Abstract
The importance of vocabulary knowledge in reading comprehension is well-recognized, and its relationship with comprehension has been widely explored in previous studies. However, there is limited research on the longitudinal relationships between them, particularly the reciprocal relations between vocabulary breadth, vocabulary depth, and reading comprehension. The present study aims to examine the contributions of vocabulary breadth and depth to reading comprehension over time as well as the reciprocal relationships between them among adolescent Chinese senior middle school students. Using structural equation modeling and a cross-lagged panel analysis, the study found that both vocabulary breadth and depth made significant contributions to reading comprehension. Vocabulary breadth was a more robust predictor of reading comprehension for 10th and 11th graders compared to vocabulary depth. However, the contribution of vocabulary depth to reading comprehension became increasingly significant as students advanced through higher grades. In addition, vocabulary breadth was reciprocally related to vocabulary depth and reading comprehension, whereas there were no reciprocal relations between vocabulary depth and reading comprehension. These findings suggested that the pattern of relationships may vary as a function of unsystematic progression in the acquisition of different aspects of vocabulary knowledge over time.
Introduction
The comprehension of written text, regardless of language, entails the utilization of a variety of linguistic processes, including accessing vocabulary, parsing grammatical structures, and constructing semantic representations (Grabe 2009; Koda 2005; Perfetti and Stafura 2014). Notably, vocabulary knowledge has been consistently recognized as one of the most influential components in reading comprehension. It is firmly established as a pivotal factor contributing significantly to this complex cognitive process. The National Reading Panel (NRP) reported: “reading comprehension…cannot be understood without examining the critical role of vocabulary learning and instruction and its development” (National Reading Panel 2000, p. 13). Research has indicated that reading comprehension impairments often stem from encountering unfamiliar words (Moghadam et al. 2012). Laufer (1997) highlighted that text comprehension was hindered without a grasp of the text’s vocabulary, whether in first or second language learning. Indeed, the importance of vocabulary in reading comprehension is also theoretically supported by the Lexical Quality Hypothesis (Perfetti 2007), which posits that efficient lexical access and the presence of high-quality lexical representations are attributed to enhancing reading comprehension. Despite the weight of vocabulary knowledge in reading comprehension, previous studies have been conducted to investigate the correlations between vocabulary knowledge and reading comprehension (Stanovich 2000) or the longitudinal development of vocabulary and reading comprehension (Sparks and Deacon 2015). In these studies, the focus has primarily been on one aspect of vocabulary knowledge, particularly on vocabulary breadth. However, acquiring word knowledge involves both a superficial understanding and a comprehensive comprehension (Qian 1999; Nation 2001) as explicitly outlined in the General Senior High School English Curriculum Standards (2017 edition)(Ministry of Education 2020) in China. During high school, students are expected to master the basics of word sound, form, and meaning, as well as to expand their vocabulary knowledge through in-depth word learning. Specifically, students are challenged to not only increase their vocabulary through word formation knowledge but also to master the precise application of multiple word meanings, encompassing the breadth and depth of vocabulary knowledge. Even though previous studies have examined the concurrent effects of vocabulary breadth and vocabulary depth on reading comprehension, there was limited research on the longitudinal relations between them, particularly the reciprocal connections between the breadth and depth of vocabulary knowledge and reading comprehension. Therefore, the current study aimed to investigate the mutual relationships between vocabulary breadth, vocabulary depth, and reading comprehension among Chinese-speaking adolescents learning English as a foreign language (EFL) in senior middle schools located in the economically underdeveloped Ningxia Hui Autonomous Region in Western China.
Dimensions of vocabulary knowledge
It is well-established that vocabulary knowledge is a rich and multifaceted construct (Koda 2005; Schmitt 2010), and its conceptualization is a challenging task, given various dimensions and facets to consider (Stahl 1999). Aitchison (1987, p. 84) aptly describes word knowledge as a “web” of words, emphasizing the interconnected and multifaceted nature of vocabulary knowledge. This metaphor suggests that vocabulary is not a linear acquisition but a network of interwoven elements. Vocabulary knowledge has been categorized into distinct types based on different approaches. In terms of component skills approach, vocabulary knowledge can be categorized into different facets. Aristotle, as early as the fourth century BC, classified it into spoken form, written form, and meaning. Palmer (1921) expanded on this by introducing the concepts of receptive and productive word knowledge, which were also known as passive and active knowledge. Furthermore, Anderson and Freebody (1981) added dimensions of vocabulary breadth and vocabulary depth to this categorization. In terms of a developmental approach, vocabulary knowledge is acquired progressively. Dale (1965) proposed a four-stage model of vocabulary acquisition: (1) unfamiliarity with the word, (2) hearing the word without understanding its meaning, (3) recognizing the word in context with partial understanding, and (4) fully knowing the word. Paribakht and Wesche (1993) extended this model with a five-stage approach: (1) complete unfamiliarity, (2) familiarity without meaning, (3) providing a correct synonym or translation, (4) using the word with semantic appropriateness, and (5) using the word with both semantic appropriateness and grammatical accuracy. Dale’s four-stage model emphasizes the foundational role of recognizing the form of words in advancing a comprehensive understanding of vocabulary. In contrast, Paribakht and Wesche’s five-stage model extends beyond the mere recognition of physical form and meaning. Their model introduces additional stages, proposing that the acquisition of knowledge about how words are used in context, as well as understanding their semantic appropriateness and grammatical functions, typically occurs in later stages of vocabulary development.
