declare no competing interests.