Vocabulary knowledge in developmental approaches is closely linked to word knowledge in component skills approach. In other words, word form and meaning as well as vocabulary breadth and vocabulary depth are acquired in different stages, and the word forms and meaning are integrated into vocabulary breadth and vocabulary depth. The study of vocabulary breadth, both in first and second languages, has long captivated researchers due to its pivotal role in effective communication and language proficiency. Initially, learners’ language skills hinged heavily on their vocabulary breadth, as expressed by McCarthy (1990), who emphasized that meaningful communication in L2 is contingent on a sufficient number of vocabulary. Nation ISP (1993) further underscored the crucial link between vocabulary breadth and skill development in language use. The previous research has consistently demonstrated a strong correlation between vocabulary breadth and language proficiency; a larger vocabulary corresponds to higher language proficiency. Studies examining vocabulary breadth (Biemiller 1999) revealed disparities among children with varying vocabulary breadth, particularly noticeable in fourth graders. Vocabulary breadth focuses on form-meaning linkage, which is usually tested through defining a given sample of words, providing L1 translations or synonyms for L2 target words, or selecting corresponding L1 translations or synonyms for target words from among several options. Studies measuring the breadth of learners’ vocabulary knowledge often solely assess whether L2 learners understand the meanings of L2 words. While this is important for gauging their potential comprehension, tests focused only on form-meaning connections can be misleading in that vocabulary knowledge encompasses various aspects beyond form and meaning, such as derivations, collocations, and associations. According to Schmitt (2010), vocabulary learning is not limited to the development of form-meaning linkage, and the deep and sufficient understanding of lexical items should also be considered. In addition to the breadth of vocabulary knowledge, the depth of vocabulary has been typically defined by the following researchers. Anderson and Freebody (1981) defined it as the depth of vocabulary; Read (1993, p. 357) described it as “the quality of the learners’ vocabulary knowledge”; Wesche and Paribakht (1993, p. 13) defined depth “in terms of kinds of knowledge of specific words or in terms of degrees of such knowledge.”; Henriksen (1999) conceptualized depth of knowledge as one of the three dimensions of lexical competence inclusion of partial to precise knowledge, depth of knowledge, and receptive to productive use ability. By the same token, Read (2004) classified the depth of knowledge into three components: precision of meaning, comprehensive word knowledge, and network knowledge. Although these definitions and classifications have different focuses, they all contain two facets: the extent to which a word’s meaning or form is known and the multiple aspects of word knowledge including semantic, orthographic, morphological, syntactic, collocational, and pragmatic characteristics. In the current study, our examination of vocabulary knowledge centered around two key aspects: vocabulary breadth and vocabulary depth.
The relations between vocabulary breadth and vocabulary depth
Vocabulary knowledge is recognized as a multidimensional construct comprising two key facets: breadth and depth (Anderson and Freebody 1981). While these two dimensions has been defined differently, extensive literature (Li and Kirby 2015; Qian 1999; Vermeer 2001) has consistently revealed a close association between them. For instance, Qian (1999, 2000) reported correlation coefficients of 0.82 and 0.70 for the relationship between the breadth and depth of vocabulary among English as a second language learners in universities. Li and Kirby (2015) reported correlations of 0.51 for Chinese high school students learning English as a second language. In Vermeer’s (2001) study, correlations of 0.85 between vocabulary breadth and vocabulary depth in Dutch monolingual kindergarteners and of 0.76 in Dutch bilingual kindergarteners were found, which led her to contend that there was essentially no distinction between the breadth and depth of vocabulary. However, Qian (2002) argued that the shared variance of 0.49 indicated that vocabulary breadth and depth captured distinct facets of vocabulary knowledge despite a robust correlation of 0.70 between them. According to Nurweni and Read (1999), the two aspects of vocabulary knowledge may coalesce when learners progress to advanced levels, whereas they tend to manifest greater differentiation at lower levels of language proficiency. Based on the above definitions for breadth and depth of vocabulary, they both were considered two distinct constructs in the present study. The two aspects of vocabulary knowledge were not only highly correlated but interconnected (Qian 2002; Perfetti 2007); therefore, it was assumed that there might be a reciprocal relations between breadth and depth of vocabulary across time even though there was few studies exploring the mutual predictive power between them. Despite the wealth of literature on the interrelations of breadth and depth, a noticeable gap exists in the exploration of the developmental relationship between these dimensions. Specifically, there is a lack of studies investigating whether vocabulary breadth and vocabulary depth at Time 1 predict the corresponding dimensions at Time 2. Addressing this gap is crucial for a comprehensive understanding of how these facets evolve over time and mutually influence each other in the developmental trajectory of vocabulary knowledge.
The relations between breadth and depth of vocabulary and reading comprehension
The significance of possessing a rich vocabulary for effective reading comprehension has garnered longstanding acknowledgment within the academic domain (Anderson and Freebody 1981; Read 2000). Anderson and Freebody’s (1981) instrumentalist hypothesis posits that vocabulary knowledge is a crucial prerequisite and a causal factor in comprehension. According to this hypothesis, readers must be acquainted with the actual words in the text to comprehend it effectively. To be specific, when an individual has a deeper understanding of the words encountered in a passage, he/she is able to grasp the message and improve reading comprehension. Perfetti’s lexical quality hypothesis (1985, 2007) complements the instrumentalist hypothesis by highlighting the significance of mastering word knowledge. It suggests that the key to achieving optimal reading comprehension lies in the acquisition of high-quality word knowledge. Empirical studies further support Perfetti’s theory, revealing a robust and consistent relationship between vocabulary knowledge and reading comprehension across various stages of development (Muter et al. 2004; Snow 2002; Verhoeven and Van Leeuwe 2008), indicating that individuals with a more extensive vocabulary exhibit superior comprehension skills. The majority of prior research examining the concurrent and longitudinal connections between vocabulary knowledge and reading comprehension predominantly employed vocabulary breadth as the indicator for assessing individuals’ lexical proficiency (Nation and Snowling 2004; Proctor et al. 2005; Quinn et al. 2015; Ricketts et al. 2007). For instance, in the study by Ricketts et al. (2007), vocabulary breadth emerged as a robust predictor, accounting for 17.8% portion of unique variance in reading comprehension. Similarly, Proctor et al. (2005) investigated the predictive role of decoding and oral language in reading comprehension for fourth-grade ESL students. It was found that vocabulary breadth was highly correlated with reading comprehension (r = 0.73) and made significant contribution to reading comprehension. In these studies, vocabulary knowledge was operationalized as vocabulary breadth, which may lead to an incomplete representation of findings. However, there has been a scarcity of studies investigating the concurrent and dynamic correlations between the breadth and depth of vocabulary and reading comprehension, with the exception of studies conducted by Qian (1999), Nation and Snowling (2004), Li and Kirby (2015), and Binder et al. (2017). For example, Qian (1999) investigated the associations between breadth and depth of vocabulary knowledge and reading comprehension for English as a second language learners in university. The results showed that vocabulary breadth and depth were closely correlated and they both were also related to reading comprehension. Furthermore, vocabulary breadth and depth made significant contributions to reading comprehension, especially the depth of vocabulary made unique contribution to reading comprehension after controlling for vocabulary breadth, suggesting that the depth of vocabulary knowledge played a more important role in reading comprehension. Similarly, Li and Kirby (2015) examined the relationships between vocabulary breadth and depth as well as their impact on reading comprehension for 46 Chinese-speaking English as a second language learners in Grade 8. The study revealed a moderate correlation between the breadth and depth of vocabulary. Vocabulary breadth demonstrated a significant predictive effect on multiple-choice reading comprehension while vocabulary depth played a role in enhancing summary writing, indicating that the distinctive contributions of vocabulary breadth and depth to diverse types of second language reading. In terms of dynamic relations between two dimensions of vocabulary knowledge and their effects on reading comprehension, Nation and Snowling (2004) conducted a concurrent and longitudinal study exploring the relationship between language skills and reading development of 72 children at age 8.5 and age 13. The finding indicated that vocabulary breadth and depth made concurrent contributions to reading comprehension, specifically, vocabulary breadth predicted 25.2% of the variance and vocabulary depth predicted 15.1% of the variance in reading comprehension. The results at Time l suggested that vocabulary breadth was more predictive than semantic depth for reading comprehension. Meanwhile, in a follow-up longitudinal analysis, it was found that vocabulary breadth and semantic depth at age 8.5 contributed 4.9% and 4.5% of the variance in reading comprehension at age 13, respectively.
Taken together, the aforementioned studies examining the concurrent and longitudinal relations between vocabulary breadth and vocabulary depth along with their effects on reading comprehension employed traditional regression analysis. However, this approach may not comprehensively address performance variances stemming from random measurement errors; consequently, resulting in overestimating the unique contributions of specific components to reading comprehension. Structural Equation Modeling (SEM) allows for the incorporation of latent variables, treating vocabulary breadth and depth as unobservable constructs inferred from multiple measured indicators. This is advantageous because it accounts for the multidimensionality and complexity inherent in these theoretical constructs. SEM also enables the simultaneous estimation of measurement and structural models, facilitating a more nuanced understanding of how vocabulary breadth and depth interact to influence reading comprehension. Additionally, SEM offers the capability to explicitly model error covariance, addressing measurement error in a way that traditional regression may overlook. All in all, SEM provides a more comprehensive and accurate representation of the contributions both vocabulary breadth and depth made to reading comprehension. Therefore, in the present study, we aimed to use SEM to examine the specific contributions of vocabulary breadth and depth to reading comprehension as well as the reciprocal relations between them. The following two research questions were addressed.
(1)
To what extend do vocabulary breadth and depth affect reading comprehension in EFL Chinese-speaking adolescents over time?
(2)
Are there reciprocal relations between vocabulary breadth, vocabulary depth and reading comprehension in EFL Chinese-speaking adolescents?
Methodology
Participants
The study involved 264 participants, comprising 148 boys and 116 girls, with an average age of 16.3 years. They were recruited from one senior public school in a city in Ningxia Hui Autonomous Region in China, and the majority of them come from families from diverse socioeconomic backgrounds. Their primary languages of communication at home are local languages. They started to learn English since the third grade in elementary schools. They have one English class for fifty minutes every day. Participants were tracked from tenth to eleventh grade for their English reading development.
Measures and instruments
Vocabulary breadth
The present study utilized the Updated Vocabulary Level Test (UVLT), initially devised by Nation ISP (1983) and subsequently revised by Schmitt et al. (2001) to evaluate participants’ vocabulary breadth within word-family levels spanning from 1000 to 5000. Considering participants’ English proficiency, the assessment focused on word-family levels ranging from 1000 to 3000 words, aiming to gauge their ability to discern form-meaning mappings of words. Participants were required to match item numbers with corresponding Chinese definitions across specific word-family levels. The UVLT included 30 questions per level, presented in a matching format. Each level comprised 10 clusters of six words, including three key words and three distractors, along with three corresponding definitions in Chinese. The distribution of nouns, verbs, and adjectives within each level remained consistent with prior versions, featuring 15 nouns, nine verbs, and six adjectives. Scoring followed a system where each item earned 3 points, resulting in a total score of 30 points for each level. An illustrative sample was provided below. The reliability (Cronbach α) of the vocabulary breadth test at two time points were 0.73 and 0.76, respectively.
game island mouth movie song yard
岛屿 ⃞
嘴巴 ⃞
歌曲 ⃞
Vocabulary depth
The depth of vocabulary knowledge was measured by the Word Associate Test (WAT) developed by Read (1993). The WAT focused on the assessment of word associations including synonymy, polysemy, and collocations. This test consisted of 40 items, each followed by two boxes with four words. The words in the left box were all adjectives with similar meanings and word class, and the words in the right box were all nouns and collocates of the target word. For example, participants were presented with a stimulus word beautiful followed by four adjectives A. enjoyable B. expensive C. free D. loud in the left column, and four nouns A. education B. face C. music D. weather in the right column. They were supposed to choose enjoyable on the left and face, music, and weather on the right. In order to counteract guessing, participants were not informed about the number of correct choices, and they were free to select any appropriate words based on their own judgment. The reliability (Cronbach α) of the vocabulary depth test at two time points were 0.77 and 0.75, respectively.
Reading comprehension
The reading comprehension test consisted of two sections: a multiple-choice section with five questions selected from a pool of seven, and a reading comprehension section with four passages—three narrative passages and one expository passage. These passages were selected from the National Matriculation English Tests administered between 2010 and 2020. Prior to the test administration, a final review was conducted by English teaching instructors and the cohort leader to ensure the test’s appropriateness for the participants’ English proficiency level. The reading comprehension test aimed to assess participants’ ability to grasp the main idea, extract specific information from the provided passages, and infer word meanings from context. The reliability (Cronbach α) of the reading comprehension test at two time points were 0.78 and 0.82, respectively.
Procedures and data collection
The data collection occurred twice with an interval of approximately one year. The tasks for vocabulary breadth, vocabulary depth, and the reading comprehension were administered during regular English classes. To ensure internal validity, tasks were conducted separately to mitigate participants’ fatigue and eliminate potential carry-over effects from preceding tasks. All tests were administered using traditional paper-and-pencil methods. Each test was conducted during different weeks, and the sequence of tasks was counterbalanced across various classes. The duration of each test varied, ranging from 15 to 40 minutes, contingent on the complexity and length of the task.
Results
Descriptive statistics and correlations
Table 1 showed the means and standard deviations of the participants’ performance on the vocabulary breadth, vocabulary depth, and reading comprehension tasks. Data were normally distributed with skewness ranging from −1.821 to 0.434, and kurtosis ranging from −0.857 to 4.631 (Kline 2016). At Time 2, participants exhibited superior performance compared to that at Time 1. The paired sample test revealed that there were significant differences in performance on standardized measures with the exception of the differences between the two aspects of depth of vocabulary at two points (p > 0.05). Table 2 showed the bivariate correlations between three levels of vocabulary breadth and two aspects of vocabulary depth through Grades 10 to 11. Overall, the correlations displayed low to moderate relationships within- and between-category concurrently and longitudinally with exception of the associations between two aspects of vocabulary depth across two time points, indicating insignificant relationship.
Table 1 Descriptive statistics for vocabulary breadth, vocabulary depth, and reading comprehension by grade (N = 264).
Full size table
Table 2 Correlations among all measures by grade (N = 264).
Full size table
Structural models for contributions of vocabulary breadth, vocabulary depth to reading comprehension across Grades 10 to 11
Figures 1 and 2 showed the contributions of vocabulary breadth and vocabulary depth to reading comprehension across Grades 10 to 11. The two models were identified and well-fitted with χ²(15, N = 264) = 41.919; CFI = 0.942; AGFI = 0.900; NFI = 0.915, RMSEA = 0.083); χ²/df = 2.759 at Time 1 and χ²(15, N = 264) = 13.824; CFI = 1.000; AGFI = 0.969; NFI = 0.978, RMSEA = 0.000); χ²/df = 0.922 at Time 2. For the model in Fig. 1 at Time 1, vocabulary breadth and vocabulary depth collectively accounted for 82.6% of the variance in reading comprehension. When considering the strength of the relationships among vocabulary breadth, vocabulary depth, and reading comprehension, it became evident that vocabulary breadth played a more influential role in predicting reading comprehension compared to vocabulary depth. Specifically, vocabulary breadth significantly contributed to reading comprehension (β̂ = 0.678, p < 0.001) whereas vocabulary depth did not have a statistically significant impact on reading comprehension (*β̂* = 0.141, *p* = 0.159 > 0.05). Meanwhile, for the model in Fig. 2 at Time 2, the explained variance in reading comprehension reduced to 30.6% in comparison to the variance explained at Time 1. However, both vocabulary breadth and depth made significant effects on reading comprehension (β̂ = 0.356, p < 0.001; β̂ = 0.231, p < 0.01), and vocabulary depth made more contribution to reading comprehension compared to vocabulary breadth. In conclusion, the influence of vocabulary depth on reading comprehension increased with grades.
Fig. 1: Structural model at Grade 10.
figure 1
This model, which explains 82.6% of the variance, illustrates the contributions of vocabulary depth and breadth to reading comprehension. Reading comprehension was significantly impacted by vocabulary breadth but not by vocabulary depth. Fit indices: RMSEA = 0.083, CFI = 0.942, χ²/df = 2.759.
Full size image
Fig. 2: Structural model at Grade 11.
figure 2
In Grade 11, both vocabulary depth and breadth demonstrated significant impacts, accounting for 30.6% of the variance in reading comprehension. Compared to vocabulary breadth, vocabulary depth was more important. Fit indices: RMSEA = 0.000, CFI = 1.000, and χ²/df = 0.922.
Full size image
Cross-lagged panel analysis between vocabulary breadth, vocabulary depth and reading comprehension
A cross-lagged panel analysis with two waves was utilized to examine the potential causal relationships between vocabulary breadth, vocabulary depth and reading comprehension. In order to obtain accurate results, the residuals for vocabulary breadth, vocabulary depth and reading comprehension at two time points were allowed to covary. Based on the previous studies, the comparative fit index (CFI) greater than 0.95 (Bentler 1990) and the root mean square error of approximation (RMSEA) less than 0.05 (Browne and Cudeck, 1992) showed a good model fit. The fit indexes indicated that the model had very good model fits: χ²(73) = 95.989, p < 0.05, CFI = 0.982, SRMR = 0.045, and RMSEA = 0.035. The estimates of unstandardized and standardized path coefficients, along with factor loadings of the individual tasks on the latent construct of vocabulary breadth, vocabulary depth, and reading comprehension in Grade 10 and Grade 11 were presented in Table 3. It was shown that all path coefficients and factor loadings were statistically significant with the exception of cross-lagged paths from vocabulary depth to reading comprehension and from reading comprehension to vocabulary depth across two grades. To attain model parsimony, the paths from English proficiency to vocabulary breadth, vocabulary depth and reading comprehension in Grade 11 were removed from Fig. 3 in that English proficiency was not significantly associated with these three variables in Grade 11. In reference to the autoregressive effects of vocabulary breadth, vocabulary depth and reading comprehension across Grades 10 and 11, vocabulary breadth and reading comprehension manifested stability over time (*β*^ = 0.455, *p* < 0.001; *β*^ = 0.288, *p* < 0.01); however, vocabulary depth did not show stable developmental pattern (*β*^ = −0.159, *p* = *0.16* > 0.05). In line with the reciprocal relationships between vocabulary breadth, vocabulary depth, and reading comprehension, Fig. 3 showed that vocabulary breadth was reciprocally related to vocabulary depth and reading comprehension through Grade 10 to Grade 11. Specifically, after controlling for English proficiency, vocabulary depth, and autoregressor, vocabulary breadth in Grade 10 had a significant consequence on reading comprehension in Grade 11 (β^ = 0.516, p < 0.001); conversely, after controlling for English proficiency, vocabulary depth, and autoregressor, reading comprehension in Grade 10 significantly predicted vocabulary breadth in Grade 11 (*β*^ = 0.229, *p* < 0.05). In the same vein, after controlling for English proficiency, autoregressor and reading comprehension, vocabulary breadth in Grade 10 had significant impact on vocabulary depth in Grade 11 (*β*^ = 0.550, *p* < 0.001); conversely, after controlling for English proficiency, autoregressor and reading comprehension, vocabulary depth in Grade 10 significantly predicted vocabulary breadth in Grade 11 (*β^* = 0.193, *p* <0.05). However, there was no reciprocal relationships between vocabulary depth and reading comprehension across Grade 10 and Grade 11. To be precise, vocabulary depth in Grade 10 failed to predict reading comprehension in Grade 11 (*β*^ = 0.001, *p* > 0.05); likewise, reading comprehension in Grade 10 did not yield a significant prediction for vocabulary depth in Grade 11 (β^ = 0.072, p > 0.05). To conclude, vocabulary breadth in Grade 10 demonstrated significant predictive power for both vocabulary depth and reading comprehension in Grade 11. However, there were no cross-lagged effects between Time 1 vocabulary depth and Time 2 reading comprehension.
Table 3 Regression weights for the relations between vocabulary breadth, vocabulary depth and reading comprehension across time.
Full size table
Fig. 3: Cross-lagged panel model for vocabulary breadth, vocabulary depth, and reading comprehension across Grades 10 and 11.
figure 3
Using a cross-lagged panel analysis, this model investigates the reciprocal relations and potential causative effects of vocabulary depth, vocabulary breadth, and reading comprehension. There was room for variation in the residuals for all three structures at both time points. χ² (73) = 95.989, p <0.05, CFI = 0.982, SRMR = 0.045, RMSEA = 0.035 were the model’s fit metrics. Both vocabulary depth (β^ =0.550, p <0.001) and reading comprehension (β^ =0.516, p <0.001) in Grade 11 were strongly predicted by vocabulary breadth in Grade 10. On the other hand, reading comprehension in Grade 10 and vocabulary depth in Grade 10 predicted vocabulary breadth and depth in Grade 11 respectively (β^ =0.229, p <0.05). For both classes, there were no significant cross-lagged effects on reading comprehension and vocabulary depth.
Full size image
Discussion
The primary aim of this study was to examine the concurrent and longitudinal connections between vocabulary breadth, vocabulary depth and reading comprehension among Chinese-speaking adolescents learning English as a Foreign Language. The first research question addressed the concurrent and developmental prediction of vocabulary breadth and vocabulary depth for reading comprehension across Grades 10 to 11 with structural equation modeling. The second research question involved the reciprocal relations between vocabulary breadth, vocabulary depth and reading comprehension over time.
To answer the first research question, we found that both vocabulary breadth and depth made significant contributions to reading comprehension across Grades 10 to 11; moreover, vocabulary breadth was more predictive than vocabulary depth for reading comprehension concurrently. In contrast, the developmental prediction of the vocabulary breadth and depth for reading comprehension varied with grades. To be specific, vocabulary breadth made stronger prediction for reading comprehension than vocabulary depth in Grade 10, whereas vocabulary depth had a stronger effect than vocabulary breadth on reading comprehension in Grade 11. The concurrent magnitude of the relationship between the breadth and depth of the vocabulary knowledge and reading comprehension in our study corroborated the findings of Nation and Snowling (2004) and Dagnaw (2023), who reported that vocabulary breadth made more contribution to reading comprehension. While considering the developmental contributions of vocabulary breadth and depth to reading comprehension, the current investigation into the respective roles of vocabulary breadth and depth in reading comprehension yielded results that were at odds with the perspective advanced by Nation and Snowling (2004), who posited that vocabulary depth played a relatively minor role in accounting for variance in reading comprehension. However, in Qian’s (1999) study, it was found that vocabulary depth exerted a more significant influence on reading comprehension compared to vocabulary breadth. The discrepant results could potentially be due to variations in the ages of the participants. For example, while Nation and Snowling conducted their research with children aged between 8 and 13, Qian’s study was carried out with university students. There appears to be an age or developmental stage component to the relationship between vocabulary knowledge and reading comprehension, suggesting that the role vocabulary knowledge plays in reading comprehension might differ depending on the age or developmental stage of the learner. This means that as a person progresses through different stages of language acquisition and cognitive development, the relative importance of having a broad vocabulary (knowing many words) versus a deep understanding of those words (including their meanings, nuances, and usages) in supporting reading comprehension may change. In other words, younger learners might benefit more from expanding the number of words they know, while older learners might gain more from deepening their understanding of known words. It also suggests that the relationship between the two dimensions of vocabulary knowledge and reading comprehension is complex and may change over the course of language development.
Regarding the second research question, it was found that vocabulary breadth in Grade 10 demonstrated significant predictive power for both vocabulary depth and reading comprehension. However, there were no cross-lagged effects between Time 1 vocabulary depth and Time 2 reading comprehension. The reciprocal and unidirectional relations between vocabulary breadth and reading comprehension were evidenced in previous studies (Biemiller and Slonim 2001; Reyonlds and Turek 2012; Quinn et al. 2015; Verhoeven and van Leeuwe 2008) because vocabulary knowledge was operationalized with the breadth of vocabulary. In terms of the reciprocal relationships between vocabulary depth and reading comprehension, little has been done in previous studies.
As is well-confirmed in prior literature, vocabulary knowledge is a multi-dimensional construct (Koda 2005; Schmitt 2010), and various aspects of a word, including its written form, form–meaning connection, collocation, and grammatical functions, may be gained to different extents during each encounter (Webb 2007). Importantly, gaining proficiency in one aspect does not guarantee a similar level of understanding in another (Webb and Nation 2017). In the current study, two aspects of vocabulary breadth and depth were explored respectively. The results of bivariate correlations indicated that Time 1 vocabulary depth was not significantly related to Time 2 vocabulary depth, which was corroborated with the findings in Currie et al.'s (2019) study. They examined the concurrent and longitudinal development of the relationships between vocabulary breadth, vocabulary depth and inferencing making, and found that vocabulary breadth developed steadily across time; vocabulary depth instead appeared erratic. Furthermore, Schmitt (1998) investigated the longitudinal development of four types of word knowledge including spelling, associations, grammatical information and meaning, and found that no developmental hierarchy of knowledge types was found. It suggested that learners did not follow a clear and systematic progression in the acquisition of different aspects of vocabulary knowledge across time. As a result, inasmuch as vocabulary knowledge did not develop in a linear fashion (Meara 1997), the nonreciprocal relationships between vocabulary depth and reading comprehension may stem from the inconsistent development of various aspects of vocabulary knowledge.
In addition, exposure to English is a critical factor in vocabulary learning. Studies have demonstrated that learners can achieve significantly higher scores on vocabulary tests even in the absence of formal English instruction (De Wilde and Eyckmans 2017). Schmitt (2008) indicated that some aspects of word knowledge were likely best acquired through repeated exposure to the lexical item in diverse contexts. However, the English learning environment in China is particularly constrained with insufficient extramural language exposure. This situation is particularly prevalent in the regions where the present study was conducted. That is, most English learning depends on formal classroom instruction, and resources and opportunities for learning English outside the classroom are relatively limited. Besides, most textbooks used in these settings do not adopt a clear systematic approach to teaching vocabulary. Instead, they are often structured around reading passages, with the vocabulary emphasized primarily reflecting the specific topic of each passage. As the topics shifts in subsequent chapters, this vocabulary is rarely revisited, leading to a somewhat random selection of words (Schmitt and Schmitt 2014). The lack of systematic repetition and varied context exposure is likely a significant contributing factor to the absence of a cross-lagged effect between vocabulary depth and reading comprehension observed in the study.
In summary, monitoring the changes and reciprocal relationships of vocabulary breadth and depth across various time points is crucial, which addresses significant gaps in existing research and contributes to the theoretical development of the field. Meanwhile, understanding how vocabulary breadth and depth evolve and interact over time is also essential for revealing the mechanisms through which these dimensions influence reading comprehension abilities, and this longitudinal perspective can help identify critical periods and patterns in vocabulary development, thereby enriching our theoretical understanding and guiding the vocabulary instruction.
Conclusions and limitations
The present study investigated the concurrent and longitudinal effects of vocabulary breadth and depth on reading comprehension among Chinese-speaking adolescents in senior middle schools. From a cross-sectional perspective, it has been found that both vocabulary breadth and depth made significant contributions to reading comprehension. Vocabulary breadth was a more robust predictor of reading comprehension for 10th and 11th graders compared to vocabulary depth. However, the contribution of vocabulary depth to reading comprehension became increasingly significant as students advanced through higher grades. The present findings suggest that educators should focus on expanding students’ vocabulary by exposing them to a wide range of words through diverse reading materials and systematic instruction. Activities such as creating vocabulary lists, providing definitions, and using words in different contexts can help build a broad vocabulary base. From a longitudinal perspective, the current study demonstrated that vocabulary breadth was reciprocally related to vocabulary depth and reading comprehension whereas there were no reciprocal relations between vocabulary depth and reading comprehension across time points. Accordingly, in Grade 11, the instructional focus should shift to deepening students’ understanding of the words they have already known, as vocabulary depth becomes more critical for reading comprehension. Advanced activities, such as exploring synonyms, antonyms, and fine-grained meanings, along with contextualized learning, can enhance vocabulary depth. Additionally, educators should regularly assess students vocabulary breadth and depth, adjust instructional strategies based on their developmental stage, and encourage extramural language exposure to supplement formal classroom learning.
The present study also has a few limitations. To begin with, it is important to take statistical limitations into account when interpreting the temporal relationships among vocabulary breadth, vocabulary depth and reading comprehension in our study. Even though we conducted a longitudinal study to examine the causal relations between them, the cross-lagged panel design in the present study only covers two-wave data, and the time interval between the two data collections is less than one year. Therefore, in order to attest to the causal relationship between variables and have a more robust developmental trajectory, more time points and longer time intervals should be considered in future studies. Another limitation results from cross-lagged panel modeling, which focuses on between-person differences over time instead of within-person variance (Selig and Little 2012). Hence, it is recommended that more suitable approaches, such as the random intercepts cross-lagged panel model (RI-CLPM) or Latent Growth Modeling, be considered for future studies. Finally, the findings of this study were interpreted on the basis of the measures that we used to assess the key constructs. The assessment of vocabulary depth was conducted using a variety of test types, potentially leading to diverse outcomes. Consequently, vocabulary depth could be evaluated through more extensive tasks.
Data availability
Data will be available on reasonable request.
References
Aitchison J (1987) Words in the mind: an introduction to the mental lexicon. Oxford, Blackwell
Anderson RC, Freebody P (1981) Vocabulary knowledge. In: Guthrie J (ed) Comprehension and teaching: research reviews. International Reading Association, Newark, DE, p 77–117
Bentler PM (1990) Comparative fit indexes in structural models. Psychol Bull 107(2):238–246. https://doi.org/10.1037/0033-2909.107.2.238
ArticleCASPubMedMATHGoogle Scholar
Biemiller A (1999) Estimating vocabulary growth for ESL children without listening comprehension instruction. Paper presented at the annual confer-ence of the American Educational Research Association, Montreal, Quebec, April 1999
Biemiller A, Slonim N (2001) Estimating root word vocabulary growth in nor-mative and advantaged populations: evidence for a common sequence of vocabulary acquisition. J Educ Psychol 93(3):498–520. https://doi.org/10.1037/0022-0663.93.3.498
ArticleMATHGoogle Scholar
Binder KS, Cote NG, Lee C, Bessette E, Vu H (2017) Beyond breadth: the contributions of vocabulary depth to reading comprehension among skilled readers. J Res Read 40:333–343. https://doi.org/10.1111/1467-9817.12069
ArticlePubMedGoogle Scholar
Browne MW, Cudeck R (1992) Alternative ways of assessing model fit. Sociol Methods Res 21(2):230–258. https://doi.org/10.1177/0049124192021002005
ArticleMATHGoogle Scholar
Currie NK, Muijselaar MML, Language Reading Research Consortium (2019) Inference making in young children: the concurrent and longitudinal con-tributions of verbal working memory and vocabulary. J Educ Psychol 111(8):1416–1431. https://doi.org/10.1037/edu0000342
ArticleGoogle Scholar
Dagnaw AT (2023) Revisiting the role of breadth and depth of vocabulary knowledge in reading comprehension. Cogent Educ 10(1). https://doi.org/10.1080/2331186X.2023.2217345
Dale E (1965) Vocabulary measurement: techniques and major findings. Elem Engl 42(8):895
MATHGoogle Scholar
De Wilde V, Eyckmans J (2017) Game on! Young learners’ incidental language learning of English prior to instruction. Stud Second Lang Learn Teach 7(4):673–694
ArticleMATHGoogle Scholar
Grabe W (2009) Reading in a second language: moving from theory to practice. Cambridge University Press, Cambridge
MATHGoogle Scholar
Henriksen B (1999) Three dimensions of vocabulary development. Stud Second Lang Acquis 21(2):303–317. https://doi.org/10.1017/S0272263199002089
ArticleMATHGoogle Scholar
Kline R (2016) Principles and practice of structural equation modeling. 4th edn.Guilford Press, New York, NY
Koda K (2005) Insights into second language reading: a cross-linguistic approach. Cambridge University Press, Cambridge
BookMATHGoogle Scholar
Li M, Kirby JR (2015) The effects of vocabulary breadth and depth on English reading. Appl Linguist 36(5):611–634. https://doi.org/10.1093/applin/amu007
ArticleMATHGoogle Scholar
McCarthy MJ (1990) Vocabulary. Oxford University Press, Oxford, UK
MATHGoogle Scholar
Meara P (1997) Towards a new approach to modelling vocabulary acquisition. In: Schmitt N, McCarthy M (eds) Vocabulary: description, acquisition and pedagogy. Cambridge: Cambridge University Press, pp 109-121
Ministry of Education (2020) General Senior High School English Curriculum Standards (2017 edition). People’s Education Press, Beijing
Google Scholar
Moghadam SH, Zainal Z, Ghaderpour M (2012) A review on the important role of vocabulary knowledge in reading comprehension performance. Procedia Soc Behav Sci 66:555–563. https://doi.org/10.1016/j.sbspro.2012.11.300
ArticleMATHGoogle Scholar
Muter V, Hulme C, Snowling MJ, Stevenson J (2004) Phonemes, rimes, vocabulary, and grammatical skills as foundations of early reading development: evidence from a longitudinal study. Dev Psychol 40(5):665–681. https://doi.org/10.1037/0012-1649.40.5.665
ArticlePubMedGoogle Scholar
Nation ISP (1983) Testing and teaching vocabulary. Guidelines 5(1):12–15
Google Scholar
Nation ISP (1993) Vocabulary breadth, growth and use. In: Schreuder R, Weltens B (eds) The bilingual lexicon. Amsterdam, John Benjamins, p 115–134
Nation ISP (2001) Learning vocabulary in another language. Cambridge Uni-versity Press, Cambridge
Nation K, Snowling MJ (2004) Beyond phonological skills: broader language skills contribute to the development of reading. J Res Read 27(4):342–356. https://doi.org/10.1111/j.1467-9817.2004.00238.x
ArticleGoogle Scholar
National Reading Panel (2000) Teaching children to read: an evidence-based assessment of the scientific literature on reading and its implications for reading instruction. National Institute of Child Health and Human Development, Bethesda, MD
Google Scholar
Nurweni A, Read J (1999) The English vocabulary knowledge of Indonesian university students. Engl Spec Purp 18(2):161–175. https://doi.org/10.1016/S0889-4906(98)00005-2
ArticleGoogle Scholar
Palmer HE (1921) The principles of language study. Harrap, London
Paribakht TS, Wesche M (1993) Reading comprehension and second language development in a comprehension-based ESL program. TESL Can J 11:9–29
ArticleMATHGoogle Scholar
Perfetti C (1985) Reading ability. Oxford University Press, New York
Google Scholar
Perfetti C (2007) Reading ability: lexical quality to comprehension. Sci Stud Read 11(4):357–383. https://doi.org/10.1080/10888430701530730
ArticleMATHGoogle Scholar
Perfetti C, Stafura J (2014) Word knowledge in a theory of reading compre-hension. Sci Stud Read 18(1):22–37. https://doi.org/10.1080/10888438.2013.827687
ArticleMATHGoogle Scholar
Proctor CP, Carlo M, August D, Snow C (2005) Native Spanish-speaking children reading in English: toward a model of comprehension. J Educ Psychol 97(2):246–256. https://doi.org/10.1037/0022-0663.97.2.246
ArticleMATHGoogle Scholar
Qian DD (1999) Assessing the roles of depth and breadth of vocabulary knowledge in ESL reading comprehension. Can Mod Lang Rev 56(2):282
ArticleMATHGoogle Scholar
Qian D (2002) Investigating the relationship between vocabulary knowledge andacademic reading performance: an assessment perspective. Lang Learn 52(3):513–536. https://doi.org/10.1111/1467-9922.00193
ArticleMATHGoogle Scholar
Quinn JM, Wagner RK, Petscher Y, Lopez D (2015) Developmental relations between vocabulary knowledge and reading comprehension: a latent change score modeling study. Child Dev 86(1):159–175
ArticlePubMedGoogle Scholar
Read J (1993) The development of a new measure of L2 vocabulary knowledge. Lang Testing 10(3):355–371. https://doi.org/10.1177/026553229301000308
ArticleMATHGoogle Scholar
Read J (2000) Assessing Vocabulary. Cambridge University Press, London
BookGoogle Scholar
Read J (2004) Plumbing the depths: how should the construct of vocabulary knowledge be defined? In: Bogaards B, Laufer B (eds) Vocabulary in a second language. John Benjamins, Amsterdam, p 209–227
Ricketts J, Nation K, Bishop DVM (2007) Vocabulary is important for some, but not all reading skills. Sci Stud Read 11(3):235–257. https://doi.org/10.1080/10888430701344306
ArticleMATHGoogle Scholar
Reynolds MR, Turek JJ (2012) A dynamic developmental link between verbal comprehension-knowledge (Gc) and reading comprehension: verbal compre-hension knowledge drives positive change in reading comprehension. J SchPsychol 50(6):841–863
Google Scholar
Schmitt N (1998) Tracking the incremental acquisition of second language vo-cabulary: a longitudinal study. Lang Learn 48:281–317
ArticleMATHGoogle Scholar
Schmitt N (2008) Instructed second language vocabulary learning. Lang Teach Res 12(3):329–363
ArticleMATHGoogle Scholar
Schmitt N (2010) Researching vocabulary: a vocabulary research manual. Palgrave Macmillan UK, London
BookMATHGoogle Scholar
Schmitt N, Schmitt D (2014) A reassessment of frequency and vocabulary size in L2 vocabulary teaching. Lang Teach 47(4):484–503
ArticleMATHGoogle Scholar
Schmitt N, Schmitt D, Clapham C (2001) Developing and exploring the be-haviour of two new versions of the vocabulary levels test. Lang Testing 18(1):55–88. https://doi.org/10.1177/026553220101800103
ArticleMATHGoogle Scholar
Selig JP, Little TD (2012) Autoregressive and cross-lagged panel analysis for longitudinal data. In: Laursen B, Little TD, Card NA (eds) Handbook of developmental research methods. Guilford Press, New York, p 265–278
Snow C (2002) Reading for understanding: toward a RandD program in reading comprehension. Rand, Santa Monica, CA
MATHGoogle Scholar
Sparks E, Deacon SH (2015) Morphological awareness and vocabulary acquisi-tion: a longitudinal examination of their relationship in English-speaking children. Appl Psycholinguist 36(2):299–321. https://doi.org/10.1017/S0142716413000246
ArticleMATHGoogle Scholar
Stanovich K (2000) Progress in understanding reading: scientific foundations and new frontiers. Guilford Press, New York
Google Scholar
Stahl S (1999) Vocabulary development. Brookline Books, Cambridge, MA
MATHGoogle Scholar
Verhoeven L, van Leeuwe J (2008) Prediction of the development of reading comprehension: a longitudinal study. Appl Cogn Psychol 22(3):407–423. https://doi.org/10.1002/acp.1414
ArticleMATHGoogle Scholar
Vermeer A (2001) Breadth and depth of vocabulary in relation to L1/L2 acqui-sition and frequency of input. Appl Psycholinguist 22(2):217–234. https://doi.org/10.1017/S0142716401002041
ArticleMATHGoogle Scholar
Webb S (2007) The effects of repetition on vocabulary knowledge. Appl Linguist 28(1):46–65. https://doi.org/10.1093/applin/aml048
ArticleMATHGoogle Scholar
Webb S, Nation P (2017) How vocabulary is learned. Oxford University Press, Oxford
MATHGoogle Scholar
Download references
Acknowledgements
Our study was supported by the First-Class Discipline Construction Project of Ningxia Higher Education Institutions (Pedagogy) in 2023, ‘A Study of the Mechanism and Pathways of EFL Reading Ability’ (Grant No.: NXYLXK2021B10). The authors also acknowledge and thank the teachers in the English Language Teaching cohort at Guyuan No. 1 Middle School in Ningxia, China, for their invaluable support in data collection.
Author information
Authors and Affiliations
Ningxia Normal University, Guyuan, China
Tuoxiong Wang
City University of Macau, Macau, China
Haomin Zhang
Authors
Tuoxiong Wang
View author publications
You can also search for this author inPubMedGoogle Scholar
2. Haomin Zhang
View author publications
You can also search for this author inPubMedGoogle Scholar
Contributions
TW and HZ were responsible for the conceptualization of the study. They both also conducted the statistical analysis. TW wrote the first draft of the manuscript. HZ polished the full text. Both authors read and approved the final manuscript.
Corresponding authors
Correspondence to Tuoxiong Wang or Haomin Zhang.
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
This study was granted an ethical exemption under Category 2 by the University Committee on Human Research Protection (UCHRP) of East China Normal University (approval No. HR281-2022) on April 29, 2022. The exemption was based on the minimal risk associated with the research and the non-identifiable nature of the data collected. The research was performed in accordance with the principles of the Declaration of Helsinki and other relevant guidelines and regulations applicable when human participants are involved.
Informed consent
As this study was classified as an ethical exemption under Category 2, no formal informed consent process was required. However, participants were informed about the study’s purpose and procedures, and their voluntary participation was assured.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
About this article
Check for updates. Verify currency and authenticity via CrossMark
Cite this article
Wang, T., Zhang, H. Reciprocal effects of vocabulary breadth, vocabulary depth, and reading comprehension: a cross-lagged panel analysis in Chinese-speaking EFL learners. Humanit Soc Sci Commun 12, 410 (2025). https://doi.org/10.1057/s41599-025-04694-2
Download citation
Received:20 September 2024
Accepted:28 February 2025
Published:22 March 2025
DOI:https://doi.org/10.1057/s41599-025-04694-2
Share this article
Anyone you share the following link with will be able to read this content:
Get shareable link
Sorry, a shareable link is not currently available for this article.
Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative